Demystifying GPU Compute Architectures

Re:Demystifying GPU Compute Architectures – By Babbage by wolletd on Tuesday February 06, 2024 @05:45PM (#39281898)

Just to be fair: as also stated at the beginning of the post, CPUs are far more complex nowadays than they represent themselves to the outside.

They don't work like a von-Neumann machine, they act like one. Granted, we know a lot more about the inner workings of modern CPUs thn GPUs, but a lot of real-life work still assumes that the CPU works "as-if" it was a computer from the 70s, just really, really fast.

[ Parent | on HN ]

Re:Demystifying GPU Compute Architectures – By Babbage by WithinReason on Wednesday February 07, 2024 @03:59AM (#39286165)

A lot of effort goes into CPU design to ensure you don't have to care, unlike GPUs.

[ Parent | on HN ]

Re:Demystifying GPU Compute Architectures – By Babbage by amelius on Wednesday February 07, 2024 @04:40AM (#39286469)

Is that really a smart choice, given that we have compilers that can deal with the gory details?

[ Parent | on HN ]

Re:Demystifying GPU Compute Architectures – By Babbage by myrmidon on Wednesday February 07, 2024 @07:34AM (#39287775)

Relying on the compiler to compensate for an unwieldy instruction set architecture is something that was tried explicitly with the Intel Itanium and failed hard.

\rant I feel that generally people often have a notion of the "compiler is unbeatable at optimizing", which is completely false: You can always start from compiler output when optimizing and you often have easy options that are inaccessible for the compiler because of language semantics/calling conventions etc. (=> but hand optimizing at instruction level is time consuming, time is expensive, and what the compiler does is basically free ;).

[ Parent | on HN ]

Re:Demystifying GPU Compute Architectures – By Babbage by WithinReason on Wednesday February 07, 2024 @10:30AM (#39289724)

Compilers can absolutely not deal with the complexity of GPUs.

[ Parent | on HN ]