Hacker News Reader

Comment by hospitalJail | Hacker News Reader

Solid work. I'm holding off for about 2 years since that kind of development time will give Nvidia time to make some consumer friendly VRAM giant. (I'm assuming the 4090 is still consumer friendly)

As long as chatGPT still lets people ask medical/healthcare questions, it doesnt seem to urgent. I'm dreading when the AMA lobbies to ban it unless under Physician supervision.

View on HN

Nvidia has no motivation to make a consumer card with lots of VRAM, that's basically the only (relevant) separator between the GeForce family and the Quadro lineup.

There are restrictions on NVENC streams with consumer cards, but that has been a solved problem for a while [0].

If they were to make a consumer card with more VRAM, it would immediately undercut their own Quadro/Tesla lineup, which cost substantially more. I don't see a reason for them to do it.

0: https://github.com/keylase/nvidia-patch

>Nvidia has no motivation to make a consumer card with lots of VRAM

Money

Presumably they're making a higher margin on the professional cards.

If NVidia doesn't, somebody else will. Maybe AMD, maybe Apple, maybe Intel.

Apple's neural engine is really interesting as a GPU alternative. It's still just a bunch of tensor cores, but the unified memory architecture allows it to use the full system RAM without a major performance hit. I think we will see more of this in the future.

Reply on news.ycombinator.com