Can GPUs really be considered VLIW just because they might share an instruction ... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		flohofwoe on Feb 4, 2023 \| parent \| context \| favorite \| on: AMD Killed the Itanium (2005) Can GPUs really be considered VLIW just because they might share an instruction pointer across a couple of 'micro cores'? AFAIK they went even back from SIMD to scalar a long time ago.

wespiser_2018 on Feb 4, 2023 | [–]

No, I don't think so. It's one instruction on multiple data (SIMD). That's the essence of a vector op!

dontlaugh on Feb 4, 2023 | [–]

I suppose. They’re still very wide machines, which to me seems like the most interesting aspect of VLIW.

Of course actual VLIW still are around as DSPs.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact