AMD Radeon Instinct MI250 vs NVIDIA Quadro K5000 SYNC
Graphics Processor
| GPU model | Aldebaran | GK104 |
|---|---|---|
| GPU variant | Aldebaran | — |
| Architecture | CDNA 2.0 | Kepler |
| Foundry | TSMC | TSMC |
| Manufacturing process | 6 nm | 28 nm |
| Number of transistors | 58,200 million | 3,540 million |
| Die size | unknown | 294 mm² |
Graphics Card
| Release date | Nov 8th, 2021 | Jan 25th, 2013 |
|---|---|---|
| Generation | Radeon Instinct | Quadro |
| Production | Active | End-of-life |
| Interface | PCIe 4.0 x16 | PCIe 2.0 x16 |
| Launch price | — | 3,299 USD |
| Reviews | — | 26 in our database |
Clocks
| Base clock | 1000 MHz | — |
|---|---|---|
| Boost clock | 1700 MHz | — |
| Memory clock | 1600 MHz 3.2 Gbps effective | 1350 MHz 5.4 Gbps effective |
| GPU clock | — | 706 MHz |
Memory Configuration
| Memory size | 128 GB | 4 GB |
|---|---|---|
| Memory type | HBM2e | GDDR5 |
| Memory bus width | 8192 bit | 256 bit |
| Bandwidth | 3,277 GB/s | 172.8 GB/s |
Render Configuration
| Shading units | 13312 | 1536 |
|---|---|---|
| TMUs | 832 | 128 |
| ROPs | 0 | 32 |
| Compute units | 208 | — |
| Cache L1 | 16 KB (per CU) | 16 KB (per SMX) |
| Cache L2 | 16 MB | 512 KB |
| gpu.details.mcm | 2 | — |
| SMX count | — | 8 |
Performance
| Pixel rate | 0 MPixel/s | 22.59 GPixel/s |
|---|---|---|
| Texture rate | 1,414 GTexel/s | 90.37 GTexel/s |
| FP16 (half) performance | 362.1 TFLOPS (8:1) | — |
| FP32 (float) performance | 45.26 TFLOPS | 2.169 TFLOPS |
| FP64 (double) performance | 45.26 TFLOPS (1:1) | 90.37 GFLOPS (1:24) |
Dimensions & Outputs
| Slot width | Dual-slot | Triple-slot |
|---|---|---|
| Length | 267 mm 10.5 inches | 267 mm 10.5 inches |
| Width | 111 mm 4.4 inches | 111 mm 4.4 inches |
| TDP | 500 W | 122 W |
| Suggested PSU | 900 W | 300 W |
| Outputs | No outputs | 2x DVI2x DisplayPort1x SDI |
| Power connectors | 2x 8-pin | 2x 6-pin |
API Support & Features
| DirectX | — | 12 (11_0) |
|---|---|---|
| OpenGL | — | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | — | 1.1 |
| Shader model | — | 5.1 |
| CUDA | — | 3.0 |
