Nom du GPU | Ellesmere | GP106 |
---|---|---|
GPU variant | Polaris 10 PRO (215-0876204) | GP106-400-A1 |
Architecture | GCN 4.0 | Pascal |
Fonderie | GlobalFoundries | TSMC |
Taille du processus | 14 nm | 16 nm |
Transistors | 5,700 million | 4,400 million |
Taille des matrices | 232 mm² | 200 mm² |
Date de publication | Aug 4th, 2016 | Jul 19th, 2016 |
---|---|---|
Génération | Arctic Islands | GeForce 10 |
Production | Active | End-of-life |
Prix de lancement | 179 USD | 299 USD |
Bus interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Critiques | 46 in our database | 117 in our database |
Prédécesseur | Pirate Islands | GeForce 900 |
Successeur | Polaris | GeForce 20 |
Base clock | 926 MHz | 1506 MHz |
---|---|---|
Boost clock | 1206 MHz | 1709 MHz |
Mémoire clock | 1650 MHz 6.6 Gbps effective | 2002 MHz 8 Gbps effective |
Taille de la mémoire | 4 GB | 6 GB |
---|---|---|
Type de mémoire | GDDR5 | GDDR5 |
Bus mémoire | 256 bit | 192 bit |
Bande passante | 211.2 GB/s | 192.2 GB/s |
Shading units | 2048 | 1280 |
---|---|---|
TMUs | 128 | 80 |
ROPs | 32 | 48 |
Compute units | 32 | — |
Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
Cache L2 | 2 MB | 1536 KB |
Comptage SM | — | 10 |
Taux de pixel | 38.59 GPixel/s | 82.03 GPixel/s |
---|---|---|
Taux de texture | 154.4 GTexel/s | 136.7 GTexel/s |
FP16 (half) performance | 4.940 TFLOPS (1:1) | 68.36 GFLOPS (1:64) |
FP32 (float) performance | 4.940 TFLOPS | 4.375 TFLOPS |
FP64 (double) performance | 308.7 GFLOPS (1:16) | 136.7 GFLOPS (1:32) |
Taille Slot | Dual-slot | Dual-slot |
---|---|---|
Longueur | 240 mm 9.4 inches | 250 mm 9.8 inches |
Taille | 95 mm 3.7 inches | 111 mm 4.4 inches |
Taille | 35 mm 1.4 inches | — |
TDP | 120 W | 120 W |
Suggestions pour le PSU | 300 W | 300 W |
Sorties | 1x HDMI3x DisplayPort | 1x DVI1x HDMI3x DisplayPort |
Connecteurs de puissance | 1x 6-pin | 1x 6-pin |
Board number | C940, D000-03, D009-07 | PG410 SKU 30 |
DirectX | 12 (12_0) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.4 |
CUDA | — | 6.1 |