ECOC Reporter's Notebook -- Day 3

Wednesday, 23 September 2009

[8:36 AM CET] Sitting in the press room before the final day of the show begins, saying last rites for my cell phone and watch batteries, both of which are about to die. Should make keeping appointments an interesting challenge. Meanwhile, here are a few stray notes from yesterday I didn't have a chance to cover...Opnext is showing off a prototype 100G CFP for LR4. They're also highlighting a 40G DQPSK module in 300-pin format and an SFF VSR module. The DQPSK module has begun shipping to customers and complements the continuously optimized DPSK technology introduced at OFC. The latter is good for applications involving lots of ROADMs, the former for links where fiber impairments are the primary impediment to reach...Along with MergeOptics (see the Day 1 blog) and Yenista (whom I neglected to mention in yesterday's test roundup) Proximion is another European company claiming to be going great guns despite the economy. The company is doing well with its tunable optical dispersion compensators for 40G, and is showing off a  compensator in a patch cord format introduced in June. The company's fiber grating expertise enables the design... 3S Photonics of France (ex-Alcatel Optronics, ex-Avanex) is highlighting its pump lasers, including a new, uncooled device for submarine applications, as well as laser technology for OFDM...Sumitomo Electric Device Innovations is quietly touting a tunable XFP-E which company sources say should beging sampling the fourth quarter of this year. Meanwhile, a 40GbE CFP module will sample in Q2 2010 and a 100GbE version the folowing quarter, if current plans hold...

[8:00 PM CET] The exhibition and the poster sessions are over, but this blog isn't...The 100G test conversations continued today.  Daniel van der Weide, vice president of engineering at Optametra, asked to meet me to make the case for the real-time approaches on which his company's offering (as well as Agilent's) is based, versus the EXFO approach he had seen highlighted in this blog yesterday. van der Weide suggested that people who are evaluating 100G test platforms should look closely at bandwidth claims and focus on informational bandwidth. He said that real-time approaches can more closely follow what's going on in a signal, since it's tracking all the bits, as opposed to analyzing optical samples. Among the benefits of the real-time approach is a greater capability to provide accuragte BER measurements, he asserted. You can bet we'll here more about this in the near future...van der Weide also expressed the hope that oscilloscope manufacturers would increase the bandwidth of their systems...Meanwhile, Aragon Photonics of Spain is also taking an optical approach to 100G and related testing. However, their approach leverages stimulated Brillouin scattering...Synthesys Research demonstrated brand new (as in "We just finished this box before we got on the plane to Vienna") capabilities to do stressed eye testing at 28.7812 Gbps, plus other capabilities for 100GbE requirements. The demonstration occurred in EXFO's booth, as the companies' equipment complements each other. Official announcement of the new capabilties (and, perhaps, details of the relationship with EXFO?) should occur around OFC next March...I neglected to mention yesterday that JDSU displayed its full range of 100GbE test gear as well.

On the transceiver front, Finisar has joined the CFP party, with a demonstration of a prototype 40G LR4. It also touted the capabilities of its Edge Wavelength Processor, a 1x2 WSS based on the company's liquid crystal on silicon technology...Gigalight of China has a range of 10G ZR (80 km) transceivers that should be available over the next two months. These include XENPAK, X2, and DWDM XFP. A company source asserted that Gigalight is one of the few Chinese module vendors with its own SFP+; the company hopes to offer an ER module in this form factor next year.

Inphi touted its differential 2811DZ 40G DQPSK modulator drivers in surface-mount packages. The company has demostrated interoperability of these devices with the SMT versions of Sierra Monolithics' 40G mux/demux devices. Loi Nguyen, vice president of broadband analog products for Inphi, expressed a sense of vindication that the company pursued DQPSK when many others were betting on DPSK for 40G. He said that the extremely small package sizes of the drivers might obviate the need for quad chips, at least from an economic perspective...Narda, meanwhile, touted its modulator drivers for 40G VSR, ODB, DPSK, and DQPSK drivers in GPPO packages. A source at the booth confirmed the obvious sense of playing in the 100G market.

And speaking of 100G, sources at the Discovery Semiconductor booth reported the company is making good progress on shrinking its KittyHawk coherent transmitter/receiver technology into something smaller than a rack unit. The hope, the sources said, is to come up with a line card that might be of interest to systems companies. Meanwhile, the company spread the word about its new 100GbE (4x25G) quad PIN-TIA optical receiver for LR and ER applications.
 

Return to top