NVIDIA GRID RTX T10-8 vs NVIDIA Tesla T10 Processor
Processeur graphique
| Nom du GPU | TU102 | GT200B |
|---|---|---|
| GPU variant | TU102-875-A1 | — |
| Architecture | Turing | Tesla 2.0 |
| Fonderie | TSMC | TSMC |
| Taille du processus | 12 nm | 55 nm |
| Transistors | 18,600 million | 1,400 million |
| Taille des matrices | 754 mm² | 470 mm² |
Carte graphique
| Date de publication | Unknown | Apr 9th, 2009 |
|---|---|---|
| Génération | GRID | Tesla |
| Production | Active | End-of-life |
| Bus interface | PCIe 3.0 x16 | PCIe 2.0 x16 |
Vitesse clock
| Base clock | 1065 MHz | — |
|---|---|---|
| Boost clock | 1395 MHz | — |
| Mémoire clock | 1750 MHz 14 Gbps effective | 800 MHz 1600 Mbps effective |
| GPU clock | — | 610 MHz |
| Shader clock | — | 1296 MHz |
Mémoire
| Taille de la mémoire | 8 GB | 4 GB |
|---|---|---|
| Type de mémoire | GDDR6 | GDDR3 |
| Bus mémoire | 384 bit | 512 bit |
| Bande passante | 672.0 GB/s | 102.4 GB/s |
Render Config
| Shading units | 4608 | 240 |
|---|---|---|
| TMUs | 288 | 80 |
| ROPs | 96 | 32 |
| Comptage SM | 72 | 30 |
| Tensor cores | 576 | — |
| RT cores | 72 | — |
| Cache L1 | 64 KB (per SM) | — |
| Cache L2 | 6 MB | 256 KB |
Performance théorique
| Taux de pixel | 133.9 GPixel/s | 19.52 GPixel/s |
|---|---|---|
| Taux de texture | 401.8 GTexel/s | 48.80 GTexel/s |
| FP16 (half) performance | 25.71 TFLOPS (2:1) | — |
| FP32 (float) performance | 12.86 TFLOPS | 622.1 GFLOPS |
| FP64 (double) performance | 401.8 GFLOPS (1:32) | 77.76 GFLOPS (1:8) |
Board Design
| Taille Slot | Dual-slot | Dual-slot |
|---|---|---|
| Longueur | 267 mm 10.5 inches | 267 mm 10.5 inches |
| TDP | 260 W | 188 W |
| Suggestions pour le PSU | 600 W | 450 W |
| Sorties | No outputs | No outputs |
| Connecteurs de puissance | 1x 6-pin + 1x 8-pin | — |
| Board number | PG150 SKU 220 | — |
Caractéristiques graphiques
| DirectX | 12 Ultimate (12_2) | 11.1 (10_0) |
|---|---|---|
| OpenGL | 4.6 | 3.3 |
| OpenCL | 3.0 | 1.1 |
| Vulkan | 1.2 | — |
| CUDA | 7.5 | 1.3 |
| Shader model | 6.6 | 4.0 |
