It appears that the Quadro based on Ampere architecture has been pictured in full. A new photo published by Moore’s Law Is Dead reveals the design of the card.
The next-generation Quadro RTX pictured
The graphics card pictured below is the upcoming flagship model for the Quadro RTX series based on Ampere architecture. NVIDIA has not yet been confirmed to be developing an RTX A8000 variant with more memory.
The RTX (A)6000 allegedly features GA102 GPU with 10752 CUDA cores. Unlike the gaming GeForce RTX series, the Quadro would sport GDDR6 non-X memory. This is because GDDR6X memory is not a manufacturer in a capacity that the Quadro series requires. It may also be a reason why Quadro 6000 launches first, as GDDR6X memory with 1GB+ capacity per module is expected next year. The memory speed is 3 Gbps lower than RTX 3080’s (16 Gbps).
The card appears to also use two fans and a short PCB, however, it remains to be seen if the PCB is shaped similarly to the GeForce RTX board.
The Quadro RTX A6000 is expected to debut on October 5th, when NVIDIA will host its GTC Digital event.
NVIDIA Quadro RTX (A)6000 Ampere, Source: Moore’s Law Is Dead
Next Gen NVIDIA Quadro RTX specifications, Source: Moore’s Law is Dead
| NVIDIA Quadro RTX Series | |||||
|---|---|---|---|---|---|
| Quadro RTX ‘Ampere’ | GeForce RTX 3090 | Quadro RTX 8000 | Quadro RTX 6000 | Quadro RTX 5000 | |
| Architecture | GA102 Ampere | GA102 Ampere | TU102 Turing | TU102 Turing | TU104 Turing |
| CUDA Cores | 10752 | 10496 | 4608 | 4608 | 3072 |
| Memory | 48GB G6 384b | 24GB G6X 384b | 48GB G6 384b | 24GB G6 384b | 16GB G6 256b |
| Memory Speed | 16 Gbps | 19.5 Gbps | 14 Gbps | 13 Gbps | 14 Gbps |
| Bandwidth | 768 GB/s | 936 GB/s | 672 GB/s | 624 GB/s | 448 GB/s |
| MSRP | TBC | 1,499 USD | 10,000 USD | 6,300 USD | 2,300 USD |
Source: Moore’s Law Is Dead