Ah, I can't wait for it to cost 140% more too then.

I mean at the top end we don't expect this to scale linearly, do we?

Nvidia's pricing is based mostly around "how much can we get away with charging"

So they're exactly like every other company?

Yes, but they don’t really have strong competition to put pressure on price.

Ah the Apple model. It's worth trillions. Literally.

I wonder how much more power will it consume... Time to start updating those fuses at this rate...

Big leaps in performance are usually tied to process node changes, like 5nm to 3nm, where the change also results in lower power consumption.

Unrelated to the article. How would I go about blocking these autoplaying videos using an adblocker? I can’t stand these websites that autoplay completely unrelated videos, not that any autoplaying video is okay.

I'm on Safari on MacOS with AdGuard and StopTheMadness. Nothing seems to have autoplayed.

There's a setting / filter in uBlock Origin to block auto playing videos.

I am due an upgrade around 5000 time...but I care less about speed and more about vram (gamer+llm enthusiast).

3090 was 24gb gddrx6 and 4090 was...exactly the same. Utter BS.

Happy to play nvidia's pricing game, but they've got to come to the party with substantially more vram

They'll never give you more VRAM at gamer prices because that would cut into workstation card sales.

I don't see them doing 3 generations of 24gb.

...we'll see I guess

I love upgrading my pc to ridiculous specs, playing in 8k, opening stupid amounts of apps and not worrying.

I was pondering about the 5090 earlier today. For the first time, I'm not sure what I could do with it. No flagship game, everything already runs fine in 8k. No mining.

I'm sure it'll be great for some people who want to upgrade, but I fail to see what would actually leverage it, better than a 4090.

Local LLMs and other AI models. Even on my 4090 I run into a lot of limitations, to the point that I've been debating whether I buy a second one or wait on the 5090 to release and move my 4090 to secondary. I've even considered going with moire then 2 cards just to get some larger models running(still cheaper then the high memory server/workstation cards).

This comment is assuming the 5090 will come with more VRAM right? I haven't heard if that will be the case for consumer cards, it would be obviously nice if nvidia did, but it doesn't seem like it is in their best interest to do so versus just offering it on their high VRAM on their datacenter/workstation cards, unless e.g. Intel or AMD starts putting pressure on them at this front.

A 4090 runs a lot of these ML models more than fast enough, the problem is just VRAM right now. I think a lot of local LLM people think even a 3090 is plenty fast enough.