Nom du GPU | Polaris 20 | TU116 |
---|---|---|
GPU variant | Polaris 20 XTX (215-0910038) | TU116-300-A1 |
Architecture | GCN 4.0 | Turing |
Fonderie | GlobalFoundries | TSMC |
Taille du processus | 14 nm | 12 nm |
Transistors | 5,700 million | 6,600 million |
Taille des matrices | 232 mm² | 284 mm² |
Date de publication | Apr 18th, 2017 | Oct 29th, 2019 |
---|---|---|
Génération | Polaris | GeForce 16 |
Production | Active | Active |
Prix de lancement | 229 USD | 229 USD |
Prix courant | Amazon / Newegg | — |
Bus interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Critiques | 52 in our database | 42 in our database |
Prédécesseur | Arctic Islands | GeForce 10 |
Successeur | Vega | GeForce 20 |
Base clock | 1257 MHz | 1530 MHz |
---|---|---|
Boost clock | 1340 MHz | 1785 MHz |
Mémoire clock | 2000 MHz 8 Gbps effective | 1750 MHz 14 Gbps effective |
Taille de la mémoire | 8 GB | 6 GB |
---|---|---|
Type de mémoire | GDDR5 | GDDR6 |
Bus mémoire | 256 bit | 192 bit |
Bande passante | 256.0 GB/s | 336.0 GB/s |
Shading units | 2304 | 1408 |
---|---|---|
TMUs | 144 | 88 |
ROPs | 32 | 48 |
Compute units | 36 | — |
Cache L1 | 16 KB (per CU) | 64 KB (per SM) |
Cache L2 | 2 MB | 1536 KB |
Comptage SM | — | 22 |
Taux de pixel | 42.88 GPixel/s | 85.68 GPixel/s |
---|---|---|
Taux de texture | 193.0 GTexel/s | 157.1 GTexel/s |
FP16 (half) performance | 6.175 TFLOPS (1:1) | 10.05 TFLOPS (2:1) |
FP32 (float) performance | 6.175 TFLOPS | 5.027 TFLOPS |
FP64 (double) performance | 385.9 GFLOPS (1:16) | 157.1 GFLOPS (1:32) |
Taille Slot | Dual-slot | Dual-slot |
---|---|---|
Longueur | 241 mm 9.5 inches | 229 mm 9 inches |
TDP | 185 W | 125 W |
Suggestions pour le PSU | 450 W | 300 W |
Sorties | 1x HDMI3x DisplayPort | 1x DVI1x HDMI1x DisplayPort |
Connecteurs de puissance | 1x 8-pin | 1x 8-pin |
Board number | C940, D009-04 | PG161 SKU 26 |
Taille | — | 111 mm 4.4 inches |
Taille | — | 35 mm 1.4 inches |
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 | 6.6 |
CUDA | — | 7.5 |