NVIDIA GeForce GTX 580 Rev. 2 vs AMD Radeon HD 6870
What is the difference between NVIDIA GeForce GTX 580 Rev. 2 and AMD Radeon HD 6870. Find out which graphics card has better performance.
Graphics Processor (GPU)
| GF110 | GPU Name | Barts |
| GF110-380-A1 | Model number | Barts XT (215-0798000) |
| Fermi 2.0 | Architecture | TeraScale 2 |
| TSMC | Foundry | TSMC |
| 40 nm | Process Size | 40 nm |
| 3,000 million | Transistors | 1,700 million |
| 520 mm² | Die Size | 255 mm² |
Graphics Card
| Jun 6th, 2011 | Release Date | Oct 21st, 2010 |
| GeForce 500 | Family | Northern Islands (HD 6800) |
| End-of-life | Production | End-of-life |
| 499 USD | Launch Price | 239 USD |
| GeForce 400 | Predecessor | Evergreen |
| PCIe 2.0 x16 | Bus Interface | PCIe 2.0 x16 |
Memory
| 1536 MB | Memory Size | 1024 MB |
| GDDR5 | Memory Type | GDDR5 |
| 384 bit | Memory Bus | 256 bit |
| 192.4 GB/s | Bandwidth | 134.4 GB/s |
Performance
| 24.70 GPixel/s | Pixel fillrate | 28.80 GPixel/s |
| 49.41 GTexel/s | Texture fillrate | 50.40 GTexel/s |
| 1.581 TFLOPS | FP32 (float) performance | 2.016 TFLOPS |
Clock Speeds
| 1002 MHz 4 Gbps effective | Memory Clock | 1050 MHz 4.2 Gbps effective |
| 772 MHz | GPU Clock | 900 MHz |
Render Config
| 512 | Shading Units | 1120 |
| 64 | TMUs | 56 |
| 48 | ROPs | 32 |
| 64 KB (per SM) | L1 Cache | 8 KB (per CU) |
| 768 KB | L2 Cache | 512 KB |
Board Design
| Dual-slot | Slot Width | Dual-slot |
| 267 mm 10.5 inches | Length | 247 mm 9.7 inches |
| 244 W | Thermal design power (TDP) | 151 W |
| 550 W | Suggested PSU | 450 W |
| 2x DVI 1x mini-HDMI | Display Connectors | 2x DVI 1x HDMI 2x mini-DisplayPort |
| 1x 6-pin + 1x 8-pin | Power Connectors | 2x 6-pin |
| P1261 SKU 12 | Board Number | C222 |
API support
| 12 (11_0) | DirectX | 11.2 (11_0) |
| 4.6 | OpenGL | 4.4 |
| 1.1 | OpenCL | 1.2 |
| N/A | Vulkan | N/A |
| 5.1 | Shader Model | 5.0 |