Intel Arc Graphics 32EU vs Intel Arc Pro B370 Comparison

Intel Arc Graphics 32EU and Intel Arc Pro B370 are both integrated graphics solutions from Intel, but they serve very different segments of the market. The 32EU is part of the Arrow Lake-S generation, while the B370 belongs to the newer Panther Lake generation. The data shows a clear performance hierarchy, but the story is more nuanced than simple compute power, as the B370 delivers significantly higher raw throughput while the 32EU holds its own in specific scenarios.

Where Each One Wins

The Intel Arc Graphics 32EU wins in the only direct benchmark comparison available in the database, where it scores 733 points in the 3dmark Steel Nomad DX12 test. This places it in the 3rd percentile of all GPUs, meaning it outperforms only a small fraction of the entire GPU landscape. Its nearest rivals include the Intel Arc Graphics 24EU and 64EU, both scoring 733 points, indicating a 0% delta, while it edges out the AMD Radeon HD 6470M by 1.4% and trails the NVIDIA GeForce GT 415M by 2.4%.

The Intel Arc Pro B370 wins in every other measurable category, but it has no recorded benchmark scores in the database. Its percentile ranking of 50th places it squarely in the middle of all GPUs, a stark contrast to the 32EU's 3rd percentile. The B370's advantage lies in its compute architecture and clock speeds, which translate to higher theoretical throughput across all pixel, texture, and floating-point operations.

For real-world usage, the 32EU is the winner in compatibility and availability scenarios tied to its older release date, while the B370 wins in any task that demands raw shader processing, ray tracing, or higher resolution rendering. The B370's 10 ray tracing cores are a dedicated feature the 32EU lacks entirely, giving it a structural advantage in DirectX 12 Ultimate workloads.

Architecture Differences

The two processors come from different manufacturing generations and foundries. The Intel Arc Graphics 32EU uses the Xe-LPG architecture built on a 3 nm process at TSMC, while the Intel Arc Pro B370 uses the Xe3-LPG architecture also on a 3 nm process but fabricated by Intel. The 32EU's chip is Arrow Lake-S, whereas the B370 is based on the Panther Lake chip.

The transistor counts tell a story of scale. The 32EU packs 17,800 million transistors into a 243 mm² die, achieving a density of 73.3 million transistors per square millimeter. The B370's transistor count and die size are listed as unknown in the database, so no direct comparison can be made on those figures. What is clear is the execution unit count: the 32EU has 256 shading units, 16 texture mapping units, and 8 raster operations pipelines, while the B370 has 1280 shading units, 40 TMUs, and 20 ROPs.

Clock behavior differs substantially. Both have a 300 MHz base clock, but the 32EU boosts to 1950 MHz while the B370 reaches 2400 MHz. This 450 MHz boost advantage, combined with the 5x shading unit count, gives the B370 a massive compute lead. The B370 also includes 10 ray tracing cores, a feature completely absent from the 32EU. Both share the same API support: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Memory configuration is identical in concept: both use System Shared memory, meaning they rely on the host system's RAM rather than dedicated VRAM. The bus width is also System Shared, and bandwidth is System Dependent for both. This means actual memory performance will vary based on the host platform's memory subsystem.

Head-to-Head Benchmarks

The database records only one direct benchmark between these two: the 3dmark Steel Nomad DX12 test. The Intel Arc Graphics 32EU scores 733 points. The Intel Arc Pro B370 has no recorded score, so the head-to-head comparison relies on the 32EU's score and the B370's theoretical specifications.

The 32EU's 733-point score places it at the 3rd percentile overall. Its nearest rival, the Intel Arc Graphics 24EU, also scores 733, showing that within the 32EU's own family, the 24EU variant delivers identical performance in this test. The 64EU variant also matches at 733 points, suggesting that the Steel Nomad test is not sensitive to execution unit scaling within this range. The 32EU beats the AMD Radeon HD 6470M by 1.4% (723 points) and loses to the NVIDIA GeForce GT 415M by 2.4% (751 points).

The B370's theoretical performance is far higher. Its pixel rate of 48.00 GPixel/s is more than triple the 32EU's 15.60 GPixel/s. Texture rate follows the same pattern: 96.00 GTexel/s versus 31.20 GTexel/s. The FP32 compute gap is even more pronounced, with the B370 delivering 6.144 TFLOPS against the 32EU's 998.4 GFLOPS, a roughly 6.2x advantage. FP16 performance scales similarly, with the B370 offering 12.29 TFLOPS versus the 32EU's 1.997 TFLOPS, both at a 2:1 ratio.

Power consumption inverts the performance relationship. The B370 draws only 25 W while the 32EU consumes 65 W. This means the B370 achieves its 6.2x FP32 advantage while using 38.5% of the power. The efficiency delta is substantial, though the database does not record an efficiency metric directly.

The Verdict

The data indicates two different use cases. The Intel Arc Graphics 32EU is a low-power integrated solution for Arrow Lake-S desktop platforms. Its 65 W TDP and 3rd percentile benchmark standing position it as a basic display and light workload adapter. The 733-point Steel Nomad score is identical to the 24EU and 64EU variants, suggesting that within this generation, the execution unit count does not materially change this specific workload result. It ties with the 24EU, beats the HD 6470M by 1.4%, and loses to the GT 415M by 2.4%.

The Intel Arc Pro B370 is positioned for professional and higher-performance mobile or compact systems. Its 25 W TDP, 2400 MHz boost clock, 1280 shading units, and 10 ray tracing cores deliver 6.144 TFLOPS of FP32 compute and 48.00 GPixel/s pixel throughput. The 50th percentile ranking, while based on no recorded benchmark scores, suggests the database estimates it as a mid-range performer. The absence of benchmark scores means direct validation is pending, but the architectural specifications point to a clear performance class above the 32EU.

For users of the 32EU, the data shows it is adequate for basic tasks but is not competitive with even low-end discrete GPUs from over a decade ago, as the GT 415M outperforms it. For users considering the B370, the specifications indicate a capable processor for ray-traced workloads, higher resolution rendering, and compute tasks, all within a 25 W power envelope. The choice is straightforward: the 32EU serves basic integrated needs, the B370 serves compute and graphics workloads in constrained power environments.

FAQ

Q: Which processor has more shading units?

A: The Intel Arc Pro B370 has 1280 shading units, while the Intel Arc Graphics 32EU has 256 shading units, a 5x difference.

Q: What is the boost clock difference between the two?

A: The Intel Arc Graphics 32EU boosts to 1950 MHz, while the Intel Arc Pro B370 boosts to 2400 MHz, a 450 MHz advantage for the B370.

Q: Does the Intel Arc Graphics 32EU support ray tracing?

A: No, the 32EU has no ray tracing cores. The Intel Arc Pro B370 includes 10 ray tracing cores.

Q: How do their power requirements compare?

A: The Intel Arc Graphics 32EU has a 65 W TDP, while the Intel Arc Pro B370 has a 25 W TDP, making the B370 significantly more power-efficient.

Q: What is the FP32 compute performance difference?

A: The Intel Arc Pro B370 delivers 6.144 TFLOPS of FP32 compute, compared to 998.4 GFLOPS for the Intel Arc Graphics 32EU, a roughly 6.2x advantage.

Q: Which processor has a higher benchmark score in the database?

A: The Intel Arc Graphics 32EU has a recorded 733-point score in the 3dmark Steel Nomad DX12 test. The Intel Arc Pro B370 has no recorded benchmark scores.

Specification Differences

| Specification | Intel Arc Graphics 32EU | Intel Arc Pro B370 |

|----------------|-------------------------|---------------------|

| Chip | Arrow Lake-S | Panther Lake |

| Architecture | Xe-LPG | Xe3-LPG |

| Generation | Arc Graphics-M (Arrow Lake) | Arc Graphics-WM (Panther Lake) |

| Foundry | TSMC | Intel |

| Process Node | 3 nm | 3 nm |

| Transistors | 17,800 million | unknown |

| Die Size | 243 mm² | unknown |

| Transistor Density | 73.3M / mm² | null |

| Base Clock | 300 MHz | 300 MHz |

| Boost Clock | 1950 MHz | 2400 MHz |

| Shading Units | 256 | 1280 |

| TMUs | 16 | 40 |

| ROPs | 8 | 20 |

| Ray Tracing Cores | null | 10 |

| Pixel Rate | 15.60 GPixel/s | 48.00 GPixel/s |

| Texture Rate | 31.20 GTexel/s | 96.00 GTexel/s |

| FP32 Performance | 998.4 GFLOPS | 6.144 TFLOPS |

| FP16 Performance | 1.997 TFLOPS (2:1) | 12.29 TFLOPS (2:1) |

| TDP | 65 W | 25 W |

| Power Connectors | null | None |

| Bus Interface | Ring Bus | IGP |

| Display Outputs | Motherboard Dependent | Portable Device Dependent |

| Release Date | 2024-10-23 | 2026-01-26 |

| Predecessor | HD Graphics-M | HD Graphics-WM |

| Production Status | Active | Active |