GeForce RTX 3080 vs Apple M4 Ultra GPU (80-core)
VS
Nvidia GeForce RTX 3080
GA102
Apple M4 Ultra GPU (80-core)
Custom
We performed a head-to-head comparison of the GeForce RTX 3080 10 GB with 68 pipelines and 8704 shaders against the 4 years and 9 months newer Apple M4 Ultra GPU (80-core) that utilizes 1280 pipelines and 10240 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.
Please note that the tests on the Apple M4 Ultra GPU (80-core) are done on an engineering sample provided by our insiders. The data will be more accurate after the final version of this GPU is available.
Key differences
Reasons to consider the GeForce RTX 3080
- Supports Nvidia DLSS 2 technology
Reasons to consider the Apple M4 Ultra GPU (80-core)
- Performs better (up to 36%) in 3DMark Steel Nomad Lite
- Manufactured using a more efficient 3 nm process technology
- Achieves 35% more points in the GeekBench 6 Compute test (218K vs 161K)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 170 114 104 80 |
- |
| The Witcher 3 | 189 165 116 63 |
- |
| Counter-Strike 2 | 274 206 149 82 |
- |
| Far Cry 6 | 145 138 114 72 |
- |
| Hogwarts Legacy | 119 98 78 43 |
- |
| CoD: Modern Warfare III | 152 145 110 73 |
- |
| Ghost of Tsushima | 109 89 74 44 |
- |
| Cyberpunk 2077 | 135 117 79 36 |
- |
| Shadow of the Tomb Raider | 210 195 151 82 |
- |
| 1080p High | 167 | - |
| 1080p Ultra | 141 | - |
| 1440p Ultra | 108 | - |
| 4K Ultra | 64 | - |
| Margin of Error | Low | High |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 17664 | - |
| Solar Bay | 85898 | - |
| Port Royal | 11597 | - |
| Fire Strike | 42472 | - |
| Wild Life Extreme | 37900 | - |
| Night Raid | 133004 | - |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 217.3 img/sec | - |
| Face Detection | 141.9 img/sec | - |
| Horizon Detection | 8.19 Gpixels/sec | - |
| Edge Detection | 11 Gpixels/sec | - |
| Gaussian Blur | 8.43 Gpixels/sec | - |
| Feature Matching | 1.29 Gpixels/sec | - |
| Stereo Matching | 780.7 Gpixels/sec | - |
| Particle Physics | 22538.5 FPS | - |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1062 | - |
| DirectX 11 | 207 FPS | - |
| DirectX 12 | 99 FPS | - |
| GPU Compute | 14058 Ops/s | - |
Recent User Tests
GeForce RTX 3080
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-08-06 (Henry) | Cinebench 2024 | 50695 |
Apple M4 Ultra GPU (80-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) | 9439 | - |
| Image Segmentation (HP) | 25471 | - |
| Image Super Resolution (Q) | 19892 | - |
| Face Detection (HP) | 46956 | - |
| Pose Estimation (Q) | 104704 | - |
| Text Classification (SP) | 2712 | - |
| Machine Translation (HP) | 4828 | - |
| Object Detection (SP) | 12153 | - |
| Depth Estimation (Q) | 36822 | - |
| Style Transfer (SP) | 246005 | - |
| Framework | ONNX | - |
| Backend | DirectML | - |
Specifications
Technical specifications of GeForce RTX 3080 and Apple M4 Ultra GPU (80-core)General
| Vendor | Nvidia | Apple |
| Build | Discrete | Integrated |
| Released | September 1, 2020 | May 1, 2025 |
| Launch price (MSRP) | $699 | - |
| Case | Desktop | Laptop |
| Purpose | Gaming | Professional |
| Segment | High-end | High-end |
| Architecture | Ampere | Apple M GPU |
| GPU Codename | GA102 | Custom |
| Rival Equivalent | - Radeon RX 6800 XT | - GeForce RTX 5080 |
| Successor | - GeForce RTX 5080 | - |
| Recommended CPU | - Intel Core i7 12700K or above | - Apple M4 Ultra or above |
| Used in CPUs | - | - Apple M4 Ultra |
Desktop GPU rating (#29th place)
Graphics Processing Unit
| Base Clock | 1440 MHz | 500 MHz |
| Boost Clock | 1710 MHz | 1578 MHz |
| Shading Units | 8704 | 10240 |
| Texture Mapping Units (TMUs) | 272 | 640 |
| Render Output Units (ROPs) | 96 | 320 |
| Compute Units (Pipelines) | 68 | 1280 |
| Tensor Cores | 272 | - |
| Ray-tracing Cores | 68 | - |
| L1 Cache | 128KB per cluster | - |
| L2 Cache | 5MB shared | - |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 164 GPixel/s | 505 GPixel/s |
| Texture Fill Rate | 465 GTexel/s | 1010 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x16 | Custom |
| TGP | 320 W | 120 W |
| Manufacturing | Samsung | TSMC |
| Fabrication Process | 8 nm | 3 nm |
| Die Size | 628 mm² | - |
| Transistor Count | 28 billion | - |
| Transistor Density | 44.59 MTr/mm² | - |
Memory
| Memory Type | GDDR6X | System Shared |
| Memory Size | 10 GB | - |
| Memory Clock | 1188 MHz | 8533 MHz |
| Effective Memory Speed | 19000 Mbps | - |
| Bus | 320-bit | 1024-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12 | - |
| Vulkan | 1.3 | - |
| OpenGL | 4.6 | - |
| OpenCL | 3.0 | - |
| CUDA | 8.6 | - |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 2 | No |
| DisplayPort | 1.4a | - |
Cast your vote
Total votes: 14