Intel Arc B390 vs Intel Arc Graphics 4 Xe Mobile Comparison

Architecture Differences

The Intel Arc B390 and Intel Arc Graphics 4 Xe Mobile share the same fundamental silicon heritage. Both are built on the Panther Lake chip, use the Xe3-LPG architecture, and belong to the Arc Graphics-M (Panther Lake) generation. Intel fabricates both on a 3 nm process node at its own foundry. Transistor counts and die sizes are recorded as unknown in the database, so no differentiation can be made on those fronts.

The two parts diverge sharply in execution resources. The Arc B390 deploys 1536 shading units, 48 texture mapping units, 24 raster operation units, and 12 ray tracing cores. The Arc Graphics 4 Xe Mobile halves or more in every category: 512 shading units, 32 TMUs, 16 ROPs, and 4 ray tracing cores. That represents a 3x advantage in shading units for the B390, a 1.5x advantage in TMUs, a 1.5x advantage in ROPs, and a 3x advantage in ray tracing cores. The B390 is not a mildly upgraded variant; it is a substantially larger configuration of the same architecture.

Clock behavior also separates the pair. The B390 boosts to 2500 MHz, while the mobile part reaches 2300 MHz. Both share a 300 MHz base clock. The B390 therefore combines a higher peak frequency with a much wider execution engine, compounding its lead in raw throughput calculations.

Both GPUs are integrated parts. The slot width is IGP for each, the bus interface is IGP, and neither uses power connectors. Display outputs are listed as portable device dependent for both, meaning the integrated nature dictates connectivity through the host system. Memory is system shared for both, with bus width and type also system shared, and bandwidth recorded as system dependent.

API support is identical. Both expose DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Feature-level parity is therefore complete; any performance gap is a matter of resource scaling and clocks, not API capability.

FAQ

Q: Are these two GPUs based on the same architecture?

A: Yes. Both the Intel Arc B390 and the Intel Arc Graphics 4 Xe Mobile use the Xe3-LPG architecture on the Panther Lake chip, fabricated on Intel's 3 nm process.

Q: Which GPU has more shading units?

A: The Intel Arc B390 has 1536 shading units, which is three times the 512 shading units found in the Intel Arc Graphics 4 Xe Mobile.

Q: How do the boost clocks compare?

A: The Arc B390 boosts to 2500 MHz, while the Arc Graphics 4 Xe Mobile boosts to 2300 MHz. Both have a 300 MHz base clock.

Q: What is the TDP difference?

A: The Arc B390 is rated at 80 W, whereas the Arc Graphics 4 Xe Mobile is rated at 25 W. The B390 draws substantially more power in exchange for its larger execution engine.

Q: Do both GPUs support the same graphics APIs?

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

Q: Which GPU has better ray tracing resources?

A: The Arc B390 has 12 ray tracing cores, three times the 4 ray tracing cores in the Arc Graphics 4 Xe Mobile.

Q: Is there a benchmark score for the Arc Graphics 4 Xe Mobile?

A: The database lists no benchmark entries for the Arc Graphics 4 Xe Mobile, so its average benchmark score is recorded as zero. The Arc B390 has a 3DMark Steel Nomad DX12 score of 1482.

Specification Differences

The recorded specifications for the two GPUs differ in the following fields:

• Boost clock: 2500 MHz for the Arc B390, 2300 MHz for the Arc Graphics 4 Xe Mobile
• Shading units: 1536 versus 512
• Texture mapping units: 48 versus 32
• Raster operation units: 24 versus 16
• Ray tracing cores: 12 versus 4
• Pixel rate: 60.00 GPixel/s versus 36.80 GPixel/s
• Texture rate: 120.0 GTexel/s versus 73.60 GTexel/s
• FP32 throughput: 7.680 TFLOPS versus 2.355 TFLOPS
• FP16 throughput: 15.36 TFLOPS (2:1) versus 4.710 TFLOPS (2:1)
• TDP: 80 W versus 25 W
• Benchmark score: 1482 in 3DMark Steel Nomad DX12 for the B390, no recorded score for the mobile part
• Percentile versus all GPUs: 9th percentile for the B390, 50th percentile for the mobile part

All other fields match: base clock, memory configuration, slot width, power connectors, bus interface, display outputs, API support, production status, and release date. Both were released on 2026-01-26 and are listed as active products.

The Verdict

The data points to a clear segmentation. The Intel Arc B390 is the high-throughput integrated solution. Its 1536 shading units, 12 ray tracing cores, and 2500 MHz boost clock deliver 7.680 TFLOPS of FP32 compute and a 3DMark Steel Nomad DX12 score of 1482. The Intel Arc Graphics 4 Xe Mobile is the low-power integrated solution, with 512 shading units, 4 ray tracing cores, a 2300 MHz boost clock, and 2.355 TFLOPS of FP32 compute. It carries no recorded benchmark score.

The B390's 80 W TDP indicates a design intended for systems that can feed and cool a larger integrated GPU. The mobile part's 25 W TDP suggests deployment in power-constrained portable devices. The specification sheet alone makes this distinction obvious: the B390 trades power for throughput, while the mobile part trades throughput for efficiency.

The percentile placement reinforces the story. The B390 sits at the 9th percentile among all GPUs in the database, with a recorded average score of 1482. The Arc Graphics 4 Xe Mobile sits at the 50th percentile, but that percentile is based on no recorded benchmark entries, so it reflects database positioning rather than measured performance. The B390's nearest rivals, the NVIDIA GeForce GT 520MX at 1463, the GeForce 800M at 1460, the GeForce GT 625 OEM at 1446, and the GeForce GT 710 at 1443, all cluster within 1.3% to 2.7% below its score.

For a system builder selecting between these two integrated options, the choice is dictated by power delivery and thermal envelope. The B390 is the only one of the pair with a measured performance record, and it outperforms its closest recorded rivals by a small but consistent margin. The Arc Graphics 4 Xe Mobile has no benchmark data, so its real-world standing cannot be quantified from the database. The selection should hinge on whether the host platform can accommodate the 80 W B390 or needs the 25 W mobile part.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark entries between these two GPUs, and the wins counters for both are zero. Direct comparison data does not exist. However, the Arc B390 has a standalone benchmark result: 1482 in 3DMark Steel Nomad DX12. That score places it 1.3% ahead of the NVIDIA GeForce GT 520MX (1463), 1.5% ahead of the NVIDIA GeForce 800M (1460), 2.5% ahead of the NVIDIA GeForce GT 625 OEM (1446), and 2.7% ahead of the NVIDIA GeForce GT 710 (1443).

The Arc Graphics 4 Xe Mobile has no benchmark entries, so no measured score exists to place against the B390 or any rival. The only quantitative comparison available is derived from the specification sheet. The B390's FP32 throughput of 7.680 TFLOPS is approximately 3.26 times the mobile part's 2.355 TFLOPS. Its texture rate of 120.0 GTexel/s is 1.63 times the mobile part's 73.60 GTexel/s. Its pixel rate of 60.00 GPixel/s is 1.63 times the mobile part's 36.80 GPixel/s. These ratios all follow from the resource counts: the B390 has three times the shading units and ray tracing cores, and 1.5 times the TMUs and ROPs, with a 200 MHz higher boost clock.

The B390's FP16 throughput of 15.36 TFLOPS (2:1) is likewise about 3.26 times the mobile part's 4.710 TFLOPS (2:1). The consistent 3.26x ratio across both FP32 and FP16 reflects the identical architecture and the pure scaling of shader count and clock. The mobile part's lower pixel and texture rates indicate proportionally reduced fillrate capacity.

Where Each One Wins

The Intel Arc B390 wins in every measured performance category. Its shading unit count triples the mobile part. Its ray tracing core count triples as well. Its boost clock is 200 MHz higher. Its pixel rate, texture rate, FP32 throughput, and FP16 throughput are all higher by the same proportional margins. It is the only one of the two with a recorded benchmark score, and that score, 1482, places it above four nearby NVIDIA rivals by margins ranging from 1.3% to 2.7%.

The Intel Arc Graphics 4 Xe Mobile wins in power efficiency. Its 25 W TDP is less than one third of the B390's 80 W TDP. For portable devices with constrained thermal budgets, that power draw is the defining attribute. The mobile part also matches the B390 on every API front, so software compatibility is identical. It shares the same DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4 support, the same 3 nm process, and the same Panther Lake chip family.

The use-case split is therefore straightforward. Systems that prioritize graphics throughput, ray tracing capacity, and raw fillrate should select the Arc B390. Systems that prioritize power draw and thermal restraint should select the Arc Graphics 4 Xe Mobile. The B390 targets scenarios where 80 W is acceptable and maximum integrated performance is required. The mobile part targets scenarios where 25 W is the limit and moderate performance suffices. The database records no measured performance for the mobile part, so its actual capability relative to the B390 cannot be stated beyond the specification-derived ratios. The B390's measured score and its narrow lead over its nearest rivals confirm that it is a low-absolute-performance part in the global GPU landscape, sitting at the 9th percentile, but it remains the stronger of the two integrated options by every recorded metric.