The first non-official benchmark of NVIDIA’s GeForce RTX 4090 graphics card has been leaked within the Geekbench 5 database forward of its launch.

NVIDIA GeForce RTX 4090 Showcases Sturdy 60% CUDA Efficiency Uplift Over RTX 3090 Ti In Geekbench 5 Benchmark

Whereas NVIDIA has been feeding avid gamers with DLSS 3 and Ray Tracing efficiency numbers, there’s been little info concerning the rasterization efficiency of their next-gen components. Now, we have got the primary non-DLSS and non-RT benchmark leak out throughout the Geekbench 5 CUDA database.

The NVIDIA GeForce RTX 4090, not realizing if this was a reference Founders Version or a customized AIB design, was examined on an AMD Ryzen 9 7950X CPU platform with 32 GB of DDR5-6000 DRAM. The graphics card provided a most reported frequency of round 2.58 GHz. When it comes to CUDA efficiency, the graphics card scored 417713 factors. For comparability, the RTX 3090 Ti scores a complete of 260,346, and the RTX 3090 scores a complete of 238,123 factors in the identical benchmark. It is a 60% and 75% enhance over these graphics playing cards, respectively.

NVIDIA GeForce RTX 4090 affords 60% larger efficiency than the RTX 3090 Ti, and 75% larger efficiency than RTX 3090 in Geekbench CUDA benchmark. (Picture Credit: Benchlife)

0

70722

141444

212166

282888

353610

424332

NVIDIA GeForce RTX 4090 ‘Official’ Specs – $1599 US Pricing

The NVIDIA GeForce RTX 4090 will use 128 SMs of the 144 SMs for a complete of 16,384 CUDA cores. The GPU will come filled with 96 MB of L2 cache and a complete of 384 ROPs which is just insane however contemplating that the RTX 4090 is a cut-down design, it might characteristic barely decrease L2 and ROP counts. The clock speeds should not confirmed but however contemplating that the TSMC 4N course of is getting used. The clock speeds are rated at as much as 2.6 GHz and NVIDIA is claiming over 3 GHz speeds with overclocking which you’ll learn extra about right here.

As for reminiscence specs, the GeForce RTX 4090 will characteristic 24 GB GDDR6X capacities that will probably be clocked at 21 Gbps speeds throughout a 384-bit bus interface. This may present as much as 1 TB/s of bandwidth. This is identical bandwidth as the present RTX 3090 Ti graphics card and so far as the facility consumption is worried, the TBP is rated at 450W. The card will probably be powered by a single 16-pin connector which delivers as much as 600W of energy. Customized fashions will probably be providing larger TBP targets.

The NVIDIA GeForce RTX 4090 GPU formally launches on October twelfth, when NVIDIA and customized card companions’ designs develop into out there to the general public. Proshop shouldn’t be taking preorders, particularly since many GPUs can be found.

Information Supply: Benchleaks

Merchandise talked about on this put up

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.