GPU model | TU106B | TU104 |
---|---|---|
GPU variant | N18E-G1-KD-A1 | TU104-875-A1 |
Architecture | Turing | Turing |
Foundry | TSMC | TSMC |
Manufacturing process | 12 nm | 12 nm |
Number of transistors | 10,800 million | 13,600 million |
Die size | 445 mm² | 545 mm² |
Release date | Jan 29th, 2019 | — |
---|---|---|
Generation | GeForce 20 Mobile | — |
Production | Active | — |
Interface | PCIe 3.0 x16 | — |
Predecessor | GeForce 10 Mobile | — |
Successor | GeForce 30 Mobile | — |
Base clock | 1005 MHz | 1620 MHz |
---|---|---|
Boost clock | 1560 MHz | 1815 MHz |
Memory clock | 1375 MHz 11 Gbps effective | 1750 MHz 14 Gbps effective |
Memory size | 6 GB | 16 GB |
---|---|---|
Memory type | GDDR6 | GDDR6 |
Memory bus width | 192 bit | 256 bit |
Bandwidth | 264.0 GB/s | 448.0 GB/s |
Shading units | 1920 | 3072 |
---|---|---|
TMUs | 120 | 192 |
ROPs | 48 | 64 |
SM count | 30 | 48 |
Tensor cores | 240 | 384 |
RT cores | 30 | 48 |
Cache L1 | 64 KB (per SM) | 64 KB (per SM) |
Cache L2 | 3 MB | 4 MB |
Pixel rate | 74.88 GPixel/s | 116.2 GPixel/s |
---|---|---|
Texture rate | 187.2 GTexel/s | 348.5 GTexel/s |
FP16 (half) performance | 11.98 TFLOPS (2:1) | 22.30 TFLOPS (2:1) |
FP32 (float) performance | 5.990 TFLOPS | 11.15 TFLOPS |
FP64 (double) performance | 187.2 GFLOPS (1:32) | 348.5 GFLOPS (1:32) |
Slot width | MXM Module | Dual-slot |
---|---|---|
TDP | 65 W | 230 W |
Outputs | No outputs | 4x DisplayPort1x USB Type-C |
Power connectors | None | 1x 6-pin + 1x 8-pin |
Board number | E4914 SKU 31 | — |
Length | — | 267 mm 10.5 inches |
Width | — | 111 mm 4.4 inches |
Suggested PSU | — | 550 W |
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 | 7.5 | 7.5 |
Shader model | 6.6 | 6.6 |
Release date | — | Aug 13th, 2018 |
---|---|---|
Generation | — | Quadro |
Production | — | Active |
Launch price | — | 2,299 USD |
Interface | — | PCIe 3.0 x16 |