GPU model | GA106 | GA104 |
---|---|---|
GPU variant | GA106-140-A1 | GA104-775-A1 |
Architecture | Ampere | Ampere |
Foundry | Samsung | Samsung |
Manufacturing process | 8 nm | 8 nm |
Number of transistors | 13,250 million | 17,400 million |
Die size | 276 mm² | 392 mm² |
Release date | 2022 | — |
---|---|---|
gpu.details.availability | 2022 | — |
Generation | GeForce 30 | — |
Predecessor | GeForce 20 | — |
Production | Unreleased | — |
Interface | PCIe 4.0 x16 | — |
Base clock | 1545 MHz | 1110 MHz |
---|---|---|
Boost clock | 1740 MHz | 1545 MHz |
Memory clock | 1750 MHz 14 Gbps effective | 1750 MHz 14 Gbps effective |
Memory size | 4 GB | 8 GB |
---|---|---|
Memory type | GDDR6 | GDDR6 |
Memory bus width | 128 bit | 256 bit |
Bandwidth | 224.0 GB/s | 448.0 GB/s |
Shading units | 2304 | 6144 |
---|---|---|
TMUs | 72 | 192 |
ROPs | 40 | 96 |
SM count | 18 | 48 |
Tensor cores | 72 | 192 |
RT cores | 18 | 48 |
Cache L1 | 128 KB (per SM) | 128 KB (per SM) |
Cache L2 | 2 MB | 4 MB |
Pixel rate | 69.60 GPixel/s | 148.3 GPixel/s |
---|---|---|
Texture rate | 125.3 GTexel/s | 296.6 GTexel/s |
FP16 (half) performance | 8.018 TFLOPS (1:1) | 18.98 TFLOPS (1:1) |
FP32 (float) performance | 8.018 TFLOPS | 18.98 TFLOPS |
FP64 (double) performance | 125.3 GFLOPS (1:64) | 296.6 GFLOPS (1:64) |
Slot width | Dual-slot | — |
---|---|---|
Length | 242 mm 9.5 inches | — |
Width | 112 mm 4.4 inches | — |
TDP | 90 W | 115 W |
Suggested PSU | 250 W | — |
Outputs | 1x HDMI3x DisplayPort | No outputs |
Power connectors | 1x 6-pin | None |
DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 8.6 | 8.6 |
Shader model | 6.6 | 6.6 |
Release date | — | Jan 12th, 2021 |
---|---|---|
gpu.details.availability | — | Feb 2nd, 2021 |
Generation | — | GeForce 30 Mobile |
Predecessor | — | GeForce 20 Mobile |
Production | — | Active |
Interface | — | PCIe 4.0 x16 |