|

RTX 4090 Outperforms RTX 6000 In Benchmarks Despite Being Over Four Times Cheaper

The Nvidia Quatro RTX 6000 Ada is a powerhouse for content creation and Professional users, but you won’t see it on any lists for the best graphics cards for gaming due to its immense price tag.

RTX 6000 Ada render showing a black shroud, gold heatsink fins and a blower-style cooler.
Image Credit: Nvidia

The RTX Quatro 6000 features Nvidia’s Ada Lovelace architecture, using the company’s flagship AD102 silicon – which is the same as the GeForce RTX 4090.

For the RTX 6000 Nvidia enabled 142 streaming multiprocessors (SMs) out of the possible 144, this makes the RTX 6000 2 SMs shy of the full AD102 die. When a graphics card core is manufactured on silicon wafers, not every core of the processor is functional, this leads to manufactures deactivating some of the cores so they can still use the silicone – this may be the reason why the RTX 6000 has 2 less SM cores.

The RTX 6000 – which was released on December 3rd 2022 – arrived with 18,176 CUDA cores, 568 Tensor cores for workloads involving AI and 142 RT cores for raytracing tasks. With a boost clock of 2,505 MHz this GPU has an impressive FP32 (Single-precision floating-point format) performance that reaches peak bandwidth at 91.06 TFLOPs (1 TFLOP equals one trillion floating-point operations per second). The card also features 48GB of GDDR6 memory with a 384-bit memory bus, this give the RTX 6000 960GB/s (gigabytes per second) of memory bandwidth. The total power draw of the RTX 6000 is capped at 300W compared to the 450W of the RTX 4090 Founders Edition.

Previously, the RTX A6000 managed 38.71 TFLOPs FP32 performance and 768 GB/s of memory bandwidth. This means the RTX 6000 offers 135% higher performance and 25% more memory bandwidth when compared to the last-gen RTX A6000. This leaves the RTX 4090, with 82.58 TFLOPs FP32 performance and 1,008 GB/s memory bandwidth, as the closest rival to the RTX 6000.

 

RTX 6000 Ada Benchmarks

Graphics Card DaVinci Resolve Studio Adobe Premiere Pro Topaz AI Suite Unreal Engine V-Ray OctaneRender Redshift Blender
RTX 6000 Ada 2,975 1,498 471.2 74.0 5,399 1,146 87 12,728
GeForce RTX 4090 2,960 1,505 465.5 92.5 5,564 1,292 85 12,779
RTX A6000 2,648 1,379 445.2 46.3 2,710 627 159 5,848

 

The above benchmarks show that the RTX 4090 comes out on top compared to the RTX 6000 in six out of 8 benchmarks, delivering between 3% and 25% higher performance.

The RTX 6000 is a solid choice for professional workloads. But the price is enough to scare off buyers. This leaves the RTX 4090 as an attractive alternative for those with limited budgets. Despite the RTX 4090 outperforming the RTX 6000 in the above benchmarks, the RTX 6000 will still benefit from Nvidia’s professional drivers while the RTX 4090 won’t, resulting in a loss of stability. Which may end up having it outperform the RTX 4090 – this would make sense given the increase in specs across the board.  Note that none of the above benchmarks benefit from Nvidia’s professional drivers.

At the end of the day, for the less serious user – if you are okay with not benefiting from Nvidia’s professional drivers and losing 24Gb of VRAM – the RTX 4090 is a solid option.

Similar Posts