NVIDIA Tesla C2075 vs NVIDIA Tesla D870
Processeur graphique
| Nom du GPU | GF110 | G80 |
|---|---|---|
| GPU variant | GF110-351-A1 | — |
| Architecture | Fermi 2.0 | Tesla |
| Fonderie | TSMC | TSMC |
| Taille du processus | 40 nm | 90 nm |
| Transistors | 3,000 million | 681 million |
| Taille des matrices | 520 mm² | 484 mm² |
Carte graphique
| Date de publication | Jul 25th, 2011 | May 2nd, 2007 |
|---|---|---|
| Génération | Tesla | Tesla |
| Production | End-of-life | End-of-life |
| Bus interface | PCIe 2.0 x16 | PCIe 1.0 x16 |
| Prix de lancement | — | 7,499 USD |
Vitesse clock
| GPU clock | 574 MHz | 600 MHz |
|---|---|---|
| Shader clock | 1147 MHz | 1350 MHz |
| Mémoire clock | 783 MHz 3.1 Gbps effective | 800 MHz 1600 Mbps effective |
Mémoire
| Taille de la mémoire | 6 GB | 1536 MB |
|---|---|---|
| Type de mémoire | GDDR5 | GDDR3 |
| Bus mémoire | 384 bit | 384 bit |
| Bande passante | 150.3 GB/s | 76.80 GB/s |
Render Config
| Shading units | 448 | 128 |
|---|---|---|
| TMUs | 56 | 32 |
| ROPs | 48 | 24 |
| Comptage SM | 14 | 16 |
| Cache L1 | 64 KB (per SM) | — |
| Cache L2 | 768 KB | 96 KB |
Performance théorique
| Taux de pixel | 16.07 GPixel/s | 14.40 GPixel/s |
|---|---|---|
| Taux de texture | 32.14 GTexel/s | 38.40 GTexel/s |
| FP32 (float) performance | 1,028 GFLOPS | 345.6 GFLOPS |
| FP64 (double) performance | 513.9 GFLOPS (1:2) | — |
Board Design
| Taille Slot | Dual-slot | Dual-slot |
|---|---|---|
| Longueur | 248 mm 9.8 inches | — |
| TDP | 247 W | 520 W |
| Suggestions pour le PSU | 550 W | 900 W |
| Sorties | 1x DVI | No outputs |
| Connecteurs de puissance | 1x 6-pin + 1x 8-pin | — |
| Board number | P1030 | — |
Caractéristiques graphiques
| DirectX | 12 (11_0) | 11.1 (10_0) |
|---|---|---|
| OpenGL | 4.6 | 3.3 |
| OpenCL | 1.1 | 1.1 (1.0) |
| Vulkan | — | — |
| CUDA | 2.0 | 1.0 |
| Shader model | 5.1 | 4.0 |
