Nvidia RTX A6000 Performance
Last October, Nvidia launched the RTX A6000 48GB professional graphics card and claimed that it would offer twice the performance of the previous-gen Quadro cards of the company. These claims are usual to hear but one will know so after testing the $4,650 RTX A6000 really for real-world benchmarks.
So Puget Systems took the responsibility to check the multiple professional-grade benchmarks on the card. And they found the results to be impressive ad fascinating. And to the cherry to the top, they found out that this was the fastest professional graphics card they have ever tested.
An Ampere with 48GB of RAM
GA102 GPU is powering the Nvidia’s RTX A6000 48GB graphics card along with 10,752 CUDA cores, 336 tensor cores, and 84 RT cores, and a 384-bit memory bus that combines with a huge 48GB slab of GDDR6 memory. We have observed a previous GeForce RTX 3090 consumer board based on the same graphics processor featuring a different GPU configuration with 10,496 CUDA cores, 328 tensor cores, 82 RT cores, and a 384-bit memory interface for its ‘mere’ 24GB of GDDR6X memory.
A slightly different approach is taken by Nvidia’s RTX A6000 with its GPU configuration as compared to GeForce RTX 3090. It makes use of a slower memory hence featuring a 768 GB/s of memory bandwidth which is 18% less than the consumer graphics card (936GB/s), making it clear that 3090 will be a better choice for gaming. But with a 48GB of DRAM onboard on RTX A6000 will perform better in memory-hungry professional workloads.
While Nvidia Studio drivers are included in all GeForce RTX graphics cards which helps in accelerating performance in some professional applications. But if we justify the rate of $4,650 for Nvidia GeForce RTX A6000, it would account for its drivers, 48GB of GDDR6, a slightly different GPU configuration, Quadro Sync support, enhanced reliability, a different display output configuration, and a blower-type cooler which is considerably higher than a $1,500 MSRP of the standard GeForce RTX 3090 Founders Edition. But a customized 3090 can be priced up to as high as $4,000.
Up to 92% Faster
The performance stats showed that RTX 3090 was close to Nvidia RTX A6000 48GB but as it was a workstation-grade graphics card, Puget decided to compare it to Nvidia’s Quadro RTX 6000 24GB, the predecessor of the new professional board priced at $4,000. The new generation is almost double the in memory is still priced at $5,000 ~ $5,500.
It is not necessary to have enormous onboard memory capacity on all professional workloads but it is handy to have which benefits GPU-accelerated rendering applications for larger projects. This is the reason why Nvidia RTX A6000 48GB outperformed its predecessor by 46.6% ~ 92.2% in all four rendering benchmarks ran by Puget. And not only than its predecessor, but it is also better than any other professional graphics card in GPU-accelerated rendering workloads. Tangible performance advantages are offered by Nvidia RTX A6000 48GB as compared to its predecessor. But these advantages don’t stop here as when this board is compared to graphics cards released several years ago.
And some modern-day applications like DaVinci Resolve 16.2.8 and Adobe Premiere Pro 14.8 which deal with video editing and color correction applications can be benefitted from these GPUs. As the board can benefit the CPU from getting bottled-neck as speeding the performance of GPU based applications.
Fastest Professional Graphics Card
Puget with no doubt declared Nvidia RTX A6000 48GB as the fastest professional graphics card they have tested to date. And with the increase in the GPU horsepower compared to the predecessor clearly shows gains in GPU rendering benchmarks as well as in DaVinci Resolve. Many such applications are under the benefits but real colors are only visible after a set amount of workloads it takes upon.