News Focus
News Focus
Followers 177
Posts 24191
Boards Moderated 0
Alias Born 09/11/2006

Re: KCCO7913 post# 199808

Thursday, 09/05/2024 8:50:43 AM

Thursday, September 05, 2024 8:50:43 AM

Post# of 232292
KCC your are implying the PIC isn’t ready and that’s why deals haven’t been made. You indicate they could try using their modulators with 4 lasers to get 800g with the current status but using four different lasers seems impractical, and that's why they need to complete a 4 lane PIC using one laser.

But then you say they’re going through the paces to prove out scalability!!

How can they prove out scalability if the PIC isn’t even ready yet? Or do you think they maybe have gotten the PIC to work in recent months but haven't announced it? Have I misunderstood what you wrote?

The PIC includes a laser, waveguides, and modulators to make up the entire module. The transceiver companies don’t just use the modulators. They use the entire PIC.

You believe there are no technical issues. How can this be if the PIC is not really ready yet? Wouldn’t it be a technical issue that is keeping the pic from having yet been developed ? Perhaps that is why they brought on Mr Lamaitre, who has long experience in the field of lasers, including VCSELS. How can you or anyone be sure there aren't any technical issues when Dr Lebby literally said there are other issues they are worried about?

Lastly you say silicon modulation is ubiquitous. But that something is needed for the future. Silicon in its current state is not sufficient. What if the industry comes up with that something and it’s not polymers?

Have you seen this?: I'm not sure how significant it is but seems another possible method to avoid needing other approaches at least for a while as it shows a new method enabling 200G lanes using just silicon. Do you have an opinion about it?:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10831040/

Optical interconnects have been recognized as the most promising solution to accelerate data transmission in the artificial intelligence era. Benefiting from their cost-effectiveness, compact dimensions, and wavelength multiplexing capability, silicon microring resonator modulators emerge as a compelling and scalable means for optical modulation. However, the inherent trade-off between bandwidth and modulation efficiency hinders the device performance. Here we demonstrate a dense wavelength division multiplexing microring modulator array on a silicon chip with a full data rate of 1 Tb/s. By harnessing the two individual p-n junctions with an optimized Z-shape doping profile, the inherent trade-off of silicon depletion-mode modulators is greatly mitigated, allowing for higher-speed modulation with energy consumption of sub-ten fJ/bit. This state-of-the-art demonstration shows that all-silicon modulators can practically enable future 200 Gb/s/lane optical interconnects.

Volume:
Day Range:
Bid:
Ask:
Last Trade Time:
Total Trades:
  • 1D
  • 1M
  • 3M
  • 6M
  • 1Y
  • 5Y
Recent LWLG News