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

361–370 of 722 posts

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

#361

Post by foobiekr »

jmward01 wrote: post #346
bayindirh wrote: post #264

Earlier quoted context omitted.

Disclosure: HPC admin who works with NIVIDA cards here. Because, no. It's not as simple as that. NVIDIA has a complete ecosystem now. They have cards. They have cards of cards (platforms), which they produce, validate and sell. They have NVLink crossbars and switches which connects these cards on their card of cards with very high speeds and low latency. For inter-server communication they have libraries which coordi…

Most of the above infra is predicated on limiting RAM so that you need so much communication between cards. Bump the RAM up and you could do single card inference and all those connections become overhead that could have gone to more ram. For training there is an argument still, but even there the more RAM you have the less all that connectivity gains you. RAM has been used to sell cards and servers for a long time n…

Correct for inference - the main use of the interconnect is RDMA requests between GPUs to fit models that wouldn't otherwise fit.

Not really correct for training - training has a lot of all-to-all problems, so hierarchical reduction is useful but doesn't really solve the incast problem - Nvlink _bandwidth_ is less of an issue than perhaps the SHARP functions in the NVLink switch ASICs.

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

#362

Post by CoastalCoder »

jsheard wrote: post #100
treprinum wrote: post #62

Earlier quoted context omitted.

M3/M4 Max MacBooks with 128GB RAM are already way better than an A6000 for very large local LLMs. So even if the GPU is as slow as the one in M3/M4 Max (<3070), and using some basic RAM like LPDDR5x it would still be way faster than anything from NVidia.

The M4 Max needs an enormous 512bit memory bus to extract enough bandwidth out of those LPDDR5x chips, while the GPUs that Intel just launched are 192/160bit and even flagships rarely exceed 384bit. They can't just slap more memory on the board, they would need to dedicate significantly more silicon area to memory IO and drive up the cost of the part, assuming their architecture would even scale that wide without hit…

> The M4 Max needs an enormous 512bit memory bus to extract enough bandwidth out of those LPDDR5x chips

Does M4 Max have 64-byte cache lines?

If they can fetch or flush an entire cache line in a single memory-bus transaction, I wonder if that opens up any additional hardware / performance optimizations.

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

#363

Post by undersuit »

sangnoir wrote: post #236
undersuit wrote: post #175

Earlier quoted context omitted.

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.

The fine article states reviews are still embargoed, and sales start next week.

The mods have thankfully changed this to a Phoronix article instead of the Intel page and the title has been reworked to not include 'launch'.

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

#364

Post by p1esk »

duskwuff wrote: post #275
p1esk wrote: post #262

Earlier quoted context omitted.

Bleeding edge processes is what Intel specializes in. Unlike Apple, they don’t need TSMC. This should have been a huge advantage for Intel. Maybe that’s why Gelsinger got the boot.

> Bleeding edge processes is what Intel specializes in. Unlike Apple, they don’t need TSMC. Intel literally outsourced their Arrow Lake manufacturing to TSMC because they couldn't fabricate the parts themselves - their 20A (2nm) process node never reached a production-ready state, and was eventually cancelled about a month ago.

OK, so the question becomes: TSMC could do it. Why can’t Intel?

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

#365

Post by phkahler »

beAbU wrote: post #222
treprinum wrote: post #70

Earlier quoted context omitted.

NVidia and AMD make $$$ on datacenter GPUs so it makes sense they don't want to discount their own high-end. Intel has nothing there so they can happily go for commodization of AI hardware like what Meta did when releasing LLaMA to the wild.

Is nVidia or AmD offering 128gb cards in any configuration?

You can run an AMD APU with 128GB of shared RAM.

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

#366

Post by suprjami »

ggregoire wrote: post #189
jmward01 wrote: post #103

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?

> 12GB max is a non-starter for ML work now. Can you even do ML work with a GPU not compatible with CUDA? (genuine question) A quick search showed me the equivalence to CUDA in the Intel world is oneAPI, but in practice, are the major Python libraries used for ML compatible with oneAPI? (Was also gonna ask if oneAPI can run inside Docker but apparently it does [1]) [1] https://hub.docker.com/r/intel/oneapi

There is ROCm and Vulkan compute.

Vulkan is especially appealing because you don't need any special GPGPU drivers and it runs on any card which supports Vulkan.

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

#367

Post by talldayo »

pjmlp wrote: post #280
heraldgeezer wrote: post #225

Earlier quoted context omitted.

You can get a clean Fractal Design case with 0 lights or RGB. You are showing your ignorance. :) Also, RGB means RGB... just set the color to white for a more clean look? But reading the manual is not for you.

Nope, because you can't get those at the mall downtown, rather have to custom build, ordering from online shops.

Considering how HN professedly struggles to use non-reversible cables, it wouldn't surprise me if someone started to complain about getting skillchecked by a Molex or SATA.

And that CPU, with the little gold tab you gotta line up? That's a whopping three ways you could go wrong! Forget about it, might as well just phone Dell or Apple and pay the idiot tax.

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

#368

Post by vundercind »

mkl wrote: post #325
vundercind wrote: post #126

Earlier quoted context omitted.

It'd be pretty weird if it were called 2k. 1080p is in an absolute sense or as a relative "distance" to the next-lowest thousand closer to 2k pixels of width than 4k is to 4k (both are under, of course, but one's under by 80 pixels, one by 160). It's got a much better claim to the label 2k than 1440p does, and arguably a somewhat better claim to 2k than 4k has to 4k. [EDIT] I mean, of course, 1080p's also not typical…

You are misunderstanding. 1080p, 1440p, 2160p refer to the number of rows of pixels, and those terms come from broadcast television and computing (the p is progressive, vs i for interlaced). 4k, 2k refer to the number of columns of pixels, and those terms come from cinema and visual effects (and originally means 4096 and 2048 pixels wide). That means 1920×1080 is both 2k and 1080p, 2560×1440 is both 2.5k and 1440p, a…

> You are misunderstanding. 1080p, 1440p, 2160p refer to the number of rows of pixels

> (the p is progressive, vs i for interlaced)

> 4k, 2k refer to the number of columns of pixels

> 2560×1440 is both 2.5k and 1440p, and 3840×2160 is both 4k and 2160p.

These parts I did not misunderstand.

> and those terms come from cinema and visual effects (and originally means 4096 and 2048 pixels wide)

OK that part I didn't know, or at least had forgotten—which are effectively the same thing, either way.

> 1920×1080 is both 2k and 1080p

Wikipedia suggests that in this particular case (unlike with 4k) application of "2k" to resolutions other than the original cinema resolution (2048x1080) is unusual; moreover, I was responding to a commenter's usage of "2k" as synonymous with "1440p", which seemed especially odd to me.

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

#369

Post by JBiserkov »

jsheard wrote: post #248
p1esk wrote: post #228

Earlier quoted context omitted.

Apple could do it. Why can’t Intel?

Because Apple isn't playing the same game as everyone else. They have the money and clout to buy out TSMCs bleeding-edge processes and leave everyone else with the scraps, and their silicon is only sold in machines with extremely fat margins that can easily absorb the BOM cost of making huge chips on the most expensive processes money can buy.

> and their silicon is only sold in machines with extremely fat margins

Like the brand new Mini that cost 600 USD and went to 500 during Black week.

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

#370

Post by BonoboIO »

p1esk wrote: post #364
duskwuff wrote: post #275

Earlier quoted context omitted.

> Bleeding edge processes is what Intel specializes in. Unlike Apple, they don’t need TSMC. Intel literally outsourced their Arrow Lake manufacturing to TSMC because they couldn't fabricate the parts themselves - their 20A (2nm) process node never reached a production-ready state, and was eventually cancelled about a month ago.

OK, so the question becomes: TSMC could do it. Why can’t Intel?

They are trying … for like 10 years
Post reply on HN