Intel Arc B390 vs Intel Arc Pro B50 Comparison

Head-to-Head Benchmarks

The database records a single shared benchmark between the Intel Arc B390 and the Intel Arc Pro B50: the 3DMark Steel Nomad DX12 test. In this test, the Intel Arc Pro B50 scores 1604 points, while the Intel Arc B390 scores 1482 points. The delta is 7.6 percent in favor of the Arc Pro B50, which is a moderate but decisive margin.

This result places the Arc Pro B50 ahead of the Arc B390 in raw DX12 performance. The Arc Pro B50's score of 1604 reflects its dedicated graphics architecture and higher shader throughput. The Arc B390, with its 1482 points, trails by 122 points, a gap that is consistent with its integrated design and shared memory subsystem.

The Arc Pro B50 also carries a broader benchmark profile in the database, including PassMark DirectX 9, 10, 11, 12, G2D, G3D, and GPU compute tests. Its PassMark G3D score of 12553 and GPU compute score of 6037 indicate strong general-purpose and compute performance. The Arc B390 has no recorded entries in these additional tests, so the comparison relies on the single DX12 result.

In terms of overall standing, the Arc Pro B50 sits at the 17th percentile among all GPUs in the database, while the Arc B390 sits at the 9th percentile. This percentile gap is substantial and reinforces the raw benchmark delta. The Arc Pro B50's nearest rivals include the NVIDIA GeForce GT 1030 (average score 2662, delta -0.1 percent), the NVIDIA Quadro K1100M (2664, -0.2 percent), and the NVIDIA GeForce GT 440 (2645, 0.6 percent). Interestingly, the NVIDIA GeForce RTX 5070 SUPER also appears with an average score of 2690, a delta of -1.1 percent, which places the Arc Pro B50 within a narrow band of that high-end card in this particular aggregate metric.

The Arc B390's nearest rivals are much lower in the performance hierarchy. The NVIDIA GeForce GT 520MX (1463, 1.3 percent), the NVIDIA GeForce 800M (1460, 1.5 percent), the NVIDIA GeForce GT 625 OEM (1446, 2.5 percent), and the NVIDIA GeForce GT 710 (1443, 2.7 percent) all sit within a few points of the Arc B390's average score of 1482. This clustering indicates that the Arc B390 performs in the entry-level integrated segment, not in the discrete card tier occupied by the Arc Pro B50.

The average benchmark score in the database further separates the two: the Arc Pro B50 averages 2660 across all recorded tests, while the Arc B390 averages 1482. This 1178-point difference is a composite measure that includes the Arc Pro B50's strong PassMark results, which are absent for the Arc B390.

Architecture Differences

The two GPUs come from different architectural families. The Intel Arc B390 uses the Xe3-LPG architecture, built on the Panther Lake chip, and belongs to the Arc Graphics-M (Panther Lake) generation. The Intel Arc Pro B50 uses the Xe2-HPG architecture, built on the BMG-G21 chip, and belongs to the Battlemage (Pro Series) generation.

The manufacturing process differs significantly. The Arc B390 is fabricated on a 3 nm node at Intel's own foundry. The Arc Pro B50 uses a 5 nm node from TSMC. The Arc Pro B50's die is 272 mm² and contains 19,600 million transistors, giving a transistor density of 72.1M per mm². The Arc B390's transistor count and die size are not recorded in the database.

The execution resources diverge as well. The Arc Pro B50 has 2048 shading units, 128 texture mapping units, 16 render output units, and 16 ray tracing cores. The Arc B390 has 1536 shading units, 48 texture mapping units, 24 render output units, and 12 ray tracing cores. The Arc Pro B50 leads in shading units, texture units, and ray tracing cores, while the Arc B390 has more render output units.

Pixel and texture throughput reflect these differences. The Arc B390 reaches 60.00 GPixel/s and 120.0 GTexel/s, while the Arc Pro B50 reaches 41.60 GPixel/s and 332.8 GTexel/s. The Arc Pro B50's texture rate is nearly triple that of the Arc B390, but its pixel rate is lower. Floating-point performance follows the shader counts: the Arc B390 delivers 7.680 TFLOPS in FP32 and 15.36 TFLOPS in FP16 (2:1), while the Arc Pro B50 delivers 10.65 TFLOPS in FP32 and 21.30 TFLOPS in FP16 (2:1).

Memory architecture is a core distinction. The Arc B390 uses system-shared memory, with its size, type, bus width, and bandwidth all marked as system dependent. The Arc Pro B50 has 16 GB of dedicated GDDR6 memory on a 128-bit bus, delivering 224.0 GB/s of bandwidth. The Arc B390's memory clock is listed as system shared, whereas the Arc Pro B50's memory runs at 1750 MHz with 14 Gbps effective.

Clock behavior also differs. The Arc B390 has a base clock of 300 MHz and a boost clock of 2500 MHz. The Arc Pro B50 has a base clock of 1700 MHz and a boost clock of 2600 MHz. The Arc Pro B50's base clock is substantially higher, while the boost clocks are close.

The form factors are entirely different. The Arc B390 is an integrated graphics processor with an IGP slot width, no power connectors, and an IGP bus interface. Its display outputs are portable device dependent. The Arc Pro B50 is a dual-slot discrete card measuring 167 mm in length, 69 mm in height, and 40 mm in width. It uses a PCIe 5.0 x8 interface, has no power connectors, and offers four mini-DisplayPort 2.1 outputs. Its suggested power supply is 250 W.

Both GPUs share the same API support: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both are marked as Active in production status.

FAQ

Q: Which GPU wins the only shared benchmark test?

A: The Intel Arc Pro B50 wins the 3DMark Steel Nomad DX12 test with a score of 1604, against the Intel Arc B390's 1482, a delta of 7.6 percent.

Q: How do the shading unit counts compare?

A: The Intel Arc Pro B50 has 2048 shading units, while the Intel Arc B390 has 1536 shading units, a difference of 512 units in favor of the Arc Pro B50.

Q: What memory configurations do the two GPUs use?

A: The Intel Arc B390 uses system-shared memory with system-dependent bandwidth. The Intel Arc Pro B50 uses 16 GB of GDDR6 on a 128-bit bus with 224.0 GB/s bandwidth.

Q: Are the architectures different?

A: Yes. The Intel Arc B390 uses the Xe3-LPG architecture on the Panther Lake chip. The Intel Arc Pro B50 uses the Xe2-HPG architecture on the BMG-G21 chip.

Q: What are the power requirements for each card?

A: The Intel Arc B390 has a TDP of 80 W and no power connectors. The Intel Arc Pro B50 has a TDP of 70 W, no power connectors, and a suggested power supply of 250 W.

Q: What is the release date for each GPU?

A: The Intel Arc B390 was released on 2026-01-26. The Intel Arc Pro B50 was released on 2025-09-04.

Specification Differences

| Specification | Intel Arc B390 | Intel Arc Pro B50 |

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

| Chip | Panther Lake | BMG-G21 |

| Architecture | Xe3-LPG | Xe2-HPG |

| Generation | Arc Graphics-M (Panther Lake) | Battlemage (Pro Series) |

| Process Node | 3 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | unknown | 19,600 million |

| Die Size | unknown | 272 mm² |

| Transistor Density | null | 72.1M / mm² |

| Base Clock | 300 MHz | 1700 MHz |

| Boost Clock | 2500 MHz | 2600 MHz |

| Memory Size | System Shared | 16 GB |

| Memory Type | System Shared | GDDR6 |

| Memory Bus Width | System Shared | 128 bit |

| Memory Bandwidth | System Dependent | 224.0 GB/s |

| Memory Clock | System Shared | 1750 MHz 14 Gbps effective |

| Shading Units | 1536 | 2048 |

| TMUs | 48 | 128 |

| ROPs | 24 | 16 |

| RT Cores | 12 | 16 |

| Pixel Rate | 60.00 GPixel/s | 41.60 GPixel/s |

| Texture Rate | 120.0 GTexel/s | 332.8 GTexel/s |

| FP32 | 7.680 TFLOPS | 10.65 TFLOPS |

| FP16 | 15.36 TFLOPS (2:1) | 21.30 TFLOPS (2:1) |

| TDP | 80 W | 70 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | None |

| Suggested PSU | null | 250 W |

| Bus Interface | IGP | PCIe 5.0 x8 |

| Display Outputs | Portable Device Dependent | 4x mini-DisplayPort 2.1 |

| Dimensions | null | 167 mm 6.6 inches, 69 mm 2.7 inches, 40 mm 1.6 inches |

| Release Date | 2026-01-26 | 2025-09-04 |

| Launch MSRP | null | 349 USD |

The Verdict

The data indicates that the Intel Arc Pro B50 is the stronger performer in the single shared benchmark. Its 3DMark Steel Nomad score of 1604 beats the Intel Arc B390's 1482 by 7.6 percent. The Arc Pro B50 also occupies a higher percentile rank (17th versus 9th) and carries a much larger average benchmark score of 2660 versus 1482.

The Arc Pro B50's advantage comes from its dedicated memory and higher execution resource counts. With 2048 shading units, 128 TMUs, and 16 RT cores, it has more parallel throughput than the Arc B390's 1536 shading units, 48 TMUs, and 12 RT cores. Its 16 GB of GDDR6 memory on a 128-bit bus with 224.0 GB/s bandwidth provides a predictable memory subsystem, whereas the Arc B390 depends entirely on system-shared memory with system-dependent bandwidth.

The Arc B390 does have advantages in specific areas. Its 60.00 GPixel/s pixel rate exceeds the Arc Pro B50's 41.60 GPixel/s. Its 24 ROPs outnumber the Arc Pro B50's 16 ROPs. Its 3 nm process node is newer than the Arc Pro B50's 5 nm node. Its TDP of 80 W is higher than the Arc Pro B50's 70 W, but both cards draw no additional power connectors.

For users selecting between these two, the Arc Pro B50 is the clear choice for applications that depend on DX12 performance, texture throughput, FP32 compute, and dedicated memory bandwidth. The Arc B390 is suited for integrated systems where the GPU shares system memory and the form factor is an IGP, with no discrete card dimensions or display outputs of its own. The Arc Pro B50's launch MSRP is 349 USD, and its release date precedes the Arc B390 by several months. The recorded data does not show any benchmark win for the Arc B390 in the shared test, so the decision rests on the Arc Pro B50's superior measured performance.