NVIDIA GRID K260Q vs NVIDIA Tesla S870
Processeur graphique
| Nom du GPU | GK104 | G80 |
|---|---|---|
| Architecture | Kepler | Tesla |
| Fonderie | TSMC | TSMC |
| Taille du processus | 28 nm | 90 nm |
| Transistors | 3,540 million | 681 million |
| Taille des matrices | 294 mm² | 484 mm² |
Carte graphique
| Date de publication | Jun 28th, 2013 | May 2nd, 2007 |
|---|---|---|
| Génération | GRID | Tesla |
| Production | End-of-life | End-of-life |
| Prix de lancement | 937 USD | 11,999 USD |
| Bus interface | PCIe 3.0 x16 | PCIe 1.0 x16 |
Vitesse clock
| GPU clock | 745 MHz | 600 MHz |
|---|---|---|
| Mémoire clock | 1250 MHz 5 Gbps effective | 800 MHz 1600 Mbps effective |
| Shader clock | — | 1350 MHz |
Mémoire
| Taille de la mémoire | 2 GB | 1536 MB |
|---|---|---|
| Type de mémoire | GDDR5 | GDDR3 |
| Bus mémoire | 256 bit | 384 bit |
| Bande passante | 160.0 GB/s | 76.80 GB/s |
Render Config
| Shading units | 1536 | 128 |
|---|---|---|
| TMUs | 128 | 32 |
| ROPs | 32 | 24 |
| Comptage SMX | 8 | — |
| Cache L1 | 16 KB (per SMX) | — |
| Cache L2 | 512 KB | 96 KB |
| Comptage SM | — | 16 |
Performance théorique
| Taux de pixel | 23.84 GPixel/s | 14.40 GPixel/s |
|---|---|---|
| Taux de texture | 95.36 GTexel/s | 38.40 GTexel/s |
| FP32 (float) performance | 2.289 TFLOPS | 345.6 GFLOPS |
| FP64 (double) performance | 95.36 GFLOPS (1:24) | — |
Board Design
| Taille Slot | IGP | Dual-slot |
|---|---|---|
| TDP | 225 W | 800 W |
| Suggestions pour le PSU | 550 W | 1200 W |
| Sorties | No outputs | No outputs |
Caractéristiques graphiques
| DirectX | 12 (11_0) | 11.1 (10_0) |
|---|---|---|
| OpenGL | 4.6 | 3.3 |
| OpenCL | 3.0 | 1.1 (1.0) |
| Vulkan | 1.1 | — |
| CUDA | 3.0 | 1.0 |
| Shader model | 5.1 | 4.0 |
