UGent, IMEC launch photonics spin-off Caliopa

9/24/2010 3:31 PM EDT

Its just a news without analysis on why would the silicon based transceivers be better that the optical ones, what are the technical challenges and why no other than Belgian organizations is interested to join in this race to dominate 2.4 Billion existing market. From job creation point of view also a job growth of 100% is good but if it adds only 7 jobs then all the excitement is gone. "13% compound growth"...we all are adults here...we can figure out that the CAGR of India and China is more than that.

9/25/2010 6:39 AM EDT

Anything that can be manufactured in scalable available silicon is much more cost-efective than using exotic materials to achieve performance. While scant information is available on Caliopa's roadmap, having access to imec's processing technology can't hurt the spinoff. Having local fund backing is normal especially if the investors are familiar with the technological prowess of their target. One other company that is commecializing silicon photonics based devices is Luxtera; their performance claim is 10 Gb/s, and promising 100 Gb/s. See the following EE Times story: http://www.eetimes.com/electronics-news/4079658/Luxtera-Sun-win-Darpa-optical-IC-contract.
Know of any other contenders in this field?

9/25/2010 7:01 AM EDT

silicon photonics offers not only lower cost but paves the way for complete CMOS integration in the all optical board/interconnect scenario in the future. There are plenty of startups in this area and heavyweights like Intel, IBM have already invested heavily.

9/26/2010 2:41 AM EDT

Oh yeah, silicon photonics! I can't wait until it becomes mainstream! I understand that the hard part is generating coherent light in silicon; its indirect bandgap makes very inefficient lasers. Detection and transmission of light in the 850, 1300, and 1550 nm telecom windows works fine in Si.

9/27/2010 10:51 AM EDT

You're right the light source would be the holy grail, but in the meantime there are plenty of alternatives to couple light onto the silicon platform. The rest of the building blocks are kind of ready and can be mass produced.

9/27/2010 2:17 PM EDT

Common, 2 Million and 7 people? Too little, too few. I worked for a number of starts-up and $10M seems to be the lowest end of anything that might eventually work. Luxtera is a market leader in silicon photonics space, Intel is trying to catch-up...Kris

9/28/2010 1:47 AM EDT

With multicore processor architechtures on the road, silicon photonics will be an essential addition in allowing high bandwidth data transfer between the processor and memory or off-chip with reduced power consumption. Not to mention , the huge cost benefit in CMOS integration.

I think UCLA and intel has already shown in 2004 Silicon based photonic lasers/sources. Latest what I know is KOTURA & SUN labs oracle has demonstrated a Silicon modulator operating at 11 Ghz with low power consumption.

http://www.kotura.com/news/020210.html

Huge progress has been made in Si based modulators and detectors and more to come. Awaiting an exciting road ahead.

The question is how much more we can push the boundaries?