AMD Radeon RX 550X vs NVIDIA Tesla P100 PCIe 12 GB
Graphics Processor
| GPU model | Lexa | GP100 |
|---|---|---|
| GPU variant | Lexa PRO (215-0904018) | — |
| Architecture | GCN 4.0 | Pascal |
| Foundry | GlobalFoundries | TSMC |
| Manufacturing process | 14 nm | 16 nm |
| Number of transistors | 2,200 million | 15,300 million |
| Die size | 103 mm² | 610 mm² |
Graphics Card
| Release date | Dec 16th, 2018 | Jun 20th, 2016 |
|---|---|---|
| Generation | Polaris | Tesla |
| Production | End-of-life | End-of-life |
| Interface | PCIe 3.0 x8 | PCIe 3.0 x16 |
| Predecessor | Polaris | — |
| Successor | Vega | — |
| Launch price | — | 4,599 USD |
Clocks
| Base clock | 1100 MHz | 1190 MHz |
|---|---|---|
| Boost clock | 1183 MHz | 1329 MHz |
| Memory clock | 1500 MHz 6 Gbps effective | 715 MHz 1430 Mbps effective |
Memory Configuration
| Memory size | 4 GB | 12 GB |
|---|---|---|
| Memory type | GDDR5 | HBM2 |
| Memory bus width | 128 bit | 3072 bit |
| Bandwidth | 96.00 GB/s | 549.1 GB/s |
Render Configuration
| Shading units | 512 | 3584 |
|---|---|---|
| TMUs | 32 | 224 |
| ROPs | 16 | 96 |
| Compute units | 8 | — |
| Cache L1 | 16 KB (per CU) | 24 KB (per SM) |
| Cache L2 | 512 KB | 3 MB |
| SM count | — | 56 |
Performance
| Pixel rate | 18.93 GPixel/s | 127.6 GPixel/s |
|---|---|---|
| Texture rate | 37.86 GTexel/s | 297.7 GTexel/s |
| FP16 (half) performance | 1,211 GFLOPS (1:1) | 19.05 TFLOPS (2:1) |
| FP32 (float) performance | 1,211 GFLOPS | 9.526 TFLOPS |
| FP64 (double) performance | 75.71 GFLOPS (1:16) | 4.763 TFLOPS (1:2) |
Dimensions & Outputs
| Slot width | Dual-slot | Dual-slot |
|---|---|---|
| Length | 145 mm 5.7 inches | 267 mm 10.5 inches |
| TDP | 50 W | 250 W |
| Suggested PSU | 250 W | 600 W |
| Outputs | 1x DVI1x HDMI1x DisplayPort | No outputs |
| Power connectors | None | 1x 8-pin |
| Board number | D090 | — |
API Support & Features
| DirectX | 12 (12_0) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| Shader model | 6.4 | 6.4 |
| CUDA | — | 6.0 |
