![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
A $1,199 5080 that goes +15% the 4090 if the 5090 is $2000 and has 24 GB of VRAM could still sell..... specially if it has over 1600 TFLOPS at FP8 / 3000 peak INT TOPS
I'd be shocked if it were more than 16 gigs. It'll sell out regardless, if it doesn't Nvidia is more than happy to sit on them until they sell. I don't expect them to lower prices until demand drops in the AI sector.
From the GTC 2024 keynote (at ~57 minutes), he shows the Inference performance increase of Blackwell over Hooper/Ada, it's 30x faster:
This is comparing FP4 operation per second with FP8 for the other gen, I think ( the other chips like the H200 are almost multi GPU) and vast part of the boost is from the complete ecosystem like you say.
H100 was 2.5-3 time the A100 on those metric for a recent reference, will have to see but if the B100 is a bigger chips and using more power... it shows that the stagnant node could mean a low raw upgrade and for Blackwell to be a very short lived stop gap and for the next gen to be in 2025 has rumoured.
I said 77% of their graphics business, not overall business. That accounts for $10 billion of that $13.5 billion in your chart.
There 2 main option, they can do like they did with Ampere Samsung 8nm for the gaming product, TSMC 7NM for the commercial product.
Same here, I just cannot get into PC gaming with top end video cards costing 2k, that used to be what the whole PC cost.
AMD eww..that card in 3 years wont have any compatible drivers.
Lol OK buddy
I actually argue if you're a "PC guy" getting the top-end parts, your experience is probably pretty terrible. Nobody who builds their own PC, who also spends tons of money on top-end parts is simply plopping them in, installing drivers and playing games happily. We tinker, test, hack, tweak, adjust... and the level of performance is never enough!
Then you want Nvidia
. Raytracing is huge in that game. Dlss helps it run with virtually no IQ loss too, unlike fsr. They did around 20ish years ago... or a bit less.
I'm fine without ray tracingThen you want Nvidia
They did around 20ish years ago... or a bit less.
It is possible that max setting more and more will just enable RT without being a different option (think Avatar or some unreal 5 game that does not even have pre-baked light to use) and it is a bit already strange to call RT off has playing with Max setting in a game like Cyberpunk
I'm a little confused.Are you saying that you're not really playing on max settings if you're not using ray tracing? I guess I could see that point of view, having never used it myself. I'm just not much of a gamer these days in terms of needing to be On the cutting edge of advanced visuals. I'm mostly a casual infrequent gamer , who likes older and more modest indy games and A and AA titles But I have been wanting to do a proper play through in cyberpunk since now they've finally finished it And polished it up. And it would be nice to play at high settings at 4K native resolution on my monitor with good responsive frames per second. But most of the other demanding triple A game titles these days really seem like trash. I think metro exodus is probably out the only other graphically demanding title I am interested in. If star Citizen ever gets finished I'd buy that too but I don't buy beta. Maybe the latest MS flight sim For some casual site seeing.
In some game the question does not really apply, Avatar for example if you set the game at ultra settings will have RT used, it is not a separate setting, that why at max setting the 4070ti beat card like a 7900xtx, it could be because it is an AMD sponsered title and they asked them to hide the RT in the menu and marketing or could be a sign for what to come:
Trying to play relatively recent title like Cyberpunk at 4K max setting is maybe not the way to describe your need, things move quite slow these days, Cyberpunk at maxuium setting at 4k getting 70fps say, is quite hard, even with RT off, you probably need a 4090 for that:
