AMD Radeon RX 560 896SP vs NVIDIA GRID K200
Graphics Processor
| GPU model | Polaris 21 | GK104 |
|---|---|---|
| GPU variant | Polaris 21 XL | — |
| Architecture | GCN 4.0 | Kepler |
| Foundry | GlobalFoundries | TSMC |
| Manufacturing process | 14 nm | 28 nm |
| Number of transistors | 3,000 million | 3,540 million |
| Die size | 123 mm² | 294 mm² |
Graphics Card
| Release date | Jul 4th, 2017 | Jun 28th, 2013 |
|---|---|---|
| Generation | Polaris | GRID |
| Production | Active | End-of-life |
| Interface | PCIe 3.0 x8 | PCIe 3.0 x16 |
| Predecessor | Arctic Islands | — |
| Successor | Vega | — |
| Launch price | — | 235 USD |
Clocks
| Base clock | 1090 MHz | — |
|---|---|---|
| Boost clock | 1175 MHz | — |
| Memory clock | 1750 MHz 7 Gbps effective | 1250 MHz 5 Gbps effective |
| GPU clock | — | 745 MHz |
Memory Configuration
| Memory size | 4 GB | 256 MB |
|---|---|---|
| Memory type | GDDR5 | GDDR5 |
| Memory bus width | 128 bit | 256 bit |
| Bandwidth | 112.0 GB/s | 160.0 GB/s |
Render Configuration
| Shading units | 896 | 1536 |
|---|---|---|
| TMUs | 56 | 128 |
| ROPs | 16 | 32 |
| Compute units | 14 | — |
| Cache L1 | 16 KB (per CU) | 16 KB (per SMX) |
| Cache L2 | 1024 KB | 512 KB |
| SMX count | — | 8 |
Performance
| Pixel rate | 18.80 GPixel/s | 23.84 GPixel/s |
|---|---|---|
| Texture rate | 65.80 GTexel/s | 95.36 GTexel/s |
| FP16 (half) performance | 2.106 TFLOPS (1:1) | — |
| FP32 (float) performance | 2.106 TFLOPS | 2.289 TFLOPS |
| FP64 (double) performance | 131.6 GFLOPS (1:16) | 95.36 GFLOPS (1:24) |
Dimensions & Outputs
| Slot width | Dual-slot | IGP |
|---|---|---|
| Length | 170 mm 6.7 inches | — |
| TDP | 45 W | 225 W |
| Suggested PSU | 200 W | 550 W |
| Outputs | 1x DVI1x HDMI1x DisplayPort | No outputs |
| Power connectors | None | — |
| Board number | C981-21 C994-71 | — |
API Support & Features
| DirectX | 12 (12_0) | 12 (11_0) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.2 | 1.1 |
| Shader model | 6.4 | 5.1 |
| CUDA | — | 3.0 |
