While we're unsure how anyone can measure 3Dmark on an unreleased card ( NVIDIA does not deliver full drivers even to board partners), some numbers are up.... Assumed NVIDIA GeForce RTX 4080 16GB 3DMark performance benchmarks leaked
@tty8k until their drivers arent all or nothing for install, and they have a shadow-play equivalent, i will stay on Nv. i dont buy model numbers, but what i have the funds for so any price difference is not really relevant to me, much more if i can get a LC version/block for it.
Guess I will be waiting for the 4090 ti or the 5000 series. Even without driver support, I don't expect much more than what was posted here. Is that performance really worth $2300 AUD.... nope not a bloody chance? Their NDAs that are not allowing reviewers to review and show us before the release just go to show Nvidia doesn't want true numbers released before the cards.
It's all 中国人 to me. I will just wait till the real results come out. No sense in going all crazy over inaccurate results. I do however agree with my European friends that the pricing is just outrageous for them. Price to performance has also lost all meaning for them. Waiting for results-------->
nVidia sure loves to manipulate numbers... They should get with intel in the same bed and manipulate numbers together, so much for 2-4x over 3090ti...
So, I don't have the numbers here for my RTX 3090 for FireStrike Ultra (have not tested it for my new setup). For Timespy Extreme, I had a 9533 graphics Score, so that would mean that this card scores 1.47 times higher than my RTX 3090 (EVGA RTX 3090 XC3 Gaming Ultra). I also have the numbers for the Xess test still in my head (as I ran it a couple a days ago). I got 56.x for the normal test and 75.x for Xess enabled one. This means that this card is about 1.33 to 1.35 times faster than my RTX 3090
Even our expectations got contaminated with NVIDIA's greed. In a prefect world RTX4090 should have cost 650$