According to rumors, NVIDIA has plans to soon unveil the workstation targeted Quadro RTX graphics cards based on the Ampere architecture as well. Twitter user Moore’s Law is Dead posted a screenshot from NVIDIA Control Panel which was later confirmed to be specifications of the upcoming Quadro RTX Ampere card and it features an insane amount of CUDA Cores.
It is rumored that the NVIDIA Quadro RTX (Ampere) graphics card will feature the GA102 GPU chip. The NVIDIA Quadro Ampere RTX card will feature 10752 CUDA cores and 48 GB of GDDR6 non-X memory. The 10752 CUDA Cores are arranged in 84 SMs. The Quadro RTX Ampere will also utilize the full-fat Ampere GA102 GPU which should be faster than the GeForce RTX 3090 since it packs 2% more cores.
Image Credits: Twitter @Moore’s Law is Dead
The leaked image that we are looking at is from the NVIDIA Control Panel which lists down several key specifications of the card itself. It is not stated what the card would be called but according to Twitter user @Kopite7kimi, the card could probably go by as Quadro RTX A6000.
The next-gen Quadro has 256 CUDA cores more than RTX 3090 graphics card and twice the memory. However, according to the Control panel image, the Quadro will feature the simple GDDR6 memory and not the newer GDDR6X memory. The data rate will be capped at 16 Gbps which is the highest speed of the GDDR6 memory.
The Quadro RTX Ampere will have a memory bus bandwidth of 384-bit and the memory bandwidth of 786 GB/s which is only 8 GB/s higher than the RTX 3080. Micron currently does not produce 16Gbit (2GB) GDDR6X but there are plans for a higher density GDDR6X module production next year. We may see that memory in the Quadro RTX A8000 in the future.
