Quick/Comparisons
The GeForce RTX 4090 outperforms the Apple M3 Max GPU (40-core) in gaming due to its significantly higher score of 92. Its excellent gaming performance makes it the clear winner in this category.
The GeForce RTX 4090 takes the lead in content creation with a score of 91, outpacing the Apple M3 Max GPU (40-core) by a considerable margin. Its strong workstation performance makes it the top choice for this use case.
The GeForce RTX 4090 excels in AI/ML compute with a score of 76, leaving the Apple M3 Max GPU (40-core) behind with a score of 39. Its capable AI/ML performance makes it the winner in this category.
33 | Gaming | 92 |
42 | Workstation | 91 |
39 | AI/ML | 76 |
82 | Energy Efficiency | 43 |
44 | Quick Comparison Final Score | 89 |
| — | Graphics Processor | AD102 |
| 5120 | Cores | 16384 |
| 320 / 160 | TMUs / ROPs | 512 / 176 |
| 3797.82 | Blender GPU | 11688.48 |
| — | Forza Horizon 5 | 203 |
| — | The Witcher 3 | 270 |
| — | Counter-Strike 2 | 245 |
| — | Far Cry 6 | 183 |
| — | Hogwarts Legacy | 165 |
| — | Call of Duty | 269 |
| — | Ghost of Tsushima | 150 |
| — | Cyberpunk 2077 | 204 |
| — | Tomb Raider | 283 |
| — | Average FPS | 219 |
| — | 1080p High | 256 |
| — | 1080p Ultra | 219 |
| — | 1440p Ultra | 176 |
| — | 4K Ultra | 107 |
| — | Margin of Error | Low |
| 94528 | GB6 Compute Score | 316301 |
| 157.6 img/sec | Background Blur | 318.9 img/sec |
| 100.9 img/sec | Face Detection | 222.8 img/sec |
| 3.99 Gpixels/sec | Horizon Detection | 14.2 Gpixels/sec |
| 6.04 Gpixels/sec | Edge Detection | 21 Gpixels/sec |
| 5 Gpixels/sec | Gaussian Blur | 23.9 Gpixels/sec |
| 0.84 Gpixels/sec | Feature Matching | 2.38 Gpixels/sec |
| 363.8 Gpixels/sec | Stereo Matching | 1950 Gpixels/sec |
| 12226.9 FPS | Particle Physics | 48090.4 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 38071 |
| — | G2D Mark | 1303 |
| — | DirectX 11 | 324 FPS |
| — | DirectX 12 | 151 FPS |
| — | GPU Compute | 26193 Ops/s |
| 20068 | GB6 ML Single Precision | 42962 |
| 22390 | GB6 ML Half Precision | 59636 |
| 20681 | GB6 ML Quantized | 31670 |
| 8685 | Image Classification (SP) | 15600 |
| 20739 | Image Segmentation (HP) | 36657 |
| 22775 | Image Super Resolution (Q) | 48114 |
| 37676 | Face Detection (HP) | 78111 |
| 91609 | Pose Estimation (Q) | 289639 |
| 2909 | Text Classification (SP) | 4074 |
| 5274 | Machine Translation (HP) | 6349 |
| 7480 | Object Detection (SP) | 21186 |
| 38212 | Depth Estimation (Q) | 75599 |
| 190681 | Style Transfer (SP) | 601428 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 221 GPixel/s | Pixel Fill Rate | 444 GPixel/s |
| 442 GTexel/s | Texture Fill Rate | 1290 GTexel/s |
| 14.1 TFLOPS | FLOPS (FP32) | 82.6 TFLOPS |
| System Shared | Memory Type | GDDR6X |
| — | Memory Size | 24 GB |
| 6400 MHz | Memory Clock | 2625 MHz |
| 12800 Mbps | Effective Memory Speed | 21000 Mbps |
| 512-bit | Bus | 384-bit |
| No | ECC | No |
| 409.6 GB/s | Memory Bandwidth | 1010 GB/s |
| Custom | Interface | PCIe 4.0 x16 |
| 60 W | TGP | 450 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 5 nm |
| — | Die Size | 609 mm² |
| 56 billion | Transistor Count | 76.3 billion |
| — | Transistor Density | 125.29 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 3 |
| — | CUDA | 8.9 |
| Yes | Ray Tracing | Yes |
| No | DLSS | DLSS 3 |
| — | DisplayPort | 1.4a |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| October 31, 2023 | Released | September 20, 2022 |
| Laptop | Case | Desktop |
| Professional | Purpose | Gaming |
| High-end | Segment | High-end |
| Apple M GPU | Architecture | Ada Lovelace |
| — | GPU Codename | AD102 |
| -Apple M3 Maxor above | Recommended CPU | -Intel Core i9 14900Kor above |
| 12391 | Steel Nomad Lite Score | 42169 |
| — | Time Spy | 36311 |
| 49656 | Solar Bay | 187386 |
| — | Port Royal | 26128 |
| — | Fire Strike | 72387 |
| 30787 | Wild Life Extreme | 85230 |
| — | Night Raid | 195880 |
| 500 MHz | Base Clock | 2235 MHz |
| 1380 MHz | Boost Clock | 2520 MHz |
| 5120 | Shading Units | 16384 |
| 320 | Texture Mapping Units (TMUs) | 512 |
| 160 | Render Output Units (ROPs) | 176 |
| 640 | Compute Units (Pipelines) | 128 |
| — | Tensor Cores | 512 |
| — | Ray-tracing Cores | 128 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 72MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |
Compare with something else?
Compare Apple M3 Max with another →