ATI Radeon HD 3850 X3 vs NVIDIA Quadro 410
Processeur graphique
| Nom du GPU | RV670 | GK107 |
|---|---|---|
| GPU variant | RV670 PRO (215-0708003) | GK107-810-A2 |
| Architecture | TeraScale | Kepler |
| Fonderie | TSMC | TSMC |
| Taille du processus | 55 nm | 28 nm |
| Transistors | 666 million | 1,270 million |
| Taille des matrices | 192 mm² | 118 mm² |
Carte graphique
| Date de publication | Never Released | Aug 7th, 2012 |
|---|---|---|
| Génération | Radeon R600 | Quadro |
| Production | End-of-life | End-of-life |
| Bus interface | PCIe 2.0 x16 | PCIe 2.0 x16 |
| Prédécesseur | Radeon R500 PCIe | — |
| Successeur | Radeon R700 | — |
| Prix de lancement | — | 149 USD |
| Critiques | — | 26 in our database |
Vitesse clock
| GPU clock | 669 MHz | 706 MHz |
|---|---|---|
| Mémoire clock | 850 MHz 1700 Mbps effective | 891 MHz 1782 Mbps effective |
Mémoire
| Taille de la mémoire | 512 MB | 512 MB |
|---|---|---|
| Type de mémoire | GDDR3 | DDR3 |
| Bus mémoire | 256 bit | 64 bit |
| Bande passante | 54.40 GB/s | 14.26 GB/s |
Render Config
| Shading units | 320 | 192 |
|---|---|---|
| TMUs | 16 | 16 |
| ROPs | 16 | 8 |
| Compute units | 4 | — |
| Cache L2 | 256 KB | 128 KB |
| Comptage SMX | — | 1 |
| Cache L1 | — | 16 KB (per SMX) |
Performance théorique
| Taux de pixel | 10.70 GPixel/s | 2.824 GPixel/s |
|---|---|---|
| Taux de texture | 10.70 GTexel/s | 11.30 GTexel/s |
| FP32 (float) performance | 428.2 GFLOPS | 271.1 GFLOPS |
| FP64 (double) performance | — | 11.30 GFLOPS (1:24) |
Board Design
| Taille Slot | Dual-slot | Single-slot |
|---|---|---|
| Longueur | 292 mm 11.5 inches | 176 mm 6.9 inches |
| TDP | 145 W | 38 W |
| Suggestions pour le PSU | 300 W | 200 W |
| Sorties | 4x DVI | 1x DVI1x DisplayPort |
| Connecteurs de puissance | 1x 8-pin | None |
| Taille | — | 69 mm 2.7 inches |
| Board number | — | P2012 |
Caractéristiques graphiques
| DirectX | 10.1 (10_1) | 12 (11_0) |
|---|---|---|
| OpenGL | 3.3 (full) 4.0 (partial) | 4.6 |
| OpenCL | — | 3.0 |
| Vulkan | — | 1.1 |
| Shader model | 4.1 | 5.1 |
| CUDA | — | 3.0 |
