Seems we’re in need of a change of subject.  Let’s talk tech for a bit: Intel vs POET in CPO.

POET’s interposer platform is the most versatile, lowest cost and and most scalable integration platform. While all the advantages POET brings to silicon photonics don’t necessarily apply to copackaged optics, many do.

Intel is actively working on copackaged optics, and they are making progress.  However, they do have some problems.  In contrast to POET, their solution for laser failure is to build in a completely redundant 32 laser system for a 16 lane device.  That’s right, they double the lasers integrated into the chip just in case one fails. Intel has to design in all of the redundant circuits, switches and energy costs just in case one of their lasers fails. Clearly, this means a more complex and expensive solution compared to POET’s customer's. Of course, the redundancy is better than replacing the entire device, but it doesn’t even come close to the simplicity of replacing the LightBar POET has developed.  Next, by Intel’s own admission their CPO solution only really comes into its own when they approach 100Tb, so while their 51.2Tb product is an improvement, it is by no means outstanding.  And the finale, Intel is anticipating they will have their CPO solution ready for mass production by the end of 2023.  That’s over 2 years for POET and their customers to make something much much better….