Intel Arc B390 vs NVIDIA RTX PRO 6000D Blackwell Max-Q Comparison

The database places two active graphics products at opposite ends of the performance spectrum: the Intel Arc B390, an integrated part from the Arc Graphics-M (Panther Lake) generation, and the NVIDIA RTX PRO 6000D Blackwell Max-Q, a discrete workstation card. The single recorded 3DMark Steel Nomad DX12 result puts the NVIDIA part far ahead, and the surrounding data shows that the two designs target entirely different segments.

Head-to-Head Benchmarks

The database contains one head-to-head test: 3dmark_3dmark_steel_nomad_dx12. The Intel Arc B390 scores 1482. The NVIDIA RTX PRO 6000D Blackwell Max-Q scores 11088. The percentage difference is -86.6%, with the NVIDIA part as the winner. In terms of head-to-head wins, the NVIDIA part takes 1 and the Intel part takes 0.

The Intel part's closest rivals in the database are all low-score cards. The GeForce GT 520MX averages 1463, putting the Arc B390 1.3% ahead. The GeForce 800M averages 1460, a 1.5% gap. The GeForce GT 625 OEM averages 1446, a 2.5% gap. The GeForce GT 710 averages 1443, a 2.7% gap. These gaps are small, which means the Arc B390 sits near the bottom of the database's overall ranking: the 9th percentile of all GPUs.

The NVIDIA part's closest rivals form a different cluster. The RTX PRO 6000 Blackwell Max-Q matches 11088 with a 0% difference. The Radeon RX 550 averages 11075, putting the RTX PRO 6000D 0.1% ahead. The GeForce GTX 1650 SUPER averages 11047, a 0.4% gap. The AMD FirePro W4300 averages 11225, which is 1.2% above the RTX PRO 6000D. The NVIDIA part sits at the 50th percentile of all GPUs.

Architecture Differences

The Intel Arc B390 uses the Xe3-LPG architecture on the Panther Lake chip, part of the Arc Graphics-M (Panther Lake) generation. It is built on a 3 nm process at Intel. The database records no transistor count, die size, or transistor density for this part.

The NVIDIA RTX PRO 6000D Blackwell Max-Q uses Blackwell 2.0 on the GB202 chip, part of the Blackwell PRO W (x000) generation. It is built on a 5 nm process at TSMC. The database records 92,200 million transistors on a 750 mm² die, for a transistor density of 122.9M per mm².

The compute resources diverge sharply. The Intel part has 1536 shading units, 48 texture mapping units, 24 ROPs, and 12 ray tracing cores. It has no tensor core count recorded. The NVIDIA part has 24064 shading units, 752 texture mapping units, 192 ROPs, 188 ray tracing cores, and 752 tensor cores. Clock behavior also differs: the Intel part runs at a 300 MHz base and 2500 MHz boost, while the NVIDIA part runs at a 1590 MHz base and 2288 MHz boost.

Memory architecture is a fundamental split. The Intel Arc B390 uses system shared memory, with a system shared bus width and system dependent bandwidth. The NVIDIA part has 96 GB of GDDR7 on a 512-bit bus, delivering 1.79 TB/s of bandwidth. Its memory clock is listed as 1750 MHz with 28 Gbps effective.

Both parts expose the same API set in the database: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The physical designs are different: the Intel part is an IGP with no power connectors and an 80 W TDP, while the NVIDIA part is a dual-slot card with one 16-pin power connector, a 300 W TDP, and a suggested power supply of 700 W. Display outputs are portable device dependent on the Intel side, while the NVIDIA side has 4x DisplayPort 2.1b. The NVIDIA part uses PCIe 5.0 x16; the Intel part uses an IGP bus interface.

Where Each One Wins

The head-to-head data gives the NVIDIA part the only win. The Intel Arc B390 has no benchmark wins in the database. The NVIDIA RTX PRO 6000D Blackwell Max-Q wins the 3DMark Steel Nomad DX12 test by a large margin: 11088 against 1482.

The throughput specifications reinforce that result. The NVIDIA part records a pixel rate of 439.3 GPixel/s, a texture rate of 1,720.6 GTexel/s, FP32 performance of 110.1 TFLOPS, and FP16 performance of 110.1 TFLOPS (1:1). The Intel part records 60.00 GPixel/s, 120.0 GTexel/s, 7.680 TFLOPS FP32, and 15.36 TFLOPS FP16 (2:1). Every measured throughput figure in the database favors the NVIDIA part.

The Intel part's distinguishing characteristics are not benchmark wins but integration. It is an 80 W IGP with no power connectors, system shared memory, and portable device dependent display outputs. The NVIDIA part is a 300 W dual-slot card with 96 GB GDDR7, a 16-pin connector, a suggested 700 W power supply, PCIe 5.0 x16, and 4x DisplayPort 2.1b. Those differences define the use cases: the Intel part belongs to integrated portable systems, while the NVIDIA part is a discrete add-in card.

Specification Differences

| Field | Intel Arc B390 | NVIDIA RTX PRO 6000D Blackwell Max-Q |

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

| Manufacturer | Intel | NVIDIA |

| Chip | Panther Lake | GB202 |

| Architecture | Xe3-LPG | Blackwell 2.0 |

| Generation | Arc Graphics-M (Panther Lake) | Blackwell PRO W (x000) |

| Process node | 3 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | Not recorded | 92,200 million |

| Die size | Not recorded | 750 mm² |

| Transistor density | Not recorded | 122.9M / mm² |

| Base clock | 300 MHz | 1590 MHz |

| Boost clock | 2500 MHz | 2288 MHz |

| Memory clock | System Shared | 1750 MHz, 28 Gbps effective |

| Memory size | System Shared | 96 GB |

| Memory type | System Shared | GDDR7 |

| Memory bus width | System Shared | 512 bit |

| Memory bandwidth | System Dependent | 1.79 TB/s |

| Shading units | 1536 | 24064 |

| TMUs | 48 | 752 |

| ROPs | 24 | 192 |

| RT cores | 12 | 188 |

| Tensor cores | Not recorded | 752 |

| Pixel rate | 60.00 GPixel/s | 439.3 GPixel/s |

| Texture rate | 120.0 GTexel/s | 1,720.6 GTexel/s |

| FP32 | 7.680 TFLOPS | 110.1 TFLOPS |

| FP16 | 15.36 TFLOPS (2:1) | 110.1 TFLOPS (1:1) |

| TDP | 80 W | 300 W |

| Slot width | IGP | Dual-slot |

| Power connectors | None | 1x 16-pin |

| Suggested PSU | Not recorded | 700 W |

| Bus interface | IGP | PCIe 5.0 x16 |

| Display outputs | Portable Device Dependent | 4x DisplayPort 2.1b |

| Dimensions | Not recorded | 267 mm (10.5 in) x 111 mm (4.4 in) x 40 mm (1.6 in) |

| Release date | 2026-01-26 | 2025-03-17 |

| Launch MSRP | Not recorded | 8,565 USD |

| Predecessor | Not recorded | Workstation Ada |

Both parts are listed as active production parts, and both have no successor recorded in the database.

FAQ

Q: Which GPU wins the recorded 3DMark Steel Nomad DX12 benchmark?

A: The NVIDIA RTX PRO 6000D Blackwell Max-Q wins with a score of 11088. The Intel Arc B390 scores 1482, a difference of -86.6% from the NVIDIA result.

Q: What memory configurations do the two GPUs use?

A: The Intel Arc B390 uses system shared memory with system dependent bandwidth. The NVIDIA RTX PRO 6000D Blackwell Max-Q uses 96 GB of GDDR7 on a 512-bit bus with 1.79 TB/s bandwidth.

Q: How do the core counts compare?

A: The Intel part has 1536 shading units, 48 TMUs, 24 ROPs, and 12 RT cores. The NVIDIA part has 24064 shading units, 752 TMUs, 192 ROPs, and 188 RT cores.

Q: What API support do both parts share?

A: Both are listed with DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What are the power requirements recorded in the database?

A: The Intel Arc B390 has an 80 W TDP and no power connectors. The NVIDIA RTX PRO 6000D Blackwell Max-Q has a 300 W TDP, one 16-pin power connector, and a suggested power supply of 700 W.

Q: Where do the two parts sit in the overall GPU ranking?

A: The Intel Arc B390 is at the 9th percentile of all GPUs, while the NVIDIA RTX PRO 6000D Blackwell Max-Q is at the 50th percentile.

The Verdict

The recorded data separates the two parts cleanly. The NVIDIA RTX PRO 6000D Blackwell Max-Q wins the only head-to-head benchmark, 11088 to 1482, and leads every throughput specification in the database. It also carries the workstation memory configuration: 96 GB GDDR7, a 512-bit bus, and 1.79 TB/s bandwidth. The Intel Arc B390 has no benchmark win and sits at the 9th percentile, but it is the only integrated option in the comparison, with an 80 W TDP, no power connectors, system shared memory, and portable device dependent display outputs. The performance profile belongs to the NVIDIA part; the integrated profile belongs to the Intel part.