Intel Arc B390 vs Intel Arc Pro B370 Comparison

Head-to-Head Benchmarks

The recorded database contains a single benchmark submission for the Intel Arc B390, the 3DMark Steel Nomad DX12 test, where it scores 1482 points. The Intel Arc Pro B370 has no benchmark entries in the database, so direct head-to-head measurements are unavailable. The B390’s score places it in the 9th percentile of all GPUs tracked, meaning 91% of recorded graphics processors score higher. Its nearest rivals in the database are all NVIDIA parts: the GeForce GT 520MX at 1463 points, the GeForce 800M at 1460 points, the GeForce GT 625 OEM at 1446 points, and the GeForce GT 710 at 1443 points.

Against those rivals, the B390’s advantage is narrow. It sits 1.3% ahead of the GT 520MX, 1.5% ahead of the GeForce 800M, 2.5% ahead of the GT 625 OEM, and 2.7% ahead of the GT 710. These deltas are small, indicating that the B390 operates in the same performance tier as those older NVIDIA parts despite being a much newer design. The B390’s average benchmark score matches its sole Steel Nomad result at 1482, so there is no variance across multiple runs.

The Arc Pro B370 has no recorded scores, an average benchmark score of zero, and no nearest rivals listed. Its percentile field reads 50, but that figure cannot be substantiated by any actual measurement in the database. The absence of data means no quantitative comparison between the two Intel parts is possible from benchmark results alone. The B390 is the only one of the two with verified performance data, and that data indicates a low-end result compared to the broader GPU landscape.

FAQ

Q: What is the only benchmark score recorded for the Intel Arc B390?

A: The database contains one score for the B390: 1482 points in 3DMark Steel Nomad DX12. This result also serves as its average benchmark score, since no other tests are logged.

Q: Does the Intel Arc Pro B370 have any benchmark results in the database?

A: No. The Arc Pro B370’s benchmark array is empty, its average score is 0, and no nearest rivals are listed. The percentile field shows 50, but there is no underlying measurement to support that value.

Q: How does the Arc B390 compare to its closest recorded rivals?

A: The B390 leads the NVIDIA GeForce GT 520MX by 1.3%, the GeForce 800M by 1.5%, the GeForce GT 625 OEM by 2.5%, and the GeForce GT 710 by 2.7%. All four rivals score between 1443 and 1463 points, placing them within a narrow band below the B390.

Q: What percentile does the Arc B390 achieve among all GPUs?

A: The B390 sits in the 9th percentile of all GPUs in the database. This indicates that its 1482 Steel Nomad score is lower than approximately 91% of recorded graphics processors.

Q: Are the two Intel GPUs from the same chip and architecture?

A: Yes. Both the Arc B390 and the Arc Pro B370 use the Panther Lake chip with the Xe3-LPG architecture, fabricated on Intel’s 3 nm process. They differ in their generation labels: the B390 is Arc Graphics-M (Panther Lake), while the B370 is Arc Graphics-WM (Panther Lake).

Q: What is the launch date for both parts?

A: Both the Intel Arc B390 and the Intel Arc Pro B370 share the same release date of January 26, 2026. Neither has a recorded launch MSRP in the database.

Architecture Differences

The two GPUs share the same fundamental architecture: Xe3-LPG on the Panther Lake chip, built at Intel’s 3 nm node. Both are integrated graphics processors (IGP) with no dedicated memory, drawing from system shared memory with system-dependent bandwidth. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Display outputs are portable device dependent for both, and neither requires a power connector or a separate slot width beyond IGP.

The execution unit counts diverge. The Arc B390 contains 1536 shading units, 48 texture mapping units, 24 raster operations pipelines, and 12 ray tracing cores. The Arc Pro B370 is smaller: 1280 shading units, 40 TMUs, 20 ROPs, and 10 RT cores. This gives the B390 a 20% advantage in shading units (1536 versus 1280), a 20% lead in TMUs (48 versus 40), and a 20% lead in ROPs (24 versus 20). The RT core count also scales by 20%, with 12 versus 10.

Clock behavior differs slightly. The base clock is identical at 300 MHz for both. The boost clock is higher on the B390 at 2500 MHz versus 2400 MHz on the B370, a 4.2% increase. Neither part has a game clock listed. Pixel and texture rates follow the hardware counts and clocks. The B390 achieves 60.00 GPixel/s and 120.0 GTexel/s. The B370 achieves 48.00 GPixel/s and 96.00 GTexel/s. The B390’s pixel rate is 25% higher, and its texture rate is also 25% higher.

Compute throughput scales accordingly. The B390 delivers 7.680 TFLOPS of FP32 and 15.36 TFLOPS of FP16 (2:1 ratio). The B370 delivers 6.144 TFLOPS FP32 and 12.29 TFLOPS FP16 (2:1 ratio). The FP32 gap is 25%, and the FP16 gap is also 25%. Both parts have unknown transistor counts and die sizes, so no density comparison is possible.

The B370 lists a predecessor, HD Graphics-WM, while the B390 has no predecessor recorded. Neither part has a successor. Both are marked as Active production status. The B390’s generation string references Arc Graphics-M, indicating a mobile-oriented integrated part, while the B370 references Arc Graphics-WM, suggesting a workstation-mobile variant. Despite these label differences, the underlying architecture and process are identical.

Specification Differences

The recorded specifications show several clear differences between the two parts. The most prominent is power draw: the Arc B390 has a TDP of 80 W, while the Arc Pro B370 has a TDP of 25 W. This is a 3.2x difference in rated power consumption, making the B370 a much lower-power integrated solution.

Clock speeds differ only at the boost level. Both start at 300 MHz base. The B390 boosts to 2500 MHz, the B370 to 2400 MHz. The B390’s higher boost contributes to its higher pixel, texture, and compute rates.

Shader and fixed-function unit counts differ as described above: 1536 versus 1280 shading units, 48 versus 40 TMUs, 24 versus 20 ROPs, and 12 versus 10 RT cores. These are uniform 20% differences in each category.

Memory configuration is identical in form: both use system shared memory, system shared type, system shared bus width, and system dependent bandwidth. Neither has dedicated VRAM. Memory size is also system shared for both.

Display outputs are listed as portable device dependent for both, indicating no fixed display interface differences are recorded. The bus interface is IGP for both. Slot width is IGP for both. Power connectors are none for both. Suggested PSU is absent for both. Dimensions are all null for both.

The B370 has a predecessor, HD Graphics-WM, while the B390 does not. Both share the same release date, January 26, 2026. Both are Active in production. Neither has a launch MSRP. The B390 has one benchmark entry; the B370 has none.

Where Each One Wins

The Arc B390 wins outright in raw performance metrics. Its 1482 Steel Nomad score is the only recorded benchmark result between the two, and its hardware specifications are uniformly higher. The B390 has more shading units, more TMUs, more ROPs, more RT cores, a higher boost clock, higher pixel rate, higher texture rate, higher FP32 throughput, and higher FP16 throughput. In every computational metric recorded, the B390 leads by 20% to 25%.

The B390’s 80 W TDP is a disadvantage in power efficiency, but the database provides no benchmark scores for the B370 to compare performance per watt. Without B370 measurements, the B390’s performance lead is the only quantifiable outcome. The B390’s 9th percentile ranking shows it is a low-tier performer overall, but it still outpaces its nearest rivals by 1.3% to 2.7%.

The Arc Pro B370 wins in power consumption. Its 25 W TDP is far below the B390’s 80 W, making it the more power-conscious choice for integrated graphics scenarios. It also has a predecessor, HD Graphics-WM, which the B390 lacks, indicating a more defined product lineage. However, with no benchmark data, its performance cannot be verified.

For tasks that rely on raw fill rate or compute throughput, the B390 is the better option based on the recorded numbers. Its 60.00 GPixel/s and 120.0 GTexel/s are 25% higher than the B370’s 48.00 GPixel/s and 96.00 GTexel/s. Its FP32 throughput of 7.680 TFLOPS exceeds the B370’s 6.144 TFLOPS by the same margin.

For power-constrained integrated designs, the B370’s 25 W TDP is a strong advantage. The B390’s 80 W TDP is 3.2x higher, which could matter in compact portable devices where thermal and power budgets are tight. The B370’s lower boost clock of 2400 MHz versus 2500 MHz likely contributes to its reduced power draw, though the database does not record a direct power efficiency metric.

In ray tracing, the B390 has 12 RT cores versus 10 on the B370, a 20% advantage. In texturing, the B390’s 48 TMUs versus 40 TMUs gives it a proportional edge. In pixel output, 24 ROPs versus 20 ROPs yields a similar lead. These hardware differences translate directly to the fill rate and compute figures already noted.

The B390’s only benchmark score, 1482 in Steel Nomad, places it in the 9th percentile of all GPUs. That is a low result, but it still beats its nearest rivals by a small margin. The B370 has no comparable data, so its performance class cannot be placed relative to the B390 or any other GPU.

The B390 is the only one of the two with a verified performance footprint. The B370 remains unmeasured, so any claims about its performance are unsupported by the database. The B370’s lower power envelope is its clear specification advantage, while the B390 leads in every recorded performance metric.