AMD Radeon HD 8670M vs NVIDIA Quadro FX 600 PCI
Graphics Processor
| GPU model | Mars | NV34 |
|---|---|---|
| GPU variant | Mars XT | NV34 GL |
| Architecture | GCN 1.0 | Rankine |
| Foundry | TSMC | TSMC |
| Manufacturing process | 28 nm | 150 nm |
| Number of transistors | 950 million | 45 million |
| Die size | 77 mm² | 124 mm² |
Mobile graphics
| Release date | Mar 1st, 2013 | — |
|---|---|---|
| Generation | Solar System (HD 8600M) | — |
| Production | End-of-life | — |
| Interface | PCIe 3.0 x8 | — |
| Predecessor | London | — |
| Successor | Crystal System | — |
Clocks
| Base clock | 775 MHz | — |
|---|---|---|
| Boost clock | 825 MHz | — |
| Memory clock | 900 MHz 1800 Mbps effective | 240 MHz 480 Mbps effective |
| GPU clock | — | 270 MHz |
Memory Configuration
| Memory size | 1024 MB | 128 MB |
|---|---|---|
| Memory type | DDR3 | DDR |
| Memory bus width | 64 bit | 128 bit |
| Bandwidth | 14.40 GB/s | 7.680 GB/s |
Render Configuration
| Shading units | 384 | — |
|---|---|---|
| TMUs | 24 | 4 |
| ROPs | 8 | 4 |
| Compute units | 6 | — |
| Cache L1 | 16 KB (per CU) | — |
| Cache L2 | 128 KB | — |
| Pixel shaders | — | 4 |
| Vertex shaders | — | 2 |
Performance
| Pixel rate | 6.600 GPixel/s | 1.080 GPixel/s |
|---|---|---|
| Texture rate | 19.80 GTexel/s | 1.080 GTexel/s |
| FP32 (float) performance | 633.6 GFLOPS | — |
| Vertex rate | — | 135.0 MVertices/s |
Dimensions & Outputs
| TDP | unknown | unknown |
|---|---|---|
| Outputs | No outputs | 2x DVI1x S-Video |
| Slot width | — | Single-slot |
| Length | — | 165 mm 6.5 inches |
| Width | — | 111 mm 4.4 inches |
| Suggested PSU | — | 200 W |
API Support & Features
| DirectX | 12 (11_1) | 9.0a |
|---|---|---|
| OpenGL | 4.6 | 1.5 |
| OpenCL | 1.2 | — |
| Vulkan | 1.2 | — |
| Shader model | 5.1 | — |
| Pixel shader | — | 2.0a |
| Vertex shader | — | 2.0a |
Other Features
Graphics Card
| Release date | — | Mar 17th, 2004 |
|---|---|---|
| Generation | — | Quadro FX |
| Production | — | End-of-life |
| Interface | — | PCI |
| Reviews | — | 26 in our database |
