ATI Radeon HD 4670 vs NVIDIA GeForce G100 OEM
Processeur graphique
| Nom du GPU | RV730 | G98 |
|---|---|---|
| GPU variant | RV730 XT (215-0719047) | G98-309-U2 |
| Architecture | TeraScale | Tesla |
| Fonderie | TSMC | TSMC |
| Taille du processus | 55 nm | 65 nm |
| Transistors | 514 million | 210 million |
| Taille des matrices | 146 mm² | 86 mm² |
Carte graphique
| Date de publication | Sep 10th, 2008 | Mar 10th, 2009 |
|---|---|---|
| Génération | Radeon R700 | GeForce 100 |
| Production | End-of-life | End-of-life |
| Prix de lancement | 67 USD | — |
| Bus interface | PCIe 2.0 x16 | PCIe 2.0 x16 |
| Critiques | 49 in our database | — |
| Prédécesseur | Radeon R600 | GeForce 9 |
| Successeur | Evergreen | GeForce 200 |
Vitesse clock
| GPU clock | 750 MHz | 540 MHz |
|---|---|---|
| Mémoire clock | 1000 MHz 2 Gbps effective | 400 MHz 800 Mbps effective |
| Shader clock | — | 1300 MHz |
Mémoire
| Taille de la mémoire | 512 MB | 256 MB |
|---|---|---|
| Type de mémoire | GDDR3 | DDR2 |
| Bus mémoire | 128 bit | 64 bit |
| Bande passante | 32.00 GB/s | 6.400 GB/s |
Render Config
| Shading units | 320 | 8 |
|---|---|---|
| TMUs | 32 | 4 |
| ROPs | 8 | 4 |
| Compute units | 4 | — |
| Cache L1 | 16 KB (per CU) | — |
| Cache L2 | 128 KB | 16 KB |
| Comptage SM | — | 1 |
Performance théorique
| Taux de pixel | 6.000 GPixel/s | 2.160 GPixel/s |
|---|---|---|
| Taux de texture | 24.00 GTexel/s | 2.160 GTexel/s |
| FP32 (float) performance | 480.0 GFLOPS | 20.80 GFLOPS |
Board Design
| Taille Slot | Single-slot | Single-slot |
|---|---|---|
| Longueur | 193 mm 7.6 inches | 168 mm 6.6 inches |
| TDP | 59 W | 35 W |
| Suggestions pour le PSU | 250 W | 200 W |
| Sorties | 2x DVI1x S-Video | 1x DVI1x S-Video |
| Connecteurs de puissance | None | None |
| Board number | B666 | P561 |
Caractéristiques graphiques
| DirectX | 10.1 (10_1) | 11.1 (10_0) |
|---|---|---|
| OpenGL | 3.3 | 3.3 |
| OpenCL | 1.1 | 1.1 |
| Vulkan | — | — |
| Shader model | 4.1 | 4.0 |
| CUDA | — | 1.1 |
