TERAYON COMMUNICATION SYSTEMS
8-K, 2000-07-18
TELEPHONE & TELEGRAPH APPARATUS
Previous: SALEM CAPITAL MANAGEMENT INC, 13F-HR, 2000-07-18
Next: MURPHY CAPITAL MANAGEMENT INC, 13F-HR, 2000-07-18



<PAGE>

                     SECURITIES EXCHANGE AND COMMISSION
                          Washington, D. C.  20549

                                  FORM 8-K

                               CURRENT REPORT

                   Pursuant to Section 13 or 15(d) of the
                       Securities Exchange Act of 1934

      Date of Report (Date of earliest event reported):  July 18, 2000

                     Terayon Communication Systems, Inc.
           (Exact name of registrant as specified in its charter)

                                  Delaware
               (State or other jurisdiction of incorporation)

        000-24647                                        77-0328533
  (Commission File No.)                    (I.R.S. Employer Identification No.)

                            2952 Bunker Hill Lane
                            Santa Clara, CA 95054
            (Address of principal executive offices and zip code)

     Registrant's telephone number, including area code: (408) 727-4400

<PAGE>

Item 5.  Other Events

Registration Statements Nos. 333-91823, 333-95357 and 333-37978

  Terayon Communication Systems, Inc. (the "Company") hereby amends Registration
Statement Nos. 333-91823, 333-95357 and 333-37978 and any prospectuses and post-
effective amendments filed thereon to revise the sections entitled Prospectus
Summary and Risk Factors.

  The section entitled Prospectus Summary should be deleted and replaced with
the following section:


                             PROSPECTUS SUMMARY

   As used in this offering memorandum, unless the context otherwise requires,
"we," "us," "our," "Company" or "Terayon" refers to Terayon Communication
Systems, Inc., the issuer of the Notes, and its subsidiaries. The following
summary contains basic information about us and this offering. It likely does
not contain all the information that is important to you. For a more complete
understanding of this offering, we encourage you to read this entire offering
memorandum and the documents to which we have referred you.

                                    Terayon

Our Business

   We develop, market and sell broadband access systems that enable cable
operators and other providers of broadband access services to cost effectively
deploy reliable broadband access services over cable, copper wire utilizing
digital subscriber line technology (or DSL) and wireless systems. We have
substantial development and marketing resources focused on cable operators.
However, through internal development and recent acquisitions of complementary
technology and businesses, we also are focusing on service providers that offer
broadband access through existing copper wire infrastructures and wireless
systems.

   In recent years, the volume of bandwidth-intensive data, voice and video
traffic across existing cable infrastructure, the Internet, corporate intranets
and other public networks has increased dramatically. International Data
Corporation estimates that the number of worldwide Internet users will increase
from approximately 200 million at the end of 2000 to more than one billion by
the end of 2005. IDC estimates that the number of homes in the United States
with broadband access will increase from two million at the end of 1999 to 20
million by the end of 2003. As a result of the rapid evolution of broadband
access, cable operators, providers of telephone services and other service
providers are providing a bundle of voice, data and video services to their
residential and commercial subscribers over existing and new infrastructures.

   Our objective is to be the leading provider of broadband access systems to
providers of broadband services that use existing cable, copper wire (DSL) and
wireless networks to offer services to residential and commercial customers.
Key elements of our strategy include the following:

  .  Build a complete portfolio of broadband products;

  .  Supply leading broadband service providers worldwide;

  .  Increase our presence in existing and new markets;

  .  Extend technology leadership and advance industry standards; and

  .  Provide superior customer support.

   Our primary product is the TeraComm system, which is based on our patented
Synchronous Code Division Multiple Access or "S-CDMA" technology. Our S-CDMA
technology enables reliable two-way broadband data communications over both
pure coaxial and hybrid fiber/coax cable infrastructure and is designed to
enable cable operators to maximize the capacity and reliability of broadband
data services over any cable plant. In furtherance of our strategy to provide
a complete portfolio of broadband products, we have

                                       2
<PAGE>

recently completed several acquisitions of complementary broadband
technologies, including digital video management systems, DSL and wireless.

   We sell our broadband access products to cable operators and other providers
of broadband access services through direct sales forces in North America,
South America, Europe and Asia. We also distribute our products via
distributors and systems integrators. Companies currently using or distributing
our TeraComm system include Rogers Communications, Inc., Shaw Communications,
Inc., TCA Cable TV, Inc. (a subsidiary of Cox Communications, Inc.), United
Pan-Europe Communications and Crossbeam Networks Corporation, a wholly owned
subsidiary of Sumitomo Corporation. Companies currently using our DSL products
include major ILEC's (Incumbent Local Exchange Carriers) in the United States,
including SBC, Bell Atlantic, Bell South, U.S. West and GTE.

   Our company was incorporated in California in January 1993 and
reincorporated in Delaware in July 1998. Our executive offices are located at
2952 Bunker Hill Lane, Santa Clara, California 95054 and our telephone number
is (408) 727-4400. Our Web site is located at www.terayon.com. Information
contained on our Web site does not constitute part of this offering memorandum.

Recent Events

   We believe that the following transactions, in combination with our own
internal development and previous acquisitions, significantly increase our
technological expertise throughout broadband access technologies including
cable modems, DSL and wireless broadband access:

  .  In January 2000, we acquired Telegate Ltd., a manufacturer of telephony
     and data access platforms that are deployed by service providers to
     deliver efficient carrier-class voice services over cable. Telegate also
     provides in-home networking capability for telephony and data, based on
     the Digital Enhanced Cordless Telephony (DECT) standard.

  .  In April 2000, we acquired the Access Network Electronics Division (ANE)
     of Tyco Electronics Corporation, a subsidiary of Tyco International Ltd.
     ANE produces DSL systems that allow for the provisioning of multiple
     phone lines over the existing copper telephony networks consisting of a
     single pair of copper wires that traditionally only support a single
     phone line.

  .  In April 2000, we acquired Combox, Ltd., a manufacturer of broadband
     data systems and satellite communications based on international
     standards. Combox's cable data access systems conform to the growing
     EuroModem international specification, based on the Digital Video
     Broadcasting (DVB) standard.

  .  In April 2000, we purchased certain assets of Internet Telecom Ltd., a
     supplier of PacketCable and other standards-based, voice-over-IP
     ("Internet Protocol") systems and technologies.

  .  In April 2000, we acquired Ultracom Communication Holdings (1995) Ltd.
     ("Ultracom"), a supplier of wireless DSL and broadband systems on
     silicon.

                                       3

<PAGE>

  The section entitled Risk Factors should be deleted and replaced with the
following section:


                                  RISK FACTORS

   Before making an investment decision, you should carefully consider the risks
described below. The risks and uncertainties described below are not the only
ones facing our company. Additional risks and uncertainties not presently known
to us or that we currently deem immaterial also may impair our business
operations. If any of the following risks actually occur, our business could be
harmed. In such case, the trading price of our common stick could decline, and
you may lose all or part of your investment.

                         Risks Related to Our Business

We have a limited operating history and a history of losses.

   We have a limited operating history, and it is difficult to predict our
future operating results. We began shipping products commercially in June 1997,
and we only have been shipping products in volume since the first quarter of
1998. As of March 31, 2000, we had an accumulated deficit of $176.8 million. We
believe that we will continue to experience net losses for the foreseeable
future. Most of our expenses are fixed in advance, and we generally are unable
to reduce our expenses significantly in the short term to compensate for any
unexpected delay or decrease in anticipated revenues. We expect to continue to
increase expenses for the foreseeable future to support increased sales and
marketing and technical support costs. Any significant delay in our anticipated
revenues or commercialization of new products would harm our business. The
revenue and profit potential of our business and our industry are unproven. We
had negative gross margins from our inception until the fourth quarter of 1998,
and any future revenue growth may not result in positive gross margins or
operating profits in future periods.

Our operating results may fluctuate.

   Our quarterly revenues are likely to fluctuate significantly in the future
due to a number of factors, many of which are outside our control. Factors that
could affect our revenues include the following:

  .  variations in the timing of orders and shipments of our products;

  .  variations in the size of the orders by our customers;

  .  new product introductions by competitors;

  .  delays in our introduction of new products;

  .  delays in our receipt of orders forecasted by our customers;

  .  delays by our customers in the completion of upgrades to their cable
     infrastructure;

  .  variations in capital spending budgets of broadband access service
     providers;

  .  adoption of industry standards and the inclusion in or compatibility of
     our technology with any such standards; and

  .  delays in obtaining regulatory approval for commercial deployment of
     cable modem systems.

   Our expenses generally will vary from quarter to quarter depending on the
level of actual and anticipated business activities. Research and development
expenses will vary as we begin development of new products and as our
development programs move to wafer fabrication and prototype development, which
results in higher engineering expenses.

                                       4
<PAGE>

   A variety of factors affect our gross margin, including the following:

  .  the sales mix of our products;

  .  the volume of products manufactured;

  .  the type of distribution channel through which we sell our products;

  .  the average selling prices or "ASPs" of our products; and

  .  the effectiveness of our cost reduction measures.

   We anticipate that unit ASPs of our products will decline in the future.
This could cause a decrease in the gross margins for these products. In
addition, the maturity of TeraComm system deployments affects our gross margin.
New deployments of the TeraComm system involve the sale of headend equipment
(which has higher margins) and generally involve smaller quantities of product.
New deployments typically are sold at higher margins than the larger volume
sales of product associated with more mature deployments of the TeraComm
system. The sales mix of TeraLink 1000 Master Controllers, TeraLink Gateways
and TeraPro cable modems also affects our gross margin. The TeraPro cable
modems have significantly lower margins than the TeraLink 1000 Master
Controller and TeraLink Gateway headend products. We expect to achieve
significantly lower margins on the TeraPro cable modems for the foreseeable
future. Further, we expect that sales of TeraPro cable modems will continue to
constitute a significant portion of our revenues for the foreseeable future.

   We also anticipate that our operating results will be impacted by sales,
gross profit and operating expenses of acquired companies. The impact of these
factors on our operating results will vary as we acquire additional companies.

We are dependent on a small number of customers.

   Four customers accounted for approximately 64% of our revenues for the
quarter ended March 31, 2000 and three customers accounted for approximately
63% of our revenues for the quarter ended March 31, 1999. We believe that a
substantial majority of our revenues will continue to be derived from sales to
a relatively small number of customers for the foreseeable future. In addition,
we believe that sales to these customers will be focused on a limited number of
projects.

   The cable industry is undergoing significant consolidation in North America
and internationally, and a limited number of cable operators controls an
increasing number of cable systems. Currently, ten cable operators in the
United States own and operate facilities passing approximately 86% of total
homes passed. In addition, the North American DSL market is concentrated with
the major ILECs, constituting a significant percentage of the market. As a
result, our sales will be largely dependent upon product acceptance by the
leading broadband service providers. Currently, the timing and size of each
customer's order is critical to our operating results. Our major customers are
likely to have significant negotiating leverage and may attempt to change the
terms, including pricing, upon which we do business with them. These customers
also may require longer payment terms than we anticipate, which could require
us to raise additional capital to meet our working capital requirements.

Acquisitions could result in dilution, operating difficulties and other adverse
consequences.

   We have acquired seven businesses since September 1999: Imedia Corporation
in September 1999; Radwiz in November 1999; Telegate in January 2000; ANE in
April 2000; ComBox in April 2000; some assets of Internet Telecom in April
2000; and Ultracom in April

                                       5

<PAGE>

2000. If appropriate opportunities present themselves, we intend to acquire
additional businesses, technologies, services or products that we believe are
strategic. The process of integrating any acquired business into our business
and operations is risky and may create unforeseen operating difficulties and
expenditures. The areas in which we may face difficulties include:

  .  diversion of management time (both ours and that of the acquired
     companies) during the period of negotiation through closing and after
     closing from the ongoing development of our businesses, issues of
     integration and future products;

  .  decline in employee morale and retention issues resulting from changes
     in compensation, reporting relationships, future prospects or the
     direction of the business;

   We typically sell products to our customers pursuant to purchase orders, and
we therefore generally do not have long-term agreements in place with our
customers.

  .  the need to integrate each company's accounting, management information,
     human resource and other administrative systems to permit effective
     management, and the lack of control if this integration is delayed or
     not implemented; and

  .  the need to implement controls, procedures and policies appropriate for
     a larger public group of companies that prior to acquisition had been
     smaller, private companies.

   We have very limited experience in managing this integration process.
Moreover, the anticipated benefits of any or all of these completed or pending
acquisitions may not be realized. Future acquisitions could result in
potentially dilutive issuances of equity securities, the incurrence of
additional debt, contingent liabilities or amortization expenses related to
goodwill and other intangible assets, any of which could harm our business.
Future acquisitions also could require us to obtain additional equity or debt
financing, which may not be available on favorable terms or at all. Even if
available, this financing may be dilutive.

The sales cycle for our products is lengthy.

   The sales cycle associated with our products typically is lengthy, often
lasting six months to a year. Our customers typically conduct significant
technical evaluations of competing technologies prior to the commitment of
capital and other resources. In addition, purchasing decisions may be delayed
because of our customers' internal budget approval procedures. Sales also
generally are subject to customer trials, which typically last three months.
Because of the lengthy sales cycle and the large size of customers' orders, if
orders forecasted for a specific customer for a particular quarter do not occur
in that quarter, our operating results for that quarter could suffer.

There are many risks associated with our participation in the establishment of
advanced physical layer specifications to be added to DOCSIS.

   In November 1998, CableLabs selected us to co-author DOCSIS 1.2 (Data Over
Cable Service Interface Specifications), an enhanced version of the DOCSIS
cable modem specification based in part on our S-CDMA technology. In September
1999, CableLabs indicated that it intended to proceed with the advanced
physical layer ("PHY") work on two parallel tracks: one for the development of
a prototype based on our S-CDMA technology and

                                       6

<PAGE>

one for the inclusion of Advanced TDMA technology (Time Division Multiple
Access), as proposed by other companies. In February 2000, CableLabs further
clarified the status of the advanced PHY project regarding a separate release
that will include TDMA technologies. In addition, CableLabs reiterated that it
is continuing to work with us on the development of a DOCSIS specification that
could include our S-CDMA technology. To that end, CableLabs has requested that
we submit a prototype of a DOCSIS system that incorporates an S-CDMA advanced
PHY capability for testing. CableLabs has stated that if the testing of this
prototype reveals that the S-CDMA advanced PHY works as claimed (including
proper backwards compatibility and coexistence with the other aspects of
DOCSIS), and if the costs for adding S-CDMA to DOCSIS products are in line with
estimates, then it is likely, but not certain, that S-CDMA advanced PHY
capabilities will be included in a future version of the DOCSIS specification.
The prototype we submit to CableLabs may fail to demonstrate the level of
performance that CableLabs seeks; even if it does meet performance expectations
there can be no guarantee that CableLabs will incorporate the technology into a
future version of DOCSIS specifications. In addition, if CableLabs does proceed
to include S-CDMA in a future DOCSIS specification, there can be no guarantee
that the DOCSIS S-CDMA specification will be the same as the specification we
incorporated in the prototype submitted for tests, which may require us to
further develop our prototype.

   Our future revenues and operating results are likely to suffer if S-CDMA is
not included in a future release of DOCSIS. We also may incur substantial
additional research and development expenditures to adapt our specifications to
the version adopted by CableLabs. CableLabs has not established a schedule for
adding the S-CDMA capabilities to the DOCSIS specifications. Delays in the
establishment of a final specification for S-CDMA in DOCSIS could harm our
plans to sell DOCSIS compatible modems and headend equipment. In particular, if
the final DOCSIS S-CDMA specification is not approved prior to the time when we
are ready to ship DOCSIS products with S-CDMA features included, then we may be
required to delay the introduction of those products until the DOCSIS S-CDMA
specification is released or to introduce the S-CDMA features as proprietary
enhancements to a standard DOCSIS product. Either one of these events could
harm revenues and operating results.

   We have already given CableLabs assurances that we will contribute some
aspects of our proprietary S-CDMA technology to a royalty-free intellectual
property pool, if S-CDMA is included in a future version of DOCSIS
specifications. This royalty-free pool has been established by CableLabs to
facilitate the participation of as many vendors as possible in providing
equipment that is compatible with the DOCSIS specifications. As a result, any
of our competitors who join the DOCSIS intellectual property pool would have
access to some aspects of our technology and would not be required to pay us
any royalties or other compensation. If a competitor is able to duplicate the
functionality and capabilities of our technology, we could lose some or all of
the time-to-market advantage we might otherwise have, which could harm our
future revenues and operating results.

   We believe the addition of advanced upstream PHY capabilities to DOCSIS will
increase the overall market for DOCSIS-compatible products, and as such will
result in increased competition in the cable modem market. This competition
could come from existing competitors or from new competitors who enter the
market as a result of the enhancements to the specifications. This increased
competition is likely to result in lower ASPs of cable modem systems and could
harm revenues and gross margins. Because our competitors will be able to
incorporate some aspects of our technology into their products, our current
customers may choose alternate suppliers or choose to purchase DOCSIS-compliant
cable modems with advanced PHY capabilities from multiple suppliers. We may be
unable to produce DOCSIS compliant cable modems with advanced PHY capabilities
more quickly or at lower cost than

                                       7

<PAGE>

our competitors. The inclusion of our S-CDMA technology in future DOCSIS
specification could result in increased competition for the services of our
existing employees who have experience with S-CDMA. The loss of these employees
to one or more competitors could harm our business.

   DOCSIS standards have not yet been accepted in Europe and Asia. An alternate
standard for cable modem systems, called the EuroModem standard, or DAVIC/DVB,
has been formalized, and some European cable system operators have embraced it.
We intend to develop and sell products that comply with the EuroModem standard
and to pursue having portions of our S-CDMA technology included in a future
version of the EuroModem standard. We may be unsuccessful in these efforts.

We need to develop new products in order to remain competitive.

   Our future success will depend in part on our ability to develop, introduce
and market new products in a timely manner. We also must respond to competitive
pressures, evolving industry standards and technological advances. Our current
S-CDMA products are not DOCSIS-compliant. We are currently developing a
prototype of a DOCSIS system that incorporates an S-CDMA advanced PHY
capability for testing and eventual inclusion in the DOCSIS standard. There is
no guarantee that we will be successful in developing the prototype or that the
prototype, if successfully developed, will be included in a future release of
the DOCSIS standard. We anticipate that during 2000, existing or potential
customers may delay purchases of our TeraComm system in order to purchase
systems that comply with the DOCSIS standard. In addition, potential new
customers could decide to purchase DOCSIS-compliant products from one or more
of our competitors rather than from us. As a result, our product sales in the
second half of 2000 may be lower than we anticipate. In order to promote sales
of our currents products we may be required to reduce our prices for sales to
existing customers. This would harm our operating results and gross margin.

   As a result of the inclusion of TDMA technology in the new DOCSIS version
announced by CableLabs in February 2000, we will have to incorporate advanced
TDMA technology into our DOCSIS-compliant products. If we are unable to do this
effectively, or in a timely manner, we will lose some or all of the time-to-
market advantage we might otherwise have had.

   Our future success will also depend on our ability to develop and market
products for broadband applications over DSL and wireless networks. The markets
for these broadband applications are also subject to evolving standards, such
as NEBS compliance in the North American DSL market, and technological advances
in these arenas. There is no guarantee that we will be successful in developing
products that are compliant with these standards or that we will be successful
in keeping pace with future technological advances in this arena.

Average selling prices of broadband access equipment typically decrease.

   The broadband access systems market has been characterized by erosion of
average selling prices. We expect this to continue. This erosion is due to a
number of factors, including competition, rapid technological change and price
performance enhancements. The ASPs for our products may be lower than expected
as a result of competitive pricing pressures, our promotional programs and
customers who negotiate price reductions in exchange for longer term purchase
commitments. We anticipate that ASPs and gross margins for our products will
decrease over product life cycles. In addition, we believe that the widespread
adoption of industry standards is likely to further erode ASPs, particularly
for cable modems and other similar consumer premise equipment. It is likely
that the widespread adoption of industry standards will result in increased
retail distribution of cable modems and other similar

                                       8

<PAGE>

consumer premise equipment, which could put further price pressure on our
products. Decreasing ASPs could result in decreased revenue even if the number
of units sold increases. As a result, we may experience substantial period-to-
period fluctuations in future operating results due to ASP erosion. Therefore,
we must continue to develop and introduce on a timely basis next-generation
products with enhanced functionalities that can be sold at higher gross
margins. Our failure to do this could cause our revenues and gross margin to
decline.

We must achieve cost reductions.

   Certain of our competitors currently offer products at prices lower than
ours. Market acceptance of our products will depend in part on reductions in
the unit cost of our products. We expect that as headend equipment becomes more
widely deployed, the price of cable modems and other similar consumer premise
equipment will decline. In particular, we believe that the widespread adoption
of industry standards such as DOCSIS will cause increased price competition for
consumer premise equipment. However, we may be unable to reduce the cost of our
products sufficiently to enable us to compete with other suppliers. Our cost
reduction efforts may not allow us to keep pace with competitive pricing
pressures or lead to gross margin improvement.

   Some of our competitors are larger and manufacture products in significantly
greater quantities than we intend to for the foreseeable future. Consequently,
these competitors have more leverage in obtaining favorable pricing from
suppliers and manufacturers. In order to remain competitive, we must
significantly reduce the cost of manufacturing our cable modems through design
and engineering changes. We may not be successful in redesigning our products.
Even if we are successful, our redesign may be delayed or may contain
significant errors and product defects. In addition, any redesign may not
result in sufficient cost reductions to allow us to significantly reduce the
list price of our products or improve our gross margin. Reductions in our
manufacturing costs will require us to use more highly integrated components in
future products and may require us to enter into high volume or long-term
purchase or manufacturing agreements. Volume purchase or manufacturing
agreements may not be available on acceptable terms. We could incur expenses
without related revenues if we enter into a high volume or long-term purchase
or manufacturing agreement and then decide that we cannot use the products or
services offered by such agreement.

We must keep pace with rapid technological change to remain competitive.

   The markets for our products are characterized by rapid technological
change, evolving industry standards, changes in end-user requirements and
frequent new product introductions and enhancements. Our future success will
depend upon our ability to enhance our existing products and to develop and
introduce new products that achieve market acceptance. Providers of broadband
access services may adopt alternative technologies or they may deploy
alternative services that are incompatible with our products.

   The demand for broadband access services has resulted in the development of
several competing modulation technologies. For example, some of our cable
products utilize a modulation technology known as S-CDMA, while several of our
competitors utilize modulation technologies known as TDMA and Frequency
Division Multiple Access or "FDMA." Our headend equipment and cable modem
products currently are not interoperable with the headend equipment and modems
of other suppliers of broadband access products. As a result, potential
customers who wish to purchase broadband access products from multiple

                                       9

<PAGE>

suppliers may be reluctant to purchase our products. Although our technology
may be incorporated into a future version of a DOCSIS specification or another
industry standard, we cannot be certain that major cable operators will adopt
these standards. Major cable operators may not adopt products or technologies
based on our current proprietary S-CDMA technology or on any future industry
standard S-CDMA technology. Further, major cable operators may adopt products
or standards technologies based on competing modulation technologies. If
competitors using other modulation technologies can incorporate functionality
and capabilities currently found in S-CDMA, the value of our S-CDMA technology
would be diminished.

Broadband access services have not achieved widespread market acceptance, and
many competing technologies exist.

   Our success will depend upon the widespread commercial acceptance of
broadband access services by service providers and end users of broadband
access services. The market for these services is not fully developed. We
cannot accurately predict the future growth rate or the ultimate size of the
market for broadband access services. Potential users of our products may have
concerns regarding the security, reliability, cost, ease of installation and
use and capability of broadband access services in general.

   The market for our products may be impacted by the development of other
technologies that enable the provisioning of broadband access services and the
deployment of services over other media. Widespread acceptance of other
technologies or deployment of services over media not supported by our products
could materially limit acceptance of our broadband access systems. Broadband
access services based on our products and technology may fail to gain
widespread commercial acceptance by providers of broadband access services and
end users. In addition, we only recently began to offer products based on
alternate technologies such as DSL. We may not be successful in marketing and
selling these products.

We need to develop additional distribution channels.

   We presently market our TeraComm system to cable operators and systems
integrators. We believe that much of the North American cable modem market may
shift to a retail distribution model. Accordingly, we may need to redirect our
future marketing efforts to sell our cable modems directly to retail
distributors and end users. This shift would require us to establish new
distribution channels for our products. We may be unable to establish these
additional distribution channels. If we do establish them, we may be unable to
hire the additional personnel necessary to foster and enhance such distribution
channels. In addition, if the cable modem market shifts to a retail
distribution model, we may not successfully establish a retail distribution
presence. To the extent that large consumer electronics companies enter the
cable modem market, their well-established retail distribution capabilities
would provide them with a significant competitive advantage.

We may be unable to market effectively to broadband access service providers.

   Our growth and future success will be substantially dependent upon our
ability to convince providers of broadband access services to adopt our
technologies, purchase our products and effectively market our products to end
users. Our potential customers are likely to prefer purchasing products from
established manufacturing companies that can demonstrate the capability to
supply large volumes of products on short notice. In addition, many of our
potential customers may be reluctant to adopt technologies that have not gained
acceptance among other providers of similar services. This reluctance could
result in lengthy product testing and acceptance cycles for our products.
Consequently, the impediments to our initial sales may be even greater than
those to later sales.

                                       10

<PAGE>

   No established distribution network in the cable modem industry exists that
would provide us with easy access to smaller or geographically diverse cable
operators. Therefore, our initial sales to larger, more established cable
operators are critical to our business. Although we intend to establish
strategic relationships with leading distributors worldwide, we may not succeed
in establishing these relationships. Even if we do establish these
relationships, the distributors may not succeed in marketing our products to
cable operators. Some of our competitors have already established relationships
with certain cable operators. These established relationships may further limit
our ability to sell products to those cable operators. We do not have long,
well-established relationships with those cable operators. If we were to sell
our products to those cable operators, it would likely not be based on long-
term contracts and those customers would be able to terminate their
relationships with us at any time. In addition, one or more of our current
customers could cancel its relationship with us at any time.

   We have recently begun marketing and selling our products to providers of
DSL and wireless broadband services and thus we have very limited experience.
We do not have long, well-established relationships with these providers, and
we may not be successful in establishing these relationships.

We are dependent on broadband service providers choosing to offer additional
services to their customers.

   We depend on cable operators to purchase our cable modem systems and to
provide our cable modems to end users. Cable operators have a limited amount of
available bandwidth over which they can offer robust data services, and they
may not choose to provide these data services to their customers. If cable
operators choose to provide these services, we also will depend upon them to
market these services to cable customers, to install our equipment and to
provide support to end users. In addition, we will be highly dependent on cable
operators to continue to maintain their cable infrastructure in a manner that
allows us to provide consistently high performance and reliable services. Our
success also will depend upon the acceptance of our products by other providers
of services to the cable industry, such as Excite@Home's @Home Network and Road
Runner, a joint venture between MediaOne Group, Inc. and Time Warner Cable.
Sales of our DSL and wireless products are also dependent on service providers
choosing to purchase our products and to provide additional services to their
end users.

Sales of our cable products are dependent on the cable industry upgrading to
two-way cable infrastructure.

   Demand for our products will depend, to a significant degree, upon the
magnitude and timing of capital spending by cable operators for implementation
of access systems for data transmission over cable networks. This involves the
enabling of two-way transmission over existing coaxial cable networks and the
eventual upgrade to HFC in areas of higher penetration of data services. If
cable operators fail to complete these upgrades of their cable infrastructures
in a timely and satisfactory manner, the market for our products could be
limited. In addition, few businesses in the United States currently have cable
access. Cable operators may not choose to upgrade existing residential cable
systems or to install new cable systems to serve business locations.

   The success and future growth of our business will be subject to economic
and other factors affecting the cable television industry generally,
particularly its ability to finance substantial capital expenditures. Capital
spending levels in the cable industry in the United

                                       11

<PAGE>

States have fluctuated significantly in the past, and we believe that such
fluctuations will occur in the future. The capital spending patterns of cable
operators are dependent on a variety of factors, including the following:

  .  the availability of financing;

  .  cable operators' annual budget cycles, as well as the typical reduction
     in upgrade projects during the winter months;

  .  the status of federal, local and foreign government regulation and
     deregulation of the telecommunications industry;

  .  overall demand for cable services;

  .  competitive pressures (including the availability of alternative data
     transmission and access technologies);

  .  discretionary consumer spending patterns; and

  .  general economic conditions.

   In recent periods, the United States cable market has been characterized by
the acquisition of smaller and independent cable operators by large cable
operators. We cannot predict the effect, if any, that consolidation in the
United States cable industry will have on overall capital spending patterns by
cable operators. The effect on our business of further industry consolidation
also is uncertain.

Supply of our products may be limited by our ability to forecast demand
accurately.

   The emerging nature of the broadband access services market makes it
difficult for us to accurately forecast demand for our products. Our inability
to accurately forecast the actual demand for our products could result in
supply, manufacturing or testing capacity constraints. These constraints could
result in delays in the delivery of our products or the loss of existing or
potential customers, either of which could have a negative impact on our
business, operating results or financial condition. In addition, we had
unconditional purchase obligations of approximately $125.7 million as of March
31, 2000, primarily to purchase minimum quantities of materials and components
used to manufacture our products. We must fulfill these obligations even if
demand for our products is lower than we anticipate.

We are dependent on key third-party suppliers.

   We manufacture all of our products using components or subassemblies
procured from third-party suppliers. Some of these components are available
from a single source and others are available from limited sources. All of our
sales are from products containing one or more components that are available
only from single supply sources. In addition, some of the components are custom
parts produced to our specifications. For example, we currently rely on Philips
Semiconductors, Inc. to supply a custom ASIC that is used in our products.
Other components, such as the radio frequency tuner and some surface acoustic
wave filters, are procured from sole source suppliers. Any interruption in the
operations of vendors of sole source parts could adversely affect our ability
to meet our scheduled product deliveries to customers. We are dependent on
semiconductor manufacturers and are affected by worldwide conditions in the
semiconductor market. If we are unable to obtain a sufficient supply of
components from our current sources, we could experience difficulties in
obtaining alternative sources or in altering product designs to use alternative
components. Resulting delays or reductions in product shipments could damage
customer relationships. Further, a significant increase in the price of one or
more of these components could harm our gross margin or operating results.

                                       12

<PAGE>

Shortages in supplies of components may impair our ability to meet customer
demands.

   Due to increasing demand for electronic and communications equipment, the
worldwide market for component parts is currently constrained. Delays in key
component or product deliveries may occur due to shortages resulting from a
limited number of suppliers, the financial or other difficulties of such
suppliers or a limitation in component product availability. Due to these
current market conditions, we face the risk of possible shortages of certain
key components that could result in product performance shortfalls and reduced
control over or delay in delivery schedules, manufacturing capability, quality
and costs, all of which could impair our ability to produce enough product to
meet our customer demand. In addition, in order to fulfill demand for our
products, we may have to purchase these components on the spot market at a
price that may be higher than we have experienced in the past.

   Although we work closely with our suppliers to avoid these types of
shortages, there can be no assurance that we will not encounter these problems
in the future. While our suppliers have performed effectively and been
relatively flexible to date, we believe that we will be faced with the
following challenges going forward:

  .  New markets in which we participate may grow quickly and consume
     component capacity; and

  .  As we continue to acquire companies and new technologies, we are
     dependent, at least initially, on unfamiliar supply chains or relatively
     small supply partners.

   Manufacturing capacity and component supply constraints could be significant
issues for us. If we were unable to obtain adequate quantities of significant
component materials on a timely basis, our business and our customer
relationships would be adversely affected. If we are unable to satisfy our
customers' demand, our customers could decide to purchase products from our
competitors. Inability to meet demand, or a decision by one or more of our
customers to purchase from our competitors, could harm our operating results.

We may be unable to migrate to new semiconductor process technologies
successfully or on a timely basis.

   Our future success will depend in part upon our ability to develop products
that utilize new semiconductor process technologies. These technologies change
rapidly and require us to spend significant amounts on research and
development. We continuously evaluate the benefits of redesigning our
integrated circuits using smaller geometry process technologies to improve
performance and reduce costs. The transition of our products to integrated
circuits with increasingly smaller geometries will be important to our
competitive position. Other companies have experienced difficulty in migrating
to new semiconductor processes and, consequently, have suffered reduced yields,
delays in product deliveries and increased expense levels. Moreover, we depend
on our relationship with our third-party manufacturers to migrate to smaller
geometry processes successfully.

Our ability to directly control product delivery schedules and product quality
is dependent on third-party contract manufacturers.

   Most of our products are assembled and tested by contract manufacturers
using testing equipment that we provide. As a result of our dependence on these
contract manufacturers for assembly and testing of our products, we do not
directly control product delivery schedules or product quality. Any product
shortages or quality assurance problems could increase the costs of
manufacture, assembly or testing of our products. In addition, as manufacturing
volume increases, we will need to procure and assemble additional testing
equipment and provide it

                                       13

<PAGE>

to our contract manufacturers. The production and assembly of testing equipment
typically requires significant time. We could experience significant delays in
the shipment of our products if we are unable to provide this testing equipment
to our contract manufacturers in a timely manner.

There are many risks associated with international operations.

   Sales to customers outside of the United States accounted for approximately
84% of our revenues in 1999 and approximately 74% of our revenues in 1998. We
expect sales to customers outside of the United States to continue to represent
a significant percentage of our revenues for the foreseeable future.
International sales are subject to a number of risks, including the following:

  .  changes in foreign government regulations and communications standards;

  .  export license requirements, tariffs and taxes;

  .  other trade barriers;

  .  difficulty in protecting intellectual property;

  .  difficulty in collecting accounts receivable;

  .  difficulty in managing foreign operations; and

  .  political and economic instability.

   If our customers are affected by currency devaluations or general economic
crises, such as the recent economic crisis affecting many Asian and Latin
American economies, their ability to purchase our products could be reduced
significantly. Payment cycles for international customers typically are longer
than those for customers in the United States. Foreign markets for our products
may develop more slowly than currently anticipated. Foreign countries may
decide not to construct cable infrastructure or may prohibit, terminate or
delay the construction of new cable plants for a variety of reasons. These
reasons include environmental issues, economic downturns, the availability of
favorable pricing for other communications services or the availability and
cost of related equipment. Any action like this by foreign countries would
reduce the market for our products.

   We anticipate that our foreign sales generally will be invoiced in U.S.
dollars, and we currently do not plan to engage in foreign currency hedging
transactions. However, as we commence and expand our international operations,
we may be paid in foreign currencies and exposure to losses in foreign currency
transactions may increase. We may choose to limit our exposure by the purchase
of forward foreign exchange contracts or through similar hedging strategies. No
currency hedging strategy can fully protect against exchange-related losses. In
addition, if the relative value of the U.S. dollar in comparison to the
currency of our foreign customers should increase, the resulting effective
price increase of our products to those foreign customers could result in
decreased sales.

We may be unable to provide adequate customer support.

   Our ability to achieve our planned sales growth and retain current and
future customers will depend in part upon the quality of our customer support
operations. Our customers generally require significant support and training
with respect to our broadband access systems, particularly in the initial
deployment and implementation stages. To date our sales have been concentrated
in a small number of customers. We have limited experience with widespread
deployment of our products to a diverse customer base. We may not have

                                       14

<PAGE>

adequate personnel to provide the levels of support that our customers may
require during initial product deployment or on an ongoing basis. Our inability
to provide sufficient support to our customers could delay or prevent the
successful deployment of our products. In addition, our failure to provide
adequate support could harm our reputation and relationship with our customers
and could prevent us from gaining new customers.

Our industry is highly competitive with many established competitors.

   The market for broadband access systems is extremely competitive and is
characterized by rapid technological change. Our direct competitors in the
cable access systems arena include Cisco Systems, Com21, General Instrument,
Matsushita Electric Industrial (which markets products under the brand name
"Panasonic"), Motorola, Nortel Networks, Vyyo, Thomson Consumer Electronics
(which markets products under the brand name "RCA"), Samsung, Scientific-
Atlanta, Sony, 3Com, Toshiba and Zenith Electronics. We also compete with
companies that develop integrated circuits for broadband access products, such
as Broadcom, Conexant and Texas Instruments. We also sell products that compete
with existing data access and transmission systems utilizing the
telecommunications networks, such as those of 3Com. Additionally, our
controller and headend system products face intense competition from well-
established companies such as Cisco, Nortel and 3Com. In addition, we compete
with companies in the DSL arena such as ECI, Charles Industries, Pairgain,
Copper Mountain, Accelerated Networks, Integral Access and VINA Technologies.
As standards, such as DOCSIS, are developed for broadband access systems, other
companies may enter the broadband access systems market. The principal
competitive factors in our market include the following:

  .  product performance, features and reliability;

  .  price;

  .  size and stability of operations;

  .  breadth of product line;

  .  sales and distribution capability;

  .  technical support and service;

  .  relationships with providers of broadband access services; and

  .  compliance with industry standards.

   Some of these factors are outside of our control. The existing conditions in
the broadband access market could change rapidly and significantly as a result
of technological advancements. The development and market acceptance of
alternative technologies could decrease the demand for our products or render
them obsolete. Our competitors may introduce broadband access products that are
less costly, provide superior performance or achieve greater market acceptance
than our products.

   Many of our current and potential competitors have significantly greater
financial, technical, marketing, distribution, customer support and other
resources, as well as greater name recognition and access to customers than we
do. The anticipated widespread adoption of DOCSIS and other industry standards
is likely to cause increased worldwide price competition, particularly in the
North American market. The adoption of DOCSIS and these other standards also
could result in lower sales of our TeraComm system, including the higher margin
headend products. Any increased price competition or reduction in sales of our
headend products would result in downward pressure on our gross margin. We
cannot accurately predict how the competitive pressures that we face will
affect our business.

                                       15

<PAGE>

Our business is dependent on the Internet and the development of the Internet
infrastructure.

   Our success will depend in large part on increased use of the Internet to
increase the need for high speed broadband access networks. Critical issues
concerning the commercial use of the Internet remain largely unresolved and are
likely to affect the development of the market for our products. These issues
include security, reliability, cost, ease of access and quality of service. Our
success also will depend on the growth of the use of the Internet by
businesses, particularly for applications that utilize multimedia content and
thus require high bandwidth. The recent growth in the use of the Internet has
caused frequent periods of performance degradation. This has required the
upgrade of routers, telecommunications links and other components forming the
infrastructure of the Internet by Internet service providers and other
organizations with links to the Internet. Any perceived degradation in the
performance of the Internet as a whole could undermine the benefits of our
products. Potentially increased performance provided by our products and the
products of others ultimately is limited by and reliant upon the speed and
reliability of the Internet backbone itself. Consequently, the emergence and
growth of the market for our products will depend on improvements being made to
the entire Internet infrastructure to alleviate overloading and congestion.

Our failure to manage growth could adversely affect us.

   The growth of our business has placed, and is expected to continue to place,
a significant strain on our limited personnel, management and other resources.
Our management, personnel, systems, procedures and controls may be inadequate
to support our existing and future operations. To manage any future growth
effectively, we will need to attract, train, motivate, manage and retain
employees successfully, to integrate new employees into our overall operations
and to continue to improve our operational, financial and management systems.

We are dependent on key personnel.

   Due to the specialized nature of our business, we are highly dependent on
the continued service of, and on the ability to attract and retain qualified
engineering, sales, marketing and senior management personnel. The competition
for personnel is intense. The loss of any of these individuals, particularly
our Chairman, President and Chief Technical Officer, Shlomo Rakib, and our
Chief Executive Officer, Zaki Rakib, would harm our business. In addition, if
we are unable to hire additional qualified personnel as needed, we may be
unable to adequately manage and complete our existing sales commitments and to
bid for and execute additional sales. Further, we must train and manage our
growing employee base, which is likely to require increased levels of
responsibility for both existing and new management personnel. Our current
management personnel and systems may be inadequate, and we may fail to
assimilate new employees successfully.

   Highly skilled employees with the education and training that we require,
especially employees with significant experience and expertise in both data
networking and radio frequency design, are in high demand. We may not be able
to continue to attract and retain the qualified personnel necessary for the
development of our business. We do not have "key person" insurance coverage for
the loss of any of our employees. Any officer or employee of our company can
terminate his or her relationship with us at any time. Our employees generally
are not bound by non-competition agreements with us.

Our business is subject to the risks of product returns, product liability and
product defects.

   Products as complex as ours frequently contain undetected errors or
failures, especially when first introduced or when new versions are released.
Despite testing, errors may occur.

                                       16

<PAGE>

The occurrence of errors could result in product returns and other losses to
our company or our customers. This occurrence also could result in the loss of
or delay in market acceptance of our products. Due to the recent introduction
of our products, we have limited experience with the problems that could arise
with this generation of products. However, the limitation of liability
provision contained in our purchase agreements may not be effective as a result
of federal, state or local laws or ordinances or unfavorable judicial decisions
in the United States or other countries. We have not experienced any product
liability claims to date, but the sale and support of our products entails the
risk of such claims. In addition, any failure by our products to properly
perform could result in claims against us by our customers. We maintain
insurance to protect against certain claims associated with the use of our
products, but our insurance coverage may not adequately cover any claim
asserted against us. In addition, even claims that ultimately are unsuccessful
could result in our expenditure of funds in litigation and management time and
resources.

We may be unable to adequately protect or enforce our intellectual property
rights.

   We rely on a combination of patent, trade secret, copyright and trademark
laws and contractual restrictions to establish and protect proprietary rights
in our products. Our pending patent applications may not be granted. Even if
they are granted, the claims covered by the patents may be reduced from those
included in our applications. Any patent might be subject to challenge in court
and, whether or not challenged, might not be broad enough to prevent third
parties from developing equivalent technologies or products without taking a
license from us. We have entered into confidentiality and invention assignment
agreements with our employees, and we enter into non-disclosure agreements with
some of our suppliers, distributors and appropriate customers so as to limit
access to and disclosure of our proprietary information. These statutory and
contractual arrangements may not prove sufficient to prevent misappropriation
of our technology or to deter independent third-party development of similar
technologies. In addition, the laws of some foreign countries might not protect
our products or intellectual property rights to the same extent as do the laws
of the United States. Protection of our intellectual property might not be
available in every country in which our products might be manufactured,
marketed or sold.

   In November 1998, CableLabs selected us to co-author DOCSIS 1.2, an enhanced
version of the DOCSIS cable modem specification based in part on our S-CDMA
technology. In September 1999, CableLabs indicated that it intended to proceed
with the advanced PHY work on two parallel tracks: one for the development of a
prototype based on our S-CDMA technology and one for the inclusion of Advanced
TDMA technology, as proposed by other companies. In February 2000, CableLabs
further clarified the status of the advanced PHY project regarding a separate
release that will include TDMA technologies. In addition, CableLabs reiterated
that it is continuing to work with us on the development of a DOCSIS
specification that could include our S-CDMA technology. To that end, we have
indicated to CableLabs that we would contribute some aspects of our S-CDMA
technology to the DOCSIS intellectual property pool if and when a DOCSIS
specification is approved that includes our S-CDMA technology.

   We would contribute our technology pursuant to a license agreement with
CableLabs that we would execute at that time, and which contains the terms that
CableLabs has established for the inclusion of any intellectual property from
any source in the DOCSIS specifications. Under the terms of the proposed
license agreement, we would grant to CableLabs a royalty-free license for those
aspects of our S-CDMA technology that are essential for compliance with the
DOCSIS cable modem standard. So-called "implementation know how" is not covered
by this license-only those aspects of the technology that are essential to
implementing a

                                       17

<PAGE>

compliant product. CableLabs would have the right to extend royalty-free
sublicenses to companies that wish to build DOCSIS-compatible products. These
sublicenses would allow participating companies to utilize and incorporate the
essential portions of the S-CDMA technology on a royalty-free basis for the
limited use of making and selling products or systems that comply with the
DOCSIS cable modem specification. Terayon has already joined the DOCSIS
intellectual property pool and, as a result, we have a royalty-free sublicense
that allows us to ship DOCSIS-compatible products which contain intellectual
property submitted by other companies. The scope of this license would not
extend to the use of the S-CDMA technology in other areas; only for products
that comply with the DOCSIS specifications. As a result, any of our competitors
who join or have joined the DOCSIS intellectual property pool will have access
to some aspects of our technology without being required to pay us any
royalties or other compensation. If and when we submit S-CDMA to the DOCSIS
Intellectual Property pool, we are in no way restricted from entering into
royalty-bearing license agreements with companies that wish to use the S-CDMA
technology for purposes other than implementing DOCSIS compatible products, or
that do not wish to enter into the DOCSIS intellectual property pool. Further,
some of our competitors have been successful in reverse engineering the
technology of other companies, and the inclusion of S-CDMA in a future DOCSIS
specification would expose some aspects of our technology to those competitors.
DOCSIS specifications are available on an open basis once they are approved,
not only to companies that are members of the DOCSIS IP Pool. If a competitor
is able to duplicate the functionality and capabilities of our technology, we
could lose all or some of the time-to-market advantage we might otherwise have.
Under the terms of the proposed license agreement, if we sue certain parties to
the proposed license agreement on claims of infringement of any copyright or
patent right or misappropriation of any trade secret, those parties may
terminate our license to the patents or copyrights they contributed to the
DOCSIS intellectual property pool. If a termination like this were to occur, we
would continue to have access to some aspects of the DOCSIS intellectual
property pool, but we would not be able to develop products that fully comply
with the DOCSIS cable modem specification. Also, even if we were to be removed
from the IP pool, we would not be prevented from developing and selling
products that fully comply with the DOCSIS specifications, but we would not be
able to do this with the benefit of a royalty-free license, which would
increase the cost of our products, assuming we were able to obtain a license
agreement for the required technology. Because of these terms, we may find it
difficult to enforce our intellectual property rights against certain
companies, even in areas that are not directly related to DOCSIS specifications
and products.

   We anticipate that developers of cable modems increasingly will be subject
to infringement claims as the number of products and competitors in our
industry segment grows. We have received letters from two individuals claiming
that our technology infringes patents held by these individuals. We have
reviewed the allegations made by these individuals and, after consulting with
our patent counsel, we do not believe that our technology infringes any valid
claim of these individuals' patents. If the issues are submitted to a court,
the court could find that our products infringe these patents. In addition,
these individuals may continue to assert infringement. If we are found to have
infringed these individuals' patents, we could be subject to substantial
damages and/or an injunction preventing us from conducting our business. In
addition, other third parties may assert infringement claims against us in the
future. An infringement claim, whether meritorious or not, could be time-
consuming, result in costly litigation, cause product shipment delays or
require us to enter into royalty or licensing agreements. These royalty or
licensing agreements may not be available on terms acceptable to us or at all.
Litigation also may be necessary to enforce our intellectual property rights.

   We pursue the registration of our trademarks in the United States and have
applications pending to register several of our trademarks. However, the laws
of certain foreign countries

                                       18

<PAGE>

might not protect our products or intellectual property rights to the same
extent as the laws of the United States. This means that effective trademark,
copyright, trade secret and patent protection might not be available in every
country in which our products might be manufactured, marketed or sold.

Our business is subject to communications industry regulations.

   Our business and our customers are subject to varying degrees of federal,
state and local regulation. The jurisdiction of the Federal Communications
Commission extends to the communications industry, including our broadband
access products. The FCC has promulgated regulations that, among other things,
set installation and equipment standards for communications systems. Although
FCC regulations and other governmental regulations have not materially
restricted our operations to date, future regulations applicable to our
business or our customers could be adopted by the FCC or other regulatory
bodies. For example, FCC regulatory policies affecting the availability of
cable services and other terms on which cable companies conduct their business
may impede our penetration of certain markets. In addition, regulation of cable
television rates may affect the speed at which cable operators upgrade their
cable infrastructures to two-way HFC. In addition, the increasing demand for
communications systems has exerted pressure on regulatory bodies worldwide to
adopt new standards for such products and services. This process generally
involves extensive investigation of and deliberation over competing
technologies. The delays inherent in this governmental approval process have in
the past, and may in the future, cause the cancellation, postponement or
rescheduling of the installation of communications systems by our customers.

   If other countries begin to regulate the cable modem industry more heavily
or introduce standards or specifications with which our products do not comply,
we will be unable to offer products in those countries until our products
comply with those standards or specifications. In addition, we may have to
incur substantial costs to comply with those standards or specifications. For
instance, should the DAVIC standards for ATM-based digital video be established
internationally, we will need to conform our cable modems to compete. Further,
many countries do not have regulations for installation of cable modem systems
or for upgrading existing cable network systems to accommodate our products.
Whether we currently operate in a country without these regulations or enter
into the market in a country these regulations do not exist, new regulations
could be proposed at any time. The imposition of regulations like this could
place limitations on a country's cable operators' ability to upgrade to support
our products. Cable operators in these countries may not be able to comply with
these regulations, and compliance with these regulations may require a long,
costly process. For example, we experienced delays in product shipments to a
customer in Brazil due to delays in certain regulatory approvals in Brazil.
Similar delays could occur in other countries in which we market or plan to
market our products. In addition, our customers in certain parts of Asia, such
as Japan, are required to obtain licenses prior to selling our products, and
delays in obtaining required licenses could harm our ability to sell products
to these customers.

Our business is subject to other regulatory approvals and certifications.

   In the United States, in addition to complying with FCC regulations, our
products are required to meet certain safety requirements. For example, we are
required to have our products certified by Underwriters Laboratory in order to
meet federal requirements relating to electrical appliances to be used inside
the home. Outside the United States, our products are subject to the regulatory
requirements of each country in which the products are manufactured or sold.
These requirements are likely to vary widely. We may be unable to obtain on a
timely basis or at all the regulatory approvals that may be required for the

                                       19

<PAGE>

manufacture, marketing and sale of our products. In addition to regulatory
compliance, some cable industry participants may require certification of
compatibility.

We are vulnerable to earthquakes and other natural disasters.

   The facility housing our corporate headquarters, the majority of our
research and development activities and our in-house manufacturing operations
is located in an area of California known for seismic activity. In addition,
the operations of some of our key suppliers are also located in this area and
in other areas know for seismic activity, such as Taiwan. An earthquake, or
other significant natural disaster, could result in an interruption in our
business or that of one or more of our key suppliers. Such an interruption
could harm our operating results.

                                       20

<PAGE>

Forms 10(K) and 10(Q)

  The Company hereby amends Form 10-K filed on March 30, 2000 and Form 10-Q
filed on May 15, 2000 and any amendments filed thereon to revise the sections,
where applicable, entitled Business, Management's Discussion and Analysis of
Financial Condition and Results of Operations and Certain Business Risk.

  The section entitled Business should be deleted and replaced with the
following section:


                                    BUSINESS

Overview

   We develop, market and sell broadband access systems that enable cable
operators and other providers of broadband access services to cost effectively
deploy reliable broadband access services over cable, copper wire utilizing DSL
and wireless systems. We have substantial development and marketing resources
focused on cable operators. However, through internal development and recent
acquisitions of complementary technology and businesses, we also are focusing
on service providers that offer broadband access through existing copper wire
infrastructures and wireless systems.

   Our primary product is the TeraComm system, which is based on our patented
S-CDMA technology. Our S-CDMA technology enables reliable two-way broadband
data communications over both pure coaxial and hybrid fiber/coax cable
infrastructure and is designed to enable cable operators to maximize the
capacity and reliability of broadband data services over any cable plant. In
furtherance of our strategy to provide a complete portfolio of broadband
products, we have recently completed several acquisitions of complementary
broadband technologies, including digital video management systems, DSL and
wireless.

   We sell our broadband access products to cable operators and other providers
of broadband access services through direct sales forces in North America,
South America Europe and Asia. We also distribute our products via distributors
and systems integrators. Companies currently using or distributing our TeraComm
system include Rogers, Shaw, TCA Cable TV (a subsidiary of Cox Communications,
Inc.), UPC and Crossbeam, a wholly owned subsidiary of Sumitomo. Companies
currently using our DSL products include major ILEC's (Incumbent Local Exchange
Carriers) in the U.S. including SBC, Bell Atlantic, Bell South, U.S. West and
GTE.

Industry Background

 Demand for Broadband Access

   In recent years, the volume of bandwidth intensive data, voice and video
traffic across cable infrastructures, the Internet, corporate intranets and
other public networks has increased dramatically. This demand has been driven
by the proliferation of residential and commercial users that are accessing
networks for a variety of applications, including voice, video and data
communication via the Internet, electronic commerce, telecommuting and
interactive television. For example, International Data Corporation, estimates
that the number of worldwide Internet users will increase from approximately
200 million at the end of 2000 to over one billion by the end of 2005. IDC
estimates that the number of homes in the United States with broadband access
will increase from two million at the end of 1999 to 20 million by the end of
2003. The rapid evolution of broadband has resulted in cable operators,
providers of telephone services and other service providers seeking to provide
a bundle of voice, data and video services to their residential and commercial
subscribers over existing and new infrastructures.

   Despite significant advances in the performance of computer processors and
data backbone networks, high speed data transmission has been limited by the
existing local access network infrastructure (also known as the last mile),
which is not optimized for distribution of data-intensive multimedia content.
Users of dial-up analog modems with

                                       21
<PAGE>

maximum data rates of only 56 Kbps often experience frustration, as they
encounter frequent and lengthy delays or complete failures in transmission. In
response to the growing demand for increased bandwidth, service providers are
now deploying new broadband access technologies, such as DSL, cable modems and
wireless devices, that can deliver performance that is from 50 to 100 times
faster than analog modems. According to industry analysts, the market for
broadband access equipment is expected to increase from approximately
$4 billion at the end of 1999 to approximately $11 billion in 2004.

 Emergence of Broadband Technologies

   As residential and commercial demand for increased bandwidth and faster
access continues to grow, particularly for applications such as streaming audio
and video, IP telephony and interactive two-way video, service providers are
investing in enabling infrastructure and technologies. Various technologies
have emerged to address the need for broadband access across cable, copper wire
and wireless platforms or networks.

   Cable operators market high speed cable modems that use the existing
television cable plant to provide subscribers access to the Internet, digital
video, telephony and other broadband services. Similarly, telephone companies
offer comparable services using DSL technology over the existing copper wire
infrastructure. Wireless operators also are establishing high speed Internet
access using both terrestrial and satellite systems to connect subscribers to
voice, video and data services.

   The competition between cable operators, telephone companies and emerging
wireless service providers is expected to intensify as the demand for broadband
access grows. Cable operators benefit from the fact that the cable
infrastructure passes over 95% of the homes in the United States and a large
number of small businesses. In addition, the cable infrastructure has the
potential to offer a wide range of broadband services, such as digital TV,
Internet access and IP telephony. Similarly, DSL providers benefit from the
fact that existing copper lines reach a substantial majority of U.S. homes.
Wireless, in contrast, does not depend on existing cable or telephone line
infrastructure and can be quickly deployed in homes and businesses that
otherwise might not have broadband access, particularly for international
markets. Although the market for broadband access services is becoming
increasingly competitive, industry analysts expect each platform to experience
significant growth over the next five years.

 Challenges in Deploying Broadband Access Services

   Service providers must overcome numerous challenges in order to successfully
meet the increasing demand for broadband access services. Among other things,
these service providers must do the following:

  .  provide end-users with a broad range of bundled services;

  .  capitalize on available bandwidth to provide a variety of services;

  .  rapidly deploy services to their customers; and

  .  provide reliable and cost-effective access.

The Terayon Solution

   Our products are designed to enable cable operators and other providers of
broadband access services to cost-effectively deploy reliable broadband
services over cable, DSL and wireless systems. Until recently, our development
and marketing efforts were focused

                                       22

<PAGE>

primarily on cable operators. As a result of some recent acquisitions of
complementary technology and businesses, we also are focusing on providers of
broadband services using existing telephone, DSL and wireless systems.

   Our products offer cable operators and other broadband providers the
following advantages:

   Enable enriched set of bundled services.  Our broadband access platforms are
designed to enable service providers to deliver a broad range of voice, video
and data services over either a cable, DSL or wireless network. Service
providers are able to reduce their total network costs because they can deliver
additional services without investing in separate network infrastructures. In
addition, because our broadband access platforms facilitate the bundling of
voice, video and data services, service providers are better able to increase
their revenues and differentiate themselves by offering tailored services and a
single point of contact to customers.

   Maximize bandwidth utilization. Our broadband access platforms allow cable
operators, telecommunications carriers and other providers of broadband
services to optimize utilization of the available bandwidth for data access,
digital video, and voice services. For example, through our proprietary S-CDMA
technology, cable operators can maximize the use of the upstream return path
using the lowest frequency ranges of the upstream spectrum, where noise is
typically too severe to operate. This use of the upstream spectrum is key to
delivering efficient interactive data services. Our Cherrypicker digital video
management systems allow cable operators to flexibly manage bandwidth by
allowing them to customize digital video programming for their target
demographics.

   Enable rapid time to market of new services. Our broadband access products
are designed to enable service providers to achieve rapid time to market and
offer additional voice, video and data services as they become available, by
leveraging existing plant and infrastructure. The rapid deployment of broadband
services is a key competitive advantage to our customers.

   Provide reliable and cost effective solutions. Our solutions are designed to
provide a high degree of reliability in a cost-effective manner. Our cable
modem system allows cable operators to deploy data services over a wide variety
of plant conditions, thus reducing ongoing maintenance costs and minimizing
service problems. Due to their reliability, our DSL systems enable service
providers to deliver mission-critical services, such as toll-quality local and
long distance voice over DSL to their customers. Wireless systems are the most
cost-effective means for delivering broadband access services to customers who
cannot otherwise access existing cable and telephone infrastructure,
particularly in rural areas and developing countries.

Strategy

   Our objective is to be the leading provider of broadband systems to cable
operators, providers of telephone services and other service providers seeking
to provide broadband services to residential and commercial end users. Key
elements of our strategy are as follows:

   Build a complete portfolio of broadband products. We are building a complete
portfolio of broadband products to support high speed delivery of voice, video
and data services over cable, copper wire and wireless infrastructures through
internal development efforts and strategic acquisitions. Until recently our
internal development efforts have focused on developing our proprietary S-CDMA
technology for cable modems and extending this technology to a standards-based
environment. Our future development efforts will be focused on the development
of new technologies for the delivery of voice, video and data over any

                                       23

<PAGE>

broadband medium. In furtherance of this strategy, during the last year we have
made several acquisitions of broadband technologies including digital video
management systems, DSL and wireless.

   Supply leading broadband service providers worldwide. We target the largest
broadband service providers in each major geographic area worldwide. Three of
the largest North American cable operators, Rogers, Shaw and TCA Cable TV (a
subsidiary of Cox Communications, Inc.) are deploying our TeraComm cable modem
system. In Japan, through our partner Sumitomo, we have established ourselves
as a market leader. In Europe, Our TeraComm system is being deployed by UPC,
one of Europe's largest broadband communications companies. We market our DSL
products to leading providers of telephony services throughout the world. The
acquisition of ANE provides us with a strong customer presence in the U.S.
communications market. Our customers include the major U.S. ILECs (Incumbent
Local Exchange Carriers), including SBC, Bell Atlantic, Bell South, U.S. West
and GTE.

   Increase presence in existing and new markets. We intend to increase our
presence in new and existing markets throughout North America, Europe, Asia and
South America. We currently sell our products in over 20 countries worldwide.
As part of our effort to increase our market penetration and to capitalize on
new opportunities, we plan to continue to expand our sales, marketing and
customer support capabilities through both our direct sales and marketing staff
as well as our relationships with distributors. We believe that a physical
presence in key customer locations provides an important advantage in
developing and maintaining our new and existing customer relationships.

   Extend technology leadership and advance industry standards. A significant
element of our strategy is to advance industry standards and to extend our
technology leadership through the development of innovative new technologies.
We intend to leverage our engineering capabilities to expand the features and
functionality of broadband products. We will strive to use our technology
leadership to achieve a significant time-to-market advantage over competing
broadband access suppliers by offering advanced features, such as IP telephony
and videoconferencing. In addition, we intend to apply S-CDMA to additional
applications such as wireless communications and Local Multipoint Distribution
Systems. We will continue to participate in industry standards organizations
and activities for DOCSIS, DVB, MPEG and PacketCable.

   Provide superior customer service and support. We believe that our ability
to provide consistent high quality service and support is a key factor in
attracting and retaining customers. We provide our customers with technical
support and training through customer support representatives and
representatives of distributors. We offer service and support to our customers
24 hours a day, seven days a week.

                                       24

<PAGE>

Products

   Our current portfolio of products complements our strategy to be the leading
provider of broadband systems to cable television operators, providers of
telephone services and other service providers seeking to provide broadband
access services to residential and commercial users.

<TABLE>
<CAPTION>
     Product Family     Product
     --------------     ---------------------------------------------
     <S>                <C>
     Cable Systems      Terayon S-CDMA Products

                        TeraLink 1000 S-CDMA Master Controller
                        TeraLink S-CDMA Gateway
                        TeraPro S-CDMA Cable Modem
                        TeraView Network Management Software

                        Terayon DOCSIS Products

                        TeraLink 2000 DOCSIS CMTS
                        TeraJet DOCSIS Cable Modem

                        Terayon DVB Cable Products

                        UCMT-2000 Interactive Network Adaptor
                        UCM-220 Cable Modem
                        HAS-2000 High Availability Server

                        Multigate Telephony System

     Video Systems      CherryPicker 500
                        CherryPicker 7000
                        CherryPicker 1000
                        CherryPicker AdSplicer

     DSL Systems        Miniplex DSL Systems
                        IPTL Converged Voice and Data Services System
                        Highlink

     Wireless Products  SatStream--Satellite modem
                        NetStream--Broadband Satellite IP Gateway
                        WebStream--Broadband Satellite IP Gateway
                        HAS-2000--High Availability Server
</TABLE>

 Cable Systems

   Our TeraComm system enables cable operators to cost-effectively deploy
reliable two-way broadband access services. The TeraComm system is comprised of
the TeraPro cable modem, the TeraLink 1000 Master Controller, the TeraLink
Gateway and the TeraView Element Management and Provisioning Software. The
price of a TeraComm system is dependent on a number of variables, including a
customer's basic cable system architecture, the type of routing equipment a
customer intends to use, the level and quality of service that a cable operator
desires to provide and the volume of products purchased by a customer.

   TeraPro Cable Modem. The TeraPro cable modem is a data communications device
installed in a subscriber's home or business. The TeraPro cable modem connects
to the subscriber's PC via a standard 10BaseT Ethernet connector and to the
cable network via a

                                       25

<PAGE>

standard coaxial cable connector. The TeraPro cable modem automatically
configures itself without user intervention, thus minimizing modem installation
time. In addition, the configuration software for the TeraPro cable modem is
downloaded remotely, allowing centralized software upgrades directly from the
headend management system.

   The TeraPro cable modem delivers full two-way communication over the cable
network, with data rates of up to 14 Mbps per 5 MHz channel in both the
upstream and the downstream direction. The TeraPro cable modem operates at full
capacity at an SNR as low as 13dB, and gradually adjusts throughput to provide
transmission at an SNR as low as -13dB. This feature will permit the TeraPro
cable modem to operate across any portion of the 5 to 42 MHz upstream RF
spectrum.

   TeraLink 1000 Master Controller. The TeraLink 1000 Master Controller is a
data channel controller and multiplexer located at the cable headend system or
distribution hub. The TeraLink 1000 Master Controller provides control,
management and data transport functions for TeraPro cable modems connected to
the cable network. It offers dynamic bandwidth management, high speed traffic
concentration, access control to data networking resources and data service
quality and integrity.

   The TeraLink 1000 Master Controller is a single channel, rack-mountable
controller that supports up to 2,000 cable modems per channel. Additional
TeraLink 1000 Master Controllers can be added to scale service as performance
and subscriber needs grow. The TeraLink 1000 Master Controller, with the
TeraLink Gateway, provides a 100 BaseT interface for direct connectivity to a
private backbone or any vendor's router or switch. Alternatively, the TeraLink
1000 Master Controller, with its built-in ATM OC-3 interface, can be connected
via an ATM switch, or directly to Cisco's 7500 series routers.

   TeraLink Gateway. The TeraLink Gateway is a rack-mountable edge concentrator
providing end-user clients with broadband access to a remote IP backbone (e.g.,
Internet) as well as efficient communication between modems. The TeraLink
Gateway includes an ATM OC-3 interface for connectivity to up to two TeraLink
1000 Master Controllers or an ATM switch. It also provides a 10/100 BaseT
Ethernet/Fast Ethernet auto-sense interface to a headend backbone or any IP
router including the Cisco Universal Broadband Router. The TeraLink Gateway
also includes a separate 10 BaseT interface, which may be connected to a
separate management network or the headend network. The TeraLink Gateway
supports up to 2,000 cable modems per RF channel when connected to a TeraLink
1000 Master Controller. When used with the TeraLink Gateway, the TeraPro cable
modems behave as an extension of the TeraLink Gateway, providing maximum
bandwidth and privacy.

   TeraView Element Management and Provisioning Software. The TeraView Element
Management and Provisioning Software is a Windows 95 and Windows NT standards-
based software application installed at the headend system or the network
operations center. The TeraView software allows cable operators to configure,
control, monitor and maintain multiple channels of the TeraComm system.

   Multigate Telephony and Data Access Systems. We produce telephony and data
access systems for deployment of efficient carrier-class voice services over
cable. Our systems also provide in-home networking capability for telephony and
data, based on the Digital Enhanced Cordless Telephony ("DECT") standard. Our
systems support voice, high-speed data, and Integrated Service Digital Network
("ISDN") services for the business telecommuting market and the SOHO market.

                                       26

<PAGE>

 Video Systems

   CherryPicker Digital Video Management Systems. CherryPicker Digital Video
Management Systems are scalable, open platforms that enable cable, satellite
and terrestrial operators to deliver customized digital programming. Patented
statistical re-multiplexing technology gives operators ultimate control over
programming and allows them to create customized video of programming by
selecting feeds (grooming) from a vast array of sources. The CherryPicker helps
operators maximize their available bandwidth while enabling them to provide
program offerings that target specific local demographics.

 DSL Systems

   Digital Subscriber Line Systems (xDSL). We produce communications access
systems based on high-speed IP routing, integrated with telephony. Our
innovative systems include both central office DSLAMs (Digital Subscriber Line
Access Multiplexers) and customer premise equipment for SOHO business broadband
services. These systems deliver high- capacity, symmetric voice and data
services to the business market, with high density port aggregation at the
central office site for maximum cost efficiency.

 Wireless Products

   Our wireless products provide high-speed Internet connectivity to the
subscriber's PC via a standard digital television services antenna. Our
SatStream Satellite modem offers residential and commercial subscribers with
connectivity to the Internet via satellite digital television services.

Customers

   We market our products to providers of broadband access services that seek
to provide broadband access services to both residential and commercial end
users. Our target market is the largest broadband service providers in each
major geographic area worldwide.

   Our principal customers for cable, DSL and wireless systems include the
following:

<TABLE>
<CAPTION>
     Cable                                   DSL
     -----                                   ---
     <S>                                     <C>
     Rogers                                  SBC
     Shaw                                    Bell Atlantic
     TCA Cable TV                            Bell South
       (a subsidiary of Cox Communications)  U.S. West
     Crossbeam                               GTE
       (a subsidiary of Sumitomo)
     UPC

<CAPTION>
     Digital Video Systems                   Wireless Systems
     ---------------------                   ----------------
     <S>                                     <C>
     Cox Communications                      Shiron
     Time Warner                             Teleo
     Adelphia                                Gilat Sat
                                             KBI e-Sat
                                             Adaptive
</TABLE>

   Four customers accounted for approximately 64% of our revenues for the
quarter ended March 31, 2000. We believe that a substantial majority of our
revenues will continue to be derived from sales to a relatively small number of
customers for the foreseeable future. In addition, we believe that sales to
these customers will be focused on a small number of projects.

                                       27

<PAGE>

Research and Development

   We believe that our future success will depend on our ability to enhance our
existing products and to develop and introduce new products that meet a wide
range of evolving broadband access service providers and end user needs. In
addition, to address competitive and pricing pressures, we expect that we will
have to reduce the unit cost of manufacturing our products through design and
engineering changes.

   We currently are designing and developing a DOCSIS system that includes S-
CDMA advanced PHY capabilities, which will include a cable modem and an
accompanying headend controller. These products are in the early stages of
development and we do not anticipate commercial availability of these products
until late 2000 or early 2001. We also are developing a common multimedia
platform for current and advanced transmission of data, voice and video over
existing broadband infrastructures, including cable, DSL and wireless. These
products are in the early stages of development, and we do not anticipate
commercial availability of these products until late 2001.

   On March 18, 1999, we entered into a one-year Product Development Assistance
Agreement ("Development Agreement") with Rogers. Under the terms of the
Development Agreement, Rogers is obligated to provide assistance with the
characterization and testing of our subscriber-end and head-end voice-over-
cable equipment. In addition, Rogers is obligated to provide technology to
assist us in connection with our efforts to develop high quality, field proven
technology solutions that are DOCSIS-compliant and PacketCable-compliant.

Sales and Marketing

   We have direct sales forces in North America, South America, Europe and
Asia. We also distribute our products via distributors and systems integrators.
We have signed a distribution agreement with Sumitomo under which Crossbeam is
distributing the TeraComm system to several of Japan's leading cable operators.
In January 2000, we signed a distribution agreement with BARCO Communication
Systems, for the sale of our CherryPicker systems to cable television operators
worldwide. In June 2000, we signed a distribution agreement with NEC, for the
sale of our IPTL DSL access system throughout Asia and Central and South
America.

   We market our products directly to providers of broadband access services
through our sales force, key distribution and technology partners, as well as
other marketing vehicles such as industry press, trade shows and the Web.
Through our marketing efforts, we strive to educate providers on the
technological and business benefits of our system solution, as well as our
ability to provide quality support and service to the customer. We participate
in the major trade shows and industry events for the broadband access industry
in the United States and we are expanding our presence in other markets through
joint participation at local events with our international sales and marketing
partners. Industry referrals and reference accounts are significant marketing
tools we develop and utilize.

Customer Service and Technical Support

   We believe that our ability to provide consistently high quality service and
support will be a key factor in attracting and retaining customers. Our TSS
(Technical Services and Support) organization, located in North America,
Europe, South America and Asia, offers support 24 hours a day, seven days per
week. Prior to deployment of our systems, each customer's needs

                                       28

<PAGE>

are assessed and proactive solutions are implemented, including various levels
of training, periodic management and coordination meetings, and problem
escalation procedures. We place a strong emphasis on technical training for our
customers. Initial training is offered at no cost, both at our headquarters in
Santa Clara and on our customers' premises.

   In addition to our TSS organization, we have developed sophisticated tools
for remote diagnosis and monitoring of the TeraComm systems deployed by cable
operators. These tools allow us to monitor cable operators' installations of
the TeraLink 1000 Master Controller to suggest solutions proactively before
problems become noticeable to end users. We are developing a Web-based
knowledge system to provide cable operators with access to the latest technical
support information.

Manufacturing

   We outsource the majority of the materials procurement, printed circuit
board assembly, and product assembly and testing to turnkey contract
manufacturers. Currently, we contract with Solectron, located in Milpitas,
California, and CalComp Electronics, located in Thailand, for the manufacture
of the majority of our products. In addition, we also manufacture our products
at subcontractors located in the U.S. and Israel. We have a limited in-house
manufacturing capability at our Santa Clara headquarters. This facility
currently is used for the assembly and final testing of TeraLink 1000 Master
Controllers and TeraLink Gateways. We also use this facility for pilot
production of new product designs, sample testing of products received from
volume manufacturers, developing the manufacturing process and documentation
for new products in preparation for outsourcing. We also perform repair
operations for some of our products returned from customers with our in-house
manufacturing resources.

   Our future success will depend in significant part on our ability to obtain
high volume manufacturing at low costs. As volume increases, we plan to engage
additional contract manufacturers, to procure additional manufacturing
facilities and equipment, to modify existing inventory procedures, to
substantially increase our personnel and to revise our quality assurance and
testing practices.

   Subcontractors supply our contract manufacturers with both standard
components and subassemblies manufactured to our specifications. We depend upon
certain key suppliers for a number of the components for our products. For
example, we currently rely on Philips Semiconductor, Inc. (formerly VLSI
Technology, Inc.) for our S-CDMA ASIC, which is used in our headend and cable
modem products. In addition, all of our products contain one or more components
that currently only are available from a single source.

Competition

   The market for broadband access systems is extremely competitive and is
characterized by rapid technological change. Our direct competitors in the
cable modem arena include Cisco, Com21, Matsushita, Motorola, Nortel, Vyyo,
RCA, Samsung, Scientific-Atlanta, Sony, 3Com, Toshiba and Zenith, and there are
many other potential market entrants. We also sell products that compete with
existing data access and transmission systems utilizing the telecommunications
networks, such as those of 3Com. Additionally, our controller and headend
system products face intense competition from well-established companies such
as Cisco, Nortel and 3Com. Our direct competitors in the DSL arena include ECI
Telecom, Charles Industries, ADC, Copper Mountain, Accelerated Networks,
Integral Access and VINA Technologies. Our direct competitors in the video
arena are Cisco and Harmonic, while our direct competitors in the voice arena
are Nortel, TelLabs, ADC and Alcatel. Our direct competitors in wireless are
Hughes, Marconi, NEC and Scientific-Atlanta.

                                       29

<PAGE>

   The principal competitive factors in the broadband access market include
product performance, features and reliability, price, size and stability of
operations, breadth of product line, sales and distribution capability,
technical support and service, relationships with providers of broadband access
services, standards compliance, and general industry and economic conditions.
Some of these factors are outside of our control. The existing conditions in
the broadband access market could change rapidly and significantly as a result
of technological changes, and the development and market acceptance of
alternative technologies could decrease the demand for our products or render
them obsolete. In addition, these companies and other competitors could
introduce broadband access products that will be less costly or provide
superior performance or achieve greater market acceptance than our products.

   Many of our current and potential competitors have significantly greater
financial, technical, marketing, distribution, customer support and other
resources, as well as greater name recognition and access to customers.

   The market for our products may be impacted by the development of other
technologies that enable the provisioning of broadband access services and the
deployment of services over other media. Widespread acceptance of other
technologies or deployment of services over media not supported by our products
could materially limit acceptance of our broadband access systems. Broadband
access services supported by our products and technology may fail to gain
widespread commercial acceptance by providers of broadband access services and
end users.

Intellectual Property

   We rely on a combination of patent, trade secret, copyright and trademark
laws and contractual restrictions to establish and protect proprietary rights
in our products. Our pending patent applications may not be granted. Even if
they are granted, the claims covered by the patents may be reduced from those
included in our applications. Any patent might be subject to challenge in court
and, whether or not challenged, might not be broad enough to prevent third
parties from developing equivalent technologies or products without taking a
license from us. We have entered into confidentiality and invention assignment
agreements with our employees, and we enter into non-disclosure agreements with
some of our suppliers, distributors and appropriate customers so as to limit
access to and disclosure of our proprietary information. These statutory and
contractual arrangements may not prove sufficient to prevent misappropriation
of our technology or to deter independent third-party development of similar
technologies. In addition, the laws of some foreign countries might not protect
our products or intellectual property rights to the same extent as do the laws
of the United States. Protection of our intellectual property might not be
available in every country in which our products might be manufactured,
marketed or sold.

   In November 1998, CableLabs selected us to co-author DOCSIS 1.2, an enhanced
version of the DOSCIS cable modem specification based in part on our S-CDMA
technology. In September 1999, CableLabs indicated that it intended to proceed
with the advanced PHY work on two parallel tracks: one for the development of a
prototype based on our S-CDMA technology and one for the inclusion of Advanced
TDMA technology, as proposed by other companies. In February 2000, CableLabs
further clarified the status of the advanced PHY project regarding a separate
release that will include TDMA technologies. In addition, CableLabs reiterated
that it is continuing to work with us on the development of a DOCSIS
specification that could include our S-CDMA technology. To that end, we have
indicated to CableLabs that we would contribute some aspects of our S-CDMA
technology to the DOCSIS intellectual property pool if and when a DOCSIS
specification is approved that includes our

                                       30

<PAGE>

S-CDMA technology. We would contribute our technology pursuant to a license
agreement with CableLabs that we would execute at that time, and which contains
the terms that CableLabs has established for the inclusion of any intellectual
property from any source in the DOCSIS specifications. Under the terms of the
proposed license agreement, we would grant to CableLabs a royalty-free license
for those aspects of our S-CDMA technology that are essential for compliance
with the DOCSIS cable modem standard. So-called "implementation know how" is
not covered by this license-only those aspects of the technology that are
essential to implementing a compliant product. CableLabs would have the right
to extend royalty-free sublicenses to companies that wish to build DOCSIS-
compatible products. These sublicenses would allow participating companies to
utilize and incorporate the essential portions of the S-CDMA technology on a
royalty-free basis for the limited use of making and selling products or
systems that comply with the DOCSIS cable modem specification Terayon has
already joined the DOCSIS intellectual property pool and, as a result, we have
a royalty-free sublicense that allows us to ship DOCSIS-compatible products
which contain intellectual property submitted by other companies. The scope of
this license would not extend to the use of the S-CDMA technology in other
areas; only for products that comply with the DOCSIS specifications. As a
result, any of our competitors who join or have joined the DOCSIS intellectual
property pool will have access to some aspects of our technology without being
required to pay us any royalties or other compensation. If and when we submit
S-CDMA to the DOCSIS Intellectual Property pool, we are in no way restricted
from entering into royalty-bearing license agreements with companies that wish
to use the S-CDMA technology for purposes other than implementing DOCSIS
compatible products, or that do not wish to enter into the DOCSIS intellectual
property pool. Further, some of our competitors have been successful in reverse
engineering the technology of other companies, and the inclusion of S-CDMA in a
future DOCSIS specification would expose some aspects of our technology to
those competitors. DOCSIS specifications are available on an open basis once
they are approved, not only to companies that are members of the DOCSIS IP
Pool. If a competitor is able to duplicate the functionality and capabilities
of our technology, we could lose all or some of the time-to-market advantage we
might otherwise have. Under the terms of the proposed license agreement, if we
sue certain parties to the proposed license agreement on claims of infringement
of any copyright or patent right or misappropriation of any trade secret, those
parties may terminate our license to the patents or copyrights they contributed
to the DOCSIS intellectual property pool. If a termination like this were to
occur, we would continue to have access to some aspects of the DOCSIS
intellectual property pool, but we would not be able to develop products that
fully comply with the DOCSIS cable modem specification. Also, even if we were
to be removed from the IP pool, we would not be prevented from developing and
selling products that fully comply with the DOCSIS specifications, but we would
not be able to do this with the benefit of a royalty-free license, which would
increase the cost of our products, assuming we were able to obtain a license
agreement for the required technology. Because of these terms, we may find it
difficult to enforce our intellectual property rights against certain
companies, even in areas that are not directly related to DOCSIS specifications
and products.

   We anticipate that developers of cable modems increasingly will be subject
to infringement claims as the number of products and competitors in our
industry segment grows. We have received letters from two individuals claiming
that our technology infringes patents held by these individuals. We have
reviewed the allegations made by these individuals and, after consulting with
our patent counsel, we do not believe that our technology infringes any valid
claim of these individuals' patents. If the issues are submitted to a court,
the court could find that our products infringe these patents. In addition,
these individuals may continue to assert infringement. If we are found to have
infringed these individuals' patents, we could be subject to substantial
damages and/or an injunction preventing us from conducting our

                                       31

<PAGE>

business. In addition, other third parties may assert infringement claims
against us in the future. An infringement claim, whether meritorious or not,
could be time-consuming, result in costly litigation, cause product shipment
delays or require us to enter into royalty or licensing agreements. These
royalty or licensing agreements may not be available on terms acceptable to us
or at all. Litigation also may be necessary to enforce our intellectual
property rights.

   We pursue the registration of our trademarks in the United States and have
applications pending to register several of our trademarks. However, the laws
of certain foreign countries might not protect our products or intellectual
property rights to the same extent as the laws of the United States. This means
that effective trademark, copyright, trade secret and patent protection might
not be available in every country in which our products might be manufactured,
marketed or sold.

Employees

   As of May 31, 2000, we had 690 employees, of which 351 were in the research
and development group, 147 were in marketing, sales and customer support,
114 were in operations and 78 were in general and administrative functions.
None of our employees is represented by a union. We believe that our relations
with our employees are good.

Properties

   Our headquarters are located in an approximately 60,000 square foot leased
facility located in Santa Clara, California. The current lease for the Santa
Clara facility expires in March 2002. We have sales offices in Atlanta,
Georgia; Sao Paulo, Brazil; and Brussels, Belgium. In addition, we lease
approximately 82,000 square feet in Tel Aviv pursuant to an agreement that
expires in 2005. It is our intention to consolidate the operations of our
Israeli operations in this space. We believe that our existing facilities are
adequate to meet our needs for the immediate future and that our future growth
can be accommodated by leasing additional or alternative space near our current
facilities.

Legal Proceedings

   In September 1999, Imedia, now our subsidiary, was named as a defendant in a
case alleging that Imedia breached its term sheet agreement with the plaintiffs
by negotiating with us while a no-shop provision was in place and refusing to
allowing the plaintiffs to invest in Imedia. The plaintiffs are seeking damages
in excess of $12.0 million. As part of the terms of the Imedia Agreement and
Plan of Merger and Reorganization, shares of our common stock to be issued to
the former shareholders of Imedia were placed in escrow to indemnify us for any
damages that are directly or indirectly suffered as a result any claim brought
by any person who was a prospective investor in Imedia and was not a
securityholder of Imedia on the closing date of the Imedia acquisition. The
value of the escrowed shares was approximately $10.0 million based on the
market value of our common stock on the closing date. The case is in its
initial stages, and no trial date has been established. We have reviewed the
allegations made by the plaintiffs and we do not believe that the outcome will
have a negative impact on our financial position, results of operations or cash
flows.

   In April 2000, a lawsuit against us and certain of our officers and
directors, entitled Birnbaum v. Terayon Communication Systems, Inc., was filed
in the United States District Court for the Central District of California. The
venue for the lawsuit has been moved to the Northern District of California.
The plaintiff purports to be suing on behalf of a class of stockholders who
purchased or obligated themselves to purchase our securities during the period
from February 2, 2000 to April 11, 2000. The complaint alleges that the
defendants

                                       32

<PAGE>

violated the federal securities laws by issuing materially false and misleading
statements and failing to disclose material information regarding our
technology. Several other lawsuits similar to the Birnbaum suit have since been
filed. The lawsuits seek an unspecified amount of damages, in addition to other
forms of relief. We consider the lawsuits to be without merit and we intend to
defend vigorously against these allegations. However, the litigation could
prove to be costly and time consuming to defend, and there can be no assurances
about the eventual outcome.

                                       33

<PAGE>

  The section entitled Management's Discussion and Analysis of Financial
Condition and Results of Operations should be deleted and replaced with the
following section:

                    MANAGEMENT'S DISCUSSION AND ANALYSIS OF
                 FINANCIAL CONDITION AND RESULTS OF OPERATIONS

Overview

   We develop, market and sell broadband access systems. Our objective is to be
the leading provider of broadband access systems to providers of services to
residential and commercial end users. Since our inception in January 1993, we
have focused on the development of our patented S-CDMA technology, as well as
certain other core technologies, to enable broadband transmission of data over
cable networks. We began the specification and design of our first ASIC in
October 1994 and produced the first version of this ASIC in September 1996. At
the same time, we developed an end-to-end broadband access system, the TeraComm
system, around the ASIC. We commenced volume shipments of our TeraComm system
in the first quarter of 1998. With some recent acquisitions of complementary
technology and businesses, in 1999 we also began to turn our attention to
providers of broadband services using existing telephone, copper wire
infrastructures and wireless systems.

   We sell our products both in North America and internationally, and we
market our products primarily to cable operators and distributors. Our strategy
is to supply the leading providers of broadband access services worldwide.
Consistent with this strategy, our initial target market consisted of the ten
largest cable companies in each major geographic area. In most markets, a small
number of large cable operators often provide services to a majority of the
subscribers in a specific region and thus influence the purchase decisions of
smaller cable operators. In North America, three of the largest cable
operators, Rogers Cable Inc. (formerly Rogers Cablesystems Limited), Shaw and
TCA Cable TV (a subsidiary of Cox Communications, Inc.), are deploying the
TeraComm system. In Japan, through our partner Sumitomo, we have established
ourselves as a market leader. In Europe, our TeraComm system is being deployed
by UPC, one of Europe's largest broadband communications companies. As a result
of the nature of the cable industry and our strategic focus, a small number of
customers have accounted for the majority of our revenues to date, and we
expect that the majority of our revenues will continue to be generated from a
small number of customers for the foreseeable future. In the three months ended
March 31, 2000, four customers accounted for approximately 64% of our revenues.
This compares to approximately 63% of our revenues from three customers in the
same period in 1999. We anticipate that the timing and maturity of these
customers' deployments of the TeraComm system will result in variations in
revenues generated from these customers.

   The rapid evolution of broadband has resulted in cable television operators,
providers of telephone service and other service providers seeking to provide a
bundle of voice, data and video services to their residential and commercial
subscribers over existing and new infrastructures. Consistent with our
objective to be a leading provider of broadband access systems, through a
series of recent acquisitions, we are building a complete portfolio of
broadband products to support high-speed delivery of voice, data and video
services over cable, DSL and wireless. In September 1999, we acquired Imedia
and the CherryPicker digital video management system. In November 1999, we
completed our acquisition of Radwiz, a provider of communication access systems
based on high-speed IP routing integrated with telephony. On January 2, 2000,
we completed our acquisition of Telegate, a developer and manufacturer of
telephony and data access platforms that are deployed by service providers to
deliver efficient carrier-class voice services over cable. In addition, we
acquired Combox, a manufacturer of broadband data systems and satellite
communications based on international standards, and certain assets of Internet
Telecom, a supplier of PacketCable and other standards-based, voice-over IP
systems and technologies, through our Israeli subsidiary, Telegate. On April
19, 2000, we acquired Ultracom, a supplier of broadband systems on silicon.
Finally, on April 22, 2000, we acquired ANE, a producer of DSL systems that
provide multiple phone lines over existing copper telephony of a single pair of
copper wires.

                                       34
<PAGE>

   The intensely competitive nature of the market for broadband access systems
has resulted in significant price erosion over time. We have experienced and
expect to continue to experience downward pressure on our unit ASP or average
selling price. A key component of our strategy is to decrease the cost of
manufacturing our products to improve our gross margins. For example, in the
fourth quarter of 1998, we introduced a cable modem using a single-board
design, which has higher gross margins than our original dual-board design.
Subsequently, we have successfully introduced several cost reduced versions of
the single-board modem that have allowed us to continue to improve gross margin
performance despite ongoing ASP declines. We intend to continue to engage in
and implement cost reduction efforts, including the further integration of ASIC
components, other design changes and manufacturing efficiencies.

   We sustained a net loss of $28.4 million for the three months ended March
31, 2000 and $5.9 million in the same period in 1999. We had an accumulated
deficit of $176.8 million as of March 31, 2000. Our operating expenses are
based in part on our expectations of future sales, and we expect that a
significant portion of our expenses will be committed in advance of sales. We
expect to continue to increase our expenditures in technical development and
sales and marketing as we engage in activities related to product enhancement,
cost reduction and increasing market penetration. In addition, we anticipate
that our expenses will increase as a result of the expenses associated with our
recent acquisitions, including amortization. We also anticipate that future
acquisitions will result in increased expense levels. Additionally, we expect
to increase our capital expenditures and other operating expenses in order to
support and expand our operations. We anticipate that we will spend
approximately $17.0 million on capital expenditures and approximately $40.0 to
$50.0 million on research and development during the 12 months ending March 31,
2001. Anticipated capital expenditures consist of purchases of additional test
equipment to support higher levels of production and computer hardware,
furniture and leasehold improvements for expansion of our facilities,
implementation of an enterprise resource planning ("ERP") system and software
and equipment for newly hired employees. As a result of these anticipated
increased operating expenses, we expect to continue to incur losses for the
foreseeable future.

Acquisition of Telegate

   In October 1999, we entered into a Share Purchase Agreement to acquire
Telegate, an Israeli company. Telegate produces telephony and data access
platforms that are deployed by service providers to deliver efficient carrier-
class voice services over cable. Telegate also provides in-home networking
capability for telephony and data, based on the Digital Enhanced Cordless
Telephony (DECT) standard. The transaction was completed on January 2, 2000.
The acquisition was accounted for under the purchase method of accounting and,
accordingly, our Report on Form 10-Q dated May 15, 2000 presents our financial
results through the entire period and combined results with Telegate for the
portion of the period following January 2, 2000.

Recent Events

   Subsequent to March 31, 2000, we completed our acquisitions of ANE, Combox,
Internet Telecom and Ultracom. The Unaudited Pro Forma Combined Condensed
Financial Statements included elsewhere in this document reflect the impact of
the ANE and Combox acquisitions. The Unaudited Pro Forma Combined Condensed
Financial Statements do not include the

                                       35

<PAGE>

impact of the Internet Telecom and Ultracom acquisitions. The purchase price
for Internet Telecom and Ultracom is estimated at approximately $94 million. We
believe that a significant portion of the purchase price will be allocated to
intangible assets, including goodwill. The intangible assets will be amortized
over lives expected to range from two to five years.

   In April 2000, we acquired certain assets and assumed certain liabilities of
ANE, which produces DSL systems that provide multiple phone lines over the
existing copper telephony network. We issued 1,404,552 shares of common stock,
valued at approximately $85.0 million based on the maximum purchase price
specified in the asset purchase agreement. We also established an employee
retention program for purposes of retaining certain identified employees of
ANE. The retention program provides for up to three annual payments to the
identified employees in a total amount of approximately $4.2 million provided
the employees remain employed by us. The retention payments will be charged to
expense over the employees' respective periods of service. We will account for
the acquisition as a purchase transaction.

   In April 2000, we also acquired Combox, an Israeli manufacturer of broadband
data systems and satellite communications based on international standards.
Combox's cable data access systems conform to the growing EuroModem
international specification, based on the Digital Video Broadcasting (DVB)
standard. We issued 1,547,770 shares of common stock valued at approximately
$92.0 million based on the fair market value of our common stock on the days
immediately preceding and following the date the acquisition was announced. In
addition, we issued options to purchase our common stock to the vested and
unvested optionholders of Combox options. We will account for the acquisition
as a purchase transaction.

   In April 2000, our subsidiary Telegate purchased certain assets of Internet
Telecom, an Israeli-based supplier of PacketCable and other standards-based,
voice-over-IP systems and technologies. We issued 377,380 shares of common
stock valued at approximately $44.0 million based on the fair market value of
our common stock on the days immediately preceding and following the
announcement of the acquisition and paid the former shareholders of Internet
Telecom approximately $1.5 million. We will account for the transaction as a
purchase transaction.

   In April 2000, we acquired Ultracom, an Israeli-based supplier of broadband
systems-on-silicon. We issued 536,766 shares of common stock valued at
approximately $50.0 million based on the fair market value of the our common
stock on the days immediately preceding and following the date the acquisition
was announced. We assumed the unvested Ultracom options, the value of which
will be included in the purchase price. We will account for the transaction as
a purchase transaction.

Results of Operations

 Three Months Ended March 31, 2000 and 1999

   Revenues.  Revenues consist primarily of sales of broadband access systems
consisting of cable modems and headend equipment to new and existing customers.
Our revenues increased to $59.3 million for the three months ended March 31,
2000 from $15.9 million for the same period in 1999. The increased revenues in
the first three months of 2000, as compared to the same quarter in 1999, were
largely attributable to continuing deployment of our TeraComm system by
existing customers. The sales of products acquired as a result of our
acquisition of Imedia, Radwiz and Telegate accounted for approximately $8.8
million of the increased revenue in the first quarter of 2000.


                                       36

<PAGE>

   We sell our products directly to broadband access service providers, system
resellers and distributors. Revenues related to product sales are generally
recognized when the products are shipped to the customer, the point at which
title passes. A provision is made for estimated product returns as product
shipments are made. Our existing agreements with our system resellers and
distributors do not contain price protection provisions and do not grant return
rights beyond those provided by our standard warranty.

   Cost of Goods Sold. Cost of goods sold consists of direct product costs as
well as the cost of our manufacturing operations group. The cost of the
manufacturing operations group includes assembly, test and quality assurance
for products, warranty costs and associated costs of personnel and equipment.
For the three months ended March 31, 2000, we incurred cost of goods sold of
$43.9 million compared to $14.0 million for same period in 1999. Cost of goods
sold for the three months ended March 31, 2000 included the cost of our
manufacturing operations group and approximately $6.3 million of amortization
of intangibles assets resulting from the acquisitions of Imedia, Radwiz and
Telegate. Our cost of goods sold increased during the three months ended March
31, 2000, compared to the same period in 1999 primarily due to increased
product shipments.

   Gross Profit. We achieved a gross profit of $15.4 million for the three
months ended March 31, 2000 compared to $1.9 million for the same period in
1999. This improvement in our gross profit was largely the result of continued
cost reduction efforts. We intend to continue to engage in and implement cost
reduction efforts, including the further integration of ASIC components, other
design changes and manufacturing efficiencies. We anticipate that continued
decreases in the average sales price of our products will partially, if not
completely, offset the benefits obtained from further cost reductions.

   Our gross profit also is influenced by the sales mix of TeraLink Master
Controllers, TeraLink Gateways and TeraPro cable modems and the maturity of
TeraComm system deployments in any quarter. TeraPro cable modems have lower
margins than the TeraLink Master Controllers and TeraLink Gateway headend
products. New deployments of TeraComm systems typically include a higher mix of
headend equipment and involve smaller quantities of product sold. Products sold
in connection with new deployments thus generally are sold at higher margins
than products associated with more mature deployments of the TeraComm system,
which generally involve larger quantities of products, primarily cable modems.
We expect that the introduction of new customers and the relationship of
revenues generated from the sale of TeraPro cable modems to our overall
revenues will result in fluctuations in our gross profit in future periods.

   We introduced a CableLabs certified DOSCIS 1.0 cable modem in third quarter
of 1999. We also are currently developing a DOCSIS system that will include a
headend controller. We believe that the widespread adoption of industry
standards will result in further price pressure. We anticipate that the
relationship of revenues generated from the sale of our proprietary TeraComm
system versus a DOCSIS compliant system will result in fluctuations in our
gross profit in future periods. The impact on our gross profit for the three
months ended March 31, 2000 resulting from the sale of our DOCSIS 1.0 cable
modem was not significant.

                                       37

<PAGE>

   Our gross profit also is influenced by the level of sales of products of
acquired companies, which produce varying gross profit results. We expect that
our gross profit will fluctuate in future periods as a result of these and
other acquisitions.

   Research and Development. Research and development expenses consist
primarily of personnel costs, as well as design expenditures, equipment and
supplies required to develop and to enhance our products. Research and
development expenses increased to $10.6 million for the three months ended
March 31, 2000 from $3.2 million in the same period of 1999, primarily as a
result of increased personnel costs. The increased personnel costs were a
result of expansion in our own employee base as we continued to focus our
efforts on developing new products and enhancing our current products. The
increased personnel costs were also the result of the acquisitions of Imedia,
Radwiz and Telegate. We intend to continue to increase investment in research
and development as a result of these and other activities.

   Cost of Product Development Assistance Agreement. In March 1999, we entered
into a one-year Product Development Assistance Agreement with Rogers. Under the
terms of the Development Agreement, Rogers is obligated to assist us with the
characterization and testing of our subscriber-end and head-end voice-over-
cable equipment. In addition, Rogers is obligated to provide us with technology
to assist us with our efforts to develop high quality, field proven technology
solutions that are DOCSIS-compliant and PacketCable-compliant. The Development
Agreement has a term of one year. In consideration of Rogers entering into the
Development Agreement, we issued Rogers two fully vested and non-forfeitable
warrants, each to purchase two million shares of common stock on a cashless
basis. One warrant had an exercise price of $0.50 per share and one warrant had
an exercise price of $18.50 per share. The fair value of the two warrants was
approximately $45.0 million and resulted in a noncash charge included in
operations over the one-year term of the Development Agreement. As a result of
the Development Agreement, our results for the three months ended March 31,
2000 include a noncash charge of $9.6 million compared to $1.6 million for the
same period in 1999. In March 2000, Rogers purchased 3,687,618 shares of our
common stock on a net exercise basis, resulting in no proceeds to us.

   Sales and Marketing. Sales and marketing expenses consist primarily of
salaries and commissions for sales, marketing and support personnel, and costs
related to trade shows, consulting and travel. Sales and marketing expenses
increased to $7.7 million for the three months ended March 31, 2000 from $2.8
million for the same period in 1999. The increase in sales and marketing
expenses was primarily due to increased payroll costs related to additional
sales and support personnel necessary to support the expansion of our customer
base. In addition, the increase in sales and marketing expenses was
attributable to the increased payroll and associated costs resulting from the
acquisitions of Imedia, Radwiz and Telegate, and increased commissions related
to higher sales. We expect sales and marketing expenses to continue to increase
as we expand our customer base.

   General and Administrative. General and administrative expenses primarily
consist of salary and benefits for administrative officers and support
personnel, travel expenses, legal, accounting and consulting fees. General and
administrative expenses increased to $4.2 million for the three months ended
March 31, 2000 from $1.2 million in the same period of 1999. The increase was
primarily due to costs associated with the increased infrastructure required to
support our expanded activities and increased personnel costs associated with
our acquisitions of Imedia, Radwiz and Telegate. We expect general and
administrative expenses will continue to increase in the near term as a result
of these factors.

   In-Process Research and Development. We incurred charges of $6.8 million for
the three months ended March 31, 2000 related to research and development
projects in process at

                                       38

<PAGE>

Telegate at the time of the acquisition. The projects identified as in-process
will require additional effort in order to establish technological feasibility.
These projects have identifiable technological risk factors that indicate that
even though successful completion is expected, it is not assured.

   In-process technology acquired consists primarily of major additions to
Telegate's core technology, which is related to Telegate's planned development
of new features. The majority of the intended functionality of these new
features is not supported by Telegate's current technology. Intended new
features include: connection on demand functionality to extend the product's
ISDN compatibility; the ability to use cordless technology for either voice or
data applications; and, a subscriber end unit that can be used in multi-
dwelling units. Notwithstanding our expectations, there remain significant
technical challenges that must be resolved in order to complete the in-process
technology.

   Goodwill Amortization. The Imedia acquisition was completed on September 16,
1999 and was accounted for as a purchase transaction. The total purchase price,
representing the fair value of our shares issued and transaction and direct
acquisition costs, was approximately $109.0 million. The purchase price was
allocated between the net tangible assets of Imedia on the date of acquisition
(approximately $645,000), in-process research and development (approximately
$11.0 million) and intangible assets acquired (approximately $97.3 million).
Intangible assets consist of developed technology (approximately $27.0
million), assembled workforce (approximately $2.5 million), trademark
(approximately $4.0 million) and goodwill (approximately $63.8 million). The
intangible assets will be amortized straight line over lives ranging from two
to six years. The amortization of developed technology will impact cost of
goods sold in future periods. The amortization of the other intangibles will
principally impact operating expenses in future periods.

   The Radwiz acquisition was completed on November 22, 1999 and was accounted
for as a purchase transaction. The total purchase price, representing the fair
value of our shares issued and transaction and direct acquisition costs, was
approximately $53.6 million. The purchase price was allocated between the net
tangible assets of Radwiz on the date of acquisition (approximately $3.1
million), in-process research and development (approximately $3.6 million) and
net intangible assets acquired (approximately $46.9 million). Intangible assets
consist of developed technology (approximately $29.9 million), assembled
workforce (approximately $2.8 million), trademark (approximately $1.1 million)
and goodwill (approximately $24.1 million). Goodwill has been increased and
deferred tax liabilities have been recorded in the amount of approximately
$11.0 million to reflect the net tax effect of the book/tax basis differences
in the acquired intangibles, excluding goodwill and in-process research and
development. The intangible assets will be amortized straight line over lives
ranging from two to six years. The amortization of developed technology will
impact cost of goods sold in future periods. The amortization of the other
intangibles will principally impact operating expenses in future periods.

   The Telegate acquisition was completed on January 2, 1999 and was accounted
for as a purchase transaction. The total purchase price, representing the fair
value of our shares issued and transaction and direct acquisition costs, was
approximately $140.0 million. The purchase price was allocated between the net
tangible assets of Telegate on the date of acquisition (approximately $6.9
million), in-process research and development (approximately $6.8 million) and
intangible assets acquired (approximately $126.3 million). Intangible assets
consist of developed technology (approximately $21.1 million), customer
relationship

                                       39

<PAGE>

(approximately $34.6 million) assembled workforce (approximately $4.2 million),
and goodwill (approximately $66.6 million). Goodwill has been increased and
deferred tax liabilities have been recorded in the amount of approximately
$167,000 to reflect the net tax effect of the book/tax basis differences in the
acquired intangibles, excluding goodwill and in-process research and
development. Deferred tax assets have been realized based on the projected
reversal of taxable temporary differences and have been netted against deferred
tax liabilities for purposes of allocating the purchase price. The intangible
assets will be amortized straight line over lives expected to range from two to
six years. The amortization of developed technology and customer relationship
will impact cost of goods sold in future periods. The amortization of the other
intangibles will principally impact operating expenses in future periods.

   We expect amortization of goodwill and other intangibles related to
acquisitions subsequent to March 31, 2000, to increase in future periods.

   Net Interest Income (Expense). Net interest income was $1.4 million for the
three months ended March 31, 2000 compared to $1.0 million for the same period
in 1999.

   Income Taxes. We have generated operating losses since our inception. Due to
our inability to recognize a benefit from these operating losses we have no
provision for income taxes in 1999 and 1998. We account for income taxes under
Statement of Financial Accounting Standards No. 109, Accounting for Income
Taxes ("FAS 109"). Under FAS No. 109, deferred tax assets and liabilities are
determined based on the difference between financial reporting and tax bases of
assets and liabilities, and are measured using the enacted tax rates and laws
that will be in effect when the differences are expected to reverse. FAS 109
provides for the recognition of deferred tax assets if realization of such
assets is more likely than not. Based upon the weight of available evidence,
which includes our historical operating performance and the reported cumulative
net losses for the prior three years, we have provided a full valuation against
our net deferred tax assets as of March 31, 2000 and 1999.

Results of Operations

 Years Ended December 31, 1999 and 1998

   Revenues. Our revenues increased to $97.0 million in 1999 from $31.7 million
in 1998. Revenues consist primarily of sales of cable modems and headend
equipment to new and existing customers. The increased revenues in 1999 were
primarily attributable to the addition of new customers in 1999 and continuing
deployments of our TeraComm system by existing customers, and, to a lesser
extent, the sales of products acquired as a result of our acquisition of
Imedia. The impact on revenues resulting from the acquisition of Radwiz was not
significant for the year ended December 31, 1999.

   We sell our products directly to broadband access service providers, system
resellers and distributors. Revenues related to product sales are generally
recognized when the products are shipped to the customer. A provision is made
for estimated product returns as product shipments are made. Our existing
agreements with our system resellers and distributors do not contain price
protection provisions and do not grant return rights beyond those provided by
the Company's standard warranty.

   Cost of Goods Sold. Cost of goods sold consists of direct product costs as
well as the cost of our manufacturing operations group. The cost of the
manufacturing operations group includes assembly, test and quality assurance
for products, warranty costs and associated costs of personnel and equipment.
In 1999, we incurred $72.0 million in cost of goods sold,

                                       40

<PAGE>

which included the cost of our manufacturing operations group and approximately
$1.9 million of amortization of intangibles assets resulting from the
acquisitions of Imedia and Radwiz. In 1998, we incurred $34.5 million in cost
of goods sold, which included the cost our manufacturing operations group and a
charge of $1.3 million relating to the write-off of obsolete inventory and the
transition of manufacturing operations.

   Gross Profit (Loss). Our gross profit was $25.0 million or 26% of revenue in
1999. This compares to a gross loss of $2.8 million in 1998. The improvement in
our gross profit in 1999 compared to 1998 was due largely to the introduction
of a cable modem using a single-board design in the fourth quarter of 1998.
This modem has a higher gross margin than our original dual board design.
Subsequent to the introduction of the single-board design, we have successfully
introduced cost reduced versions that have allowed us to continue to improve
gross margin. We intend to continue to engage in and implement cost reduction
efforts, including the further integration of ASIC components, other design
changes and manufacturing efficiencies. We anticipate that continued decreases
in the average sales price of our products will partially, if not completely,
offset the benefits obtained from further cost reductions.

   Our gross profit also is influenced by the sales mix of TeraLink Master
Controllers, TeraLink Gateways and TeraPro cable modems and the maturity of
TeraComm system deployments in any quarter.  TeraPro cable modems have lower
margins than the TeraLink Master Controllers and TeraLink Gateway headend
products. New deployments of TeraComm systems typically include a higher mix of
headend equipment and involve smaller quantities of product sold. Products sold
in connection with new deployments thus generally are sold at higher margins
than products associated with more mature deployments of the TeraComm system,
which generally involve larger quantities of products, primarily cable modems.
We expect that the introduction of new customers and the relationship of
revenues generated from the sale of TeraPro cable modems to our overall revenue
will result in fluctuations in our gross profit in future periods.

   We introduced a CableLabs certified DOSCIS 1.0 cable modem to the market in
third quarter of 1999. We also are currently developing a DOCSIS system that
will include a headend controller. We believe that the widespread adoption of
industry standards will result in further price pressure. We anticipate that
the relationship of revenues generated from the sale of our proprietary
TeraComm system versus a DOCSIS compliant system will result in fluctuations in
our gross profit in future periods. The impact on our gross margin in 1999
resulting from the sale of our DOCSIS 1.0 cable modem was not significant.

   Our gross margin also is influenced by the level of sales of products of
acquired companies, which produce varying gross margin results. We expect that
our gross margin will fluctuate in future periods as a result of these and
other acquisitions.

   Research and Development. Research and development expenses consist
primarily of personnel costs, as well as design expenditures, equipment and
supplies required to develop and to enhance our products. Research and
development expenses increased to $17.6 million in 1999 from $10.7 million in
1998, primarily as a result of increased personnel costs. The increased
personnel costs are a result of expansion in our own employee base as we
continue to focus our efforts on developing new products and enhancing our
current products. The increased personnel costs are also the result of the
acquisitions of Imedia and Radwiz. We intend to continue to increase investment
in research and development as a result of these and other activities.

                                       41

<PAGE>

   Cost of Product Development Assistance Agreement. In March 1999, we entered
into a one-year Product Development Assistance Agreement with Rogers. Under the
terms of the Development Agreement, Rogers is obligated to assist us with the
characterization and testing of our subscriber-end and head-end voice-over-
cable equipment. In addition, Rogers is obligated to provide us with technology
to assist us with our efforts to develop high quality, field proven technology
solutions that are DOCSIS-compliant and PacketCable-compliant. The Development
Agreement has a term of one year. In consideration of Rogers entering into the
Development Agreement, we issued Rogers two fully vested and non-forfeitable
warrants, each to purchase 2.0 million shares of common stock on a cashless
basis. One warrant has an exercise price of $0.50 per share and one warrant had
an exercise price of $18.50 per share. The warrants may be exercised at any
time in full or in part through March 31, 2000. The fair value of the two
warrants was approximately $45.0 million and resulted in a noncash charge
included in operations over the one-year term of the Development Agreement. As
a result of the Development Agreement, our results for 1999 include a noncash
charge of $35.1 million. In March 2000, Rogers purchased 3,687,618 shares of
our common stock on a net exercise basis, resulting in no proceeds to us.

   In-Process Research and Development. We incurred charges of $11.0 million
for the year ended December 31, 1999 related to research and development
projects in process at Imedia at the time of the acquisition and $3.6 million
related to research and development project in process at Radwiz at the time of
the acquisition. The projects identified as in-process will require additional
effort in order to establish technological feasibility. These projects have
identifiable technological risk factors that indicate that even though
successful completion is expected, it is not assured.

   In-process technology resulting from the acquisition of Imedia consists
primarily of major additions to Imedia's core technology, which is related to
Imedia's planned development of new features. The majority of the intended
functionality of these new features is not supported by Imedia's current
technology. Intended new features include offering high quality video service
over the Internet and multiplexing data with video. We expect that in-process
technology will be successfully developed and that initial benefits from these
projects will begin in 2001. However, there remain significant technical
challenges that must be resolved in order to complete the in-process
technology.

   In-process technology resulting from the acquisition of Radwiz consists
primarily of major additions to Radwiz's core technology, which is related to
Radwiz's planned development of new features. The majority of the intended
functionality of these new features is not supported by Radwiz's current
technology. Intended new features include offering: end-to-end carrier quality
of service; allowing access via an ATM network; and, providing ISDN line
functionality. We expect that in-process technology will be successfully
developed and that initial benefits from these projects will begin in mid-year
2000. However, significant technical challenges must be resolved in order to
complete the in-process technology.

   Sales and Marketing. Sales and marketing expenses consist primarily of
salaries and commissions for sales, marketing and support personnel, and costs
related to trade shows, consulting and travel. Sales and marketing expenses
increased to $15.7 million in 1999 from $6.9 million in 1998, primarily due to
increased payroll costs related to additional sales and support personnel
necessary to support the expansion of our customer base and resulting from the
acquisitions of Imedia and Radwiz, and increased commissions related to higher
sales. We expect sales and marketing expenses to continue to increase as we
expand our customer base.

                                       42

<PAGE>

   General and Administrative. General and administrative expenses primarily
consist of salary and benefits for administrative officers and support
personnel, travel expenses, legal, accounting and consulting fees. General and
administrative expenses increased to $7.5 million in 1999 from $3.2 million in
1998. The increase was primarily a result of costs associated with the
increased infrastructure required to support our expanded activities and due to
increased personnel costs associated with our acquisitions of Imedia and
Radwiz. We expect that general and administrative expenses will continue to
increase in the near term as a result of these factors.

   Amortization of Goodwill and Other Intangible Assets. The Imedia acquisition
was completed on September 16, 1999 and was accounted for as a purchase
transaction. The total purchase price, representing the fair value of our
shares issued and transaction and direct acquisition costs, was approximately
$109.0 million. The purchase price was allocated between the net tangible
assets of Imedia on the date of acquisition (approximately $645,000), in-
process research and development (approximately $11.0 million) and intangible
assets acquired (approximately $97.3 million). Intangible assets consist of
developed technology (approximately $27.0 million), assembled workforce
(approximately $2.5 million), trademark (approximately $4.0 million) and
goodwill (approximately $63.8 million). The intangible assets will be amortized
straight line over lives ranging from two to six years. The amortization of
developed technology will impact cost of goods sold in future periods. The
amortization of the other intangibles will principally impact operating
expenses in future periods.

   The Radwiz acquisition was completed on November 22, 1999 and was accounted
for as a purchase transaction. The total purchase price, representing the fair
value of our shares issued and transaction and direct acquisition costs, was
approximately $53.6 million. The purchase price was allocated (based on an
independent appraisal) between the net tangible assets of Imedia on the date of
acquisition (approximately $3.1 million), in-process research and development
(approximately $3.6 million) and net intangible assets acquired (approximately
$46.9 million). Intangible assets consist of developed technology
(approximately $29.9 million), assembled workforce (approximately $2.8
million), trademark (approximately $1.1 million) and goodwill (approximately
$24.1 million). Goodwill has been increased and deferred tax liabilities have
been recorded in the amount of approximately $11.0 million to reflect the net
tax effect of the book/tax basis differences in the acquired intangibles,
excluding goodwill and in-process research and development. Deferred tax assets
have been realized based on the projected reversal of taxable temporary
differences and have been netted against deferred tax liabilities for purposes
of allocating the purchase price. The intangible assets will be amortized
straight line over lives ranging from two to six years. The amortization of
developed technology will impact cost of goods sold in future periods. The
amortization of the other intangibles will principally impact operating
expenses in future periods.

   We expect amortization of goodwill and other intangibles related to the
acquisitions of Imedia and Radwiz, as well as subsequent acquisitions, to
increase in future periods.

   Net Interest Income. Net interest income increased to $5.0 million in 1999
from $449,000 in 1998, primarily as a result of higher average cash balances
subsequent to our public offering in January 1999.

   Series F Convertible Preferred Stock Dividend. We recorded a dividend of
$23.9 million in 1998. The dividend represented the fair value of a warrant to
purchase 6,000,000 shares of our common stock (the "Shaw Warrant"), and was
recorded under the provision of EITF No. 96-13, "Accounting for Derivative
Financial Instruments Indexed to, and Potentially Settled in, a Company's Own
Stock." The warrant was issued in connection with the sale of $5.0 million of
convertible preferred stock to Shaw. Shaw purchased 3,000,000 shares of common
stock in

                                       43

<PAGE>

March 1999 and 3,000,000 shares of our common stock in November 1999 under the
terms of the Shaw Warrant resulting in net proceeds to us of $19.5 million.

   Income Taxes. We have generated operating losses since our inception. Due to
our inability to recognize a benefit from these operating losses we have no
provision for income taxes in 1999 and 1998, We account for income taxes under
FAS 109. Under FAS 109, deferred tax assets and liabilities are determined
based on the difference between financial reporting and tax bases of assets and
liabilities, and are measured using the enacted tax rates and laws that will be
in effect when the differences are expected to reverse. FAS 109 provides for
the recognition of deferred tax assets if realization of such assets is more
likely than not. Based upon the weight of available evidence, which includes
the Company's historical operating performance and the reported cumulative net
losses for the prior three years, the Company has provided a full valuation
against its net deferred tax assets as of December 31, 1999 and 1998. We intend
to evaluate our ability to realize the benefit of the deferred tax assets on a
quarterly basis. See Note 12 to the Consolidated Financial Statements for the
year ended December 31,1999.

   Stock-Based Compensation. As of December 31, 1999, we had recorded deferred
compensation of $2.6 million related to certain stock options granted in 1998
and 1997. We amortized approximately $631,000 of the deferred compensation in
1999, $421,000 in 1998 and $12,000 in 1997. The remainder will be amortized
over the related vesting period of the stock options. See Note 11 of Notes to
Consolidated Financial Statements for the year ended December 31, 1999.

 Years Ended December 31, 1998 and 1997

   Revenues. Our revenues increased to $31.7 million in 1998 from $2.1 million
in 1997. Revenues consist primarily of sales of cable modems and headend
equipment to new and existing customers. We did not commence selling our
products until June 1997 and did not commence selling our products in volume
until the first quarter of 1998.

   Cost of Goods Sold. In 1998, we incurred $34.5 million in cost of goods
sold, which included the costs our manufacturing operations group and a charge
of $1.3 million relating to the write-off of obsolete inventory and the
transition of manufacturing operations. In 1997, we incurred $6.5 million in
cost of goods sold, which included the cost of the manufacturing group for the
entire period as we readied our products for commercialization.

   Gross Loss. We incurred a gross loss of $2.8 million in 1998. In 1997, we
incurred a gross loss of $4.3 million due to costs associated with our
manufacturing operations group. The decrease in gross loss in 1998 compared to
1997 was primarily due to a reduced negative margin on cable modems, which was
partially offset by the $1.3 million write-off relating to obsolete inventory
and the transition of manufacturing operations to Solectron. We completed the
transition to a lower cost, single-board modem product in the fourth quarter of
1998. Primarily as a result of this lower cost product, we achieved a positive
gross margin of $425,000 in the fourth quarter of 1998.

   Research and Development. Research and development expenses decreased to
$10.7 million in 1998 from $11.3 million in 1997 as a result of the timing of
our development projects.

   Sales and Marketing. Sales and marketing expenses increased to $6.9 million
in 1998 from $4.5 million in 1997, primarily due to increased payroll costs
related to additional sales

                                       44

<PAGE>

and support personnel for commercial trials and deployment of our products and
increased commissions related to higher sales in 1998.

   General and Administrative. General and administrative expenses increased to
$3.2 million in 1998 from $2.5 million in 1997, primarily a result of the
additional reporting requirements imposed on us as a public company and
increased infrastructure to support our expanded activities.

   Net Interest Income. Net interest income increased to $449,000 in 1998 from
$128,000 in 1997, primarily as a result of higher average cash balances
subsequent to our initial public offering in August 1998.

   Series F Convertible Preferred Stock Dividend. We recorded a dividend of
$23.9 million in 1998. The dividend represents the fair value the Shaw Warrant.
The Shaw Warrant was issued in connection with the sale of $5.0 million of
convertible preferred stock to Shaw (the "Shaw Financing"). Our accounting
conclusion was that the sale of preferred stock was, in substance, a financing
transaction and not the issuance of equity instruments in exchange for goods or
services.

   Income Taxes. We have not generated operating losses since inception. Due to
our inability to recognize a benefit from these operating losses we have no
provision for income taxes in 1998 and 1997. We account for income taxes under
Statement of Financial Accounting Standards No. 109. Realization of deferred
tax assets is dependent on future earnings, if any, the timing and amount of
which are uncertain. Accordingly, valuation allowances in amounts equal to the
net deferred tax assets as of December 31, 1998 and 1997 have been established
to reflect these uncertainties.

   Stock-Based Compensation. As of December 31, 1998, we had recorded deferred
compensation of $2.6 million related to certain stock options granted in 1998
and 1997. We amortized approximately $421,000 of the deferred compensation in
1998 and $12,000 in 1997. The remainder will be amortized over the related
vesting period of the stock options. See Note 10 of Notes to Consolidated
Financial Statements for the year ended December 31, 1999.

 Liquidity and Capital Resources

   At March 31, 2000, we had approximately $54.7 million in cash and cash
equivalents, $63.2 million in short-term investments and a $2.5 million
revolving line of credit. There were no outstanding borrowings under the line
of credit and approximately $2.5 million was available.

   Cash provided by operating activities in the first three months of 2000 was
$5.8 million. This compares to cash used in operating activities of $3.7
million in the first three months of 1999. Cash provided by investing
activities was $13.2 million in the first three months of 2000 compared to cash
used in investing activities of $55.5 million in the first three months of
1999. Investment activities consisted primarily of the purchase and sale of
short-term investments in 2000 and 1999. Cash provided by financing activities
was $3.2 million for the three months ended March 31, 2000 compared to $85.6
million for the three months ended March 31, 1999. Financing activities
consisted primarily of proceeds from the exercise of options and the issuance
of common stock in 2000 and from the sale and issuance of common stock in a
public offering in 1999.

   As of March 31, 2000, we had approximately $125.7 million of unconditional
purchase obligations. We anticipate that we will pay approximately $69.1
million of these obligations by June 30, 2000. We intend to make these payments
out of available working capital.

                                       45

<PAGE>

   We believe that our current cash and cash equivalent balances, together with
cash anticipated to be generated from operations and anticipated financing
arrangements, including this offering, will satisfy our projected working
capital and capital expenditure requirements, at a minimum, for the next twelve
months.

 Impact of Year 2000

   In prior years, we discussed the nature and progress of our plans to become
year 2000 ready. In late 1999, we completed the remediation and testing of our
systems. As a result of those planning and implementation efforts, we
experienced no significant disruptions in mission critical information
technology and non-information technology systems and we believe those systems
successfully responded to the year 2000 date change. We incurred no significant
costs in 1999 in connection with the remediation of our systems. We are not
aware of any material problems resulting from year 2000 issues, either with our
products, our internal systems, or the products and services of third parties.
We will continue to monitor our mission critical computer applications and
those of our suppliers and vendors throughout the rest of the year to ensure
that any latent year 2000 matters that may arise are addressed promptly.

 Recent Financial Accounting Pronouncement

   In September 1998, the Financial Accounting Standards Board issued Statement
of Financial Accounting Standards No. 133, "Accounting for Derivative
Instruments and Hedging Activities" ("FAS 133"). FAS 133 provides a
comprehensive and consistent standard for the recognition and measurement of
derivatives and hedging activities. FAS 133 was effective for fiscal years
beginning after September 15, 1999. In July 1999, the Financial Accounting
Standards Board issued Statement of Financial Accounting Standards No. 137,
"Accounting for Derivative Instruments and Hedging Activities--Deferral of the
Effective Date of FASB Statement No. 133" (FAS 137). FAS 137 defers for one
year the effective date of FAS 133 which will now apply to all fiscal quarters
of all fiscal years beginning after June 15, 2000. We believe that the adoption
of FAS 133 will not have a significant impact on our operating results or cash
flows.

   In December 1999, the Securities and Exchange Commission issued Staff
Accounting Bulletin No. 101, "Revenue Recognition in Financial Statements"
("SAB 101"). SAB 101 provides guidance on the recognition, presentation and
disclosure of revenue in the financial statements of public companies. The
accounting profession is currently exploring practical means of implementing
the guidelines established by SAB 101. To the extent that SAB 101 is applicable
to us, we will be required to change our revenue recognition policy. In June
2000 the Securities and Exchange Commission issued SAB 101 B which delays the
effective date of SAB 101 until no later than the fourth fiscal quarter
beginning after December 15, 1999. We will be required to adopt SAB 101 by the
fourth quarter of 2000. Financial information for prior quarters would be
restated with the cumulative effect adjustment reflected in the first quarter
of 2000 in accordance with FASB Statement No. 3, Reporting Accounting Changes
in Interim Financial Statements. In subsequent periods, we would disclose the
amount of revenue, if material, recognized in those periods that were included
in the cumulative effect adjustment. We currently cannot determine what effect,
if any, that SAB 101 will have on our financial statements. We believe that SAB
101, to the extent it is applicable to us, will not effect the underlying
strength or weakness of our business operations as measured by the dollar value
of our shipments and cash flows.

                                       46

<PAGE>

  The section entitled Certain Business Risk should be deleted and replaced with
the following section:


                            CERTAIN BUSINESS RISK

   Before making an investment decision, you should carefully consider the risks
described below. The risks and uncertainties described below are not the only
ones facing our company. Additional risks and uncertainties not presently known
to us or that we currently deem immaterial also may impair our business
operations. If any of the following risks actually occur, our business could be
harmed. In such case, the trading price of our common stock could decline, and
you may lose all or part of your investment.

                         Risks Related to Our Business

We have a limited operating history and a history of losses.

   We have a limited operating history, and it is difficult to predict our
future operating results. We began shipping products commercially in June 1997,
and we only have been shipping products in volume since the first quarter of
1998. As of March 31, 2000, we had an accumulated deficit of $176.8 million. We
believe that we will continue to experience net losses for the foreseeable
future. Most of our expenses are fixed in advance, and we generally are unable
to reduce our expenses significantly in the short term to compensate for any
unexpected delay or decrease in anticipated revenues. We expect to continue to
increase expenses for the foreseeable future to support increased sales and
marketing and technical support costs. Any significant delay in our anticipated
revenues or commercialization of new products would harm our business. The
revenue and profit potential of our business and our industry are unproven. We
had negative gross margins from our inception until the fourth quarter of 1998,
and any future revenue growth may not result in positive gross margins or
operating profits in future periods.

Our operating results may fluctuate.

   Our quarterly revenues are likely to fluctuate significantly in the future
due to a number of factors, many of which are outside our control. Factors that
could affect our revenues include the following:

  .  variations in the timing of orders and shipments of our products;

  .  variations in the size of the orders by our customers;

  .  new product introductions by competitors;

  .  delays in our introduction of new products;

  .  delays in our receipt of orders forecasted by our customers;

  .  delays by our customers in the completion of upgrades to their cable
     infrastructure;

  .  variations in capital spending budgets of broadband access service
     providers;

  .  adoption of industry standards and the inclusion in or compatibility of
     our technology with any such standards; and

  .  delays in obtaining regulatory approval for commercial deployment of
     cable modem systems.

   Our expenses generally will vary from quarter to quarter depending on the
level of actual and anticipated business activities. Research and development
expenses will vary as we begin development of new products and as our
development programs move to wafer fabrication and prototype development, which
results in higher engineering expenses.

                                       47
<PAGE>

   A variety of factors affect our gross margin, including the following:

  .  the sales mix of our products;

  .  the volume of products manufactured;

  .  the type of distribution channel through which we sell our products;

  .  the average selling prices or "ASPs" of our products; and

  .  the effectiveness of our cost reduction measures.

   We anticipate that unit ASPs of our products will decline in the future.
This could cause a decrease in the gross margins for these products. In
addition, the maturity of TeraComm system deployments affects our gross margin.
New deployments of the TeraComm system involve the sale of headend equipment
(which has higher margins) and generally involve smaller quantities of product.
New deployments typically are sold at higher margins than the larger volume
sales of product associated with more mature deployments of the TeraComm
system. The sales mix of TeraLink 1000 Master Controllers, TeraLink Gateways
and TeraPro cable modems also affects our gross margin. The TeraPro cable
modems have significantly lower margins than the TeraLink 1000 Master
Controller and TeraLink Gateway headend products. We expect to achieve
significantly lower margins on the TeraPro cable modems for the foreseeable
future. Further, we expect that sales of TeraPro cable modems will continue to
constitute a significant portion of our revenues for the foreseeable future.

   We also anticipate that our operating results will be impacted by sales,
gross profit and operating expenses of acquired companies. The impact of these
factors on our operating results will vary as we acquire additional companies.

We are dependent on a small number of customers.

   Four customers accounted for approximately 64% of our revenues for the
quarter ended March 31, 2000 and three customers accounted for approximately
63% of our revenues for the quarter ended March 31, 1999. We believe that a
substantial majority of our revenues will continue to be derived from sales to
a relatively small number of customers for the foreseeable future. In addition,
we believe that sales to these customers will be focused on a limited number of
projects.

   The cable industry is undergoing significant consolidation in North America
and internationally, and a limited number of cable operators controls an
increasing number of cable systems. Currently, ten cable operators in the
United States own and operate facilities passing approximately 86% of total
homes passed. In addition, the North American DSL market is concentrated with
the major ILECs, constituting a significant percentage of the market. As a
result, our sales will be largely dependent upon product acceptance by the
leading broadband service providers. Currently, the timing and size of each
customer's order is critical to our operating results. Our major customers are
likely to have significant negotiating leverage and may attempt to change the
terms, including pricing, upon which we do business with them. These customers
also may require longer payment terms than we anticipate, which could require
us to raise additional capital to meet our working capital requirements.

Acquisitions could result in dilution, operating difficulties and other adverse
consequences.

   We have acquired seven businesses since September 1999: Imedia Corporation
in September 1999; Radwiz in November 1999; Telegate in January 2000; ANE in
April 2000; ComBox in April 2000; some assets of Internet Telecom in April
2000; and Ultracom in April

                                       48

<PAGE>

2000. If appropriate opportunities present themselves, we intend to acquire
additional businesses, technologies, services or products that we believe are
strategic. The process of integrating any acquired business into our business
and operations is risky and may create unforeseen operating difficulties and
expenditures. The areas in which we may face difficulties include:

  .  diversion of management time (both ours and that of the acquired
     companies) during the period of negotiation through closing and after
     closing from the ongoing development of our businesses, issues of
     integration and future products;

  .  decline in employee morale and retention issues resulting from changes
     in compensation, reporting relationships, future prospects or the
     direction of the business;

   We typically sell products to our customers pursuant to purchase orders, and
we therefore generally do not have long-term agreements in place with our
customers.

  .  the need to integrate each company's accounting, management information,
     human resource and other administrative systems to permit effective
     management, and the lack of control if this integration is delayed or
     not implemented; and

  .  the need to implement controls, procedures and policies appropriate for
     a larger public group of companies that prior to acquisition had been
     smaller, private companies.

   We have very limited experience in managing this integration process.
Moreover, the anticipated benefits of any or all of these completed or pending
acquisitions may not be realized. Future acquisitions could result in
potentially dilutive issuances of equity securities, the incurrence of
additional debt, contingent liabilities or amortization expenses related to
goodwill and other intangible assets, any of which could harm our business.
Future acquisitions also could require us to obtain additional equity or debt
financing, which may not be available on favorable terms or at all. Even if
available, this financing may be dilutive.

The sales cycle for our products is lengthy.

   The sales cycle associated with our products typically is lengthy, often
lasting six months to a year. Our customers typically conduct significant
technical evaluations of competing technologies prior to the commitment of
capital and other resources. In addition, purchasing decisions may be delayed
because of our customers' internal budget approval procedures. Sales also
generally are subject to customer trials, which typically last three months.
Because of the lengthy sales cycle and the large size of customers' orders, if
orders forecasted for a specific customer for a particular quarter do not occur
in that quarter, our operating results for that quarter could suffer.

There are many risks associated with our participation in the establishment of
advanced physical layer specifications to be added to DOCSIS.

   In November 1998, CableLabs selected us to co-author DOCSIS 1.2 (Data Over
Cable Service Interface Specifications), an enhanced version of the DOCSIS
cable modem specification based in part on our S-CDMA technology. In September
1999, CableLabs indicated that it intended to proceed with the advanced
physical layer ("PHY") work on two parallel tracks: one for the development of
a prototype based on our S-CDMA technology and

                                       49

<PAGE>

one for the inclusion of Advanced TDMA technology (Time Division Multiple
Access), as proposed by other companies. In February 2000, CableLabs further
clarified the status of the advanced PHY project regarding a separate release
that will include TDMA technologies. In addition, CableLabs reiterated that it
is continuing to work with us on the development of a DOCSIS specification that
could include our S-CDMA technology. To that end, CableLabs has requested that
we submit a prototype of a DOCSIS system that incorporates an S-CDMA advanced
PHY capability for testing. CableLabs has stated that if the testing of this
prototype reveals that the S-CDMA advanced PHY works as claimed (including
proper backwards compatibility and coexistence with the other aspects of
DOCSIS), and if the costs for adding S-CDMA to DOCSIS products are in line with
estimates, then it is likely, but not certain, that S-CDMA advanced PHY
capabilities will be included in a future version of the DOCSIS specification.
The prototype we submit to CableLabs may fail to demonstrate the level of
performance that CableLabs seeks; even if it does meet performance expectations
there can be no guarantee that CableLabs will incorporate the technology into a
future version of DOCSIS specifications. In addition, if CableLabs does proceed
to include S-CDMA in a future DOCSIS specification, there can be no guarantee
that the DOCSIS S-CDMA specification will be the same as the specification we
incorporated in the prototype submitted for tests, which may require us to
further develop our prototype.

   Our future revenues and operating results are likely to suffer if S-CDMA is
not included in a future release of DOCSIS. We also may incur substantial
additional research and development expenditures to adapt our specifications to
the version adopted by CableLabs. CableLabs has not established a schedule for
adding the S-CDMA capabilities to the DOCSIS specifications. Delays in the
establishment of a final specification for S-CDMA in DOCSIS could harm our
plans to sell DOCSIS compatible modems and headend equipment. In particular, if
the final DOCSIS S-CDMA specification is not approved prior to the time when we
are ready to ship DOCSIS products with S-CDMA features included, then we may be
required to delay the introduction of those products until the DOCSIS S-CDMA
specification is released or to introduce the S-CDMA features as proprietary
enhancements to a standard DOCSIS product. Either one of these events could
harm revenues and operating results.

   We have already given CableLabs assurances that we will contribute some
aspects of our proprietary S-CDMA technology to a royalty-free intellectual
property pool, if S-CDMA is included in a future version of DOCSIS
specifications. This royalty-free pool has been established by CableLabs to
facilitate the participation of as many vendors as possible in providing
equipment that is compatible with the DOCSIS specifications. As a result, any
of our competitors who join the DOCSIS intellectual property pool would have
access to some aspects of our technology and would not be required to pay us
any royalties or other compensation. If a competitor is able to duplicate the
functionality and capabilities of our technology, we could lose some or all of
the time-to-market advantage we might otherwise have, which could harm our
future revenues and operating results.

   We believe the addition of advanced upstream PHY capabilities to DOCSIS will
increase the overall market for DOCSIS-compatible products, and as such will
result in increased competition in the cable modem market. This competition
could come from existing competitors or from new competitors who enter the
market as a result of the enhancements to the specifications. This increased
competition is likely to result in lower ASPs of cable modem systems and could
harm revenues and gross margins. Because our competitors will be able to
incorporate some aspects of our technology into their products, our current
customers may choose alternate suppliers or choose to purchase DOCSIS-compliant
cable modems with advanced PHY capabilities from multiple suppliers. We may be
unable to produce DOCSIS compliant cable modems with advanced PHY capabilities
more quickly or at lower cost than

                                       50

<PAGE>

our competitors. The inclusion of our S-CDMA technology in future DOCSIS
specification could result in increased competition for the services of our
existing employees who have experience with S-CDMA. The loss of these employees
to one or more competitors could harm our business.

   DOCSIS standards have not yet been accepted in Europe and Asia. An alternate
standard for cable modem systems, called the EuroModem standard, or DAVIC/DVB,
has been formalized, and some European cable system operators have embraced it.
We intend to develop and sell products that comply with the EuroModem standard
and to pursue having portions of our S-CDMA technology included in a future
version of the EuroModem standard. We may be unsuccessful in these efforts.

We need to develop new products in order to remain competitive.

   Our future success will depend in part on our ability to develop, introduce
and market new products in a timely manner. We also must respond to competitive
pressures, evolving industry standards and technological advances. Our current
S-CDMA products are not DOCSIS-compliant. We are currently developing a
prototype of a DOCSIS system that incorporates an S-CDMA advanced PHY
capability for testing and eventual inclusion in the DOCSIS standard. There is
no guarantee that we will be successful in developing the prototype or that the
prototype, if successfully developed, will be included in a future release of
the DOCSIS standard. We anticipate that during 2000, existing or potential
customers may delay purchases of our TeraComm system in order to purchase
systems that comply with the DOCSIS standard. In addition, potential new
customers could decide to purchase DOCSIS-compliant products from one or more
of our competitors rather than from us. As a result, our product sales in the
second half of 2000 may be lower than we anticipate. In order to promote sales
of our currents products we may be required to reduce our prices for sales to
existing customers. This would harm our operating results and gross margin.

   As a result of the inclusion of TDMA technology in the new DOCSIS version
announced by CableLabs in February 2000, we will have to incorporate advanced
TDMA technology into our DOCSIS-compliant products. If we are unable to do this
effectively, or in a timely manner, we will lose some or all of the time-to-
market advantage we might otherwise have had.

   Our future success will also depend on our ability to develop and market
products for broadband applications over DSL and wireless networks. The markets
for these broadband applications are also subject to evolving standards, such
as NEBS compliance in the North American DSL market, and technological advances
in these arenas. There is no guarantee that we will be successful in developing
products that are compliant with these standards or that we will be successful
in keeping pace with future technological advances in this arena.

Average selling prices of broadband access equipment typically decrease.

   The broadband access systems market has been characterized by erosion of
average selling prices. We expect this to continue. This erosion is due to a
number of factors, including competition, rapid technological change and price
performance enhancements. The ASPs for our products may be lower than expected
as a result of competitive pricing pressures, our promotional programs and
customers who negotiate price reductions in exchange for longer term purchase
commitments. We anticipate that ASPs and gross margins for our products will
decrease over product life cycles. In addition, we believe that the widespread
adoption of industry standards is likely to further erode ASPs, particularly
for cable modems and other similar consumer premise equipment. It is likely
that the widespread adoption of industry standards will result in increased
retail distribution of cable modems and other similar

                                       51

<PAGE>

consumer premise equipment, which could put further price pressure on our
products. Decreasing ASPs could result in decreased revenue even if the number
of units sold increases. As a result, we may experience substantial period-to-
period fluctuations in future operating results due to ASP erosion. Therefore,
we must continue to develop and introduce on a timely basis next-generation
products with enhanced functionalities that can be sold at higher gross
margins. Our failure to do this could cause our revenues and gross margin to
decline.

We must achieve cost reductions.

   Certain of our competitors currently offer products at prices lower than
ours. Market acceptance of our products will depend in part on reductions in
the unit cost of our products. We expect that as headend equipment becomes more
widely deployed, the price of cable modems and other similar consumer premise
equipment will decline. In particular, we believe that the widespread adoption
of industry standards such as DOCSIS will cause increased price competition for
consumer premise equipment. However, we may be unable to reduce the cost of our
products sufficiently to enable us to compete with other suppliers. Our cost
reduction efforts may not allow us to keep pace with competitive pricing
pressures or lead to gross margin improvement.

   Some of our competitors are larger and manufacture products in significantly
greater quantities than we intend to for the foreseeable future. Consequently,
these competitors have more leverage in obtaining favorable pricing from
suppliers and manufacturers. In order to remain competitive, we must
significantly reduce the cost of manufacturing our cable modems through design
and engineering changes. We may not be successful in redesigning our products.
Even if we are successful, our redesign may be delayed or may contain
significant errors and product defects. In addition, any redesign may not
result in sufficient cost reductions to allow us to significantly reduce the
list price of our products or improve our gross margin. Reductions in our
manufacturing costs will require us to use more highly integrated components in
future products and may require us to enter into high volume or long-term
purchase or manufacturing agreements. Volume purchase or manufacturing
agreements may not be available on acceptable terms. We could incur expenses
without related revenues if we enter into a high volume or long-term purchase
or manufacturing agreement and then decide that we cannot use the products or
services offered by such agreement.

We must keep pace with rapid technological change to remain competitive.

   The markets for our products are characterized by rapid technological
change, evolving industry standards, changes in end-user requirements and
frequent new product introductions and enhancements. Our future success will
depend upon our ability to enhance our existing products and to develop and
introduce new products that achieve market acceptance. Providers of broadband
access services may adopt alternative technologies or they may deploy
alternative services that are incompatible with our products.

   The demand for broadband access services has resulted in the development of
several competing modulation technologies. For example, some of our cable
products utilize a modulation technology known as S-CDMA, while several of our
competitors utilize modulation technologies known as TDMA and Frequency
Division Multiple Access or "FDMA." Our headend equipment and cable modem
products currently are not interoperable with the headend equipment and modems
of other suppliers of broadband access products. As a result, potential
customers who wish to purchase broadband access products from multiple

                                       52

<PAGE>

suppliers may be reluctant to purchase our products. Although our technology
may be incorporated into a future version of a DOCSIS specification or another
industry standard, we cannot be certain that major cable operators will adopt
these standards. Major cable operators may not adopt products or technologies
based on our current proprietary S-CDMA technology or on any future industry
standard S-CDMA technology. Further, major cable operators may adopt products
or standards technologies based on competing modulation technologies. If
competitors using other modulation technologies can incorporate functionality
and capabilities currently found in S-CDMA, the value of our S-CDMA technology
would be diminished.

Broadband access services have not achieved widespread market acceptance, and
many competing technologies exist.

   Our success will depend upon the widespread commercial acceptance of
broadband access services by service providers and end users of broadband
access services. The market for these services is not fully developed. We
cannot accurately predict the future growth rate or the ultimate size of the
market for broadband access services. Potential users of our products may have
concerns regarding the security, reliability, cost, ease of installation and
use and capability of broadband access services in general.

   The market for our products may be impacted by the development of other
technologies that enable the provisioning of broadband access services and the
deployment of services over other media. Widespread acceptance of other
technologies or deployment of services over media not supported by our products
could materially limit acceptance of our broadband access systems. Broadband
access services based on our products and technology may fail to gain
widespread commercial acceptance by providers of broadband access services and
end users. In addition, we only recently began to offer products based on
alternate technologies such as DSL. We may not be successful in marketing and
selling these products.

We need to develop additional distribution channels.

   We presently market our TeraComm system to cable operators and systems
integrators. We believe that much of the North American cable modem market may
shift to a retail distribution model. Accordingly, we may need to redirect our
future marketing efforts to sell our cable modems directly to retail
distributors and end users. This shift would require us to establish new
distribution channels for our products. We may be unable to establish these
additional distribution channels. If we do establish them, we may be unable to
hire the additional personnel necessary to foster and enhance such distribution
channels. In addition, if the cable modem market shifts to a retail
distribution model, we may not successfully establish a retail distribution
presence. To the extent that large consumer electronics companies enter the
cable modem market, their well-established retail distribution capabilities
would provide them with a significant competitive advantage.

We may be unable to market effectively to broadband access service providers.

   Our growth and future success will be substantially dependent upon our
ability to convince providers of broadband access services to adopt our
technologies, purchase our products and effectively market our products to end
users. Our potential customers are likely to prefer purchasing products from
established manufacturing companies that can demonstrate the capability to
supply large volumes of products on short notice. In addition, many of our
potential customers may be reluctant to adopt technologies that have not gained
acceptance among other providers of similar services. This reluctance could
result in lengthy product testing and acceptance cycles for our products.
Consequently, the impediments to our initial sales may be even greater than
those to later sales.

                                       53

<PAGE>

   No established distribution network in the cable modem industry exists that
would provide us with easy access to smaller or geographically diverse cable
operators. Therefore, our initial sales to larger, more established cable
operators are critical to our business. Although we intend to establish
strategic relationships with leading distributors worldwide, we may not succeed
in establishing these relationships. Even if we do establish these
relationships, the distributors may not succeed in marketing our products to
cable operators. Some of our competitors have already established relationships
with certain cable operators. These established relationships may further limit
our ability to sell products to those cable operators. We do not have long,
well-established relationships with those cable operators. If we were to sell
our products to those cable operators, it would likely not be based on long-
term contracts and those customers would be able to terminate their
relationships with us at any time. In addition, one or more of our current
customers could cancel its relationship with us at any time.

   We have recently begun marketing and selling our products to providers of
DSL and wireless broadband services and thus we have very limited experience.
We do not have long, well-established relationships with these providers, and
we may not be successful in establishing these relationships.

We are dependent on broadband service providers choosing to offer additional
services to their customers.

   We depend on cable operators to purchase our cable modem systems and to
provide our cable modems to end users. Cable operators have a limited amount of
available bandwidth over which they can offer robust data services, and they
may not choose to provide these data services to their customers. If cable
operators choose to provide these services, we also will depend upon them to
market these services to cable customers, to install our equipment and to
provide support to end users. In addition, we will be highly dependent on cable
operators to continue to maintain their cable infrastructure in a manner that
allows us to provide consistently high performance and reliable services. Our
success also will depend upon the acceptance of our products by other providers
of services to the cable industry, such as Excite@Home's @Home Network and Road
Runner, a joint venture between MediaOne Group, Inc. and Time Warner Cable.
Sales of our DSL and wireless products are also dependent on service providers
choosing to purchase our products and to provide additional services to their
end users.

Sales of our cable products are dependent on the cable industry upgrading to
two-way cable infrastructure.

   Demand for our products will depend, to a significant degree, upon the
magnitude and timing of capital spending by cable operators for implementation
of access systems for data transmission over cable networks. This involves the
enabling of two-way transmission over existing coaxial cable networks and the
eventual upgrade to HFC in areas of higher penetration of data services. If
cable operators fail to complete these upgrades of their cable infrastructures
in a timely and satisfactory manner, the market for our products could be
limited. In addition, few businesses in the United States currently have cable
access. Cable operators may not choose to upgrade existing residential cable
systems or to install new cable systems to serve business locations.

   The success and future growth of our business will be subject to economic
and other factors affecting the cable television industry generally,
particularly its ability to finance substantial capital expenditures. Capital
spending levels in the cable industry in the United

                                       54

<PAGE>

States have fluctuated significantly in the past, and we believe that such
fluctuations will occur in the future. The capital spending patterns of cable
operators are dependent on a variety of factors, including the following:

  .  the availability of financing;

  .  cable operators' annual budget cycles, as well as the typical reduction
     in upgrade projects during the winter months;

  .  the status of federal, local and foreign government regulation and
     deregulation of the telecommunications industry;

  .  overall demand for cable services;

  .  competitive pressures (including the availability of alternative data
     transmission and access technologies);

  .  discretionary consumer spending patterns; and

  .  general economic conditions.

   In recent periods, the United States cable market has been characterized by
the acquisition of smaller and independent cable operators by large cable
operators. We cannot predict the effect, if any, that consolidation in the
United States cable industry will have on overall capital spending patterns by
cable operators. The effect on our business of further industry consolidation
also is uncertain.

Supply of our products may be limited by our ability to forecast demand
accurately.

   The emerging nature of the broadband access services market makes it
difficult for us to accurately forecast demand for our products. Our inability
to accurately forecast the actual demand for our products could result in
supply, manufacturing or testing capacity constraints. These constraints could
result in delays in the delivery of our products or the loss of existing or
potential customers, either of which could have a negative impact on our
business, operating results or financial condition. In addition, we had
unconditional purchase obligations of approximately $125.7 million as of March
31, 2000, primarily to purchase minimum quantities of materials and components
used to manufacture our products. We must fulfill these obligations even if
demand for our products is lower than we anticipate.

We are dependent on key third-party suppliers.

   We manufacture all of our products using components or subassemblies
procured from third-party suppliers. Some of these components are available
from a single source and others are available from limited sources. All of our
sales are from products containing one or more components that are available
only from single supply sources. In addition, some of the components are custom
parts produced to our specifications. For example, we currently rely on Philips
Semiconductors, Inc. to supply a custom ASIC that is used in our products.
Other components, such as the radio frequency tuner and some surface acoustic
wave filters, are procured from sole source suppliers. Any interruption in the
operations of vendors of sole source parts could adversely affect our ability
to meet our scheduled product deliveries to customers. We are dependent on
semiconductor manufacturers and are affected by worldwide conditions in the
semiconductor market. If we are unable to obtain a sufficient supply of
components from our current sources, we could experience difficulties in
obtaining alternative sources or in altering product designs to use alternative
components. Resulting delays or reductions in product shipments could damage
customer relationships. Further, a significant increase in the price of one or
more of these components could harm our gross margin or operating results.

                                       55

<PAGE>

Shortages in supplies of components may impair our ability to meet customer
demands.

   Due to increasing demand for electronic and communications equipment, the
worldwide market for component parts is currently constrained. Delays in key
component or product deliveries may occur due to shortages resulting from a
limited number of suppliers, the financial or other difficulties of such
suppliers or a limitation in component product availability. Due to these
current market conditions, we face the risk of possible shortages of certain
key components that could result in product performance shortfalls and reduced
control over or delay in delivery schedules, manufacturing capability, quality
and costs, all of which could impair our ability to produce enough product to
meet our customer demand. In addition, in order to fulfill demand for our
products, we may have to purchase these components on the spot market at a
price that may be higher than we have experienced in the past.

   Although we work closely with our suppliers to avoid these types of
shortages, there can be no assurance that we will not encounter these problems
in the future. While our suppliers have performed effectively and been
relatively flexible to date, we believe that we will be faced with the
following challenges going forward:

  .  New markets in which we participate may grow quickly and consume
     component capacity; and

  .  As we continue to acquire companies and new technologies, we are
     dependent, at least initially, on unfamiliar supply chains or relatively
     small supply partners.

   Manufacturing capacity and component supply constraints could be significant
issues for us. If we were unable to obtain adequate quantities of significant
component materials on a timely basis, our business and our customer
relationships would be adversely affected. If we are unable to satisfy our
customers' demand, our customers could decide to purchase products from our
competitors. Inability to meet demand, or a decision by one or more of our
customers to purchase from our competitors, could harm our operating results.

We may be unable to migrate to new semiconductor process technologies
successfully or on a timely basis.

   Our future success will depend in part upon our ability to develop products
that utilize new semiconductor process technologies. These technologies change
rapidly and require us to spend significant amounts on research and
development. We continuously evaluate the benefits of redesigning our
integrated circuits using smaller geometry process technologies to improve
performance and reduce costs. The transition of our products to integrated
circuits with increasingly smaller geometries will be important to our
competitive position. Other companies have experienced difficulty in migrating
to new semiconductor processes and, consequently, have suffered reduced yields,
delays in product deliveries and increased expense levels. Moreover, we depend
on our relationship with our third-party manufacturers to migrate to smaller
geometry processes successfully.

Our ability to directly control product delivery schedules and product quality
is dependent on third-party contract manufacturers.

   Most of our products are assembled and tested by contract manufacturers
using testing equipment that we provide. As a result of our dependence on these
contract manufacturers for assembly and testing of our products, we do not
directly control product delivery schedules or product quality. Any product
shortages or quality assurance problems could increase the costs of
manufacture, assembly or testing of our products. In addition, as manufacturing
volume increases, we will need to procure and assemble additional testing
equipment and provide it

                                       56

<PAGE>

to our contract manufacturers. The production and assembly of testing equipment
typically requires significant time. We could experience significant delays in
the shipment of our products if we are unable to provide this testing equipment
to our contract manufacturers in a timely manner.

There are many risks associated with international operations.

   Sales to customers outside of the United States accounted for approximately
84% of our revenues in 1999 and approximately 74% of our revenues in 1998. We
expect sales to customers outside of the United States to continue to represent
a significant percentage of our revenues for the foreseeable future.
International sales are subject to a number of risks, including the following:

  .  changes in foreign government regulations and communications standards;

  .  export license requirements, tariffs and taxes;

  .  other trade barriers;

  .  difficulty in protecting intellectual property;

  .  difficulty in collecting accounts receivable;

  .  difficulty in managing foreign operations; and

  .  political and economic instability.

   If our customers are affected by currency devaluations or general economic
crises, such as the recent economic crisis affecting many Asian and Latin
American economies, their ability to purchase our products could be reduced
significantly. Payment cycles for international customers typically are longer
than those for customers in the United States. Foreign markets for our products
may develop more slowly than currently anticipated. Foreign countries may
decide not to construct cable infrastructure or may prohibit, terminate or
delay the construction of new cable plants for a variety of reasons. These
reasons include environmental issues, economic downturns, the availability of
favorable pricing for other communications services or the availability and
cost of related equipment. Any action like this by foreign countries would
reduce the market for our products.

   We anticipate that our foreign sales generally will be invoiced in U.S.
dollars, and we currently do not plan to engage in foreign currency hedging
transactions. However, as we commence and expand our international operations,
we may be paid in foreign currencies and exposure to losses in foreign currency
transactions may increase. We may choose to limit our exposure by the purchase
of forward foreign exchange contracts or through similar hedging strategies. No
currency hedging strategy can fully protect against exchange-related losses. In
addition, if the relative value of the U.S. dollar in comparison to the
currency of our foreign customers should increase, the resulting effective
price increase of our products to those foreign customers could result in
decreased sales.

We may be unable to provide adequate customer support.

   Our ability to achieve our planned sales growth and retain current and
future customers will depend in part upon the quality of our customer support
operations. Our customers generally require significant support and training
with respect to our broadband access systems, particularly in the initial
deployment and implementation stages. To date our sales have been concentrated
in a small number of customers. We have limited experience with widespread
deployment of our products to a diverse customer base. We may not have

                                       57

<PAGE>

adequate personnel to provide the levels of support that our customers may
require during initial product deployment or on an ongoing basis. Our inability
to provide sufficient support to our customers could delay or prevent the
successful deployment of our products. In addition, our failure to provide
adequate support could harm our reputation and relationship with our customers
and could prevent us from gaining new customers.

Our industry is highly competitive with many established competitors.

   The market for broadband access systems is extremely competitive and is
characterized by rapid technological change. Our direct competitors in the
cable access systems arena include Cisco Systems, Com21, General Instrument,
Matsushita Electric Industrial (which markets products under the brand name
"Panasonic"), Motorola, Nortel Networks, Vyyo, Thomson Consumer Electronics
(which markets products under the brand name "RCA"), Samsung, Scientific-
Atlanta, Sony, 3Com, Toshiba and Zenith Electronics. We also compete with
companies that develop integrated circuits for broadband access products, such
as Broadcom, Conexant and Texas Instruments. We also sell products that compete
with existing data access and transmission systems utilizing the
telecommunications networks, such as those of 3Com. Additionally, our
controller and headend system products face intense competition from well-
established companies such as Cisco, Nortel and 3Com. In addition, we compete
with companies in the DSL arena such as ECI, Charles Industries, Pairgain,
Copper Mountain, Accelerated Networks, Integral Access and VINA Technologies.
As standards, such as DOCSIS, are developed for broadband access systems, other
companies may enter the broadband access systems market. The principal
competitive factors in our market include the following:

  .  product performance, features and reliability;

  .  price;

  .  size and stability of operations;

  .  breadth of product line;

  .  sales and distribution capability;

  .  technical support and service;

  .  relationships with providers of broadband access services; and

  .  compliance with industry standards.

   Some of these factors are outside of our control. The existing conditions in
the broadband access market could change rapidly and significantly as a result
of technological advancements. The development and market acceptance of
alternative technologies could decrease the demand for our products or render
them obsolete. Our competitors may introduce broadband access products that are
less costly, provide superior performance or achieve greater market acceptance
than our products.

   Many of our current and potential competitors have significantly greater
financial, technical, marketing, distribution, customer support and other
resources, as well as greater name recognition and access to customers than we
do. The anticipated widespread adoption of DOCSIS and other industry standards
is likely to cause increased worldwide price competition, particularly in the
North American market. The adoption of DOCSIS and these other standards also
could result in lower sales of our TeraComm system, including the higher margin
headend products. Any increased price competition or reduction in sales of our
headend products would result in downward pressure on our gross margin. We
cannot accurately predict how the competitive pressures that we face will
affect our business.

                                       58

<PAGE>

Our business is dependent on the Internet and the development of the Internet
infrastructure.

   Our success will depend in large part on increased use of the Internet to
increase the need for high speed broadband access networks. Critical issues
concerning the commercial use of the Internet remain largely unresolved and are
likely to affect the development of the market for our products. These issues
include security, reliability, cost, ease of access and quality of service. Our
success also will depend on the growth of the use of the Internet by
businesses, particularly for applications that utilize multimedia content and
thus require high bandwidth. The recent growth in the use of the Internet has
caused frequent periods of performance degradation. This has required the
upgrade of routers, telecommunications links and other components forming the
infrastructure of the Internet by Internet service providers and other
organizations with links to the Internet. Any perceived degradation in the
performance of the Internet as a whole could undermine the benefits of our
products. Potentially increased performance provided by our products and the
products of others ultimately is limited by and reliant upon the speed and
reliability of the Internet backbone itself. Consequently, the emergence and
growth of the market for our products will depend on improvements being made to
the entire Internet infrastructure to alleviate overloading and congestion.

Our failure to manage growth could adversely affect us.

   The growth of our business has placed, and is expected to continue to place,
a significant strain on our limited personnel, management and other resources.
Our management, personnel, systems, procedures and controls may be inadequate
to support our existing and future operations. To manage any future growth
effectively, we will need to attract, train, motivate, manage and retain
employees successfully, to integrate new employees into our overall operations
and to continue to improve our operational, financial and management systems.

We are dependent on key personnel.

   Due to the specialized nature of our business, we are highly dependent on
the continued service of, and on the ability to attract and retain qualified
engineering, sales, marketing and senior management personnel. The competition
for personnel is intense. The loss of any of these individuals, particularly
our Chairman, President and Chief Technical Officer, Shlomo Rakib, and our
Chief Executive Officer, Zaki Rakib, would harm our business. In addition, if
we are unable to hire additional qualified personnel as needed, we may be
unable to adequately manage and complete our existing sales commitments and to
bid for and execute additional sales. Further, we must train and manage our
growing employee base, which is likely to require increased levels of
responsibility for both existing and new management personnel. Our current
management personnel and systems may be inadequate, and we may fail to
assimilate new employees successfully.

   Highly skilled employees with the education and training that we require,
especially employees with significant experience and expertise in both data
networking and radio frequency design, are in high demand. We may not be able
to continue to attract and retain the qualified personnel necessary for the
development of our business. We do not have "key person" insurance coverage for
the loss of any of our employees. Any officer or employee of our company can
terminate his or her relationship with us at any time. Our employees generally
are not bound by non-competition agreements with us.

Our business is subject to the risks of product returns, product liability and
product defects.

   Products as complex as ours frequently contain undetected errors or
failures, especially when first introduced or when new versions are released.
Despite testing, errors may occur.

                                       59

<PAGE>

The occurrence of errors could result in product returns and other losses to
our company or our customers. This occurrence also could result in the loss of
or delay in market acceptance of our products. Due to the recent introduction
of our products, we have limited experience with the problems that could arise
with this generation of products. However, the limitation of liability
provision contained in our purchase agreements may not be effective as a result
of federal, state or local laws or ordinances or unfavorable judicial decisions
in the United States or other countries. We have not experienced any product
liability claims to date, but the sale and support of our products entails the
risk of such claims. In addition, any failure by our products to properly
perform could result in claims against us by our customers. We maintain
insurance to protect against certain claims associated with the use of our
products, but our insurance coverage may not adequately cover any claim
asserted against us. In addition, even claims that ultimately are unsuccessful
could result in our expenditure of funds in litigation and management time and
resources.

We may be unable to adequately protect or enforce our intellectual property
rights.

   We rely on a combination of patent, trade secret, copyright and trademark
laws and contractual restrictions to establish and protect proprietary rights
in our products. Our pending patent applications may not be granted. Even if
they are granted, the claims covered by the patents may be reduced from those
included in our applications. Any patent might be subject to challenge in court
and, whether or not challenged, might not be broad enough to prevent third
parties from developing equivalent technologies or products without taking a
license from us. We have entered into confidentiality and invention assignment
agreements with our employees, and we enter into non-disclosure agreements with
some of our suppliers, distributors and appropriate customers so as to limit
access to and disclosure of our proprietary information. These statutory and
contractual arrangements may not prove sufficient to prevent misappropriation
of our technology or to deter independent third-party development of similar
technologies. In addition, the laws of some foreign countries might not protect
our products or intellectual property rights to the same extent as do the laws
of the United States. Protection of our intellectual property might not be
available in every country in which our products might be manufactured,
marketed or sold.

   In November 1998, CableLabs selected us to co-author DOCSIS 1.2, an enhanced
version of the DOCSIS cable modem specification based in part on our S-CDMA
technology. In September 1999, CableLabs indicated that it intended to proceed
with the advanced PHY work on two parallel tracks: one for the development of a
prototype based on our S-CDMA technology and one for the inclusion of Advanced
TDMA technology, as proposed by other companies. In February 2000, CableLabs
further clarified the status of the advanced PHY project regarding a separate
release that will include TDMA technologies. In addition, CableLabs reiterated
that it is continuing to work with us on the development of a DOCSIS
specification that could include our S-CDMA technology. To that end, we have
indicated to CableLabs that we would contribute some aspects of our S-CDMA
technology to the DOCSIS intellectual property pool if and when a DOCSIS
specification is approved that includes our S-CDMA technology.

   We would contribute our technology pursuant to a license agreement with
CableLabs that we would execute at that time, and which contains the terms that
CableLabs has established for the inclusion of any intellectual property from
any source in the DOCSIS specifications. Under the terms of the proposed
license agreement, we would grant to CableLabs a royalty-free license for those
aspects of our S-CDMA technology that are essential for compliance with the
DOCSIS cable modem standard. So-called "implementation know how" is not covered
by this license-only those aspects of the technology that are essential to
implementing a

                                       60
<PAGE>

compliant product. CableLabs would have the right to extend royalty-free
sublicenses to companies that wish to build DOCSIS-compatible products. These
sublicenses would allow participating companies to utilize and incorporate the
essential portions of the S-CDMA technology on a royalty-free basis for the
limited use of making and selling products or systems that comply with the
DOCSIS cable modem specification. Terayon has already joined the DOCSIS
intellectual property pool and, as a result, we have a royalty-free sublicense
that allows us to ship DOCSIS-compatible products which contain intellectual
property submitted by other companies. The scope of this license would not
extend to the use of the S-CDMA technology in other areas; only for products
that comply with the DOCSIS specifications. As a result, any of our competitors
who join or have joined the DOCSIS intellectual property pool will have access
to some aspects of our technology without being required to pay us any
royalties or other compensation. If and when we submit S-CDMA to the DOCSIS
Intellectual Property pool, we are in no way restricted from entering into
royalty-bearing license agreements with companies that wish to use the S-CDMA
technology for purposes other than implementing DOCSIS compatible products, or
that do not wish to enter into the DOCSIS intellectual property pool. Further,
some of our competitors have been successful in reverse engineering the
technology of other companies, and the inclusion of S-CDMA in a future DOCSIS
specification would expose some aspects of our technology to those competitors.
DOCSIS specifications are available on an open basis once they are approved,
not only to companies that are members of the DOCSIS IP Pool. If a competitor
is able to duplicate the functionality and capabilities of our technology, we
could lose all or some of the time-to-market advantage we might otherwise have.
Under the terms of the proposed license agreement, if we sue certain parties to
the proposed license agreement on claims of infringement of any copyright or
patent right or misappropriation of any trade secret, those parties may
terminate our license to the patents or copyrights they contributed to the
DOCSIS intellectual property pool. If a termination like this were to occur, we
would continue to have access to some aspects of the DOCSIS intellectual
property pool, but we would not be able to develop products that fully comply
with the DOCSIS cable modem specification. Also, even if we were to be removed
from the IP pool, we would not be prevented from developing and selling
products that fully comply with the DOCSIS specifications, but we would not be
able to do this with the benefit of a royalty-free license, which would
increase the cost of our products, assuming we were able to obtain a license
agreement for the required technology. Because of these terms, we may find it
difficult to enforce our intellectual property rights against certain
companies, even in areas that are not directly related to DOCSIS specifications
and products.

   We anticipate that developers of cable modems increasingly will be subject
to infringement claims as the number of products and competitors in our
industry segment grows. We have received letters from two individuals claiming
that our technology infringes patents held by these individuals. We have
reviewed the allegations made by these individuals and, after consulting with
our patent counsel, we do not believe that our technology infringes any valid
claim of these individuals' patents. If the issues are submitted to a court,
the court could find that our products infringe these patents. In addition,
these individuals may continue to assert infringement. If we are found to have
infringed these individuals' patents, we could be subject to substantial
damages and/or an injunction preventing us from conducting our business. In
addition, other third parties may assert infringement claims against us in the
future. An infringement claim, whether meritorious or not, could be time-
consuming, result in costly litigation, cause product shipment delays or
require us to enter into royalty or licensing agreements. These royalty or
licensing agreements may not be available on terms acceptable to us or at all.
Litigation also may be necessary to enforce our intellectual property rights.

   We pursue the registration of our trademarks in the United States and have
applications pending to register several of our trademarks. However, the laws
of certain foreign countries

                                       61

<PAGE>

might not protect our products or intellectual property rights to the same
extent as the laws of the United States. This means that effective trademark,
copyright, trade secret and patent protection might not be available in every
country in which our products might be manufactured, marketed or sold.

Our business is subject to communications industry regulations.

   Our business and our customers are subject to varying degrees of federal,
state and local regulation. The jurisdiction of the Federal Communications
Commission extends to the communications industry, including our broadband
access products. The FCC has promulgated regulations that, among other things,
set installation and equipment standards for communications systems. Although
FCC regulations and other governmental regulations have not materially
restricted our operations to date, future regulations applicable to our
business or our customers could be adopted by the FCC or other regulatory
bodies. For example, FCC regulatory policies affecting the availability of
cable services and other terms on which cable companies conduct their business
may impede our penetration of certain markets. In addition, regulation of cable
television rates may affect the speed at which cable operators upgrade their
cable infrastructures to two-way HFC. In addition, the increasing demand for
communications systems has exerted pressure on regulatory bodies worldwide to
adopt new standards for such products and services. This process generally
involves extensive investigation of and deliberation over competing
technologies. The delays inherent in this governmental approval process have in
the past, and may in the future, cause the cancellation, postponement or
rescheduling of the installation of communications systems by our customers.

   If other countries begin to regulate the cable modem industry more heavily
or introduce standards or specifications with which our products do not comply,
we will be unable to offer products in those countries until our products
comply with those standards or specifications. In addition, we may have to
incur substantial costs to comply with those standards or specifications. For
instance, should the DAVIC standards for ATM-based digital video be established
internationally, we will need to conform our cable modems to compete. Further,
many countries do not have regulations for installation of cable modem systems
or for upgrading existing cable network systems to accommodate our products.
Whether we currently operate in a country without these regulations or enter
into the market in a country these regulations do not exist, new regulations
could be proposed at any time. The imposition of regulations like this could
place limitations on a country's cable operators' ability to upgrade to support
our products. Cable operators in these countries may not be able to comply with
these regulations, and compliance with these regulations may require a long,
costly process. For example, we experienced delays in product shipments to a
customer in Brazil due to delays in certain regulatory approvals in Brazil.
Similar delays could occur in other countries in which we market or plan to
market our products. In addition, our customers in certain parts of Asia, such
as Japan, are required to obtain licenses prior to selling our products, and
delays in obtaining required licenses could harm our ability to sell products
to these customers.

Our business is subject to other regulatory approvals and certifications.

   In the United States, in addition to complying with FCC regulations, our
products are required to meet certain safety requirements. For example, we are
required to have our products certified by Underwriters Laboratory in order to
meet federal requirements relating to electrical appliances to be used inside
the home. Outside the United States, our products are subject to the regulatory
requirements of each country in which the products are manufactured or sold.
These requirements are likely to vary widely. We may be unable to obtain on a
timely basis or at all the regulatory approvals that may be required for the

                                       62

<PAGE>

manufacture, marketing and sale of our products. In addition to regulatory
compliance, some cable industry participants may require certification of
compatibility.

We are vulnerable to earthquakes and other natural disasters.

   The facility housing our corporate headquarters, the majority of our
research and development activities and our in-house manufacturing operations
is located in an area of California known for seismic activity. In addition,
the operations of some of our key suppliers are also located in this area and
in other areas know for seismic activity, such as Taiwan. An earthquake, or
other significant natural disaster, could result in an interruption in our
business or that of one or more of our key suppliers. Such an interruption
could harm our operating results.


                                       63

<PAGE>


                                   Signature

  Pursuant to the requirements of the Securities Exchange Act of 1934, the
registrant has duly caused this report to be signed on its behalf by the
undersigned, thereunto duly authorized.

                                    Terayon Communication Systems, Inc.


Dated:  July 18, 2000                By: /s/ Ray M. Fritz
                                         ----------------------------
                                         Ray M. Fritz
                                         Chief Financial Officer


                                      64


© 2022 IncJournal is not affiliated with or endorsed by the U.S. Securities and Exchange Commission