The Itanic Has Sunk

Re:The Itanic Has Sunk by jabl on Friday July 30, 2021 @09:49AM (#28007931)

I suspect if you want a HW architecture for running array operations you'll end up with something like a vector machine (e.g. ARM SVE(2) ) or a GPU rather than a VLIW CPU?

[ Parent | on HN ]

Re:The Itanic Has Sunk by marcosdumay on Friday July 30, 2021 @11:31AM (#28009060)

VLIW is basically a more flexible kind of vector machine.

[ Parent | on HN ]

Re:The Itanic Has Sunk by jabl on Friday July 30, 2021 @01:08PM (#28010053)

And a traditional scalar architecture is more flexible still. The trick is to pick the correct set of tradeoffs for the targeted applications. I claim that for most array style workloads vector/GPU architectures are flexible enough, and offer better perf/watt and perf/chip area.

[ Parent | on HN ]