AMD Radeon 660M vs Apple M1 GPU (8-core)
VS
AMD Radeon 660M
Rembrandt
We compared two integrated laptop graphics cards: the Radeon 660M with 6 pipelines and 384 shaders against the 1 year and 2 months older Apple M1 GPU (8-core) that utilizes 128 pipelines and 1024 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
Key differences
Reasons to consider the Radeon 660M
- Ray-tracing support with 6 dedicated RT cores
Reasons to consider the Apple M1 GPU (8-core)
- Performs significantly better (up to 48%) in 3DMark Steel Nomad Lite
- 73% higher maximum theoretical performance (2.6 vs 1.5 TFLOPS)
- Manufactured using a more efficient 5 nm process technology
- Achieves 48% more points in the GeekBench 6 Compute test (20K vs 14K)
- Has 2.7x more shading units (1024 vs 384)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 27 13 11 - |
- |
| The Witcher 3 | 9 12 - - |
- |
| Counter-Strike 2 | 28 17 13 - |
- |
| Far Cry 6 | 20 16 9 - |
- |
| Hogwarts Legacy | 10 - - - |
- |
| CoD: Modern Warfare III | 18 15 13 - |
- |
| Ghost of Tsushima | 13 - - - |
- |
| Cyberpunk 2077 | 9 9 - - |
- |
| Shadow of the Tomb Raider | 17 14 11 - |
- |
| 1080p High | 17 | - |
| 1080p Ultra | 12 | - |
| 1440p Ultra | 9 | - |
| 4K Ultra | - | - |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 1514 | - |
| Solar Bay | 6396 | 6317 |
| Port Royal | 586 | - |
| Fire Strike | 4879 | - |
| Wild Life Extreme | 2676 | 4966 |
| Night Raid | 18958 | - |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 36 img/sec | 38.3 img/sec |
| Face Detection | 16.4 img/sec | 23 img/sec |
| Horizon Detection | 0.48 Gpixels/sec | 0.76 Gpixels/sec |
| Edge Detection | 0.65 Gpixels/sec | 1.33 Gpixels/sec |
| Gaussian Blur | 0.53 Gpixels/sec | 0.87 Gpixels/sec |
| Feature Matching | 0.18 Gpixels/sec | 0.27 Gpixels/sec |
| Stereo Matching | 46.5 Gpixels/sec | 69.9 Gpixels/sec |
| Particle Physics | 2056.5 FPS | 2766.4 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 654 | - |
| DirectX 11 | 47 FPS | - |
| DirectX 12 | 23 FPS | - |
| GPU Compute | 1321 Ops/s | - |
Blender
Blender GPU
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) | 1465 | 1400 |
| Image Segmentation (HP) | 1706 | 2280 |
| Image Super Resolution (Q) | 4266 | 3402 |
| Face Detection (HP) | 5115 | 5638 |
| Pose Estimation (Q) | 12992 | 12027 |
| Text Classification (SP) | 1116 | 2324 |
| Machine Translation (HP) | 1859 | 4296 |
| Object Detection (SP) | 1574 | 1530 |
| Depth Estimation (Q) | 6538 | 7675 |
| Style Transfer (SP) | 26656 | 17293 |
| Framework | ONNX | Core ML |
| Backend | DirectML | CPU |
Results were compared using different ML frameworks
Specifications
Technical specifications of Radeon 660M and Apple M1 GPU (8-core)General
| Vendor | Amd | Apple |
| Build | Integrated | Integrated |
| Released | January 1, 2022 | November 20, 2020 |
| Case | Laptop | Laptop |
| Purpose | Generic | Professional |
| Segment | Entry-level | Mid-range |
| Architecture | RDNA 2.0 | Apple M GPU |
| GPU Codename | Rembrandt | - |
| Rival Equivalent | - | - GeForce GTX 1050 Ti Mobile |
| Successor | - Radeon 8040S | - Apple M5 GPU (10-Core) |
| Recommended CPU | - AMD Ryzen 5 6600H or above | - Apple M1 or above |
| Used in CPUs | - AMD Ryzen 5 150 - AMD Ryzen 5 7535U - AMD Ryzen 5 7535HS - AMD Ryzen 5 Pro 6650U - AMD Ryzen 5 6600U - AMD Ryzen 5 6600H |
- Apple M1 |
Laptop GPU ranking (115th and 103rd place)
Graphics Processing Unit
| Base Clock | 400 MHz | 450 MHz |
| Boost Clock | 1800-1900 MHz (configurable) | 1278 MHz |
| Shading Units | 384 | 1024 |
| Texture Mapping Units (TMUs) | 24 | 64 |
| Render Output Units (ROPs) | 16 | 32 |
| Compute Units (Pipelines) | 6 | 128 |
| Ray-tracing Cores | 6 | No |
| L1 Cache | 128KB per cluster | - |
| L2 Cache | 1024KB shared | - |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 29-30 GPixel/s | 41 GPixel/s |
| Texture Fill Rate | 43-46 GTexel/s | 82 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x8 | Custom |
| TGP | 15 W | 15 W |
| Manufacturing | TSMC | Intel |
| Fabrication Process | 6 nm | 5 nm |
| Die Size | 208 mm² | - |
| Transistor Count | 13.1 billion | 11.6 billion |
| Transistor Density | 62.98 MTr/mm² | - |
| Max. Temperature | - | 94°C |
Memory
| Memory Type | System Shared | System Shared |
| Memory Clock | - | 4266 MHz |
| Effective Memory Speed | - | 8532 Mbps |
| Bus | - | 128-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12 | - |
| Vulkan | 1.2 | - |
| OpenGL | 4.6 | - |
| OpenCL | 2.0 | - |
| Ray Tracing | Yes | No |
| DLSS | No | No |
Cast your vote
Total votes: 5