AMD Radeon Vega Frontier Edition Watercooled vs NVIDIA Quadro P620
Graphics Processor
| GPU model | Vega 10 | GP107 |
|---|---|---|
| GPU variant | Vega 10 XTX LCS | — |
| Architecture | GCN 5.0 | Pascal |
| Foundry | GlobalFoundries | Samsung |
| Manufacturing process | 14 nm | 14 nm |
| Number of transistors | 12,500 million | 3,300 million |
| Die size | 495 mm² | 132 mm² |
Graphics Card
| Release date | Jul 13th, 2017 | Feb 1st, 2018 |
|---|---|---|
| Generation | Radeon Pro | Quadro |
| Production | Active | End-of-life |
| Launch price | 1,489 USD | — |
| Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Clocks
| Base clock | 1382 MHz | 1266 MHz |
|---|---|---|
| Boost clock | 1600 MHz | 1354 MHz |
| Memory clock | 945 MHz 1890 Mbps effective | 1252 MHz 5 Gbps effective |
Memory Configuration
| Memory size | 16 GB | 2 GB |
|---|---|---|
| Memory type | HBM2 | GDDR5 |
| Memory bus width | 2048 bit | 128 bit |
| Bandwidth | 483.8 GB/s | 80.13 GB/s |
Render Configuration
| Shading units | 4096 | 512 |
|---|---|---|
| TMUs | 256 | 32 |
| ROPs | 64 | 16 |
| Compute units | 64 | — |
| Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
| Cache L2 | 4 MB | 1024 KB |
| SM count | — | 4 |
Performance
| Pixel rate | 102.4 GPixel/s | 21.66 GPixel/s |
|---|---|---|
| Texture rate | 409.6 GTexel/s | 43.33 GTexel/s |
| FP16 (half) performance | 26.21 TFLOPS (2:1) | 21.66 GFLOPS (1:64) |
| FP32 (float) performance | 13.11 TFLOPS | 1,386 GFLOPS |
| FP64 (double) performance | 819.2 GFLOPS (1:16) | 43.33 GFLOPS (1:32) |
Dimensions & Outputs
| Slot width | Dual-slot | Single-slot |
|---|---|---|
| Length | 267 mm 10.5 inches | 145 mm 5.7 inches |
| Width | 111 mm 4.4 inches | 69 mm 2.7 inches |
| TDP | 375 W | 40 W |
| Suggested PSU | 750 W | 200 W |
| Outputs | 1x HDMI3x DisplayPort | 4x mini-DisplayPort |
| Power connectors | 2x 8-pin | None |
| Board number | D050-12 | — |
API Support & Features
| DirectX | 12 (12_1) | 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 | — | 6.1 |
