GPU model | Navi 12 | TU106 |
---|---|---|
GPU variant | Navi 12 | N18E-G1-KD-A1 |
Architecture | RDNA 1.0 | Turing |
Foundry | TSMC | TSMC |
Manufacturing process | 7 nm | 12 nm |
Number of transistors | unknown | 10,800 million |
Die size | unknown | 445 mm² |
Release date | Dec 1st, 2020 | — |
---|---|---|
Generation | Radeon Pro | — |
Production | Active | — |
Interface | PCIe 4.0 x16 | — |
Base clock | 1000 MHz | 960 MHz |
---|---|---|
Boost clock | 1600 MHz | 1200 MHz |
Memory clock | 1000 MHz 2 Gbps effective | 1750 MHz 14 Gbps effective |
Memory size | 8 GB | 6 GB |
---|---|---|
Memory type | HBM2 | GDDR6 |
Memory bus width | 2048 bit | 192 bit |
Bandwidth | 512.0 GB/s | 336.0 GB/s |
Shading units | 2304 | 1920 |
---|---|---|
TMUs | 144 | 120 |
ROPs | 64 | 48 |
Compute units | 36 | — |
Cache L2 | 4 MB | 3 MB |
SM count | — | 30 |
Tensor cores | — | 240 |
RT cores | — | 30 |
Cache L1 | — | 64 KB (per SM) |
Pixel rate | 102.4 GPixel/s | 57.60 GPixel/s |
---|---|---|
Texture rate | 230.4 GTexel/s | 144.0 GTexel/s |
FP16 (half) performance | 14.75 TFLOPS (2:1) | 9.216 TFLOPS (2:1) |
FP32 (float) performance | 7.373 TFLOPS | 4.608 TFLOPS |
FP64 (double) performance | 460.8 GFLOPS (1:16) | 144.0 GFLOPS (1:32) |
Slot width | Dual-slot | MXM Module |
---|---|---|
Length | 267 mm 10.5 inches | — |
Width | 111 mm 4.4 inches | — |
TDP | 225 W | 115 W |
Suggested PSU | 550 W | — |
Outputs | No outputs | No outputs |
Power connectors | 1x 8-pin | None |
Board number | — | E4914 SKU 30 |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.2 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.5 | 6.6 |
CUDA | — | 7.5 |
Release date | — | Jan 29th, 2019 |
---|---|---|
Generation | — | GeForce 20 Mobile |
Production | — | Active |
Interface | — | PCIe 3.0 x16 |
Predecessor | — | GeForce 10 Mobile |
Successor | — | GeForce 30 Mobile |