AMD Radeon Sky 900 vs NVIDIA Tesla P100 DGXS
Graphics Processor
| GPU model | Tahiti | GP100 |
|---|---|---|
| GPU variant | Orthrus | — |
| Architecture | GCN 1.0 | Pascal |
| Foundry | TSMC | TSMC |
| Manufacturing process | 28 nm | 16 nm |
| Number of transistors | 4,313 million | 15,300 million |
| Die size | 352 mm² | 610 mm² |
Graphics Card
| Release date | Mar 27th, 2013 | Apr 5th, 2016 |
|---|---|---|
| Generation | Radeon Sky | Tesla |
| Production | End-of-life | End-of-life |
| Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Clocks
| Base clock | 825 MHz | 1328 MHz |
|---|---|---|
| Boost clock | 950 MHz | 1480 MHz |
| Memory clock | 1250 MHz 5 Gbps effective | 715 MHz 1430 Mbps effective |
Memory Configuration
| Memory size | 3 GB | 16 GB |
|---|---|---|
| Memory type | GDDR5 | HBM2 |
| Memory bus width | 384 bit | 4096 bit |
| Bandwidth | 240.0 GB/s | 732.2 GB/s |
Render Configuration
| Shading units | 1792 | 3584 |
|---|---|---|
| TMUs | 112 | 224 |
| ROPs | 32 | 96 |
| Compute units | 28 | — |
| Cache L1 | 16 KB (per CU) | 24 KB (per SM) |
| Cache L2 | 768 KB | 4 MB |
| SM count | — | 56 |
Performance
| Pixel rate | 30.40 GPixel/s | 142.1 GPixel/s |
|---|---|---|
| Texture rate | 106.4 GTexel/s | 331.5 GTexel/s |
| FP32 (float) performance | 3.405 TFLOPS | 10.61 TFLOPS |
| FP64 (double) performance | 851.2 GFLOPS (1:4) | 5.304 TFLOPS (1:2) |
| FP16 (half) performance | — | 21.22 TFLOPS (2:1) |
Dimensions & Outputs
| Slot width | Dual-slot | — |
|---|---|---|
| Length | 305 mm 12 inches | — |
| Width | 111 mm 4.4 inches | — |
| TDP | 300 W | 300 W |
| Suggested PSU | 700 W | 700 W |
| Outputs | 1x DVI1x mini-DisplayPort | No outputs |
| Power connectors | 2x 8-pin | None |
| Board number | C476 | — |
API Support & Features
| DirectX | 12 (11_1) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 1.2 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| Shader model | 5.1 | 6.4 |
| CUDA | — | 6.0 |
