Post #4818399 · Elektrine

@wholookshere@lemmy.blahaj.zone

2026-09-23 12:33 UTC

Gonna take GPS example to explain general processors vs specific ones. GPS works by doing some simple math around triangulation from satalites. When they first came out there were no GPS chips. It was done by a general processors. Now they can do it, but because they can do Amy Hong they’re not as fast as newer GPS chips. How they work is the silicon can only handle the GPS calculations. But because of that, it can do it way faster. LLMs run on matrix math and probabilities. We can 100% come up with specialised silicon to handle this math. Intact, thats what a lot of AI specialised chips are.

Replies (3)

@piccolo@sh.itjust.works 2026-09-23 13:43

NPUs are already a thing, specialized cpus for nerual networks. The problem is the models are just databases, and you need lots of fast memory in order to feed the NPUs data, and thats the current bottleneck.

Open ##4818958
@trebach@sh.itjust.works 2026-09-23 13:51

They aren’t comparable. GPS is deterministic and simple so as you said it could be reduced to an FPGA or ASIC. LLMs are partially matrix math and probabilities but they’re also more complex than that and require a large amount of RAM to run even the first time. Each query added to the context increases the RAM needed further.

Open ##4819015
@merc@sh.itjust.works 2026-09-23 18:03

LLMs run on matrix math and probabilities Yes, and the silicon to handle that math already exists. It’s the “GPUs” being pumped out by nVidia. Those are basically now highly specialized matrix math machines.

Open ##4821267