GPU model | TU102 | GM107 |
---|---|---|
GPU variant | TU102-875-A1 | N16P-Q3-A2 |
Architecture | Turing | Maxwell |
Foundry | TSMC | TSMC |
Manufacturing process | 12 nm | 28 nm |
Number of transistors | 18,600 million | 1,870 million |
Die size | 754 mm² | 148 mm² |
Release date | Unknown | — |
---|---|---|
Generation | GRID | — |
Production | Active | — |
Interface | PCIe 3.0 x16 | — |
Base clock | 1065 MHz | 1029 MHz |
---|---|---|
Boost clock | 1395 MHz | 1098 MHz |
Memory clock | 1750 MHz 14 Gbps effective | 1253 MHz 5 Gbps effective |
Memory size | 8 GB | 4 GB |
---|---|---|
Memory type | GDDR6 | GDDR5 |
Memory bus width | 384 bit | 128 bit |
Bandwidth | 672.0 GB/s | 80.19 GB/s |
Shading units | 4608 | 640 |
---|---|---|
TMUs | 288 | 40 |
ROPs | 96 | 16 |
SM count | 72 | — |
Tensor cores | 576 | — |
RT cores | 72 | — |
Cache L1 | 64 KB (per SM) | 64 KB (per SMM) |
Cache L2 | 6 MB | 2 MB |
SMM count | — | 5 |
Pixel rate | 133.9 GPixel/s | 17.57 GPixel/s |
---|---|---|
Texture rate | 401.8 GTexel/s | 43.92 GTexel/s |
FP16 (half) performance | 25.71 TFLOPS (2:1) | — |
FP32 (float) performance | 12.86 TFLOPS | 1,405 GFLOPS |
FP64 (double) performance | 401.8 GFLOPS (1:32) | 43.92 GFLOPS (1:32) |
Slot width | Dual-slot | MXM Module |
---|---|---|
Length | 267 mm 10.5 inches | — |
TDP | 260 W | 55 W |
Suggested PSU | 600 W | — |
Outputs | No outputs | No outputs |
Power connectors | 1x 6-pin + 1x 8-pin | None |
Board number | PG150 SKU 220 | — |
DirectX | 12 Ultimate (12_2) | 12 (11_0) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.1 |
CUDA | 7.5 | 5.0 |
Shader model | 6.6 | 5.1 |
Release date | — | Dec 3rd, 2015 |
---|---|---|
Generation | — | Quadro Mobile (Mx000M) |
Production | — | End-of-life |
Interface | — | MXM-A (3.0) |
Reviews | — | 26 in our database |