NVIDIA GeForce RTX 4090 CUDA benchmark shows 60% improvement over RTX 3090 Ti

NVIDIA GeForce RTX 4090 CUDA benchmark shows 60% improvement over RTX 3090 Ti

The first unofficial benchmark of NVIDIA’s GeForce RTX 4090 graphics card leaked to the Geekbench 5 database ahead of its launch.

NVIDIA GeForce RTX 4090 shows strong 60% CUDA performance improvement over RTX 3090 Ti in Geekbench 5 Benchmark

While NVIDIA has provided gamers with DLSS 3 and Ray Tracing performance figures, there has been little information regarding the rasterization performance of their next-gen parts. Now we have the first non-DLSS, non-RT benchmark leak in the Geekbench 5 CUDA database.

The NVIDIA GeForce RTX 4090, unsure if it was a reference Founders Edition or a custom AIB design, was tested on an AMD Ryzen 9 7950X CPU platform with 32GB of DDR5 DRAM- 6000. The graphics card offered a reported maximum frequency of around 2.58 GHz. In terms of CUDA performance, the graphics card scored 417,713 points. For comparison, the RTX 3090 Ti scores a total of 260,346 points and the RTX 3090 a total of 238,123 points in the same benchmark. This is a 60% and 75% increase over those graphics cards, respectively.

NVIDIA GeForce RTX 4090 delivers 60% better performance than RTX 3090 Ti and 75% better performance than RTX 3090 in Geekbench CUDA benchmark. (Image credits: Benchlife)

0

70722

141444

212166

282888

353610

424332

NVIDIA GeForce RTX 4090 “Official” Specs – Priced at US$1599

The NVIDIA GeForce RTX 4090 will use 128 SM out of the 144 SM for a total of 16,384 CUDA cores. The GPU will come with 96MB of L2 cache and a total of 384 ROPs, which is just insane, but given that the RTX 4090 is a scaled-down design, it may feature slightly lower L2 and ROP numbers. Clock speeds are not yet confirmed, but given that the TSMC 4N process is used. Clock speeds are rated up to 2.6GHz and NVIDIA claims speeds of over 3GHz with overclocking which you can read more about here.

As far as memory specs go, the GeForce RTX 4090 will feature 24GB GDDR6X capacities that will be clocked at 21Gbps speeds on a 384-bit bus interface. This will provide up to 1 TB/s of bandwidth. This is the same bandwidth as the existing RTX 3090 Ti graphics card and when it comes to power consumption, the TBP is rated at 450W. The card will be powered by a single 16-pin connector that delivers up to at 600W of power. Custom models will offer higher TBP targets.

Official launch of the NVIDIA GeForce RTX 4090 GPU October 12, when designs from NVIDIA and custom board partners will be publicly available. Proshop does not take pre-orders, especially since many GPUs are available.

News source: Bench leaks

Products mentioned in this article

Share This

Leave a Comment

Your email address will not be published.