GPU model | Navi 10 | GA106 |
---|---|---|
GPU variant | Navi 10 XTX | GA106-300-A1 |
Architecture | RDNA 1.0 | Ampere |
Foundry | TSMC | Samsung |
Manufacturing process | 7 nm | 8 nm |
Number of transistors | 10,300 million | 13,250 million |
Die size | 251 mm² | 276 mm² |
Release date | Jul 7th, 2019 | Jan 12th, 2021 |
---|---|---|
Generation | Navi | GeForce 30 |
Predecessor | Vega | GeForce 20 |
Production | Active | Active |
Launch price | 449 USD | 329 USD |
Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
Reviews | 121 in our database | 60 in our database |
Successor | Navi II | — |
gpu.details.availability | — | Feb 25th, 2021 |
Base clock | 1680 MHz | 1320 MHz |
---|---|---|
Game clock | 1830 MHz | — |
Boost clock | 1980 MHz | 1777 MHz |
Memory clock | 1750 MHz 14 Gbps effective | 1875 MHz 15 Gbps effective |
Memory size | 8 GB | 12 GB |
---|---|---|
Memory type | GDDR6 | GDDR6 |
Memory bus width | 256 bit | 192 bit |
Bandwidth | 448.0 GB/s | 360.0 GB/s |
Shading units | 2560 | 3584 |
---|---|---|
TMUs | 160 | 112 |
ROPs | 64 | 48 |
Compute units | 40 | — |
Cache L2 | 4 MB | 3 MB |
SM count | — | 28 |
Tensor cores | — | 112 |
RT cores | — | 28 |
Cache L1 | — | 128 KB (per SM) |
Pixel rate | 126.7 GPixel/s | 85.30 GPixel/s |
---|---|---|
Texture rate | 316.8 GTexel/s | 199.0 GTexel/s |
FP16 (half) performance | 20.28 TFLOPS (2:1) | 12.74 TFLOPS (1:1) |
FP32 (float) performance | 10.14 TFLOPS | 12.74 TFLOPS |
FP64 (double) performance | 633.6 GFLOPS (1:16) | 199.0 GFLOPS (1:64) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
Length | 272 mm 10.7 inches | 242 mm 9.5 inches |
Width | 111 mm 4.4 inches | 112 mm 4.4 inches |
Height | 40 mm 1.6 inches | — |
TDP | 225 W | 170 W |
Suggested PSU | 550 W | 450 W |
Outputs | 1x HDMI3x DisplayPort | 1x HDMI3x DisplayPort |
Power connectors | 1x 6-pin + 1x 8-pin | 1x 12-pin |
Part number | 102-D18206 | — |
Board number | 109-D18237 | PG190 SKU 60 |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.5 | 6.6 |
CUDA | — | 8.6 |