NVIDIA GeForce RTX 3090 vs NVIDIA A100 SXM4 40 GB
What is the difference between NVIDIA GeForce RTX 3090 and NVIDIA A100 SXM4 40 GB. Find out which graphics card has better performance.
Graphics Processor (GPU)
| GA102 | GPU Name | GA100 |
| Ampere | Architecture | Ampere |
| Samsung | Foundry | TSMC |
| 8 nm | Process Size | 7 nm |
| 28,300 million | Transistors | 54,200 million |
| 628 mm² | Die Size | 826 mm² |
Graphics Card
| Sep 1st, 2020 | Release Date | May 14th, 2020 |
| GeForce 30 | Family | Tesla (Axx) |
| Active | Production | Active |
| PCIe 4.0 x16 | Bus Interface | PCIe 4.0 x16 |
Memory
| 24 GB | Memory Size | 40 GB |
| GDDR6X | Memory Type | HBM2e |
| 384 bit | Memory Bus | 5120 bit |
| 936.2 GB/s | Bandwidth | 1,555 GB/s |
Performance
| 189.8 GPixel/s | Pixel fillrate | 225.6 GPixel/s |
| 556.0 GTexel/s | Texture fillrate | 609.1 GTexel/s |
| 35.58 TFLOPS (1:1) | FP16 (half) performance | 77.97 TFLOPS (4:1) |
| 35.58 TFLOPS | FP32 (float) performance | 19.49 TFLOPS |
| 556.0 GFLOPS (1:64) | FP64 (double) performance | 9.746 TFLOPS (1:2) |
Clock Speeds
| 1395 MHz | Base Clock | 1095 MHz |
| 1695 MHz | Boost Clock | 1410 MHz |
| 1219 MHz 19.5 Gbps effective | Memory Clock | 1215 MHz 2.4 Gbps effective |
Render Config
| 10496 | Shading Units | 6912 |
| 328 | TMUs | 432 |
| 112 | ROPs | 160 |
| 128 KB (per SM) | L1 Cache | 192 KB (per SM) |
| 6 MB | L2 Cache | 40 MB |
| 82 | SM Count | 108 |
| 328 | Tensor Cores | 432 |
Board Design
| Triple-slot | Slot Width | IGP |
| 350 W | Thermal design power (TDP) | 400 W |
| 750 W | Suggested PSU | 800 W |
| 1x HDMI 3x DisplayPort | Display Connectors | No outputs |
| 1x 12-pin | Power Connectors | None |
API support
| 12 Ultimate (12_2) | DirectX | N/A |
| 4.6 | OpenGL | N/A |
| 3.0 | OpenCL | 3.0 |
| 1.2 | Vulkan | N/A |
| 6.6 | Shader Model | N/A |
| 8.6 | CUDA | 8.0 |