NVIDIA GeForce RTX 3090 SUPER vs AMD Radeon Pro W6900X
What is the difference between NVIDIA GeForce RTX 3090 SUPER and AMD Radeon Pro W6900X. Find out which graphics card has better performance.
Graphics Processor (GPU)
| GA102 | GPU Name | Navi 21 |
| GA102-350-A1 | Model number | Navi 21 Pro-XTA |
| Ampere | Architecture | RDNA 2.0 |
| Samsung | Foundry | TSMC |
| 8 nm | Process Size | 7 nm |
| 28,300 million | Transistors | 26,800 million |
| 628 mm² | Die Size | 520 mm² |
Graphics Card
| Unknown | Release Date | Aug 3rd, 2021 |
| GeForce 30 | Family | Radeon Pro Mac (Navi II Series) |
| Active | Production | Active |
| PCIe 4.0 x16 | Bus Interface | PCIe 4.0 x16 |
Memory
| 24 GB | Memory Size | 32 GB |
| GDDR6X | Memory Type | GDDR6 |
| 384 bit | Memory Bus | 256 bit |
| 1,018 GB/s | Bandwidth | 512.0 GB/s |
Performance
| 189.8 GPixel/s | Pixel fillrate | 275.2 GPixel/s |
| 569.5 GTexel/s | Texture fillrate | 688.0 GTexel/s |
| 36.45 TFLOPS (1:1) | FP16 (half) performance | 44.03 TFLOPS (2:1) |
| 36.45 TFLOPS | FP32 (float) performance | 22.02 TFLOPS |
| 569.5 GFLOPS (1:64) | FP64 (double) performance | 1,376 GFLOPS (1:16) |
Clock Speeds
| 1395 MHz | Base Clock | 1825 MHz |
| 1695 MHz | Boost Clock | 2150 MHz |
| 1325 MHz 21.2 Gbps effective | Memory Clock | 2000 MHz 16 Gbps effective |
Render Config
| 10752 | Shading Units | 5120 |
| 336 | TMUs | 320 |
| 112 | ROPs | 128 |
| 84 | RT Cores | 80 |
| 128 KB (per SM) | L1 Cache | 128 KB per Array |
| 6 MB | L2 Cache | 4 MB |
Board Design
| Triple-slot | Slot Width | Quad-slot |
| 336 mm 13.2 inches | Length | 267 mm 10.5 inches |
| 140 mm 5.5 inches | Width | 120 mm 4.7 inches |
| 400 W | Thermal design power (TDP) | 300 W |
| 800 W | Suggested PSU | 700 W |
| 1x HDMI 3x DisplayPort | Display Connectors | 1x HDMI 4x Thunderbolt |
| 1x 12-pin | Power Connectors | None |
API support
| 12 Ultimate (12_2) | DirectX | 12 Ultimate (12_2) |
| 4.6 | OpenGL | 4.6 |
| 3.0 | OpenCL | 2.1 |
| 1.2 | Vulkan | 1.2 |
| 6.6 | Shader Model | 6.5 |