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

I'm not sure you needed to predict AI.

GPUs exist because CPUs just aren't fast enough. Whether or not people are going to start making GPU-only computers is debatable (although there has clearly been a lot of CPU+GPU single chips).



What a frankly weird oversimplification? GPUs don't exist because CPUs aren't fast enough. GPUs exist because CPUs aren't parallel enough. And to achieve that parallelism, they sacrifice massive amounts of performance to get it.

A GPU-only computer would be absolutely horrendous to use. It'd be incredibly slow and unresponsive as GPUs just absolutely suck at running single-threaded code or branchy code.


I mean you can splits hairs about the difference between CPU and GPUs all you want.

The overall point is that work is being increasingly done _not_ on the CPU. If your business is CPUs-only then you're going to have rough times as the world moves away from a CPU-centric world. You don't need to predict AI; you just need to see that alternatives are being looked at by competitors and you'll lose an edge if you don't also look at them.

It's not going to matter much if you have a crappy CPU if most of the work is done on the GPU. Its like how iPhones don't advertise themselves as surround sound; phones aren't about calling people anymore so no reason to advertise legacy features.


> The overall point is that work is being increasingly done _not_ on the CPU.

Eh? GPGPU has been a thing for decades and yet barely made a dent in the demand for CPUs. Heck, CUDA is 17 years old!

The world has not budged from being CPU-centric and it isn't showing any signs of doing so. GPUs remain an accelerator for specialized workloads and are going to continue to be just that.


Hmm, this isn't power-efficient thinking. iPhones have lots of accelerators beyond GPUs - not because they're more performant but because they're better at performance/power.

Of course they also add GPU-like things to the CPU too for the same reason: https://developer.arm.com/documentation/109246/0100/SME-Over...


That's orthogonal. Fixed-function hardware has a power efficiency advantage over programmable hardware. This isn't new or surprising, but it's also unrelated to CPU vs. GPU. Both CPU & GPU are programmable, so that's not relevant.


Those aren't useful distinctions. You might as well say the ALUs in a CPU aren't programmable because they aren't the instruction decoders.


Huh? Weren't you referring to things like the video decoding blocks? Those are genuinely fixed-function things, they have no aspects of programmability to them. It's why their capabilities are fixed for all time (eg, AV1 support requires a hardware change). It's like a pre-shader GPU.

It's a very useful distinction to make.




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

Search: