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

281–290 of 722 posts

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

#281

Post by wtallis »

p1esk wrote: post #262
jsheard wrote: post #248

Earlier quoted context omitted.

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.

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.

These days, Intel merely specializes in bleeding processes. They spent far too many years believing the unrealistic promises from their fab division, and in the past few years they've been suffering the consequences as the problems are too big to be covered up by the cost savings of vertical integration.

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

#282

Post by jsheard »

hughesjj wrote: post #274
jsheard wrote: post #100

Earlier quoted context omitted.

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…

Man, I'm old enough to remember when 512 was a thing for consumer cards back when we had 4-8gb memory Sure that was only gddr5 and not gddr6 or lpddr5, but I would have bet we'd be up to 512bit again 10 years down the line.. (I mean supposedly hbm3 has done 1024-2048bit busses but that seems more research or super high end cards, not consumer)

Rumor is the 5090 will be bringing back the 512bit bus, for a whopping 1.5TB/sec bandwidth.

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

#283

Post by dgfitz »

ben_w wrote: post #188
PittleyDunkin wrote: post #156

Earlier quoted context omitted.

> ML is about hit another winter. I agree ML is about to hit (or has likely already hit) some serious constraints compared to breathless predictions of two years ago. I don't think there's anything equivalent to the AI winter on the horizon, though—LLMs even operated by people who have no clue how the underlying mechanism functions are still far more empowered than anything like the primitives of the 80s enabled.

What we had in the 80s was barely able to perform spell-check, free downloadable LLMs today are mind-blowing even in comparison to GPT-2.

I think the only good thing that came out of the 80s was the 90s. I’d leave that decade alone so we can forget about it.

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

#284

Post by latchkey »

yumraj wrote: post #82
m00x wrote: post #73

Earlier quoted context omitted.

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.

Yes, and inference is a huge market in itself and potentially larger than training (gut feeling haven’t run numbers) Keep NVIDIA for training and Intel/AMD/Cerebras/… for interference.

The funny thing about Cerebras is that it doesn't scale well at all for inference and if you talk to them in person, they are currently making all their money on training workloads.

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

#285

Post by dragontamer »

p1esk wrote: post #228
jsheard wrote: post #100

Earlier quoted context omitted.

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…

Apple could do it. Why can’t Intel?

Because LPDDR5x is soldered on RAM.

Everyone else wants configurable RAM that scales both down (to 16GB) and up (to 2TB), to cover smaller laptops and bigger servers.

GPUs with soldered on RAM has 500GB/sec bandwidths, far in excess of Apples chips. So the 8GB or 16GB offered by NVidia or AMD is just far superior at vid o game graphics (where textures are the priority)

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

#286

Post by dgfitz »

bayindirh wrote: post #264
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...

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…

How does any of this make money?

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

#287

Post by wtallis »

treprinum wrote: post #162
wtallis wrote: post #146

Earlier quoted context omitted.

You're asking for a GPU die at least as large as NVIDIA's TU102 that was $1k in 2018 when paired with only 11GB of RAM (because $1k couldn't get you a fully-enabled die to use 12GB of RAM). I think you're off by at least a factor of two in your cost estimates.

Intel has Xeon Phi which was a spin-off of their first attempt at GPU so they have a lot of tech in place they can reuse already. They don't need to go with GDDRx/HBMx designs that require large dies.

Even LPDDR requires a large die. It only takes things out of the realm of technologically impossible to merely economically impractical. A 512-bit bus is still very inconveniently large for a single die.

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

#288

Post by throwaway48476 »

bayindirh wrote: post #264
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...

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…

All of that is highly relevant for training but what the poster was asking for is a desktop inference card.

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

#289

Post by jsheard »

dragontamer wrote: post #285
p1esk wrote: post #228

Earlier quoted context omitted.

Apple could do it. Why can’t Intel?

Because LPDDR5x is soldered on RAM. Everyone else wants configurable RAM that scales both down (to 16GB) and up (to 2TB), to cover smaller laptops and bigger servers. GPUs with soldered on RAM has 500GB/sec bandwidths, far in excess of Apples chips. So the 8GB or 16GB offered by NVidia or AMD is just far superior at vid o game graphics (where textures are the priority)

> GPUs with soldered on RAM has 500GB/sec bandwidths, far in excess of Apples chips.

Apple is doing 800GB/sec on the M2 Ultra and should reach about 1TB/sec with the M4 Ultra, but that's still lagging behind GPUs. The 4090 was already at the 1TB/sec mark two years ago, the 5090 is supposedly aiming for 1.5TB/sec, and the H200 is doing 5TB/sec.

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

#290

Post by vel0city »

bryanlarsen wrote: post #19
jeffbee wrote: post #15

Earlier quoted context omitted.

What is the newest platform that lacks resizable BAR? It was standardized in 2006. Is 4060-level graphics performance useful in whatever old computer has that problem?

Sandy Bridge (2009) is still a very usable CPU with a modern GPU. In theory Sandy Bridge supported resizable BAR but in practice they didn't. I think the problem was BIOS's.

My old Ice Lake CPU was very much a bottleneck in lots of games in 2018 when I finally replaced it. It was a noticeable improvement across the board making the jump to a Zen+ CPU at the time, even with the same GPU.
Post reply on HN