NVIDIA A100 PCIe vs NVIDIA Quadro P5000
Graphics Processor
| GPU model | GA100 | GP104 |
|---|---|---|
| Architecture | Ampere | Pascal |
| Foundry | TSMC | TSMC |
| Manufacturing process | 7 nm | 16 nm |
| Number of transistors | 54,200 million | 7,200 million |
| Die size | 826 mm² | 314 mm² |
| GPU variant | — | GP104-875-A1 |
Graphics Card
| Release date | Jun 22nd, 2020 | Oct 1st, 2016 |
|---|---|---|
| Generation | Tesla | Quadro |
| Production | Active | End-of-life |
| Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
| Launch price | — | 2,499 USD |
Clocks
| GPU clock | 1410 MHz | — |
|---|---|---|
| Memory clock | 1215 MHz 2.4 Gbps effective | 1127 MHz 9 Gbps effective |
| Base clock | 765 MHz | 1607 MHz |
| Boost clock | 1410 MHz | 1733 MHz |
Memory Configuration
| Memory size | 40 GB | 16 GB |
|---|---|---|
| Memory type | HBM2E | GDDR5X |
| Memory bus width | 5120 bit | 256 bit |
| Bandwidth | 1,555 GB/s | 288.5 GB/s |
Render Configuration
| Shading units | 6912 | 2560 |
|---|---|---|
| TMUs | 432 | 160 |
| ROPs | 160 | 64 |
| SM count | 108 | 20 |
| Tensor cores | 432 | — |
| Cache L1 | 192 KB (per SM) | 48 KB (per SM) |
| Cache L2 | 40 MB | 2 MB |
Performance
| Pixel rate | 225.6 GPixel/s | 110.9 GPixel/s |
|---|---|---|
| Texture rate | 609.1 GTexel/s | 277.3 GTexel/s |
| FP16 (half) performance | 77.97 TFLOPS (4:1) | 138.6 GFLOPS (1:64) |
| FP32 (float) performance | 19.49 TFLOPS | 8.873 TFLOPS |
| FP64 (double) performance | 9.746 TFLOPS (1:2) | 277.3 GFLOPS (1:32) |
Dimensions & Outputs
| Slot width | Dual-slot | Dual-slot |
|---|---|---|
| Length | 267 mm 10.5 inches | 267 mm 10.5 inches |
| TDP | 250 W | 180 W |
| Suggested PSU | 600 W | 450 W |
| Outputs | No outputs | 1x DVI4x DisplayPort |
| Power connectors | 8-pin EPS | 1x 8-pin |
| Board number | P1001 SKU 200 | PG413 SKU 500 |
| Width | — | 111 mm 4.4 inches |
API Support & Features
| DirectX | — | 12 (12_1) |
|---|---|---|
| OpenGL | — | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | — | 1.2 |
| CUDA | 8.0 | 6.1 |
| Shader model | — | 6.4 |
