Tuesday, April 23, 2024
HomeComputer HardwareNVIDIA GeForce RTX 4090 Benchmark Leak Reveals A 60 % Acquire Over...

NVIDIA GeForce RTX 4090 Benchmark Leak Reveals A 60 % Acquire Over RTX 3090 Ti


rtx 4090 slide

It is the primary of October, which suggests there’s solely 11 days left earlier than you should purchase your individual GeForce RTX 4090—assuming your pockets are deep sufficient, anyway. If you happen to’ve acquired the dosh however are on the fence concerning whether or not to dole it out for a dominant GPU, here is yet another knowledge level for you: a Geekbench 5 leak.

4090 geekbench 5 compute cuda

Immediately’s leak comes courtesy of the tearless retina of the Benchleaks bot on Twitter. Any individual with an ASRock X670E Taichi Carrera and a Ryzen 9 7950X (already fairly an enviable setup) slapped a GeForce RTX 4090 into the machine and ran Geekbench’s compute benchmark utilizing CUDA. They did it a minimum of twice, and the 2 outcomes have been very shut, with a most efficiency of 424,332 factors.

We do not usually check Geekbench Compute as a result of we’ve got different benchmarks for compute workloads. Nonetheless, a fast perusal by means of the Geekbench database finds us the end result beneath for a GeForce RTX 3090 Ti:

3090 ti geekbench 5 compute cuda

There are larger RTX 3090 Ti outcomes, however this one appears to be within the vary the place a lot of the CUDA exams fall. Evaluating this end result to the RTX 4090 provides us roughly a 60% uplift over the previous-generation half. That is a fully loopy uplift in efficiency, and excellent news for folk that use CUDA-based software program for manufacturing work.

Naturally, an enormous portion of the uplift comes from the truth that the RTX 4090 is a a lot wider GPU than the RTX 3090 Ti, with barely greater than half-again the CUDA core depend. Nonetheless, the RTX 4090 can also be operating at a a lot larger clock fee: 2.58 GHz, in comparison with the 1.92 GHz clock of the previous-generation half.

ampere vs ada sep 21 22
Provided that theoretical compute more-than doubled, possibly this end result is not so nice in any case?

Take into account that these are CUDA-based compute exams, and have little-to-no correlation with graphics efficiency. Lots of the improvements within the Ada Lovelace structure, resembling Shader Execution Reordering, will not profit the brand new chip in this sort of check. Likewise for the tensor and RT core enhancements.

How spectacular these outcomes really are additionally relies upon considerably on the ability consumption of the elements, which we’ll have to check for ourselves. Search for our evaluate of the RTX 4090 within the coming weeks!



RELATED ARTICLES

LEAVE A REPLY

Please enter your comment!
Please enter your name here

- Advertisment -
Google search engine

Most Popular

Recent Comments