AMD Radeon Pro V340 vs NVIDIA Tesla D870
Processeur graphique
| Nom du GPU | Vega 10 | G80 |
|---|---|---|
| GPU variant | Vega 10 XL GL (215-0894304) | — |
| Architecture | GCN 5.0 | Tesla |
| Fonderie | GlobalFoundries | TSMC |
| Taille du processus | 14 nm | 90 nm |
| Transistors | 12,500 million | 681 million |
| Taille des matrices | 495 mm² | 484 mm² |
Carte graphique
| Date de publication | Aug 26th, 2018 | May 2nd, 2007 |
|---|---|---|
| Génération | Radeon Pro | Tesla |
| Production | Active | End-of-life |
| Bus interface | PCIe 3.0 x16 | PCIe 1.0 x16 |
| Prix de lancement | — | 7,499 USD |
Vitesse clock
| Base clock | 852 MHz | — |
|---|---|---|
| Boost clock | 1500 MHz | — |
| Mémoire clock | 945 MHz 1890 Mbps effective | 800 MHz 1600 Mbps effective |
| GPU clock | — | 600 MHz |
| Shader clock | — | 1350 MHz |
Mémoire
| Taille de la mémoire | 16 GB | 1536 MB |
|---|---|---|
| Type de mémoire | HBM2 | GDDR3 |
| Bus mémoire | 2048 bit | 384 bit |
| Bande passante | 483.8 GB/s | 76.80 GB/s |
Render Config
| Shading units | 3584 | 128 |
|---|---|---|
| TMUs | 224 | 32 |
| ROPs | 64 | 24 |
| Compute units | 56 | — |
| Cache L1 | 16 KB (per CU) | — |
| Cache L2 | 4 MB | 96 KB |
| Comptage SM | — | 16 |
Performance théorique
| Taux de pixel | 96.00 GPixel/s | 14.40 GPixel/s |
|---|---|---|
| Taux de texture | 336.0 GTexel/s | 38.40 GTexel/s |
| FP16 (half) performance | 21.50 TFLOPS (2:1) | — |
| FP32 (float) performance | 10.75 TFLOPS | 345.6 GFLOPS |
| FP64 (double) performance | 672.0 GFLOPS (1:16) | — |
Board Design
| Taille Slot | Dual-slot | Dual-slot |
|---|---|---|
| Longueur | 267 mm 10.5 inches | — |
| Taille | 111 mm 4.4 inches | — |
| TDP | 230 W | 520 W |
| Suggestions pour le PSU | 550 W | 900 W |
| Sorties | 1x mini-DisplayPort | No outputs |
| Connecteurs de puissance | 2x 8-pin | — |
Caractéristiques graphiques
| DirectX | 12 (12_1) | 11.1 (10_0) |
|---|---|---|
| OpenGL | 4.6 | 3.3 |
| OpenCL | 2.1 | 1.1 (1.0) |
| Vulkan | 1.2 | — |
| Shader model | 6.4 | 4.0 |
| CUDA | — | 1.0 |
