ATI Radeon HD 3870 X2 vs NVIDIA GeForce 9600 GS OEM
Graphics Processor
| GPU model | R680 | G94 |
|---|---|---|
| GPU variant | R680 XT (215-0708017) | G94-300-A1 |
| Architecture | TeraScale | Tesla |
| Foundry | TSMC | TSMC |
| Manufacturing process | 55 nm | 65 nm |
| Number of transistors | 666 million | 505 million |
| Die size | 192 mm² | 240 mm² |
Graphics Card
| Release date | Jan 26th, 2008 | Jul 29th, 2008 |
|---|---|---|
| Generation | Radeon R600 | GeForce 9 |
| Production | End-of-life | End-of-life |
| Launch price | 449 USD | — |
| Interface | PCIe 2.0 x16 | PCIe 2.0 x16 |
| Reviews | 72 in our database | 151 in our database |
| Predecessor | Radeon R500 PCIe | GeForce 8 |
| Successor | Radeon R700 | GeForce 200 |
Clocks
| GPU clock | 825 MHz | 500 MHz |
|---|---|---|
| Memory clock | 901 MHz 1802 Mbps effective | 1000 MHz 2 Gbps effective |
| Shader clock | — | 1250 MHz |
Memory Configuration
| Memory size | 512 MB | 768 MB |
|---|---|---|
| Memory type | GDDR3 | DDR2 |
| Memory bus width | 256 bit | 192 bit |
| Bandwidth | 57.66 GB/s | 48.00 GB/s |
Render Configuration
| Shading units | 320 | 48 |
|---|---|---|
| TMUs | 16 | 24 |
| ROPs | 16 | 12 |
| Compute units | 4 | — |
| Cache L2 | 256 KB | 48 KB |
| SM count | — | 6 |
Performance
| Pixel rate | 13.20 GPixel/s | 6.000 GPixel/s |
|---|---|---|
| Texture rate | 13.20 GTexel/s | 12.00 GTexel/s |
| FP32 (float) performance | 528.0 GFLOPS | 120.0 GFLOPS |
| FP64 (double) performance | 105.6 GFLOPS (1:5) | — |
Dimensions & Outputs
| Slot width | Dual-slot | Single-slot |
|---|---|---|
| Length | 267 mm 10.5 inches | — |
| Width | 113 mm 4.4 inches | — |
| Height | 39 mm 1.5 inches | — |
| TDP | 165 W | unknown |
| Suggested PSU | 450 W | 200 W |
| Outputs | 2x DVI1x S-Video | 1x DVI1x VGA1x S-Video |
| Power connectors | 1x 6-pin + 1x 8-pin | 1x 6-pin |
| Board number | B400 | P545 |
API Support & Features
| DirectX | 10.1 (10_1) | 11.1 (10_0) |
|---|---|---|
| OpenGL | 3.3 (full) 4.0 (partial) | 3.3 |
| OpenCL | — | 1.1 |
| Vulkan | — | — |
| Shader model | 4.1 | 4.0 |
| CUDA | — | 1.1 |
