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

Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM

phoronix.com

Post reply on HN
Print view View original

101–110 of 131 posts

Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM

#101

Post by Aurornis »

mft_ wrote: post #16

I have no idea of the likely price, but (IMO) this is the sort of disruption that Intel needs to aim at if it's going to make some sort of dent in this market. If they could release this for around the price of a 5090, it would be very interesting.

> If they could release this for around the price of a 5090

This is not targeted at consumers. It’s competing with nVidia’s high RAM workstation cards. Think $10K price range, not $1-2K.

The 160GB of LPDDR5X chips alone is expensive enough that they couldn’t release this at the $2K price point unless they felt like giving it away (which they don’t)

Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM

#102

Post by hnuser123456 »

cwillu wrote: post #81

Any discussion of an intel entry to discrete graphics cards needs to at least _mention_ intel's repeated history of abandoning discrete graphics cards.

At least larrabee's cancellation resulted in the Offset engine going to the Firefall (2014) devs, which was a really great F2P MMO game for a while.

Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM

#103

Post by Aurornis »

wmf wrote: post #45
baq wrote: post #23

Earlier quoted context omitted.

With this much ram don’t expect anything remotely affordable by civilians.

160 GB LPDDR5 is ~$1,200 retail so the card could be sold for $2,000. The price will depend on how desperate Intel is. Intel probably can't copy Nvidia's pricing.

> 160 GB LPDDR5 is ~$1,200 retail so the card could be sold for $2,000.

Prices are set by what the market will bear, not the lowest possible price where they could break even on the BOM and manufacturing costs.

The high cost of the LPDDR5X should be a clue that this is going to be in the $10K range, not the $2K range.

Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM

#104

Post by vardump »

Aurornis wrote: post #100
tonetegeatinst wrote: post #96

What price is this sitting at? Because if its software support is decent then Intel might have just managed to break into the hardware for AI on the edge. Examples like self hosted LLM finetuning and RAG on a old dell or HP server with these type of cards on them.

> Examples like self hosted LLM finetuning and RAG on an old dell or HP server with these type of cards on them. This won’t be in the price range of an old Dell server or a fun impulse buy for a hobbyist. 160GB of raw LPDDR5X chips alone is not cheap. This is a server/workstation grade card and the price is going where the market will allow. Consider that an nVidia card with almost half the RAM is going to cost $8K o…

That nVidia card is going to have 5x the memory bandwidth. LPDDR5X is going to be rather low bandwidth.

(My guess is Intel's card is only going to have about 400 GB/s bandwidth.)

Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM

#105

Post by gary_0 »

Sesse__ wrote: post #75
gary_0 wrote: post #59

Earlier quoted context omitted.

Yes. The earliest consumer PC 3D graphics cards just rasterized pre-transformed triangles and that's it; the CPU had to do pretty much all the math (but drawing the pixels was considered the hard part). Later, "Hardware Transform and Lighting (T&L)" was introduced circa 2000 by cards like the GeForce 256.

And even then, you couldn't really get any sort of serious matmul out of it; they were per-vertex, not per-pixel. Per-pixel matmul (which is what you really need for anything resembling GPGPU) came with Shader Model 2.0, circa 2002; Radeon 9700, the GeForce FX series and the likes. CUDA didn't exist (nor really any other form of compute shaders), but you could wrangle it with pixel shaders, and some of us did.

Oh man, I forgot about doing vector math using OpenGL textures as "hardware acceleration". And it would be many more years before it was reasonable to require a GPU with programmable shaders; having to support fixed-function was a fact of life for most of the 2000's.

Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM

#106

Post by kobalsky »

cwillu wrote: post #81

Any discussion of an intel entry to discrete graphics cards needs to at least _mention_ intel's repeated history of abandoning discrete graphics cards.

the GPU market is not what it used to be, it's not some checkbox some executive needs to check to say "we are doing something".

the chips are so valuable now NVIDIA will end up owning a chunk of every major tech company, everyone is throwing cash and shares at them as fast as they can.

Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM

#107

Post by toast0 »

knowitnone3 wrote: post #3

Any business people here that can explain why companies announce products a year before their release? I can understand getting consumers excited but it also tells competitors what you are doing giving them time to make changes of their own. What's the advantage here?

Adding on to everyone else. It might help with sales for those with long procurement cycles.

If you're planning a supercomputer to be built in 2027, you want to look at what's on the roadmap.

Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM

#109

Post by incomingpain »

A year out, in that time nvidia and amd; not to mention huawei and others are going to hit the market as well. Intel are quite behind.

To me, the price point is what matters. It's going to be slow with ddr5. The 5090 today is much faster. But sure big ram.

RTX pro 6000 with 96gb of ram will be much faster.

So I'm thinking price point is below the 6000, above the 5090.

Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM

#110

Post by woleium »

schmorptron wrote: post #68
Neywiny wrote: post #20

Earlier quoted context omitted.

Certainly an interesting choice. Dramatically worse performance but dramatically larger only time will tell how it actually goes

Rumor has it (according to MLID, so no one knows whether it's accurate) that AMD is also looking to use regular LPDDR memory for some of it's lower end next gen GPUs to not have to contend with nvidia over limited and cartelled GDDR7 supply. Maybe they're going to increase parallel bandwidth to compensate it? Or have wholly different tricks up their sleeve.

probably just a lot more of it, to capture that consumer ai market
Post reply on HN