ATI FireGL V8650 vs NVIDIA Quadro 600
Processeur graphique
| Nom du GPU | R600 | GF108 |
|---|---|---|
| GPU variant | R600 V8650 (215RGIDKA13FG) | — |
| Architecture | TeraScale | Fermi |
| Fonderie | TSMC | TSMC |
| Taille du processus | 80 nm | 40 nm |
| Transistors | 720 million | 585 million |
| Taille des matrices | 420 mm² | 116 mm² |
Carte graphique
| Date de publication | Aug 6th, 2007 | Dec 13th, 2010 |
|---|---|---|
| Génération | FireGL | Quadro |
| Production | End-of-life | End-of-life |
| Prix de lancement | 2,799 USD | 179 USD |
| Bus interface | PCIe 1.0 x16 | PCIe 2.0 x16 |
| Critiques | 5 in our database | 26 in our database |
Vitesse clock
| GPU clock | 688 MHz | 640 MHz |
|---|---|---|
| Mémoire clock | 868 MHz 1736 Mbps effective | 800 MHz 1600 Mbps effective |
| Shader clock | — | 1280 MHz |
Mémoire
| Taille de la mémoire | 2 GB | 1024 MB |
|---|---|---|
| Type de mémoire | GDDR4 | DDR3 |
| Bus mémoire | 512 bit | 128 bit |
| Bande passante | 111.1 GB/s | 25.60 GB/s |
Render Config
| Shading units | 320 | 96 |
|---|---|---|
| TMUs | 16 | 16 |
| ROPs | 16 | 8 |
| Compute units | 4 | — |
| Cache L2 | 256 KB | 256 KB |
| Comptage SM | — | 2 |
| Cache L1 | — | 64 KB (per SM) |
Performance théorique
| Taux de pixel | 11.01 GPixel/s | 2.560 GPixel/s |
|---|---|---|
| Taux de texture | 11.01 GTexel/s | 10.24 GTexel/s |
| FP32 (float) performance | 440.3 GFLOPS | 245.8 GFLOPS |
| FP64 (double) performance | — | 20.48 GFLOPS (1:12) |
Board Design
| Taille Slot | Dual-slot | Single-slot |
|---|---|---|
| Longueur | 254 mm 10 inches | 168 mm 6.6 inches |
| Taille | 111 mm 4.4 inches | 69 mm 2.7 inches |
| TDP | unknown | 40 W |
| Suggestions pour le PSU | 200 W | 200 W |
| Sorties | 2x DVI1x S-Video | 1x DVI1x DisplayPort |
| Connecteurs de puissance | 1x 6-pin + 1x 8-pin | None |
| Board number | B313 | P1033 |
Caractéristiques graphiques
| DirectX | 10.0 (10_0) | 12 (11_0) |
|---|---|---|
| OpenGL | 3.3 (full) 4.0 (partial) | 4.6 |
| OpenCL | — | 1.1 |
| Vulkan | — | — |
| Shader model | 4.0 | 5.1 |
| CUDA | — | 2.1 |
