GPU model | TU106 | GP106 |
---|---|---|
GPU variant | TU106-200A-KA-A1 | — |
Architecture | Turing | Pascal |
Foundry | TSMC | TSMC |
Manufacturing process | 12 nm | 16 nm |
Number of transistors | 10,800 million | 4,400 million |
Die size | 445 mm² | 200 mm² |
Release date | Jan 7th, 2019 | Jun 10th, 2019 |
---|---|---|
Generation | GeForce 20 | Quadro |
Predecessor | GeForce 10 | — |
Production | Active | End-of-life |
Launch price | 349 USD | — |
Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Reviews | 85 in our database | — |
Successor | GeForce 30 | — |
Base clock | 1365 MHz | 1000 MHz |
---|---|---|
Boost clock | 1680 MHz | 1493 MHz |
Memory clock | 1750 MHz 14 Gbps effective | 1251 MHz 10 Gbps effective |
Memory size | 6 GB | 5 GB |
---|---|---|
Memory type | GDDR6 | GDDR5X |
Memory bus width | 192 bit | 160 bit |
Bandwidth | 336.0 GB/s | 200.2 GB/s |
Shading units | 1920 | 1280 |
---|---|---|
TMUs | 120 | 80 |
ROPs | 48 | 40 |
SM count | 30 | 10 |
Tensor cores | 240 | — |
RT cores | 30 | — |
Cache L1 | 64 KB (per SM) | 48 KB (per SM) |
Cache L2 | 3 MB | 1280 KB |
Pixel rate | 80.64 GPixel/s | 59.72 GPixel/s |
---|---|---|
Texture rate | 201.6 GTexel/s | 119.4 GTexel/s |
FP16 (half) performance | 12.90 TFLOPS (2:1) | 59.72 GFLOPS (1:64) |
FP32 (float) performance | 6.451 TFLOPS | 3.822 TFLOPS |
FP64 (double) performance | 201.6 GFLOPS (1:32) | 119.4 GFLOPS (1:32) |
Slot width | Dual-slot | Single-slot |
---|---|---|
Length | 229 mm 9 inches | 201 mm 7.9 inches |
Width | 113 mm 4.4 inches | 111 mm 4.4 inches |
Height | 35 mm 1.4 inches | — |
TDP | 160 W | 75 W |
Suggested PSU | 450 W | 250 W |
Outputs | 1x DVI1x HDMI2x DisplayPort1x USB Type-C | 4x DisplayPort |
Power connectors | 1x 8-pin | None |
Board number | PG160 SKU 42 | PG410 SKU 502 |
DirectX | 12 Ultimate (12_2) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 7.5 | 6.1 |
Shader model | 6.6 | 6.4 |