ATI Radeon HD 5770 X2 vs NVIDIA GeForce GTS 450
Graphics Processor
| GPU model | Juniper | GF106 |
|---|---|---|
| GPU variant | Juniper XT (215-0754013) | GF106-250-KA-A1 |
| Architecture | TeraScale 2 | Fermi |
| Foundry | TSMC | TSMC |
| Manufacturing process | 40 nm | 40 nm |
| Number of transistors | 1,040 million | 1,170 million |
| Die size | 166 mm² | 238 mm² |
Graphics Card
| Release date | Oct 13th, 2010 | Sep 13th, 2010 |
|---|---|---|
| Generation | Evergreen | GeForce 400 |
| Production | End-of-life | End-of-life |
| Interface | PCIe 2.0 x16 | PCIe 2.0 x16 |
| Reviews | 142 in our database | 80 in our database |
| Predecessor | Radeon R700 | GeForce 200 |
| Successor | Northern Islands | GeForce 500 |
| Launch price | — | 129 USD |
Clocks
| GPU clock | 850 MHz | 783 MHz |
|---|---|---|
| Memory clock | 1200 MHz 4.8 Gbps effective | 902 MHz 3.6 Gbps effective |
| Shader clock | — | 1566 MHz |
Memory Configuration
| Memory size | 1024 MB | 1024 MB |
|---|---|---|
| Memory type | GDDR5 | GDDR5 |
| Memory bus width | 128 bit | 128 bit |
| Bandwidth | 76.80 GB/s | 57.73 GB/s |
Render Configuration
| Shading units | 800 | 192 |
|---|---|---|
| TMUs | 40 | 32 |
| ROPs | 16 | 16 |
| Compute units | 10 | — |
| Cache L1 | 8 KB (per CU) | 64 KB (per SM) |
| Cache L2 | 256 KB | 256 KB |
| SM count | — | 4 |
Performance
| Pixel rate | 13.60 GPixel/s | 6.264 GPixel/s |
|---|---|---|
| Texture rate | 34.00 GTexel/s | 25.06 GTexel/s |
| FP32 (float) performance | 1,360 GFLOPS | 601.3 GFLOPS |
| FP64 (double) performance | — | 50.11 GFLOPS (1:12) |
Dimensions & Outputs
| Slot width | Dual-slot | Dual-slot |
|---|---|---|
| TDP | unknown | 106 W |
| Suggested PSU | 200 W | 300 W |
| Outputs | 2x DVI1x HDMI1x DisplayPort | 2x DVI1x mini-HDMI |
| Length | — | 210 mm 8.3 inches |
| Width | — | 111 mm 4.4 inches |
| Power connectors | — | 1x 6-pin |
| Board number | — | P1060 |
API Support & Features
| DirectX | 11.2 (11_0) | 12 (11_0) |
|---|---|---|
| OpenGL | 4.4 | 4.6 |
| OpenCL | 1.2 | 1.1 |
| Vulkan | — | — |
| Shader model | 5.0 | 5.1 |
| CUDA | — | 2.1 |
