Intel Arc B770 vs Intel Arc Pro B390 Comparison

Intel Arc B770 and Intel Arc Pro B390 represent two distinct interpretations of Intel’s graphics architecture, separated by a full generation of design philosophy and intended deployment. The B770 is a discrete, dual-slot Battlemage part built for raw throughput, while the B390 is an integrated graphics processor embedded into the Panther Lake platform. The recorded data shows no direct head-to-head benchmark results, so the analysis relies entirely on the architectural specifications and derived performance metrics.

Head-to-Head Benchmarks

The database contains no shared benchmark scores for these two processors. The Intel Arc B770 has an average benchmark score of 0 and holds a 50th percentile position against all GPUs. The Intel Arc Pro B390 also records an average benchmark score of 0 and a 50th percentile standing. With no wins recorded for either side, the head-to-head comparison must instead be built from the raw compute and rendering figures available.

The largest single gap appears in pixel throughput. The B770 delivers 307.2 GPixel/s, which is 5.12 times the 60.00 GPixel/s of the B390. That ratio indicates the B770 can fill framebuffers and handle high-resolution rasterization work at a substantially higher rate. The texture rate follows a similar pattern: 614.4 GTexel/s versus 120.0 GTexel/s, a 5.12x advantage for the discrete card. These two metrics alone place the B770 in a different performance class for any workload that stresses fill rate.

Compute throughput shows a narrower but still decisive gap. The B770 reaches 19.66 TFLOPS in FP32, while the B390 manages 7.680 TFLOPS. That works out to a 2.56x advantage for the B770. In FP16 with a 2:1 ratio, the B770 hits 39.32 TFLOPS against 15.36 TFLOPS, again a 2.56x margin. The consistency of these ratios across FP32 and FP16 suggests the architectural scaling between the two parts is uniform, with the B770 simply carrying more execution units at similar clock behavior.

Clock speeds tell a nuanced story. The B770 has a base clock of 2100 MHz and a boost of 2400 MHz. The B390 has a much lower base of 300 MHz but a higher boost of 2500 MHz. The B390’s boost advantage of 100 MHz over the B770 does not compensate for the massive difference in shading units: 4096 versus 1536. The B770 carries 2.67 times the shading units, 5.33 times the texture mapping units (256 versus 48), and 5.33 times the raster operation units (128 versus 24). Ray tracing cores are also unevenly distributed, with 32 on the B770 and 12 on the B390.

Memory bandwidth creates the most extreme separation. The B770 uses 16 GB of GDDR6 on a 256-bit bus, delivering 512.0 GB/s. The B390 uses system shared memory, with bandwidth listed as system dependent. In practical terms, the B770’s dedicated memory subsystem provides a fixed, high-bandwidth path that the integrated part cannot match, though the exact magnitude depends on the host system’s memory configuration.

Architecture Differences

The two processors come from different manufacturing nodes and foundries. The Intel Arc B770 uses the BMG-G31 chip built on a 5 nm process at TSMC. The Intel Arc Pro B390 uses the Panther Lake chip built on a 3 nm process at Intel. The die size is 368 mm² for the B770, while the B390’s die size is unknown. Transistor counts are not recorded for either part.

The architecture generations are not directly comparable. The B770 belongs to Xe2-HPG, part of the Battlemage (Arc 7) generation. The B390 belongs to Xe3-LPG, part of the Arc Graphics-WM (Panther Lake) generation. The B770’s predecessor is Alchemist, while the B390’s predecessor is HD Graphics-WM. This places the B390 one architectural step ahead in terms of generation numbering, but the Xe3-LPG design is optimized for low-power integrated use rather than peak discrete performance.

The B770 is a dual-slot discrete card with a 225 W TDP and requires a 6-pin and an 8-pin power connector. The suggested power supply is 550 W. The B390 is an integrated graphics processor with an 80 W TDP, no power connectors, and no suggested PSU rating. The B390’s slot width is listed as IGP, meaning it occupies no expansion slot. The bus interface for the B770 is PCIe 4.0 x16, while the B390 uses an integrated graphics path.

Memory architecture is fundamentally different. The B770 has 16 GB of GDDR6 with a 256-bit bus and 512.0 GB/s bandwidth. The B390 uses system shared memory for size, type, bus width, and bandwidth. The B770’s memory clock is 2000 MHz with 16 Gbps effective speed. The B390 has no dedicated memory clock because it relies on the host’s memory subsystem.

Display outputs also differ. The B770 provides 1x HDMI 2.1a and 3x DisplayPort 2.1. The B390’s display outputs are listed as portable device dependent. The B770 is a standalone add-in card, while the B390 is integrated into a processor package, meaning its output capabilities are tied to the host device.

Both processors share the same API support: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. This means software compatibility at the API level is identical, and the differences in performance come down to raw hardware resources rather than feature support.

Where Each One Wins

The Intel Arc B770 wins in every fixed hardware metric recorded. It has more shading units, more texture mapping units, more raster operation units, more ray tracing cores, higher pixel rate, higher texture rate, higher FP32 throughput, higher FP16 throughput, and dedicated memory bandwidth. The B770 also has a higher base clock and a larger die. For any workload that scales with compute units or memory bandwidth, the B770 is the clear choice.

The Intel Arc Pro B390 wins in power efficiency and physical integration. Its 80 W TDP is 145 W lower than the B770’s 225 W TDP. It requires no power connectors and no separate slot, making it suitable for compact or portable systems. The B390’s 3 nm process node at Intel is one generation ahead of the B770’s 5 nm TSMC node, which likely contributes to the lower power draw despite the higher boost clock of 2500 MHz versus 2400 MHz.

The B390’s boost clock advantage of 100 MHz suggests that, per clock, the Xe3-LPG design can reach higher frequencies. However, the B770’s much larger execution resource pool means the per-clock advantage does not translate into a throughput win. The B390 also has a later release date of 2026-01-26 compared to the B770’s 2025-12-31, and it holds an active production status, while the B770’s production status is not recorded.

For use cases, the B770 is positioned for desktop systems with a PCIe 4.0 x16 slot, a 550 W power supply, and a need for 16 GB of VRAM. The B390 is positioned for integrated systems where the GPU shares system memory and power is constrained to 80 W. The B390’s system dependent bandwidth means its real-world performance will vary with the host platform, whereas the B770’s 512.0 GB/s is a fixed characteristic.

FAQ

Q: Which processor has higher raw compute performance?

A: The Intel Arc B770 delivers 19.66 TFLOPS FP32 and 39.32 TFLOPS FP16, while the Intel Arc Pro B390 delivers 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16. The B770 is 2.56x faster in both precision formats.

Q: How do the memory systems compare?

A: The B770 uses 16 GB of GDDR6 on a 256-bit bus with 512.0 GB/s bandwidth. The B390 uses system shared memory with system dependent bandwidth. The B770’s dedicated memory provides a fixed bandwidth advantage.

Q: What is the power consumption difference?

A: The B770 has a 225 W TDP and requires a 6-pin and 8-pin power connector with a 550 W suggested PSU. The B390 has an 80 W TDP, no power connectors, and no suggested PSU.

Q: Are the API features the same?

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

Q: Which processor has a higher boost clock?

A: The B390 boosts to 2500 MHz, which is 100 MHz higher than the B770’s 2400 MHz boost. The B770 has a much higher base clock at 2100 MHz versus 300 MHz.

Q: What are the physical form factors?

A: The B770 is a dual-slot discrete card using PCIe 4.0 x16. The B390 is an integrated graphics processor with an IGP slot width and no expansion slot requirement.

The Verdict

The recorded data indicates the Intel Arc B770 is the superior performer in every measured physical capability. It offers 2.56x the FP32 throughput, 5.12x the pixel rate, 5.12x the texture rate, and 5.33x the ROPs and TMUs of the B390. Its 512.0 GB/s dedicated bandwidth versus system dependent memory further widens the gap. The B770 is designed for workloads that demand sustained throughput: high-resolution rendering, compute-heavy tasks, and applications that benefit from 16 GB of VRAM.

The Intel Arc Pro B390 is the more efficient and portable option. Its 80 W TDP fits into power-constrained platforms, and its integrated nature removes the need for external power connectors or a dedicated slot. The 3 nm Intel process node gives it a manufacturing advantage and a higher boost clock of 2500 MHz, but the smaller resource pool of 1536 shading units versus 4096 limits its ceiling.

Users who need maximum graphics throughput should select the B770 based on its 19.66 TFLOPS FP32 output and 307.2 GPixel/s pixel rate. Users who need an integrated graphics solution with a 2500 MHz boost clock and 80 W power envelope should select the B390. The B770’s 2025-12-31 release date and the B390’s 2026-01-26 release date place them in adjacent product cycles, but the architectural intent diverges sharply: one is a discrete performance part, the other is a low-power integrated component.

Specification Differences

| Specification | Intel Arc B770 | Intel Arc Pro B390 |

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

| Chip | BMG-G31 | Panther Lake |

| Architecture | Xe2-HPG | Xe3-LPG |

| Generation | Battlemage (Arc 7) | Arc Graphics-WM (Panther Lake) |

| Process Node | 5 nm | 3 nm |

| Foundry | TSMC | Intel |

| Die Size | 368 mm² | Unknown |

| Base Clock | 2100 MHz | 300 MHz |

| Boost Clock | 2400 MHz | 2500 MHz |

| Memory Size | 16 GB | System Shared |

| Memory Type | GDDR6 | System Shared |

| Memory Bus Width | 256 bit | System Shared |

| Memory Bandwidth | 512.0 GB/s | System Dependent |

| Shading Units | 4096 | 1536 |

| TMUs | 256 | 48 |

| ROPs | 128 | 24 |

| RT Cores | 32 | 12 |

| Pixel Rate | 307.2 GPixel/s | 60.00 GPixel/s |

| Texture Rate | 614.4 GTexel/s | 120.0 GTexel/s |

| FP32 | 19.66 TFLOPS | 7.680 TFLOPS |

| FP16 | 39.32 TFLOPS | 15.36 TFLOPS |

| TDP | 225 W | 80 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 1x 6-pin + 1x 8-pin | None |

| Suggested PSU | 550 W | None |

| Bus Interface | PCIe 4.0 x16 | IGP |

| Display Outputs | 1x HDMI 2.1a, 3x DisplayPort 2.1 | Portable Device Dependent |

| Release Date | 2025-12-31 | 2026-01-26 |

| Predecessor | Alchemist | HD Graphics-WM |

| Production Status | Unknown | Active |