ATI Radeon HD 3870 vs NVIDIA GeForce 8800 GTS 640
Processeur graphique
| Nom du GPU | RV670 | G80 |
|---|---|---|
| GPU variant | RV670 XT (215-0708005) | G80-100-K0-A2 |
| Architecture | TeraScale | Tesla |
| Fonderie | TSMC | TSMC |
| Taille du processus | 55 nm | 90 nm |
| Transistors | 666 million | 681 million |
| Taille des matrices | 192 mm² | 484 mm² |
Carte graphique
| Date de publication | Nov 19th, 2007 | Nov 8th, 2006 |
|---|---|---|
| Génération | Radeon R600 | GeForce 8 |
| Production | End-of-life | End-of-life |
| Prix de lancement | 269 USD | 449 USD |
| Bus interface | PCIe 2.0 x16 | PCIe 1.0 x16 |
| Critiques | 80 in our database | 403 in our database |
| Prédécesseur | Radeon R500 PCIe | GeForce 7 PCIe |
| Successeur | Radeon R700 | GeForce 9 |
Vitesse clock
| GPU clock | 777 MHz | 513 MHz |
|---|---|---|
| Mémoire clock | 1126 MHz 2.3 Gbps effective | 792 MHz 1584 Mbps effective |
| Shader clock | — | 1188 MHz |
Mémoire
| Taille de la mémoire | 512 MB | 640 MB |
|---|---|---|
| Type de mémoire | GDDR4 | GDDR3 |
| Bus mémoire | 256 bit | 320 bit |
| Bande passante | 72.06 GB/s | 63.36 GB/s |
Render Config
| Shading units | 320 | 96 |
|---|---|---|
| TMUs | 16 | 24 |
| ROPs | 16 | 20 |
| Compute units | 4 | — |
| Cache L2 | 256 KB | 80 KB |
| Comptage SM | — | 12 |
Performance théorique
| Taux de pixel | 12.43 GPixel/s | 10.26 GPixel/s |
|---|---|---|
| Taux de texture | 12.43 GTexel/s | 24.62 GTexel/s |
| FP32 (float) performance | 497.3 GFLOPS | 228.1 GFLOPS |
| FP64 (double) performance | 99.46 GFLOPS (1:5) | — |
Board Design
| Taille Slot | Dual-slot | Dual-slot |
|---|---|---|
| Longueur | 241 mm 9.5 inches | 267 mm 10.5 inches |
| TDP | 106 W | 143 W |
| Suggestions pour le PSU | 300 W | 300 W |
| Sorties | 2x DVI1x S-Video | 2x DVI1x S-Video |
| Connecteurs de puissance | 1x 6-pin | 1x 6-pin |
| Board number | B339 | P356 |
Caractéristiques graphiques
| DirectX | 10.1 (10_1) | 11.1 (10_0) |
|---|---|---|
| OpenGL | 3.3 (full) 4.0 (partial) | 3.3 |
| OpenCL | — | 1.1 (1.0) |
| Vulkan | — | — |
| Shader model | 4.1 | 4.0 |
| CUDA | — | 1.0 |
