Intel Arc B770 vs Intel Arc Pro B60 Comparison

Head-to-Head Benchmarks

The Intel Arc B770 and Intel Arc Pro B60 share the Xe2-HPG architecture and the same 5 nm TSMC process node, but the recorded data shows they are positioned for very different workloads. The most direct comparison comes from the Arc Pro B60's benchmark entries, which place it in a specific performance tier within the database.

The Arc Pro B60 delivers a 3DMark Steel Nomad DX12 score of 2646, while its Passmark G3D score reaches 14580. The Arc B770 has no recorded benchmark scores in the database, so direct head-to-head numbers are unavailable. What the database does provide, however, is a clear picture of where the Arc Pro B60 sits relative to other GPUs: its average benchmark score of 3182 places it at the 20th percentile among all GPUs, meaning roughly 80 percent of recorded graphics cards score higher on average.

The nearest rival data for the Arc Pro B60 shows a tight cluster. It sits 0.6 percent ahead of the NVIDIA Quadro P1000, which averages 3163. It trails the NVIDIA GeForce GT 640 by 0.9 percent (that card averages 3210) and the NVIDIA GeForce 920M by 3.2 percent (3287 average). Interestingly, it holds a 3.5 percent advantage over the NVIDIA GeForce RTX 5080 SUPER, which averages 3075. This grouping suggests the Arc Pro B60 performs in a narrow band near these cards, despite the architectural gulf between a professional workstation GPU and consumer parts from different generations.

The Arc B770, by contrast, offers substantially higher raw specifications on paper. Its FP32 throughput of 19.66 TFLOPS is roughly 60 percent higher than the Arc Pro B60's 12.29 TFLOPS. Texture rate follows a similar pattern: 614.4 GTexel/s versus 384.0 GTexel/s. Pixel rate measures 307.2 GPixel/s against 192.0 GPixel/s. These figures indicate the Arc B770 should handle compute-heavy and fill-rate-heavy workloads with considerably more headroom, but the database contains no measured scores to confirm this in practice.

Where Each One Wins

The Arc Pro B60 wins in memory capacity decisively. It carries 24 GB of GDDR6 across a 192 bit bus, while the Arc B770 carries 16 GB across a wider 256 bit bus. For datasets that exceed 16 GB, the Arc Pro B60 is the only option between the two that can hold the entire working set in VRAM. Its memory operates at 2375 MHz (19 Gbps effective), compared to 2000 MHz (16 Gbps effective) on the Arc B770. Even so, the Arc B770's wider bus gives it higher total bandwidth: 512.0 GB/s versus 456.0 GB/s.

The Arc B770 wins in every compute and rendering throughput category recorded in the database. Its shading unit count of 4096 is 60 percent higher than the Arc Pro B60's 2560. TMUs number 256 versus 160, and ROPs number 128 versus 80. Ray tracing cores follow the same ratio: 32 on the Arc B770 versus 20 on the Arc Pro B60. The FP16 figures mirror FP32 with a 2:1 ratio on both cards, giving the Arc B770 39.32 TFLOPS and the Arc Pro B60 24.58 TFLOPS.

The Arc Pro B60 wins on physical footprint and interface flexibility. Its dimensions are recorded as 167 mm by 69 mm by 40 mm, while the Arc B770 has no listed dimensions. The Arc Pro B60 uses a single 8 pin power connector and draws 200 W, whereas the Arc B770 requires one 6 pin plus one 8 pin connector and draws 225 W. Both cards suggest a 550 W power supply. The Arc Pro B60 also uses PCIe 5.0 x8, while the Arc B770 uses PCIe 4.0 x16; the newer interface can be an advantage in systems with limited lane allocation, even though the x8 width reduces total lanes.

Architecture Differences

Both GPUs are built on Xe2-HPG architecture, Intel's second generation of the Battlemage family. The Arc B770 belongs to the Arc 7 tier and uses the BMG-G31 chip, while the Arc Pro B60 uses the smaller BMG-G21 die. The process node is identical: 5 nm at TSMC. Die size differs markedly, with the BMG-G31 measuring 368 mm² against 272 mm² for the BMG-G21. Transistor counts are not recorded for the Arc B770, but the Arc Pro B60 lists 19,600 million transistors at a density of 72.1 million per square millimeter.

The Arc B770 carries the full complement of execution resources: 4096 shading units, 256 TMUs, 128 ROPs, and 32 ray tracing cores. The Arc Pro B60 scales down to 2560 shading units, 160 TMUs, 80 ROPs, and 20 ray tracing cores. Neither card records tensor core counts, and both expose the same API surface: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Clock behavior differs slightly. The Arc B770 has a base clock of 2100 MHz and a boost clock of 2400 MHz. The Arc Pro B60 runs a 2000 MHz base with the same 2400 MHz boost. That 100 MHz base clock gap is small in absolute terms but contributes to the Arc B770's higher peak throughput at stock settings.

Display outputs separate the two as well. The Arc B770 provides one HDMI 2.1a port and three DisplayPort 2.1 outputs. The Arc Pro B60 provides four mini-DisplayPort 2.1 outputs. The pro card's compact output configuration aligns with workstation mounting patterns, while the consumer card offers a standard HDMI plus DisplayPort mix.

The Arc Pro B60's production status is listed as Active, and its release date is recorded as September 4, 2025. The Arc B770 lists a release date of December 31, 2025, with no production status. The Arc B770's predecessor is listed as Alchemist, reflecting the prior architecture generation; the Arc Pro B60 has no predecessor recorded. The Arc B770's launch MSRP is not present in the database, while the Arc Pro B60 carries a launch MSRP of 499 USD.

The average benchmark score of 3182 for the Arc Pro B60, combined with its 20th percentile ranking, positions it as a modest performer in the overall GPU landscape. Its nearest rivals include the NVIDIA Quadro P1000, GeForce GT 640, GeForce 920M, and GeForce RTX 5080 SUPER, all within a 3.5 percent band. That clustering suggests the Arc Pro B60 does not compete at the top tier of raw graphics performance, but its 24 GB frame buffer gives it a niche that those rivals do not match.

The Verdict

The recorded data supports a clear split. The Arc Pro B60 is a professional workstation card defined by memory capacity and a compact, low-power design. Its 24 GB VRAM is the largest memory allocation between the two, and its 200 W power draw with a single 8 pin connector makes it easier to fit into dense or constrained systems. Its measured performance, however, is modest: the 20th percentile ranking and the tight grouping with the Quadro P1000, GT 640, 920M, and RTX 5080 SUPER indicate it is not built for peak frame rates or raw compute throughput.

The Arc B770 is the stronger compute and rendering part on paper. Every throughput metric in the database favors it: FP32, FP16, texture rate, pixel rate, shading units, TMUs, ROPs, and ray tracing cores. Its 512.0 GB/s memory bandwidth exceeds the Arc Pro B60's 456.0 GB/s despite the smaller 16 GB capacity. The absence of recorded benchmark scores for the Arc B770 means the database cannot confirm how those specifications translate into measured performance, but the specification advantage is unambiguous.

The choice depends on workload. The Arc Pro B60 suits tasks where 24 GB of VRAM matters more than raw speed, such as large model inference, high-resolution image processing, or rendering scenes that exceed 16 GB. The Arc B770 suits tasks that demand maximum throughput per frame or per compute pass, where its higher shading unit count and greater bandwidth provide measurable headroom. The Arc Pro B60's PCIe 5.0 x8 interface may benefit systems with limited PCIe lanes, while the Arc B770's PCIe 4.0 x16 remains a standard, well-supported configuration.

The launch MSRP of 499 USD for the Arc Pro B60 is the only price data in the database; the Arc B770 has no recorded launch price. The Arc Pro B60 is listed as Active in production and was released earlier, on September 4, 2025, while the Arc B770's release date is December 31, 2025. No successor is recorded for either card, and the Arc B770's predecessor is Alchemist.

FAQ

Q: Which GPU has more memory bandwidth?

A: The Intel Arc B770, with 512.0 GB/s from a 256 bit bus, compared to 456.0 GB/s on the Arc Pro B60's 192 bit bus.

Q: How much VRAM does each card have?

A: The Arc B770 has 16 GB of GDDR6, and the Arc Pro B60 has 24 GB of GDDR6.

Q: What is the Arc Pro B60's average benchmark score and percentile?

A: It averages 3182 across its recorded benchmarks and sits at the 20th percentile among all GPUs.

Q: Which card has more ray tracing cores?

A: The Arc B770 has 32 ray tracing cores, while the Arc Pro B60 has 20.

Q: What are the power connector requirements?

A: The Arc B770 uses one 6 pin and one 8 pin connector at 225 W, while the Arc Pro B60 uses a single 8 pin connector at 200 W. Both recommend a 550 W power supply.

Q: Do the cards share the same API support?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Specification Differences

| Field | Intel Arc B770 | Intel Arc Pro B60 |

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

| Chip | BMG-G31 | BMG-G21 |

| Generation | Battlemage (Arc 7) | Battlemage (Pro Series) |

| Die Size | 368 mm² | 272 mm² |

| Transistors | unknown | 19,600 million |

| Transistor Density | null | 72.1M / mm² |

| Base Clock | 2100 MHz | 2000 MHz |

| Boost Clock | 2400 MHz | 2400 MHz |

| Memory Clock | 2000 MHz, 16 Gbps effective | 2375 MHz, 19 Gbps effective |

| Memory Size | 16 GB | 24 GB |

| Memory Type | GDDR6 | GDDR6 |

| Memory Bus Width | 256 bit | 192 bit |

| Memory Bandwidth | 512.0 GB/s | 456.0 GB/s |

| Shading Units | 4096 | 2560 |

| TMUs | 256 | 160 |

| ROPs | 128 | 80 |

| Ray Tracing Cores | 32 | 20 |

| Pixel Rate | 307.2 GPixel/s | 192.0 GPixel/s |

| Texture Rate | 614.4 GTexel/s | 384.0 GTexel/s |

| FP32 | 19.66 TFLOPS | 12.29 TFLOPS |

| FP16 | 39.32 TFLOPS (2:1) | 24.58 TFLOPS (2:1) |

| TDP | 225 W | 200 W |

| Power Connectors | 1x 6-pin + 1x 8-pin | 1x 8-pin |

| Bus Interface | PCIe 4.0 x16 | PCIe 5.0 x8 |

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

| Production Status | null | Active |

| Release Date | 2025-12-31 | 2025-09-04 |

| Predecessor | Alchemist | null |

| Launch MSRP | null | 499 USD |

| Dimensions | null | 167 mm, 69 mm, 40 mm |

| Percentile vs All GPUs | 50 | 20 |

| Average Benchmark Score | 0 | 3182 |