GPU model | GP104 | GA104 |
---|---|---|
GPU variant | N17E-Q5-A1 | — |
Architecture | Pascal | Ampere |
Foundry | TSMC | Samsung |
Manufacturing process | 16 nm | 8 nm |
Number of transistors | 7,200 million | 17,400 million |
Die size | 314 mm² | 392 mm² |
Release date | Jan 11th, 2017 | Apr 12th, 2021 |
---|---|---|
Generation | Quadro Mobile (Px000) | Quadro Mobile (Ax000) |
Production | End-of-life | Active |
Interface | MXM-B (3.0) | PCIe 4.0 x16 |
Base clock | 1278 MHz | 1140 MHz |
---|---|---|
Boost clock | 1582 MHz | 1680 MHz |
Memory clock | 1502 MHz 6 Gbps effective | 1500 MHz 12 Gbps effective |
Memory size | 16 GB | 8 GB |
---|---|---|
Memory type | GDDR5 | GDDR6 |
Memory bus width | 256 bit | 256 bit |
Bandwidth | 192.3 GB/s | 384.0 GB/s |
Shading units | 2048 | 5120 |
---|---|---|
TMUs | 128 | 160 |
ROPs | 64 | 80 |
SM count | 16 | 40 |
Cache L1 | 48 KB (per SM) | 128 KB (per SM) |
Cache L2 | 2 MB | 4 MB |
Tensor cores | — | 160 |
RT cores | — | 40 |
Pixel rate | 101.2 GPixel/s | 134.4 GPixel/s |
---|---|---|
Texture rate | 202.5 GTexel/s | 268.8 GTexel/s |
FP16 (half) performance | 101.2 GFLOPS (1:64) | 17.20 TFLOPS (1:1) |
FP32 (float) performance | 6.480 TFLOPS | 17.20 TFLOPS |
FP64 (double) performance | 202.5 GFLOPS (1:32) | 537.6 GFLOPS (1:32) |
Slot width | MXM Module | — |
---|---|---|
TDP | 100 W | 140 W |
Outputs | No outputs | No outputs |
Power connectors | None | None |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 6.1 | 8.6 |
Shader model | 6.4 | 6.6 |