Live data from Hacker News

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

phoronix.com

171–180 of 722 posts

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

#172

12GB memory -.- I feel like _anyone_ who can pump out GPU's with 24GB+ of memory that are capable to use for py-stuff would benefit greatly. Even if it's not as performant as the NVIDIA options - just to be able to get the models to run, at whatever speed. They would fly off the shelves.

I don't know a single person in real life that has any desire to run local LLMs. Even amongst my colleagues and tech friends, not very many use LLMs period. It's still very niche outside AI enthusiasts. GPT is better than anything I can run locally anyway. It's not as popular as you think it is.

The only consumer demand for local AI models is for generating pornography

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

#173

12GB memory -.- I feel like _anyone_ who can pump out GPU's with 24GB+ of memory that are capable to use for py-stuff would benefit greatly. Even if it's not as performant as the NVIDIA options - just to be able to get the models to run, at whatever speed. They would fly off the shelves.

You can get that on mac mini and it will probably cost you less than equivalent PC setup. Should also perform better than low end Intel GPU and be better supported. Will use less power as well.

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

#174

Earlier quoted context omitted.

GDDR isnt like the ram that connects to cpu, it's much more difficult and expensive to add more. You can get up to 48GB with some expensive stacked gddr, but if you wanted to add more stacks you'd need to solve some serious signal timing related headaches that most users wouldn't benefit from. I think the high memory local inference stuff is going to come from "AI enabled" cpus that share the memory in your computer.…

They can use LPDDR5x, it would still massively accelerate inference of large local LLMs that need more than 48GB RAM. Any tensor swapping between CPU RAM and GPU RAM kills the performance.

I think we don't really disagree, I just think that this shape isn't really a gpu its just a cpu because it isn't very good for graphics at that point.

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

#175

Too late, and it has a bad rep. This effort from Intel to sell discrete GPUs is just inertia from old aspirations, won't really help noticeably to save it, as there is not much money in it. Most probably the whole Intel ARC effort will be mothballed, and probably many more will.

No reviews and when you click on the reseller links in the press announcement they're still selling A750s with no B-Series in sight. Strong paper launch.

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

#176
I wanted Intel to do well so I purchased an ARC card. The problem is not the hardware. For some games, it worked fine, but in others, it kept crashing left and right. After updates to drivers, crashing was reduced, but it still happened. Driver software is not easy to develop thoroughly. Even AMD had problems when compared to Nvidia when AMD really started to enter the GPU game after buying ATI. AMD has long since solved their driver woes, but years after ARC's launch, Intel still has not.

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

#178
post #109

12GB memory -.- I feel like _anyone_ who can pump out GPU's with 24GB+ of memory that are capable to use for py-stuff would benefit greatly. Even if it's not as performant as the NVIDIA options - just to be able to get the models to run, at whatever speed. They would fly off the shelves.

100% - this could be Intel's ticket to capture the hearts of developers and then everything else that flows downstream. They have nothing to lose here -- just do it Intel!

They could lose a lot of money?

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

#179
post #157
post #151

Earlier quoted context omitted.

The ML dev world isn’t a consumer mass market like PC gaming is.

Launching a new SKU for $500-1000 with 48gb of RAM seems like a profitable idea. The GPU isn't top-of-the-line, but the RAM would be unmatched for running a lot of models locally.

give me 48gb with reasonable power consumption so I can dev locally and I will buy it in a heartbeat. Anyone that is fine-tuning would want a setup like that to test things before pushing to real GPUs. And in reality if you can fine-tune on a card like that in two days instead of a few hours it would totally be worth it.

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

#180

12GB max is a non-starter for ML work now. Why not come out with a reasonably priced 24gb card even if it isn't the fastest and target it at the ML dev world? Am I missing something here?

These are $200 low end cards, the B5X0 cards. Presumably they have B7X0 and perhaps even B9X0 cards in the pipeline as well.

There has been no hint or evidence (beyond hope) Intel will add a 900 class this generation.

B770 was rumoured to match the 16 GB of the A770 (and to be the top end offering for Battlemage) but it is said to not have even been taped out yet with rumour it may end up having been cancelled completely.

I.e. don't hold your breath for anything consumer from Intel this generation better for AI than tha A770 you could have bought 2 years ago. Even if something slightly better is coming at all there is no hint it will be soon.

Post reply on HN