Nvidia RTX Pro 6000 Blackwell vs Intel Arc Pro B70
VS
Nvidia RTX Pro 6000 Blackwell
GB202-300-A1
Intel Arc Pro B70
BMG-G31
We compared two discrete desktop professional GPUs: the Nvidia RTX Pro 6000 Blackwell 96 GB with 188 pipelines and 24064 shaders against the Intel Arc Pro B70 32 GB that utilizes 32 pipelines and 4096 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.
Value for money (Beta)
Key differences
Reasons to consider the Nvidia RTX Pro 6000 Blackwell
- Performs significantly better (up to 3.3x) in 3DMark Steel Nomad Lite
- 5.5x higher maximum theoretical performance (126 vs 22.9 TFLOPS)
- Shows 2.4x higher average frame rate in modern games at QHD resolution – 196 vs 81 FPS
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 309 235 215 175 |
140 94 84 64 |
| The Witcher 3 | 343 311 216 116 |
184 156 114 58 |
| Counter-Strike 2 | 362 263 203 93 |
177 131 93 43 |
| Far Cry 6 | 221 198 185 112 |
127 115 92 52 |
| Hogwarts Legacy | 235 189 150 89 |
86 72 57 28 |
| CoD: Modern Warfare III | 305 296 252 176 |
111 107 79 51 |
| Ghost of Tsushima | 206 170 158 107 |
83 68 51 28 |
| Cyberpunk 2077 | 253 221 160 71 |
104 93 66 30 |
| Shadow of the Tomb Raider | 319 298 224 121 |
137 119 92 50 |
| 1080p High | 284 | 128 |
| 1080p Ultra | 242 | 106 |
| 1440p Ultra | 196 | 81 |
| 4K Ultra | 118 | 45 |
| Margin of Error | High | High |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 42614 | 18321 |
| Solar Bay | 255911 | 76101 |
| Port Royal | 39888 | 10688 |
| Fire Strike | 75325 | 46723 |
| Wild Life Extreme | 110070 | 37771 |
| Night Raid | 214655 | 182972 |
GeekBench 6 GPU Compute
GB6 Compute Score
| Background Blur | 366.1 img/sec | 413.1 img/sec |
| Face Detection | 248 img/sec | 226.4 img/sec |
| Horizon Detection | 23.3 Gpixels/sec | 5.05 Gpixels/sec |
| Edge Detection | 35.2 Gpixels/sec | 9.66 Gpixels/sec |
| Gaussian Blur | 45 Gpixels/sec | 10.6 Gpixels/sec |
| Feature Matching | 2.47 Gpixels/sec | 1.67 Gpixels/sec |
| Stereo Matching | 3040 Gpixels/sec | 712.1 Gpixels/sec |
| Particle Physics | 42574.6 FPS | 20998.9 FPS |
| API | OpenCL | Vulkan |
Passmark Graphics
G3D Mark Score
| G2D Mark | 1391 | 739 |
| DirectX 11 | 339 FPS | 149 FPS |
| DirectX 12 | 154 FPS | 91 FPS |
| GPU Compute | 26607 Ops/s | 9676 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) | 16903 | - |
| Image Segmentation (HP) | 48518 | - |
| Image Super Resolution (Q) | 54618 | - |
| Face Detection (HP) | 103224 | - |
| Pose Estimation (Q) | 338587 | - |
| Text Classification (SP) | 3894 | - |
| Machine Translation (HP) | 7291 | - |
| Object Detection (SP) | 26498 | - |
| Depth Estimation (Q) | 90853 | - |
| Style Transfer (SP) | 635289 | - |
| Framework | ONNX | - |
| Backend | DirectML | - |
Specifications
Technical specifications of Nvidia RTX Pro 6000 Blackwell and Intel Arc Pro B70General
| Vendor | Nvidia | Intel |
| Build | Discrete | Discrete |
| Released | March 18, 2026 | March 25, 2026 |
| Launch price (MSRP) | $8559 | $949 |
| Case | Desktop | Desktop |
| Purpose | Professional | Professional |
| Segment | High-end | High-end |
| Architecture | Blackwell 2.0 | Battlemage Pro Series |
| GPU Codename | GB202-300-A1 | BMG-G31 |
| Recommended CPU | - AMD Ryzen 9 9955HX3D or above | - |
Desktop GPU rating (1st and 27th place)
Graphics Processing Unit
| Base Clock | 1590 MHz | 2280 MHz |
| Boost Clock | 2617 MHz | 2800 MHz |
| Shading Units | 24064 | 4096 |
| Texture Mapping Units (TMUs) | 752 | 256 |
| Render Output Units (ROPs) | 190 | 128 |
| Compute Units (Pipelines) | 188 | 32 |
| Tensor Cores | 752 | 256 |
| Ray-tracing Cores | 188 | 32 |
| L1 Cache | 128KB per cluster | 256KB per cluster |
| L2 Cache | 128MB shared | 16MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 497 GPixel/s | 358 GPixel/s |
| Texture Fill Rate | 1968 GTexel/s | 717 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 5.0 x16 | PCIe 5.0 x16 |
| TGP | 600 W | 230 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 5 nm | 5 nm |
| Die Size | 750 mm² | - |
| Transistor Count | 92.2 billion | - |
| Transistor Density | 122.93 MTr/mm² | - |
Memory
| Memory Type | GDDR7 | GDDR6 |
| Memory Size | 96 GB | 32 GB |
| Memory Clock | 1750 MHz | 2375 MHz |
| Effective Memory Speed | 28000 Mbps | 19000 Mbps |
| Bus | 512-bit | 256-bit |
| ECC | Yes | No |
Memory Bandwidth
API
| DirectX | 12.2 | 12.2 |
| Vulkan | 1.4 | 1.3 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| CUDA | 12 | No |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | No |
| DisplayPort | 2.1b | 2.1a |
Cast your vote
Total votes: 2