Nvidia N1X (48SM) vs Intel Arc B370
VS
Nvidia N1X (48SM)
GB20B
Intel Arc B370
Panther Lake
We compared two integrated laptop graphics cards: the Nvidia N1X (48SM) with 48 pipelines and 6144 shaders against the 5 months older Intel Arc B370 that utilizes 10 pipelines and 1280 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
Scores marked with a red asterisk (**) represent initial estimates from early testing.
Key differences
Reasons to consider the Nvidia N1X (48SM)
- 5.1x higher maximum theoretical performance (31.2 vs 6.1 TFLOPS)
- Supports Nvidia DLSS 4 technology
- Has 4.8x more shading units (6144 vs 1280)
- Features 112 more tensor cores for effective ML and AI workloads
Reasons to consider the Intel Arc B370
- Performs significantly better (up to 69%) in 3DMark Steel Nomad Lite
- Achieves 16.5x more points in the GeekBench 6 Compute test (48K vs 2K)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | - | 71 41 32 18 |
| The Witcher 3 | - | 65 55 39 19 |
| Counter-Strike 2 | - | 85 62 41 19 |
| Far Cry 6 | - | 56 50 36 17 |
| Hogwarts Legacy | - | 31 22 17 9 |
| CoD: Modern Warfare III | - | 36 32 24 14 |
| Ghost of Tsushima | - | 35 27 21 11 |
| Cyberpunk 2077 | - | 38 32 20 - |
| Shadow of the Tomb Raider | - | 61 50 34 20 |
| 1080p High | - | 53 |
| 1080p Ultra | - | 41 |
| 1440p Ultra | - | 29 |
| 4K Ultra | - | 15 |
| Margin of Error | High | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | - | 6028 |
| Solar Bay | - | 23177 |
| Port Royal | - | 3498 |
| Fire Strike | - | 15198 |
| Wild Life Extreme | - | 11657 |
| Night Raid | - | 69941 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | - | 117.4 img/sec |
| Face Detection | - | 60.6 img/sec |
| Horizon Detection | - | 1.39 Gpixels/sec |
| Edge Detection | - | 1.95 Gpixels/sec |
| Gaussian Blur | - | 2.45 Gpixels/sec |
| Feature Matching | - | 0.66 Gpixels/sec |
| Stereo Matching | - | 144.8 Gpixels/sec |
| Particle Physics | - | 7272.6 FPS |
| API | OpenCL | OpenCL |
Passmark Graphics
G3D Mark Score
| G2D Mark | - | 832 |
| DirectX 11 | - | 65 FPS |
| DirectX 12 | - | 45 FPS |
| GPU Compute | - | 3369 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) | - | 5389 |
| Image Segmentation (HP) | - | 18899 |
| Image Super Resolution (Q) | - | 8326 |
| Face Detection (HP) | - | 25377 |
| Pose Estimation (Q) | - | 16174 |
| Text Classification (SP) | - | 1763 |
| Machine Translation (HP) | - | 4140 |
| Object Detection (SP) | - | 7138 |
| Depth Estimation (Q) | - | 20034 |
| Style Transfer (SP) | - | 84880 |
| Framework | - | ONNX |
| Backend | - | DirectML |
Specifications
Technical specifications of Nvidia N1X (48SM) and Intel Arc B370General
| Vendor | Nvidia | Intel |
| Build | Integrated | Integrated |
| Released | June 1, 2026 | January 5, 2026 |
| Case | Laptop | Laptop |
| Purpose | Professional | Generic |
| Segment | Mid-range | Mid-range |
| Architecture | Blackwell (N1x) | Celestial |
| GPU Codename | GB20B | Panther Lake |
| Rival Equivalent | - Apple M5 Max GPU (40-core) | - |
| Recommended CPU | - Nvidia RTX Spark N1X (20-Core) or above | - Intel Core Ultra 7 356H or above |
| Used in CPUs | - Nvidia RTX Spark N1X (20-Core) | - Intel Core Ultra 5 338H |
Laptop GPU ranking (38th and 72nd place)
Graphics Processing Unit
| Base Clock | 1665 MHz | 300 MHz |
| Boost Clock | 2535 MHz | 2400 MHz |
| Shading Units | 6144 | 1280 |
| Texture Mapping Units (TMUs) | 192 | 40 |
| Render Output Units (ROPs) | 48 | 20 |
| Compute Units (Pipelines) | 48 | 10 |
| Tensor Cores | 192 | 80 |
| Ray-tracing Cores | 48 | 10 |
| L1 Cache | 128KB per cluster | 64KB per cluster |
| L2 Cache | 24MB shared | 16MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 122 GPixel/s | 48 GPixel/s |
| Texture Fill Rate | 487 GTexel/s | 96 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 5.0 x16 | IGP |
| TGP | - | 25 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 3 nm | 3 nm |
| Die Size | 173.6 mm² | - |
| Max. Temperature | - | 100°C |
Memory
| Memory Type | System Shared | System Shared |
| ECC | No | No |
API
| DirectX | - | 12.1 |
| Vulkan | 1.4 | 1.4 |
| OpenGL | - | 4.6 |
| OpenCL | 3.0 | 3.0 |
| CUDA | 12.1 | No |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | No |
| DisplayPort | - | 2.1 |
Cast your vote
Total votes: < 1