Nvidia GeForce RTX 5070 vs Intel Arc B390
61
out 100
VS
33
out 100
Compared 2 processors: Nvidia GeForce RTX 5070, vs Intel Arc B390. Below are specs, tests, pros and cons for each device
Review
Games
Performance in DirectX, OpenCL, Vulkan
AI/ML
AI workload testing
Professional Apps
3D modeling, video editing, rendering
Energy Efficiency
Power usage in various scenarios
Tests & specs
Specs & test comparisonGames
| Game title | 1080p (High) 1080p (Ultra) 1440p (Ultra) 4K (Ultra) | |
 Forza Horizon 5 | 192 140 127 98 | 70 38 29 20 |
 The Witcher 3 | 212 180 134 70 | 68 55 39 19 |
 Counter-Strike 2 | 284 215 157 82 | 84 60 42 17 |
 Far Cry 6 | 161 148 127 82 | 55 49 37 14 |
 Hogwarts Legacy | 137 115 92 50 | 29 21 16 n/a |
 COD Modern Warfare 3 | 181 172 133 88 | 35 32 26 15 |
 Ghost of Tsushima | 123 107 88 56 | 33 25 21 10 |
 Cyberpunk 2077 | 161 141 99 44 | 37 33 18 n/a |
 Shadow of the Tomb Raider | 250 241 181 101 | 57 52 35 19 |
| 1080p High | 189 | 52 |
| 1080p Ultra | 162 | 41 |
| 1440p Ultra | 126 | 29 |
| 4K Ultra | 75 | 14 |
| Deviation | Low | Low |
Benchmarks
| Time Spy | 22613 | 6595 |
| Solar Bay | 103768 | 26399 |
| Port Royal | 14312 | 3886 |
| Fire Strike | 57793 | 16770 |
| Night Raid | 172986 | 68498 |
| Background Blur | 254.4 frames/s | n/a |
| Face Detection | 180.6 frames/s | n/a |
| Horizon Detection | 7.91 Gpix/sec | n/a |
| Edge Detection | 10.2 Gpix/sec | n/a |
| Gaussian Blur | 10.7 Gpix/sec | n/a |
| Feature Matching | 1.84 Gpix/sec | n/a |
| Stereo Matching | 975.7 Gpix/sec | n/a |
| Particle Physics | 27816.5 fps | n/a |
| API | OpenCL | n/a |
| G2D Mark | 1297 | 786 |
| DirectX 11 | 277 fps | 70 fps |
| DirectX 12 | 108 fps | 50 fps |
| GPU Compute | 15779 Ops/s | 3702 Ops/s |
| Image Classification (SP) | 12270 | 5487 |
| Image Segmentation (HP) | 30237 | 16864 |
| Horizon Detection | 35585 | 8788 |
| Face Detection (HP) | 62662 | 25029 |
| Pose Estimation (Q) | 145190 | 18852 |
| Text Classification (SP) | 4198 | 2115 |
| Machine Translation (HP) | 6233 | 4266 |
| Object Detection (SP) | 16709 | 7449 |
| Depth Estimation (Q) | 58300 | 22207 |
| Style Transfer (SP) | 246598 | 95039 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Blender
Tech Specs
Specs & test comparisonGeneral Info
| Manufacturer | Nvidia | Intel |
| Build | Discrete | Integrated |
| Announced | January, 7, 2025 | January, 5, 2026 |
| Form Factor | PC use | Laptop use |
| Use case | Gaming | Gaming |
| Segment | Mid-range | Mid-range |
| Architecture | Blackwell 2.0 | Celestial |
| Codename | GB205-300 | Panther Lake |
GPU
| GPU Frequency | 2325 MHz | 300 MHz |
| Boost GPU Frequency | 2512 MHz | 2500 MHz |
| Shader Units | 6144 | 1536 |
| TMUs | 192 | 48 |
| ROPs | 80 | 24 |
| Execution Units | 48 | 12 |
| Ray Tracing Cores | 48 | n/a |
| Tensor Cores | 192 | n/a |
| L1 cache | 128 KB/core | 64 KB/core |
| L2 cache | 48000 KB/core | 16000 KB/core |
| Instructions per Clock | 2 IPC | 2 IPC |
Performance
| Pixel Fill Rate | 201 Gpix/sec | 60 Gpix/sec |
| Texture Fill Rate | 482 GTex/s | 120 GTex/s |
| FLOPS | 30.9 TFLOPS | 7.7 TFLOPS |
Process
| Interface | PCIe 5.0 x16 | PCIe 5.0 |
| TGP | 250 W | n/a |
| Process | TSMC | TSMC |
| Process tech. | 4 nm | 3 nm |
| Max temperature | n/a | 100 °C |
| Transistor Size | 263 mm² | n/a |
| Transistor count | 31.1 billion | n/a |
| Transistor density | 118.25 MTr/mm² | n/a |
Memory
| Interface | GDDR7 | System Shared |
| Memory Frequency | 1750 MHz | n/a |
| Bus | 192 bit | n/a |
| ECC support | No | No |
| Effective Memory Speed | 28000 MB/s | n/a |
| Memory Size | 12 GB | n/a |
Max Throughput
API
| DirectX | 12.2 | 12.1 |
| Vulkan | 1.4 | 1.4 |
| OpenGL | 4.6 | 4.6 |
| CUDA | 12.0 | n/a |
| DisplayPort | 2.1b | 2.1 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | No |
Results
After analyzing the data, we consider Nvidia GeForce RTX 5070 the clear winner in this comparison.
Survey
Total Votes: 0
