Nom du GPU | Vega 20 | Polaris 20 |
---|---|---|
GPU variant | Vega 20 XT | Polaris 20 XL (215-0910052) |
Architecture | GCN 5.1 | GCN 4.0 |
Fonderie | TSMC | GlobalFoundries |
Taille du processus | 7 nm | 14 nm |
Transistors | 13,230 million | 5,700 million |
Taille des matrices | 331 mm² | 232 mm² |
Date de publication | Unknown | Unknown |
---|---|---|
Génération | Radeon Pro Mac | Mining GPUs |
Production | Active | Active |
Bus interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Critiques | — | 29 in our database |
Base clock | 1400 MHz | 1120 MHz |
---|---|---|
Boost clock | 1720 MHz | 1206 MHz |
Mémoire clock | 1000 MHz 2 Gbps effective | 2100 MHz 8.4 Gbps effective |
Taille de la mémoire | 32 GB | 8 GB |
---|---|---|
Type de mémoire | HBM2 | GDDR5 |
Bus mémoire | 4096 bit | 256 bit |
Bande passante | 1,024 GB/s | 268.8 GB/s |
Shading units | 4096 | 2048 |
---|---|---|
TMUs | 256 | 128 |
ROPs | 64 | 32 |
Compute units | 64 | 32 |
Cache L1 | 16 KB (per CU) | 16 KB (per CU) |
Cache L2 | 4 MB | 2 MB |
Taux de pixel | 110.1 GPixel/s | 38.59 GPixel/s |
---|---|---|
Taux de texture | 440.3 GTexel/s | 154.4 GTexel/s |
FP16 (half) performance | 28.18 TFLOPS (2:1) | 4.940 TFLOPS (1:1) |
FP32 (float) performance | 14.09 TFLOPS | 4.940 TFLOPS |
FP64 (double) performance | 880.6 GFLOPS (1:16) | 308.7 GFLOPS (1:16) |
Taille Slot | Quad-slot | Dual-slot |
---|---|---|
TDP | 475 W | 180 W |
Suggestions pour le PSU | 850 W | 450 W |
Sorties | 1x HDMI4x Thunderbolt | 1x HDMI |
Connecteurs de puissance | None | 2x 8-pin |
Board number | MW732ZM/A | P010 |
Longueur | — | 241 mm 9.5 inches |
DirectX | 12 (12_1) | 12 (12_0) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 2.1 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.4 |