Nom du GPU | GM107 | TU102 |
---|---|---|
GPU variant | N16P-GT | TU102-875-A1 |
Architecture | Maxwell | Turing |
Fonderie | TSMC | TSMC |
Taille du processus | 28 nm | 12 nm |
Transistors | 1,870 million | 18,600 million |
Taille des matrices | 148 mm² | 754 mm² |
Date de publication | Mar 13th, 2015 | — |
---|---|---|
Génération | GeForce 900M | — |
Production | End-of-life | — |
Bus interface | PCIe 3.0 x8 | — |
Prédécesseur | GeForce 800M | — |
Successeur | GeForce 10 Mobile | — |
Base clock | 993 MHz | 1395 MHz |
---|---|---|
Boost clock | 1124 MHz | 1770 MHz |
Mémoire clock | 900 MHz 1800 Mbps effective | 1750 MHz 14 Gbps effective |
Taille de la mémoire | 4 GB | 48 GB |
---|---|---|
Type de mémoire | DDR3 | GDDR6 |
Bus mémoire | 128 bit | 384 bit |
Bande passante | 28.80 GB/s | 672.0 GB/s |
Shading units | 640 | 4608 |
---|---|---|
TMUs | 40 | 288 |
ROPs | 16 | 96 |
Comptage SMM | 5 | — |
Cache L1 | 64 KB (per SMM) | 64 KB (per SM) |
Cache L2 | 2 MB | 6 MB |
Comptage SM | — | 72 |
Tensor cores | — | 576 |
RT cores | — | 72 |
Taux de pixel | 17.98 GPixel/s | 169.9 GPixel/s |
---|---|---|
Taux de texture | 44.96 GTexel/s | 509.8 GTexel/s |
FP32 (float) performance | 1,439 GFLOPS | 16.31 TFLOPS |
FP64 (double) performance | 44.96 GFLOPS (1:32) | 509.8 GFLOPS (1:32) |
FP16 (half) performance | — | 32.62 TFLOPS (2:1) |
Taille Slot | IGP | Dual-slot |
---|---|---|
TDP | 75 W | 260 W |
Sorties | No outputs | 4x DisplayPort1x USB Type-C |
Longueur | — | 267 mm 10.5 inches |
Taille | — | 111 mm 4.4 inches |
Suggestions pour le PSU | — | 600 W |
Connecteurs de puissance | — | 1x 6-pin + 1x 8-pin |
DirectX | 12 (11_0) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.1 | 1.2 |
CUDA | 5.0 | 7.5 |
Shader model | 5.1 | 6.6 |
Date de publication | — | Aug 13th, 2018 |
---|---|---|
Génération | — | Quadro |
Production | — | Active |
Prix de lancement | — | 9,999 USD |
Bus interface | — | PCIe 3.0 x16 |