GeForce RTX 5090 vs Apple M3 Ultra GPU (80-core)
VS
Nvidia GeForce RTX 5090
GB202-300
Apple M3 Ultra GPU (80-core)
Custom
We performed a head-to-head comparison of the GeForce RTX 5090 32 GB with 170 pipelines and 21760 shaders against the Apple M3 Ultra GPU (80-core) that utilizes 1280 pipelines and 10240 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 GeForce RTX 5090
- Performs significantly better (up to 2.7x) in 3DMark Steel Nomad Lite
- 3.7x higher maximum theoretical performance (104.8 vs 28.3 TFLOPS)
- Supports Nvidia DLSS 4 technology
- Achieves 2.8x more points in the GeekBench 6 Compute test (419K vs 147K)
- Has 2.2x higher memory bandwidth: 1792 vs 819.3 GB/s
- Has 2.1x more shading units (21760 vs 10240)
Reasons to consider the Apple M3 Ultra GPU (80-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 | 336 247 233 187 |
- |
| The Witcher 3 | 364 327 237 123 |
- |
| Counter-Strike 2 | 363 265 202 99 |
- |
| Far Cry 6 | 237 202 190 121 |
- |
| Hogwarts Legacy | 249 199 166 97 |
- |
| CoD: Modern Warfare III | 334 322 260 184 |
- |
| Ghost of Tsushima | 221 185 175 113 |
- |
| Cyberpunk 2077 | 261 233 160 77 |
- |
| Shadow of the Tomb Raider | 329 302 238 128 |
- |
| 1080p High | 299 | - |
| 1080p Ultra | 254 | - |
| 1440p Ultra | 207 | - |
| 4K Ultra | 125 | - |
| Margin of Error | Low | High |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 46873 | - |
| Solar Bay | 236689 | 63597 |
| Port Royal | 38202 | - |
| Fire Strike | 87569 | - |
| Wild Life Extreme | 108751 | 47814 |
| Night Raid | 206648 | - |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 333 img/sec | 204.5 img/sec |
| Face Detection | 298.5 img/sec | 126 img/sec |
| Horizon Detection | 21.7 Gpixels/sec | 6.53 Gpixels/sec |
| Edge Detection | 33.7 Gpixels/sec | 10.7 Gpixels/sec |
| Gaussian Blur | 34.5 Gpixels/sec | 9.5 Gpixels/sec |
| Feature Matching | 2.53 Gpixels/sec | 1.05 Gpixels/sec |
| Stereo Matching | 2850 Gpixels/sec | 560.3 Gpixels/sec |
| Particle Physics | 59146.9 FPS | 25267.8 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1410 | - |
| DirectX 11 | 341 FPS | - |
| DirectX 12 | 179 FPS | - |
| GPU Compute | 24360 Ops/s | - |
Blender
Blender GPU
Recent User Tests
GeForce RTX 5090
No benchmark results yet
Apple M3 Ultra GPU (80-core)
| Date | Benchmark | Result |
|---|---|---|
| 📘 2025-03-16 (Tyler) | GeekBench | 145706 |
| 📘 2025-03-16 (Tyler) | Cinebench 2024 | 19448 |
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) | 18948 | 8094 |
| Image Segmentation (HP) | 55109 | 24595 |
| Image Super Resolution (Q) | 58478 | 25190 |
| Face Detection (HP) | 114685 | 38337 |
| Pose Estimation (Q) | 344729 | 111021 |
| Text Classification (SP) | 4464 | 2786 |
| Machine Translation (HP) | 8184 | 4339 |
| Object Detection (SP) | 27802 | 7668 |
| Depth Estimation (Q) | 85572 | 37488 |
| Style Transfer (SP) | 687080 | 244483 |
| Framework | ONNX | Core ML |
| Backend | DirectML | GPU |
Results were compared using different ML frameworks
Specifications
Technical specifications of GeForce RTX 5090 and Apple M3 Ultra GPU (80-core)General
| Vendor | Nvidia | Apple |
| Build | Discrete | Integrated |
| Released | January 7, 2025 | March 5, 2025 |
| Launch price (MSRP) | $1999 | - |
| Case | Desktop | Desktop |
| Purpose | Gaming | Professional |
| Segment | High-end | High-end |
| Architecture | Blackwell 2.0 | Apple M GPU |
| GPU Codename | GB202-300 | Custom |
| Rival Equivalent | - | - GeForce RTX 5070 Ti |
| Successor | - | - Apple M4 Ultra GPU (80-core) |
| Recommended CPU | - Intel Core Ultra 9 285K or above | - Apple M3 Ultra or above |
| Used in CPUs | - | - Apple M3 Ultra |
Desktop GPU rating (2nd and 30th place)
Graphics Processing Unit
| Base Clock | 2017 MHz | - |
| Boost Clock | 2407 MHz | 1380 MHz |
| Shading Units | 21760 | 10240 |
| Texture Mapping Units (TMUs) | 680 | 640 |
| Render Output Units (ROPs) | 176 | 320 |
| Compute Units (Pipelines) | 170 | 1280 |
| Tensor Cores | 680 | - |
| Ray-tracing Cores | 170 | - |
| L1 Cache | 128KB per cluster | - |
| L2 Cache | 96MB shared | - |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 424 GPixel/s | 442 GPixel/s |
| Texture Fill Rate | 1637 GTexel/s | 883 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 5.0 x16 | Custom |
| TGP | 575 W | 140 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 4 nm | 3 nm |
| Die Size | 750 mm² | - |
| Transistor Count | 92.2 billion | 184 billion |
| Transistor Density | 122.93 MTr/mm² | - |
| Max. Temperature | - | 100°C |
Memory
| Memory Type | GDDR7 | System Shared |
| Memory Size | 32 GB | - |
| Memory Clock | 1750 MHz | 6400 MHz |
| Effective Memory Speed | 28000 Mbps | - |
| Bus | 512-bit | 1024-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12.2 | - |
| Vulkan | 1.4 | - |
| OpenGL | 4.6 | - |
| OpenCL | 3.0 | - |
| CUDA | 12.0 | - |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | No |
| DisplayPort | 2.1b | - |
Cast your vote
Total votes: 99
