Nom du GPU | G80 | GV100 |
---|---|---|
Architecture | Tesla | Volta |
Fonderie | TSMC | TSMC |
Taille du processus | 90 nm | 12 nm |
Transistors | 681 million | 21,100 million |
Taille des matrices | 484 mm² | 815 mm² |
Date de publication | Mar 5th, 2007 | Mar 27th, 2018 |
---|---|---|
Génération | Quadro FX | Tesla |
Production | End-of-life | End-of-life |
Prix de lancement | 2,999 USD | — |
Bus interface | PCIe 1.0 x16 | PCIe 3.0 x16 |
Critiques | 26 in our database | — |
GPU clock | 600 MHz | — |
---|---|---|
Shader clock | 1440 MHz | — |
Mémoire clock | 800 MHz 1600 Mbps effective | 876 MHz 1752 Mbps effective |
Base clock | — | 1290 MHz |
Boost clock | — | 1530 MHz |
Taille de la mémoire | 1536 MB | 32 GB |
---|---|---|
Type de mémoire | GDDR3 | HBM2 |
Bus mémoire | 384 bit | 4096 bit |
Bande passante | 76.80 GB/s | 897.0 GB/s |
Shading units | 128 | 5120 |
---|---|---|
TMUs | 32 | 320 |
ROPs | 24 | 128 |
Comptage SM | 16 | 80 |
Cache L2 | 96 KB | 6 MB |
Tensor cores | — | 640 |
Cache L1 | — | 128 KB (per SM) |
Taux de pixel | 14.40 GPixel/s | 195.8 GPixel/s |
---|---|---|
Taux de texture | 38.40 GTexel/s | 489.6 GTexel/s |
FP32 (float) performance | 368.6 GFLOPS | 15.67 TFLOPS |
FP16 (half) performance | — | 31.33 TFLOPS (2:1) |
FP64 (double) performance | — | 7.834 TFLOPS (1:2) |
Taille Slot | Dual-slot | Dual-slot |
---|---|---|
Longueur | 312 mm 12.3 inches | — |
Taille | 111 mm 4.4 inches | — |
TDP | 171 W | 250 W |
Suggestions pour le PSU | 450 W | 600 W |
Sorties | 2x DVI | No outputs |
Connecteurs de puissance | 2x 6-pin | None |
DirectX | 11.1 (10_0) | 12 (12_1) |
---|---|---|
OpenGL | 3.3 | 4.6 |
OpenCL | 1.1 (1.0) | 3.0 |
Vulkan | — | 1.2 |
CUDA | 1.0 | 7.0 |
Shader model | 4.0 | 6.6 |