Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Intel announces Arc B-series "Battlemage" discrete graphics with Linux support

phoronix.com

Post reply on HN
Print view View original

71–80 of 722 posts

Re: Intel announces Arc B-series "Battlemage" discrete graphics with Linux support

#71

Post by karmakaze »

oidar wrote: post #38
karmakaze wrote: post #37

I wanted to have alternative choices than Nvidia for high power GPUs. Then the more I thought about it, the more it made sense to rent cloud services for AI/ML workloads and lesser powered ones for gaming. The only use cases I could come up with for wanting high-end cards are 4k gaming (a luxury I can't justify for infrequent use) or for PC VR which may still be valid if/when a decent OLED (or mini-OLED) headset is a…

Which video card are you using for PSVR?

I haven't decided/pulled-the-trigger but the Intel ARC series are giving the AMD parts a good run for the money.

The only concern is how well the new Intel drivers work (full support for DX12) with older titles which are continuously being improved (for DX11, 10, and some for 9 others via emulation).

There's likely some deep discounting of Intel cards because of how bad the drivers were at launch and the prices may not stay so low once things are working much better.

Re: Intel announces Arc B-series "Battlemage" discrete graphics with Linux support

#73

Post by m00x »

treprinum wrote: post #60
Muskyinhere wrote: post #58

Earlier quoted context omitted.

Because if they could just do that and it would rival what NVidia has, they would just do it. But obvoiusly they don't. And for reasons: NVidia has worked on CUDA for ages, do you believe they just replace this whole thing in no time?

llama.cpp and its derivatives say yes.

This is the most script kiddy comment I've seen in a while.

llama.cpp is just inference, not training, and the CUDA backend is still the fastest one by far. No one is even close to matching CUDA on either training or inference. The closest is AMD with ROCm, but there's likely a decade of work to be done to be competitive.

Re: Intel announces Arc B-series "Battlemage" discrete graphics with Linux support

#74

Post by Wytwwww »

Muskyinhere wrote: post #58
treprinum wrote: post #26

Why don't they just release a basic GPU with 128GB RAM and eat NVidia's local generative AI lunch? The networking effect of all devs porting their LLMs etc. to that card would instantly put them as a major CUDA threat. But beancounters running the company would never get such an idea...

Because if they could just do that and it would rival what NVidia has, they would just do it. But obvoiusly they don't. And for reasons: NVidia has worked on CUDA for ages, do you believe they just replace this whole thing in no time?

Does CUDA even matter than much for LLMs? Especially inference? I don't think software would be the limiting factor for this hypothetical GPU. Afterall it would be competing with Apple's M chips not with the 4090 or Nvidia's enterprise GPUs.

Re: Intel announces Arc B-series "Battlemage" discrete graphics with Linux support

#75

Post by treprinum »

m00x wrote: post #73
treprinum wrote: post #60

Earlier quoted context omitted.

llama.cpp and its derivatives say yes.

This is the most script kiddy comment I've seen in a while. llama.cpp is just inference, not training, and the CUDA backend is still the fastest one by far. No one is even close to matching CUDA on either training or inference. The closest is AMD with ROCm, but there's likely a decade of work to be done to be competitive.

Inference on very large LLMs where model + backprop exceed 48GB is already way faster on a 128GB MacBook than on NVidia unless you have one of those monstrous Hx00s with lots of RAM which most devs don't.

Re: Intel announces Arc B-series "Battlemage" discrete graphics with Linux support

#76

Post by bryanlarsen »

beAbU wrote: post #56
treprinum wrote: post #26

Why don't they just release a basic GPU with 128GB RAM and eat NVidia's local generative AI lunch? The networking effect of all devs porting their LLMs etc. to that card would instantly put them as a major CUDA threat. But beancounters running the company would never get such an idea...

They are probably held back by same reason thats preventing AMD and nVidia from doing it either.

The reason is AMD and Nvidia don't is that they don't want to cannibalize their high end AI market. Intel doesn't have a high end AI market to protect.

Re: Intel announces Arc B-series "Battlemage" discrete graphics with Linux support

#78

Post by FuriouslyAdrift »

treprinum wrote: post #26

Why don't they just release a basic GPU with 128GB RAM and eat NVidia's local generative AI lunch? The networking effect of all devs porting their LLMs etc. to that card would instantly put them as a major CUDA threat. But beancounters running the company would never get such an idea...

HBM3E memory is at least 3x the price of DDR5 (it requires 3x the wafer as DDR5) and capacity is sold out for all of 2025 already... that's the price and production bottleneck.

High speed, low latency server grade DDR5 is around $800-$1600 for 128GB. Triple that for $2400 - $4800 just for the memory. Still need the GPUs/APUs, card, VRMs, etc.

Even the nVidia H100 with "only" 94GB starts at $30k...

Re: Intel announces Arc B-series "Battlemage" discrete graphics with Linux support

#79

Post by m00x »

treprinum wrote: post #72
pjmlp wrote: post #68

Earlier quoted context omitted.

A fraction of CUDA capabilities.

Sufficient for LLMs and image/video gen.

FLUX.1 D generation is about a minute at 20 steps on a 4080, but takes 35 minutes on the CPU.
Post reply on HN