AMD Radeon RX 7900 XTX vs GeForce RTX 3090 Ti
VS
AMD Radeon RX 7900 XTX
Navi 31
We compared two discrete desktop gaming GPUs: the Radeon RX 7900 XTX with 6144 shaders against the 8 months older GeForce RTX 3090 Ti that utilizes 84 pipelines and 10752 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.
Review
Gaming
Performance in DirectX, OpenCL, and Vulkan games
Workstation
Perf. in 3D modeling, video editing and rendering apps
AI/ML
Capabilities for machine learning and AI-related tasks
Energy Efficiency
Power consumption efficiency in different scenarios
NanoReview Final Score
Overall video card score
Value for money (Beta)
Key differences
Reasons to consider the Radeon RX 7900 XTX
- Performs better (up to 23%) in 3DMark Steel Nomad Lite
- Manufactured using a more efficient 5 nm process technology
- 53% higher maximum theoretical performance (61.4 vs 40 TFLOPS)
- Shows 25% higher average frame rate in modern games at QHD resolution – 159 vs 127 FPS
Reasons to consider the GeForce RTX 3090 Ti
- Supports Nvidia DLSS 2 technology
- Has 75% more shading units (10752 vs 6144)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 253 175 156 122 |
200 136 126 99 |
| The Witcher 3 | 280 251 177 96 |
214 185 130 71 |
| Counter-Strike 2 | 333 227 166 85 |
286 213 154 85 |
| Far Cry 6 | 212 181 170 109 |
160 151 133 83 |
| Hogwarts Legacy | 167 143 113 64 |
141 116 93 52 |
| CoD: Modern Warfare III | 264 251 192 128 |
186 174 139 90 |
| Ghost of Tsushima | 168 140 122 77 |
129 105 89 57 |
| Cyberpunk 2077 | 203 181 126 59 |
158 140 98 44 |
| Shadow of the Tomb Raider | 302 274 208 112 |
254 237 183 101 |
| 1080p High | 242 | 192 |
| 1080p Ultra | 203 | 162 |
| 1440p Ultra | 159 | 127 |
| 4K Ultra | 95 | 76 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 30804 | 21855 |
| Solar Bay | 123985 | 109239 |
| Port Royal | 16759 | 14890 |
| Fire Strike | 73279 | 51935 |
| Wild Life Extreme | 58739 | 48861 |
| Night Raid | 183191 | 156537 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 423.9 img/sec | 362.7 img/sec |
| Face Detection | 270.8 img/sec | 227.9 img/sec |
| Horizon Detection | 9.23 Gpixels/sec | 11.3 Gpixels/sec |
| Edge Detection | 13.7 Gpixels/sec | 15.9 Gpixels/sec |
| Gaussian Blur | 11.1 Gpixels/sec | 12.1 Gpixels/sec |
| Feature Matching | 1.74 Gpixels/sec | 1.56 Gpixels/sec |
| Stereo Matching | 871 Gpixels/sec | 1080 Gpixels/sec |
| Particle Physics | 29891.2 FPS | 31024.6 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1289 | 1226 |
| DirectX 11 | 355 FPS | 238 FPS |
| DirectX 12 | 133 FPS | 121 FPS |
| GPU Compute | 17772 Ops/s | 16494 Ops/s |
Blender
Blender GPU
Artificial Intelligence Tests
Performance in machine learning and artificial intelligence tasksGeekBench 6 ML
GB6 ML Single Precision
GB6 ML Half Precision
GB6 ML Quantized
| Image Classification (SP) | 18608 | 10246 |
| Image Segmentation (HP) | 25385 | 22830 |
| Image Super Resolution (Q) | 45641 | 23532 |
| Face Detection (HP) | 71321 | 48657 |
| Pose Estimation (Q) | 257258 | 147241 |
| Text Classification (SP) | 5240 | 2917 |
| Machine Translation (HP) | 7425 | 4427 |
| Object Detection (SP) | 21335 | 13323 |
| Depth Estimation (Q) | 66140 | 40055 |
| Style Transfer (SP) | 483867 | 298683 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of Radeon RX 7900 XTX and GeForce RTX 3090 TiGeneral
| Vendor | Amd | Nvidia |
| Build | Discrete | Discrete |
| Released | November 3, 2022 | March 29, 2022 |
| Launch price (MSRP) | $999 | $1999 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | High-end | High-end |
| Architecture | RDNA 3.0 | Ampere |
| GPU Codename | Navi 31 | GA102 |
| Rival Equivalent | - GeForce RTX 4080 | - |
| Recommended CPU | - Intel Core i7 13700K or above | - Intel Core i9 12900K or above |
Desktop GPU rating (6th and 14th place)
Graphics Processing Unit
| Base Clock | 1929 MHz | 1560 MHz |
| Boost Clock | 2498 MHz | 1860 MHz |
| Shading Units | 6144 | 10752 |
| Texture Mapping Units (TMUs) | 384 | 336 |
| Render Output Units (ROPs) | 192 | 112 |
| Compute Units (Pipelines) | - | 84 |
| Tensor Cores | - | 336 |
| Ray-tracing Cores | 96 | 84 |
| L1 Cache | 256KB per cluster | 128KB per cluster |
| L2 Cache | 6MB shared | 6MB shared |
| Instructions Per Cycle | 4 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 480 GPixel/s | 208 GPixel/s |
| Texture Fill Rate | 959 GTexel/s | 625 GTexel/s |
FLOPS (FP32)
Differences in GPUs' instructions per clock make TFLOPS values not directly comparable.
Physical
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| TGP | 355 W | 450 W |
| Manufacturing | TSMC | Samsung |
| Fabrication Process | 5 nm | 8 nm |
| Die Size | 529 mm² | 628 mm² |
| Transistor Count | 57 billion | 28 billion |
| Transistor Density | 107.75 MTr/mm² | 44.59 MTr/mm² |
Memory
| Memory Type | GDDR6 | GDDR6X |
| Memory Size | 24 GB | 24 GB |
| Memory Clock | 2500 MHz | 1313 MHz |
| Effective Memory Speed | 20000 Mbps | 21000 Mbps |
| Bus | 384-bit | 384-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12 | 12 |
| Vulkan | 1.3 | 1.3 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.2 | 3.0 |
| CUDA | - | 8.6 |
| Ray Tracing | Yes | Yes |
| DLSS | No | DLSS 2 |
| DisplayPort | 2.1 | 1.4a |
Cast your vote
Total votes: 21