GeForce RTX 4090 vs Apple M4 Pro GPU (20-core)
VS
Nvidia GeForce RTX 4090
AD102
Apple M4 Pro GPU (20-core)
Custom
We performed a head-to-head comparison of the GeForce RTX 4090 24 GB with 128 pipelines and 16384 shaders against the 2 years and 2 months newer Apple M4 Pro GPU (20-core) that utilizes 320 pipelines and 2560 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
DeviceFacts Final Score
Overall video card score
The "Energy Efficiency" metric has less impact on the DeviceFacts Score for desktop GPU.
Key differences
Reasons to consider the GeForce RTX 4090
- Performs significantly better (up to 5.2x) in 3DMark Steel Nomad Lite
- 10.2x higher maximum theoretical performance (82.6 vs 8.1 TFLOPS)
- Achieves 4.5x more points in the GeekBench 6 Compute test (316K vs 70K)
- Has 3.7x higher memory bandwidth: 1010 vs 273 GB/s
- Has 6.4x more shading units (16384 vs 2560)
- Supports Nvidia DLSS 3 technology
Reasons to consider the Apple M4 Pro GPU (20-core)
- Manufactured using a more efficient 3 nm process technology
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 275 199 193 155 |
- |
| The Witcher 3 | 292 268 191 101 |
- |
| Counter-Strike 2 | 339 238 188 90 |
- |
| Far Cry 6 | 203 179 170 104 |
- |
| Hogwarts Legacy | 200 163 132 74 |
- |
| CoD: Modern Warfare III | 266 263 219 151 |
- |
| Ghost of Tsushima | 188 152 137 91 |
- |
| Cyberpunk 2077 | 226 208 143 66 |
- |
| Shadow of the Tomb Raider | 302 284 222 118 |
- |
| 1080p High | 255 | - |
| 1080p Ultra | 217 | - |
| 1440p Ultra | 177 | - |
| 4K Ultra | 106 | - |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 36299 | - |
| Solar Bay | 187525 | 33418 |
| Port Royal | 26127 | - |
| Fire Strike | 72432 | - |
| Wild Life Extreme | 85218 | 19535 |
| Night Raid | 195813 | - |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 318.9 img/sec | 125.1 img/sec |
| Face Detection | 222.8 img/sec | 80 img/sec |
| Horizon Detection | 14.2 Gpixels/sec | 2.88 Gpixels/sec |
| Edge Detection | 21 Gpixels/sec | 4.22 Gpixels/sec |
| Gaussian Blur | 23.9 Gpixels/sec | 2.98 Gpixels/sec |
| Feature Matching | 2.38 Gpixels/sec | 0.8 Gpixels/sec |
| Stereo Matching | 1950 Gpixels/sec | 235 Gpixels/sec |
| Particle Physics | 48090.4 FPS | 9516.1 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1303 | - |
| DirectX 11 | 323 FPS | - |
| DirectX 12 | 151 FPS | - |
| GPU Compute | 25991 Ops/s | - |
Blender
Blender GPU
Recent User Tests
GeForce RTX 4090
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-10-06 (Devansh) | Geekbench 6 OpenCL | 394674 |
| 📘 2026-10-06 (Devansh) | Geekbench 7 OpenCL | 338129 |
| 📘 2026-04-03 (afroman420IU) | Cinebench 2024 | 32578 |
Apple M4 Pro GPU (20-core)
No benchmark results yet
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) | 15600 | 8046 |
| Image Segmentation (HP) | 36657 | 15443 |
| Image Super Resolution (Q) | 48114 | 20807 |
| Face Detection (HP) | 78111 | 30860 |
| Pose Estimation (Q) | 289639 | 56783 |
| Text Classification (SP) | 4074 | 2888 |
| Machine Translation (HP) | 6349 | 6167 |
| Object Detection (SP) | 21186 | 7592 |
| Depth Estimation (Q) | 75599 | 33607 |
| Style Transfer (SP) | 601428 | 126726 |
| Framework | ONNX | Core ML |
| Backend | DirectML | GPU |
Results were compared using different ML frameworks
Specifications
Technical specifications of GeForce RTX 4090 and Apple M4 Pro GPU (20-core)General
| Vendor | Nvidia | Apple |
| Build | Discrete | Integrated |
| Released | September 20, 2022 | October 29, 2024 |
| Launch price (MSRP) | $1599 | - |
| Case | Desktop | Laptop |
| Purpose | Gaming | Professional |
| Segment | High-end | Mid-range |
| Architecture | Ada Lovelace | Apple M GPU |
| GPU Codename | AD102 | Custom |
| Rival Equivalent | - | - GeForce RTX 4060 Laptop |
| Successor | - GeForce RTX 5090 | - Apple M5 Pro GPU (20-core) |
| Recommended CPU | - Intel Core i9 14900K or above | - Apple M4 Pro (14-Core) or above |
| Used in CPUs | - | - Apple M4 Pro (14-Core) |
Desktop GPU rating (#3rd place)
Laptop GPU ranking (#38th place)
Graphics Processing Unit
| Base Clock | 2235 MHz | 500 MHz |
| Boost Clock | 2520 MHz | 1578 MHz |
| Shading Units | 16384 | 2560 |
| Texture Mapping Units (TMUs) | 512 | 160 |
| Render Output Units (ROPs) | 176 | 80 |
| Compute Units (Pipelines) | 128 | 320 |
| Tensor Cores | 512 | - |
| Ray-tracing Cores | 128 | - |
| L1 Cache | 128KB per cluster | - |
| L2 Cache | 72MB shared | - |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 444 GPixel/s | 126 GPixel/s |
| Texture Fill Rate | 1290 GTexel/s | 252 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x16 | Custom |
| TGP | 450 W | 26 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 5 nm | 3 nm |
| Die Size | 609 mm² | - |
| Transistor Count | 76.3 billion | - |
| Transistor Density | 125.29 MTr/mm² | - |
| Max. Temperature | - | 100°C |
Memory
| Memory Type | GDDR6X | System Shared |
| Memory Size | 24 GB | - |
| Memory Clock | 2625 MHz | 8533 MHz |
| Effective Memory Speed | 21000 Mbps | 15000 Mbps |
| Bus | 384-bit | 192-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12 | - |
| Vulkan | 1.3 | - |
| OpenGL | 4.6 | - |
| OpenCL | 3.0 | - |
| CUDA | 8.9 | - |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 3 | No |
| DisplayPort | 1.4a | - |
Cast your vote
Total votes: 13