Nom du GPU | Vesuvius | GP102 |
---|---|---|
GPU variant | Vesuvius XT (215-0852022) | — |
Architecture | GCN 2.0 | Pascal |
Fonderie | TSMC | TSMC |
Taille du processus | 28 nm | 16 nm |
Transistors | 6,200 million | 11,800 million |
Taille des matrices | 438 mm² | 471 mm² |
Date de publication | Apr 29th, 2014 | Never Released |
---|---|---|
Génération | Volcanic Islands | GeForce 10 |
Production | End-of-life | End-of-life |
Prix de lancement | 1,499 USD | — |
Bus interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Critiques | 33 in our database | 149 in our database |
Prédécesseur | Sea Islands | GeForce 900 |
Successeur | Pirate Islands | GeForce 20 |
GPU clock | 1018 MHz | — |
---|---|---|
Mémoire clock | 1250 MHz 5 Gbps effective | 1376 MHz 11 Gbps effective |
Base clock | — | 1557 MHz |
Boost clock | — | 1670 MHz |
Taille de la mémoire | 4 GB | 10 GB |
---|---|---|
Type de mémoire | GDDR5 | GDDR5X |
Bus mémoire | 512 bit | 320 bit |
Bande passante | 320.0 GB/s | 440.3 GB/s |
Shading units | 2816 | 3200 |
---|---|---|
TMUs | 176 | 200 |
ROPs | 64 | 80 |
Compute units | 44 | — |
Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
Cache L2 | 1024 KB | 2.5 MB |
Comptage SM | — | 25 |
Taux de pixel | 65.15 GPixel/s | 133.6 GPixel/s |
---|---|---|
Taux de texture | 179.2 GTexel/s | 334.0 GTexel/s |
FP32 (float) performance | 5.733 TFLOPS | 10.69 TFLOPS |
FP64 (double) performance | 716.7 GFLOPS (1:8) | 334.0 GFLOPS (1:32) |
FP16 (half) performance | — | 167.0 GFLOPS (1:64) |
Taille Slot | Dual-slot | Dual-slot |
---|---|---|
Longueur | 307 mm 12.1 inches | 267 mm 10.5 inches |
Taille | 114 mm 4.5 inches | 112 mm 4.4 inches |
Taille | 42 mm 1.7 inches | 40 mm 1.6 inches |
TDP | 500 W | 250 W |
Suggestions pour le PSU | 900 W | 600 W |
Sorties | 1x DVI4x mini-DisplayPort | 1x HDMI3x DisplayPort |
Connecteurs de puissance | 3x 8-pin | 1x 6-pin + 1x 8-pin |
Board number | C673-A7 | PG611 SKU 10 |
DirectX | 12 (12_0) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.3 | 6.4 |
CUDA | — | 6.1 |