AMD Radeon Instinct MI100 vs NVIDIA A100 PCIe
Graphics Processor
| GPU model | Arcturus | GA100 |
|---|---|---|
| GPU variant | Arcturus XL | — |
| Architecture | CDNA 1.0 | Ampere |
| Foundry | TSMC | TSMC |
| Manufacturing process | 7 nm | 7 nm |
| Number of transistors | 25,600 million | 54,200 million |
| Die size | 750 mm² | 826 mm² |
Graphics Card
| Release date | Nov 16th, 2020 | Jun 22nd, 2020 |
|---|---|---|
| Generation | Radeon Instinct | Tesla |
| Production | Active | Active |
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
Clocks
| Base clock | 1000 MHz | 765 MHz |
|---|---|---|
| Boost clock | 1502 MHz | 1410 MHz |
| SOC clock | 1091 MHz | — |
| UVD clock | 1403 MHz | — |
| Memory clock | 1200 MHz 2.4 Gbps effective | 1215 MHz 2.4 Gbps effective |
| GPU clock | — | 1410 MHz |
Memory Configuration
| Memory size | 32 GB | 40 GB |
|---|---|---|
| Memory type | HBM2 | HBM2E |
| Memory bus width | 4096 bit | 5120 bit |
| Bandwidth | 1,229 GB/s | 1,555 GB/s |
Render Configuration
| Shading units | 7680 | 6912 |
|---|---|---|
| TMUs | 480 | 432 |
| ROPs | 64 | 160 |
| Compute units | 120 | — |
| Cache L1 | 16 KB (per CU) | 192 KB (per SM) |
| Cache L2 | 8 MB | 40 MB |
| SM count | — | 108 |
| Tensor cores | — | 432 |
Performance
| Pixel rate | 96.13 GPixel/s | 225.6 GPixel/s |
|---|---|---|
| Texture rate | 721.0 GTexel/s | 609.1 GTexel/s |
| FP16 (half) performance | 184.6 TFLOPS (8:1) | 77.97 TFLOPS (4:1) |
| FP32 (float) performance | 23.07 TFLOPS | 19.49 TFLOPS |
| FP64 (double) performance | 11.54 TFLOPS (1:2) | 9.746 TFLOPS (1:2) |
Dimensions & Outputs
| Slot width | Dual-slot | Dual-slot |
|---|---|---|
| Length | 267 mm 10.5 inches | 267 mm 10.5 inches |
| Width | 111 mm 4.4 inches | — |
| TDP | 300 W | 250 W |
| Suggested PSU | 700 W | 600 W |
| Outputs | No outputs | No outputs |
| Power connectors | 2x 8-pin | 8-pin EPS |
| Board number | 109-D05187 | P1001 SKU 200 |
| gpu.details.bios-number | 113-D3430300-025 | — |
API Support & Features
| DirectX | — | — |
|---|---|---|
| OpenGL | — | — |
| OpenCL | 2.1 | 3.0 |
| Vulkan | — | — |
| Shader model | — | — |
| CUDA | — | 8.0 |
