Quick/Comparisons
The GeForce RTX 4090's exceptional gaming performance is driven by its 24GB GDDR6X memory and Ada Lovelace architecture, outpacing the Radeon RX 6800 by a significant margin.
The GeForce RTX 4090's strong workstation performance and 24GB GDDR6X memory make it the better choice for content creation tasks, outperforming the Radeon RX 6800.
The GeForce RTX 4090's capable AI/ML compute capabilities and 24GB GDDR6X memory give it an edge over the Radeon RX 6800 in AI and machine learning tasks.
92 | Gaming | 45 |
91 | Workstation | 47 |
76 | AI/ML | 44 |
43 | Energy Efficiency | 36 |
89 | Quick Comparison Final Score | 45 |
| AD102 | Graphics Processor | Navi 21 |
| 16384 | Cores | 3840 |
| 512 / 176 | TMUs / ROPs | 240 / 96 |
| 11688.48 | Blender GPU | 1894.91 |
| 203 | Forza Horizon 5 | 116 |
| 270 | The Witcher 3 | 152 |
| 245 | Counter-Strike 2 | 186 |
| 183 | Far Cry 6 | 125 |
| 165 | Hogwarts Legacy | 88 |
| 269 | Call of Duty | 134 |
| 150 | Ghost of Tsushima | 90 |
| 204 | Cyberpunk 2077 | 98 |
| 283 | Tomb Raider | 168 |
| 219 | Average FPS | 129 |
| 256 | 1080p High | 157 |
| 219 | 1080p Ultra | 129 |
| 176 | 1440p Ultra | 95 |
| 107 | 4K Ultra | 55 |
| Low | Margin of Error | Low |
| 2.38 Gpixels/sec | Feature Matching | 1.16 Gpixels/sec |
| 1950 Gpixels/sec | Stereo Matching | 579 Gpixels/sec |
| 48090.4 FPS | Particle Physics | 18755.1 FPS |
| OpenCL | API | OpenCL |
| 316301 | GB6 Compute Score | 135060 |
| 318.9 img/sec | Background Blur | 298.6 img/sec |
| 222.8 img/sec | Face Detection | 186.2 img/sec |
| 14.2 Gpixels/sec | Horizon Detection | 4.75 Gpixels/sec |
| 21 Gpixels/sec | Edge Detection | 6.68 Gpixels/sec |
| 23.9 Gpixels/sec | Gaussian Blur | 5.58 Gpixels/sec |
| 38071 | G3D Mark Score | 22023 |
| 1303 | G2D Mark | 998 |
| 324 FPS | DirectX 11 | 212 FPS |
| 151 FPS | DirectX 12 | 89 FPS |
| 26193 Ops/s | GPU Compute | 10764 Ops/s |
| 42962 | GB6 ML Single Precision | 25097 |
| 59636 | GB6 ML Half Precision | 27988 |
| 31670 | GB6 ML Quantized | 19116 |
| 15600 | Image Classification (SP) | 11245 |
| 36657 | Image Segmentation (HP) | 13617 |
| 48114 | Image Super Resolution (Q) | 27747 |
| 78111 | Face Detection (HP) | 38124 |
| 289639 | Pose Estimation (Q) | 119493 |
| 4074 | Text Classification (SP) | 2425 |
| 6349 | Machine Translation (HP) | 5228 |
| 21186 | Object Detection (SP) | 11986 |
| 75599 | Depth Estimation (Q) | 46083 |
| 601428 | Style Transfer (SP) | 281860 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 444 GPixel/s | Pixel Fill Rate | 202 GPixel/s |
| 1290 GTexel/s | Texture Fill Rate | 505 GTexel/s |
| 82.6 TFLOPS | FLOPS (FP32) | 16.2 TFLOPS |
| GDDR6X | Memory Type | GDDR6 |
| 24 GB | Memory Size | 16 GB |
| 2625 MHz | Memory Clock | 2000 MHz |
| 21000 Mbps | Effective Memory Speed | 16000 Mbps |
| 384-bit | Bus | 256-bit |
| No | ECC | No |
| 1010 GB/s | Memory Bandwidth | 512 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 450 W | TGP | 250 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 7 nm |
| 609 mm² | Die Size | 520 mm² |
| 76.3 billion | Transistor Count | 26 billion |
| 125.29 MTr/mm² | Transistor Density | 50 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 8.9 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | 1.4a |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| September 20, 2022 | Released | October 28, 2020 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Ada Lovelace | Architecture | RDNA 2.0 |
| AD102 | GPU Codename | Navi 21 |
| -Intel Core i9 14900Kor above | Recommended CPU | -Intel Core i5 12600or above |
| 42169 | Steel Nomad Lite Score | 14548 |
| 36311 | Time Spy | 16220 |
| 187386 | Solar Bay | 64755 |
| 26128 | Port Royal | 8137 |
| 72387 | Fire Strike | 44377 |
| 85230 | Wild Life Extreme | 29764 |
| 195880 | Night Raid | 142721 |
| 2235 MHz | Base Clock | 1700 MHz |
| 2520 MHz | Boost Clock | 2105 MHz |
| 16384 | Shading Units | 3840 |
| 512 | Texture Mapping Units (TMUs) | 240 |
| 176 | Render Output Units (ROPs) | 96 |
| 128 | Compute Units (Pipelines) | — |
| 512 | Tensor Cores | No |
| 128 | Ray-tracing Cores | 60 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 72MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |
Compare with something else?
Compare GeForce RTX 4090 with another →