AMD Stream Processor vs NVIDIA GRID RTX T10-16
Processeur graphique
| Nom du GPU | R580 | TU102 |
|---|---|---|
| GPU variant | R580 XTX | TU102-875-A1 |
| Architecture | R500 | Turing |
| Fonderie | TSMC | TSMC |
| Taille du processus | 90 nm | 12 nm |
| Transistors | 384 million | 18,600 million |
| Taille des matrices | 352 mm² | 754 mm² |
Carte graphique
| Date de publication | 2006 | Unknown |
|---|---|---|
| Génération | FireStream | GRID |
| Production | End-of-life | Active |
| Bus interface | PCIe 1.0 x16 | PCIe 3.0 x16 |
Vitesse clock
| GPU clock | 594 MHz | — |
|---|---|---|
| Mémoire clock | 648 MHz 1296 Mbps effective | 1575 MHz 12.6 Gbps effective |
| Base clock | — | 1065 MHz |
| Boost clock | — | 1395 MHz |
Mémoire
| Taille de la mémoire | 1024 MB | 16 GB |
|---|---|---|
| Type de mémoire | GDDR3 | GDDR6 |
| Bus mémoire | 256 bit | 384 bit |
| Bande passante | 41.47 GB/s | 604.8 GB/s |
Render Config
| Pixel shaders | 48 | — |
|---|---|---|
| Vertex shaders | 8 | — |
| TMUs | 16 | 288 |
| ROPs | 16 | 96 |
| Shading units | — | 4608 |
| Comptage SM | — | 72 |
| Tensor cores | — | 576 |
| RT cores | — | 72 |
| Cache L1 | — | 64 KB (per SM) |
| Cache L2 | — | 6 MB |
Performance théorique
| Taux de pixel | 9.504 GPixel/s | 133.9 GPixel/s |
|---|---|---|
| Vertex rate | 1.188 GVertices/s | — |
| Taux de texture | 9.504 GTexel/s | 401.8 GTexel/s |
| FP16 (half) performance | — | 25.71 TFLOPS (2:1) |
| FP32 (float) performance | — | 12.86 TFLOPS |
| FP64 (double) performance | — | 401.8 GFLOPS (1:32) |
Board Design
| Taille Slot | Dual-slot | Dual-slot |
|---|---|---|
| Longueur | 241 mm 9.5 inches | 267 mm 10.5 inches |
| Taille | 111 mm 4.4 inches | — |
| TDP | 165 W | 260 W |
| Suggestions pour le PSU | 450 W | 600 W |
| Sorties | 2x DVI1x S-Video | No outputs |
| Connecteurs de puissance | 1x 6-pin | 1x 6-pin + 1x 8-pin |
| Board number | — | PG150 SKU 215 |
Caractéristiques graphiques
| DirectX | 9.0c (9_3) | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 2.1 (full) 3.0 (partial) | 4.6 |
| OpenCL | — | 3.0 |
| Vulkan | — | 1.2 |
| Shader model | 3.0 | 6.6 |
| CUDA | — | 7.5 |
