Nom du GPU | Vega 20 | GP106 |
---|---|---|
GPU variant | Vega 20 XT | GP106-400-A1 |
Architecture | GCN 5.1 | Pascal |
Fonderie | TSMC | TSMC |
Taille du processus | 7 nm | 16 nm |
Transistors | 13,230 million | 4,400 million |
Taille des matrices | 331 mm² | 200 mm² |
Date de publication | Unknown | Jan 5th, 2018 |
---|---|---|
Génération | Radeon Pro Mac | GeForce 10 |
Production | Active | End-of-life |
Bus interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Prédécesseur | — | GeForce 900 |
Successeur | — | GeForce 20 |
Prix de lancement | — | 299 USD |
Critiques | — | 117 in our database |
Base clock | 1400 MHz | 1506 MHz |
---|---|---|
Boost clock | 1720 MHz | 1709 MHz |
Mémoire clock | 1000 MHz 2 Gbps effective | 2002 MHz 8 Gbps effective |
Taille de la mémoire | 32 GB | 6 GB |
---|---|---|
Type de mémoire | HBM2 | GDDR5 |
Bus mémoire | 4096 bit | 192 bit |
Bande passante | 1,024 GB/s | 192.2 GB/s |
Shading units | 4096 | 1280 |
---|---|---|
TMUs | 256 | 80 |
ROPs | 64 | 48 |
Compute units | 64 | — |
Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
Cache L2 | 4 MB | 1536 KB |
Comptage SM | — | 10 |
Taux de pixel | 110.1 GPixel/s | 82.03 GPixel/s |
---|---|---|
Taux de texture | 440.3 GTexel/s | 136.7 GTexel/s |
FP16 (half) performance | 28.18 TFLOPS (2:1) | 68.36 GFLOPS (1:64) |
FP32 (float) performance | 14.09 TFLOPS | 4.375 TFLOPS |
FP64 (double) performance | 880.6 GFLOPS (1:16) | 136.7 GFLOPS (1:32) |
Taille Slot | Quad-slot | Dual-slot |
---|---|---|
TDP | 475 W | 120 W |
Suggestions pour le PSU | 850 W | 300 W |
Sorties | 1x HDMI4x Thunderbolt | 1x DVI1x HDMI3x DisplayPort |
Connecteurs de puissance | None | 1x 6-pin |
Board number | MW732ZM/A | PG410 SKU 31 |
Longueur | — | 250 mm 9.8 inches |
DirectX | 12 (12_1) | 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 |