Intel Arc Pro B60 vs Intel Arc Pro B70 Comparison

Head-to-Head Benchmarks

The Intel Arc Pro B60 and Intel Arc Pro B70 share the same Xe2-HPG architecture and Battlemage Pro Series generation, but the recorded data places them in very different competitive tiers. The B60 has a complete set of benchmark results in the database, while the B70 currently has no recorded benchmark scores, an average benchmark score of zero, and no nearest rivals listed. This makes a direct numerical comparison impossible, but the available specifications and the B60’s measured performance against its own rivals provide the basis for analysis.

The B60’s average benchmark score is 3182. Its nearest rival in the database is the NVIDIA Quadro P1000, which scores 3163, placing the B60 0.6% ahead. The B60 also sits 3.5% above the NVIDIA GeForce RTX 5080 SUPER, which scores 3075. On the other side, the NVIDIA GeForce GT 640 scores 3210, putting it 0.9% ahead of the B60, and the NVIDIA GeForce 920M scores 3287, which is 3.2% ahead. The B60’s percentile ranking against all GPUs is 20, meaning it outperforms only a fifth of the database’s recorded graphics cards.

Looking at individual benchmark results, the B60 delivers its strongest showing in the Passmark G3D test with a score of 14580. Its Passmark GPU compute score is 7029, and its Passmark G2D score is 763. In the DirectX legacy tests, the B60 scores 179 in DirectX 9, 122 in DirectX 11, and 76 in DirectX 12. The 3DMark Steel Nomad DX12 test yields a score of 2646. These figures indicate a card that handles older DirectX workloads more comfortably than newer ones, with the DirectX 9 score exceeding the DirectX 12 score by a factor of roughly 2.4.

The B70, by contrast, has no recorded benchmark data. Its percentile ranking of 50 is a placeholder value, not a measured result, and its average benchmark score of zero confirms that no tests have been logged. The database therefore cannot produce head-to-head wins for either card. The B60 wins zero head-to-head comparisons, and the B70 also wins zero, because no comparative benchmark entries exist.

What can be established from the data is the B60’s position relative to its listed rivals. Against the Quadro P1000, the B60 holds a narrow but positive delta of 0.6%, a margin that suggests near-parity in overall average score. Against the RTX 5080 SUPER, the B60 leads by 3.5%, a more comfortable gap. However, the GT 640 and GeForce 920M both sit above the B60, with negative deltas of -0.9% and -3.2% respectively, indicating that the B60 trails those two cards in the database’s average scoring.

The Verdict

The data presents a straightforward conclusion: the Intel Arc Pro B70 is not yet benchmarked, so it cannot be judged on measured performance. The Intel Arc Pro B60, however, has a complete profile and sits in the lower percentile range of the database, ranked at 20th percentile overall. Its average score of 3182 places it in the company of older or lower-tier NVIDIA cards, specifically the Quadro P1000, GT 640, GeForce 920M, and RTX 5080 SUPER.

For users selecting between these two Intel cards, the B70 offers substantially larger hardware resources on paper: 4096 shading units versus 2560, 32 GB of GDDR6 memory versus 24 GB, a 256-bit memory bus versus 192-bit, and 22.94 TFLOPS of FP32 compute versus 12.29 TFLOPS. The B70 also boosts to 2800 MHz against the B60’s 2400 MHz. These specification differences suggest the B70 is designed for heavier workloads, but without benchmark scores, the database cannot confirm how that hardware translates into real-world performance.

The B60, meanwhile, is a measured product with a known competitive position. Its 0.6% lead over the Quadro P1000 and 3.5% lead over the RTX 5080 SUPER are the only positive deltas in its rival list. The database shows the B60 as a mid-to-low performer in its current standings, and any purchase decision should weight the B70’s unmeasured potential against the B60’s recorded, modest results.

Architecture Differences

Both cards use the Intel Xe2-HPG architecture and belong to the Battlemage Pro Series generation. The manufacturing process is identical: 5 nm at TSMC. The B60 uses the BMG-G21 chip, while the B70 uses the BMG-G31 chip, a larger die. The B60’s die size is 272 mm² with 19,600 million transistors and a transistor density of 72.1 million per mm². The B70’s die size is 368 mm², but its transistor count and density are listed as unknown in the database.

The B70’s larger die supports a substantially larger execution resource pool. The B70 has 4096 shading units, 256 texture mapping units, 128 render output units, and 32 ray tracing cores. The B60 has 2560 shading units, 160 TMUs, 80 ROPs, and 20 RT cores. Pixel rate on the B70 is 358.4 GPixel/s versus 192.0 GPixel/s on the B60. Texture rate on the B70 is 716.8 GTexel/s versus 384.0 GTexel/s. FP32 throughput is 22.94 TFLOPS on the B70 and 12.29 TFLOPS on the B60. FP16 throughput follows the same 2:1 ratio, with the B70 at 45.88 TFLOPS and the B60 at 24.58 TFLOPS.

Memory configuration differs in capacity and bandwidth. Both use GDDR6 memory at 2375 MHz with 19 Gbps effective speed. The B60 has 24 GB on a 192-bit bus, delivering 456.0 GB/s. The B70 has 32 GB on a 256-bit bus, delivering 608.0 GB/s. The B70’s bandwidth advantage is roughly one-third higher than the B60’s, which aligns with the wider bus.

The B70 also carries a higher power envelope at 230 W TDP versus 200 W for the B60, and both require a single 8-pin power connector with a suggested 550 W PSU. The B60 uses a PCIe 5.0 x8 interface, while the B70 uses PCIe 5.0 x16. Display outputs differ as well: the B60 provides four mini-DisplayPort 2.1 connectors, while the B70 provides one HDMI 2.1a and three DisplayPort 2.1 connectors.

FAQ

Q: Which card has more shading units?

A: The Intel Arc Pro B70 has 4096 shading units, while the Intel Arc Pro B60 has 2560.

Q: What is the memory bandwidth difference?

A: The B70 delivers 608.0 GB/s over a 256-bit bus with 32 GB of GDDR6. The B60 delivers 456.0 GB/s over a 192-bit bus with 24 GB of GDDR6.

Q: Are there any benchmark scores for the B70?

A: No. The database lists an empty benchmark array for the B70, an average benchmark score of zero, and no nearest rivals.

Q: How does the B60 compare to the NVIDIA Quadro P1000?

A: The B60 has an average benchmark score of 3182, which is 0.6% higher than the Quadro P1000’s average score of 3163.

Q: What is the B60’s best and worst benchmark result?

A: The B60’s highest recorded score is 14580 in Passmark G3D. Its lowest is 61 in Passmark DirectX 10.

Q: Do both cards use the same process node?

A: Yes, both are built on a 5 nm process at TSMC and use the Xe2-HPG architecture.

Where Each One Wins

The Intel Arc Pro B60 wins in the domain of measured data. It has a complete benchmark profile, including a 3DMark Steel Nomad DX12 score of 2646, a Passmark G3D score of 14580, and a Passmark GPU compute score of 7029. It also holds positive deltas against two of its four nearest rivals: 0.6% over the NVIDIA Quadro P1000 and 3.5% over the NVIDIA GeForce RTX 5080 SUPER. For any workload where a verified performance number matters, the B60 is the only card in this comparison with evidence.

The Intel Arc Pro B70 wins on paper specifications across every compute and memory metric. It offers 60% more shading units (4096 versus 2560), double the ROPs (128 versus 80), and 60% more ray tracing cores (32 versus 20). Its FP32 throughput of 22.94 TFLOPS is 86.7% higher than the B60’s 12.29 TFLOPS. Its memory bandwidth of 608.0 GB/s exceeds the B60’s 456.0 GB/s by 33.3%, and its 32 GB capacity is one-third larger. The B70 also boosts to 2800 MHz, 400 MHz higher than the B60’s 2400 MHz boost clock.

The B70’s bus interface is also wider, using PCIe 5.0 x16 versus the B60’s PCIe 5.0 x8, which may matter for data transfer in compute-heavy scenarios. Its pixel rate of 358.4 GPixel/s and texture rate of 716.8 GTexel/s are both roughly 87% higher than the B60’s corresponding rates. The B70’s larger die, 368 mm² versus 272 mm², provides the physical substrate for these additional resources.

For legacy DirectX performance, the B60’s scores show a pattern: DirectX 9 at 179, DirectX 11 at 122, DirectX 12 at 76, and DirectX 10 at 61. This suggests the B60 handles older API workloads with relative ease, but the database has no equivalent figures for the B70 to establish a comparison.

Specification Differences

| Specification | Intel Arc Pro B60 | Intel Arc Pro B70 |

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

| Chip | BMG-G21 | BMG-G31 |

| Die Size | 272 mm² | 368 mm² |

| Transistors | 19,600 million | unknown |

| Transistor Density | 72.1M / mm² | null |

| Base Clock | 2000 MHz | 2280 MHz |

| Boost Clock | 2400 MHz | 2800 MHz |

| Memory Size | 24 GB | 32 GB |

| Memory Bus Width | 192 bit | 256 bit |

| Memory Bandwidth | 456.0 GB/s | 608.0 GB/s |

| Shading Units | 2560 | 4096 |

| TMUs | 160 | 256 |

| ROPs | 80 | 128 |

| Ray Tracing Cores | 20 | 32 |

| Pixel Rate | 192.0 GPixel/s | 358.4 GPixel/s |

| Texture Rate | 384.0 GTexel/s | 716.8 GTexel/s |

| FP32 | 12.29 TFLOPS | 22.94 TFLOPS |

| FP16 | 24.58 TFLOPS (2:1) | 45.88 TFLOPS (2:1) |

| TDP | 200 W | 230 W |

| Bus Interface | PCIe 5.0 x8 | PCIe 5.0 x16 |

| Display Outputs | 4x mini-DisplayPort 2.1 | 1x HDMI 2.1a, 3x DisplayPort 2.1 |

| Dimensions (L x H x W) | 167 mm x 69 mm x 40 mm | 267 mm x 110 mm x 39 mm |

| Release Date | 2025-09-04 | 2026-03-25 |

| Launch MSRP | 499 USD | 949 USD |

The B60 and B70 share the same architecture, process node, foundry, memory type, memory clock, power connector requirement, suggested PSU wattage, slot width, DirectX support, OpenGL support, and Vulkan support. Both are dual-slot cards with a single 8-pin power connector and a 550 W suggested PSU. The B70 is physically larger in length and height, while the B60 is slightly thicker. The B70’s launch MSRP of 949 USD is nearly double the B60’s 499 USD, and its release date is later in the database’s timeline.