Any discussion of an intel entry to discrete graphics cards needs to at least _mention_ intel's repeated history of abandoning discrete graphics cards.
Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM
81–90 of 131 posts
Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM
#82Earlier quoted context omitted.
Certainly an interesting choice. Dramatically worse performance but dramatically larger only time will tell how it actually goes
With 160GB, surely they can add more channels to compensate?
Is there anything preventing them from using heterogeneous memory chips, like 1/4 GDDR7 and 3/4 LPDDR? It could enable new MEO-like architectures with finer-grained performance tuning for long contexts.
Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM
#83Any discussion of an intel entry to discrete graphics cards needs to at least _mention_ intel's repeated history of abandoning discrete graphics cards.
You’re saying it’s like the Google of graphics cards?
Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM
#84Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM
#85Honestly, Intel just has to build a GPU with insane amount of VRAM. It doesn't even have to be the fastest to compete... just a ton of vram for dirt cheap
Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM
#86Honestly, Intel just has to build a GPU with insane amount of VRAM. It doesn't even have to be the fastest to compete... just a ton of vram for dirt cheap
It’s LPDDR5x
It’s gonna be slowwww
It’s gonna be what, 273GB/sec vram bandwidth at most? Might as well as buy an AND 395+ 128GB right now for the same inference performance and slightly less VRAM.
Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM
#87Honestly, Intel just has to build a GPU with insane amount of VRAM. It doesn't even have to be the fastest to compete... just a ton of vram for dirt cheap
Isn't this exactly that?
Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM
#88Honestly, Intel just has to build a GPU with insane amount of VRAM. It doesn't even have to be the fastest to compete... just a ton of vram for dirt cheap
It’s LPDDR5x It’s gonna be slowwww It’s gonna be what, 273GB/sec vram bandwidth at most? Might as well as buy an AND 395+ 128GB right now for the same inference performance and slightly less VRAM.
How can you tell without knowing the bus width?
Re: Intel Announces Inference-Optimized Xe3P Graphics Card with 160GB VRAM
#89Earlier quoted context omitted.
How does LPDDR5 (This Xe3P) compare with GDDR7 (Nvidia's flagships) when it comes to inference performance? Local inference is an interesting proposition because today in real life, the NV H300 and AMD MI-300 clusters are operated by OpenAI and Anthropic in batching mode, which slows users down as they're forced to wait for enough similar sized queries to arrive. For local inference, no waiting is required - so you c…
I think the better comparison, for consumers, is how fast is LPDDR5 compared to the normal DDR5 attached to your CPU? Or, to be more specific, what is the speed when your GPU is out of RAM and it's reading from main memory over the PCI-E bus? PCI-E 5.0: 64GB/s @ 16x or 32GB/s @ 8x 2x 48GB (96GB) of DDR5 in an AM5 rig: ~50GB/s Versus the ~300GB/s+ possible with a card like this, it's a lot faster for large 'dense' mod…
This doesn’t matter at all, if the resulting tokens/sec is still too slow for interactive use.