AMD Radeon HD 6870 vs NVIDIA GeForce GT 140 OEM
Processeur graphique
| Nom du GPU | Barts | G94B |
|---|---|---|
| GPU variant | Barts XT (215-0798000) | — |
| Architecture | TeraScale 2 | Tesla |
| Fonderie | TSMC | TSMC |
| Taille du processus | 40 nm | 55 nm |
| Transistors | 1,700 million | 505 million |
| Taille des matrices | 255 mm² | 196 mm² |
Carte graphique
| Date de publication | Oct 21st, 2010 | Mar 10th, 2009 |
|---|---|---|
| Génération | Northern Islands | GeForce 100 |
| Production | End-of-life | End-of-life |
| Prix de lancement | 239 USD | — |
| Bus interface | PCIe 2.0 x16 | PCIe 2.0 x16 |
| Critiques | 170 in our database | — |
| Prédécesseur | Evergreen | GeForce 9 |
| Successeur | Southern Islands | GeForce 200 |
Vitesse clock
| GPU clock | 900 MHz | 650 MHz |
|---|---|---|
| Mémoire clock | 1050 MHz 4.2 Gbps effective | 900 MHz 1800 Mbps effective |
| Shader clock | — | 1625 MHz |
Mémoire
| Taille de la mémoire | 1024 MB | 1024 MB |
|---|---|---|
| Type de mémoire | GDDR5 | GDDR3 |
| Bus mémoire | 256 bit | 256 bit |
| Bande passante | 134.4 GB/s | 57.60 GB/s |
Render Config
| Shading units | 1120 | 64 |
|---|---|---|
| TMUs | 56 | 32 |
| ROPs | 32 | 16 |
| Compute units | 14 | — |
| Cache L1 | 8 KB (per CU) | — |
| Cache L2 | 512 KB | 64 KB |
| Comptage SM | — | 8 |
Performance théorique
| Taux de pixel | 28.80 GPixel/s | 10.40 GPixel/s |
|---|---|---|
| Taux de texture | 50.40 GTexel/s | 20.80 GTexel/s |
| FP32 (float) performance | 2.016 TFLOPS | 208.0 GFLOPS |
Board Design
| Taille Slot | Dual-slot | Single-slot |
|---|---|---|
| Longueur | 247 mm 9.7 inches | 229 mm 9 inches |
| Taille | 111 mm 4.4 inches | — |
| Taille | 36 mm 1.4 inches | — |
| TDP | 151 W | 105 W |
| Suggestions pour le PSU | 450 W | 300 W |
| Sorties | 2x DVI1x HDMI2x mini-DisplayPort | 2x DVI1x S-Video |
| Connecteurs de puissance | 2x 6-pin | 1x 6-pin |
| Board number | C222 | P545 |
Caractéristiques graphiques
| DirectX | 11.2 (11_0) | 11.1 (10_0) |
|---|---|---|
| OpenGL | 4.4 | 3.3 |
| OpenCL | 1.2 | 1.1 |
| Vulkan | — | — |
| Shader model | 5.0 | 4.0 |
| CUDA | — | 1.1 |
