Press Inquiries Contact:
Lisa Frieden (for LightCounting)
(510) 525-7945
press@lightcounting.com

Analyst Relations Contact:
Vladimir G. Kozlov
LightCounting
1 (408) 962-4851
vladimir.kozlov@lightcounting.com

LightCounting News: July 2010

40 Gbps-Good for Headlines, Not so Much for Profit.

By Vladimir Kozlov, Founder and CEO, LightCounting, LLC

The optical communications community has been waiting for wide scale deployment of 40 Gbps transport systems for a decade. Now, with 40G on its third technology iteration, itÂs not surprising the excitement is wearing off, and valuations of 40 Gbps module vendors are coming down. Recent acquisitions averaged only two times sales and near or below VC-invested capital:

-Stratalight was acquired by Opnext in January 2009 and valued at $170 million, on annual sales estimated at under $80 million.

-Cisco acquired CoreOptics in May 2010 for $99 million, on estimated annual sales of $50 million.

-Oclaro acquired Mintera in July 2010 for $12 million plus an additional amount based on Mintera's revenue over the next 18 months. If Mintera generates $70 million in revenue over the next 18 months, the maximum additional payment would be $20 million.

Even in the best-case scenario, with Mintera generating $70 million in sales over the next 18 months, the acquisition valuation of Mintera is well below annual revenues. Meanwhile, forecasts for 40 Gbps deployments remain as upbeat as ever, despite emergence of 100 Gbps technology that some expect will eclipse them.

Why are 40 Gbps module vendors struggling? As Niall Robinson, Mintera's VP of Marketing, commented in one of his presentations, the road for 40 Gbps products has been a long and winding one. Striving to optimize performance of 40 Gbps transmission systems, vendors developed several generations of transponder modules: DPSK, DQPSK, and coherent modules. Mintera was on the leading edge of product development cycle for DPSK modules but fell behind as the industry moved to DQPSK and coherent modulation formats. CoreOptics and Stratalight also had problems developing a profitable business, as sales volumes of 40 Gbps modules remained limited and the economic downturn of late 2008 and early 2009 slowed deployments of 40 Gbps systems. Market data collected by LightCounting shows that sales of 40 Gbps DWDM transponders reached only $54 million in 2009 (for a detailed market analysis, see LightCounting's most recent Optical Communications Market Update Report at http://www.lightcounting.com/reports/OpticalCommunicationsMarketRepor_abstract.html.)

While the economic downturn was a temporary factor limiting shipments of 40 Gbps modules, the main problem is that leading system vendors, including Alcatel-Lucent, Ciena, and Huawei, design and manufacture 40 Gbps DWDM interfaces internally instead of relying on transponder vendors. LightCounting estimates that more than 60% of 40G DWDM ports shipped in 2009 were based on interfaces manufactured internally by system vendors. While transponders sales are likely to increase in 2010, they will probably account for only 30% of the total 40G DWDM ports, marginalizing the market opportunity for transponder vendors. Transition to 100 Gbps products is unlikely to improve this situation, as even higher percentages of 100 Gbps ports will be made internally.

Developing technology for high-data-rate interfaces in house gives equipment manufacturers more opportunities to differentiate their products on cost and performance and to maintain higher margins. Alcatel-Lucent and Nortel (now Ciena) invested heavily in the development of 40G systems starting in the late 1990s and managed to preserve at least some of this expertise in houseenough to internally develop 40 and 100 Gbps coherent technology. Huawei entered the development race years later, but it seems to be leading in sales of 40 Gbps products today. Cisco surprised the industry by taking CoreOptics operations in house and thus gaining a strong position at 40G and a head start on 100G product development.

In these circumstances, it would make more sense for Mintera to find a buyer among other system houses focused on high-data-rate transport, such as Fujitsu Networks or Nokia Siemens. Apparently, Mintera just could not find the right fit. It also possible that not all system vendors are willing to commit to internal developments of 40G and 100G interfaces and are planning to rely on products offered by transponder suppliers. This is certainly the case for midsize equipment suppliers, who simply cannot afford or do not have infrastructure to support internal developments of high-data-rate DWDM interfaces.

Oclaro clearly took advantage of the situation, acquiring Mintera's top-notch engineering team and existing customer relationships. If everything goes well, this acquisition could help Oclaro to sort out its 40G and, more importantly, 100G product offerings. Inheriting Lithium Niobate technology from Avanex and Indium Phosphide modulators from Bookham, there is plenty to choose from. As always, execution and timing will be the key factors.