Quick/Comparisons
The Nvidia RTX 6000 Ada's superior gaming performance earns it the win, with a 47-point gap over the Apple M3 Max GPU (40-core).
The Nvidia RTX 6000 Ada's strong workstation performance and high score of 94 make it the clear winner in content creation tasks.
Although the Nvidia RTX 6000 Ada only leads by 36 points, its higher score in AI/ML tasks earns it the win.
33 | Gaming | 80 |
42 | Workstation | 94 |
39 | AI/ML | 75 |
82 | Energy Efficiency | 44 |
44 | Quick Comparison Final Score | 83 |
| — | Graphics Processor | AD102 |
| 5120 | Cores | 18176 |
| 320 / 160 | TMUs / ROPs | 568 / 192 |
| 3797.82 | Blender GPU | 11038.91 |
| — | Forza Horizon 5 | 165 |
| — | The Witcher 3 | 219 |
| — | Counter-Strike 2 | 221 |
| — | Far Cry 6 | 164 |
| — | Hogwarts Legacy | 133 |
| — | Call of Duty | 216 |
| — | Ghost of Tsushima | 123 |
| — | Cyberpunk 2077 | 172 |
| — | Tomb Raider | 267 |
| — | Average FPS | 187 |
| — | 1080p High | 219 |
| — | 1080p Ultra | 187 |
| — | 1440p Ultra | 148 |
| — | 4K Ultra | 89 |
| — | Margin of Error | High |
| 94528 | GB6 Compute Score | 389848 |
| 157.6 img/sec | Background Blur | 618.3 img/sec |
| 100.9 img/sec | Face Detection | 399.9 img/sec |
| 3.99 Gpixels/sec | Horizon Detection | 15.1 Gpixels/sec |
| 6.04 Gpixels/sec | Edge Detection | 22.1 Gpixels/sec |
| 5 Gpixels/sec | Gaussian Blur | 26.3 Gpixels/sec |
| 0.84 Gpixels/sec | Feature Matching | 2.33 Gpixels/sec |
| 363.8 Gpixels/sec | Stereo Matching | 2210 Gpixels/sec |
| 12226.9 FPS | Particle Physics | 53494.5 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 28613 |
| — | G2D Mark | 1005 |
| — | DirectX 11 | 284 FPS |
| — | DirectX 12 | 102 FPS |
| — | GPU Compute | 22439 Ops/s |
| 20068 | GB6 ML Single Precision | 37270 |
| 22390 | GB6 ML Half Precision | 53469 |
| 20681 | GB6 ML Quantized | 27850 |
| 8685 | Image Classification (SP) | 13367 |
| 20739 | Image Segmentation (HP) | 32477 |
| 22775 | Image Super Resolution (Q) | 42056 |
| 37676 | Face Detection (HP) | 69004 |
| 91609 | Pose Estimation (Q) | 230965 |
| 2909 | Text Classification (SP) | 3491 |
| 5274 | Machine Translation (HP) | 5545 |
| 7480 | Object Detection (SP) | 18668 |
| 38212 | Depth Estimation (Q) | 68370 |
| 190681 | Style Transfer (SP) | 520389 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 221 GPixel/s | Pixel Fill Rate | 481 GPixel/s |
| 442 GTexel/s | Texture Fill Rate | 1423 GTexel/s |
| 14.1 TFLOPS | FLOPS (FP32) | 91.1 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 48 GB |
| 6400 MHz | Memory Clock | 2500 MHz |
| 12800 Mbps | Effective Memory Speed | 20000 Mbps |
| 512-bit | Bus | 384-bit |
| No | ECC | No |
| 409.6 GB/s | Memory Bandwidth | 960 GB/s |
| Custom | Interface | PCIe 4.0 x16 |
| 60 W | TGP | 300 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 5 nm |
| — | Die Size | 609 mm² |
| 56 billion | Transistor Count | 76 billion |
| — | Transistor Density | 124.79 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 | December 3, 2022 |
| Laptop | Case | Desktop |
| Professional | Purpose | Professional |
| 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 | 32284 |
| — | Time Spy | 26686 |
| 49656 | Solar Bay | 145429 |
| — | Port Royal | 19557 |
| — | Fire Strike | 62803 |
| 30787 | Wild Life Extreme | 53640 |
| — | Night Raid | 148225 |
| 500 MHz | Base Clock | 915 MHz |
| 1380 MHz | Boost Clock | 2505 MHz |
| 5120 | Shading Units | 18176 |
| 320 | Texture Mapping Units (TMUs) | 568 |
| 160 | Render Output Units (ROPs) | 192 |
| 640 | Compute Units (Pipelines) | 142 |
| — | Tensor Cores | 568 |
| — | Ray-tracing Cores | 142 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 96MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |
Compare with something else?
Compare Apple M3 Max with another →