Hacker News new | ask | show | jobs
by saas_co_de 2854 days ago
> it's very difficult to identify which layout is most efficient

Don't worry. The compiler will figure it out. And this time the compiler will have AI™.

> the killer is that dead silicon tends to be a LOT of silicon for any given program

Part of the idea here is that the cost dynamic has changed. Silicon is cheap compared to power so even if you have lots of chips not being used at any given time as long as they can be fully powered off total system cost (capex+opex) is still better.

> why it would work this time

The difference is scale. If you are running millions of CPUs and adding 100k's per month then something like this could work, assuming the AI™ magic that figures out which new chips to build.

Intel is talking up their grandiose vision but practically this is the same as AMD's chiplets on active interposers (https://spectrum.ieee.org/tech-talk/semiconductors/design/am...).

The physical technology is real and probably coming soon but it will be limited to incremental improvements to the existing CPU/GPU compute architecture from increasing bandwidth and decreasing latency until the magic compilers arrive.

1 comments

If I'm understanding you right, what you're suggesting is that the individual silicon modules will be fixed, but they will be contected in a single package in lot's of different ways.

If that's correct I'd love to see the cost of making the trip between the modules. I've got to imagine the cost of that is just huge. The interconnect is also crazy - it's easy to do complex routing tasks on silicon at high performance. I don't know how you acheive that in a scalable fashion between silicon.