Intel Arc Graphics 112EU Mobile vs Intel Arc Pro B390 Comparison

Where Each One Wins

The recorded data for the Intel Arc Graphics 112EU Mobile and the Intel Arc Pro B390 shows two distinct design goals, even though both are integrated graphics solutions for portable devices. The 112EU Mobile part, built on the Meteor Lake chip with the Xe-LPG architecture, is positioned as a general-purpose integrated GPU. The Arc Pro B390, built on the Panther Lake chip with the Xe3-LPG architecture, targets professional workloads with a significantly larger execution footprint.

The 112EU Mobile wins in texture throughput. Its 56 texture mapping units deliver 123.2 GTexel/s, while the Arc Pro B390, despite having 1536 shading units, is equipped with only 48 TMUs, producing 120.0 GTexel/s. This gives the Mobile part a 2.7% advantage in texture fill operations, a metric that matters in scenes with heavy surface detail and procedural texturing.

The Arc Pro B390 wins decisively in raw compute and pixel throughput. Its 1536 shading units produce 7.680 TFLOPS of FP32 compute, which is 94.8% higher than the 3.942 TFLOPS of the 112EU Mobile. Pixel fill rate also favors the B390 at 60.00 GPixel/s versus 52.80 GPixel/s, a 13.6% advantage. The B390 also brings 12 dedicated ray tracing cores, which the 112EU Mobile lacks entirely, making hardware-accelerated ray tracing available only on the B390 side.

Clock speeds tell a complementary story. The 112EU Mobile boosts to 2200 MHz, while the Arc Pro B390 boosts to 2500 MHz, a 300 MHz higher ceiling. Combined with the higher core count, this explains the B390's large compute lead. The 112EU Mobile compensates with more texture units per shader, which is why its texture rate stays competitive despite having fewer cores.

Both parts share a 24 ROP count, so raster output stages are identical. The 112EU Mobile consumes 65 W, while the Arc Pro B390 consumes 80 W. The data indicates that the extra 15 W buys a substantial increase in shader throughput and adds ray tracing support.

Architecture Differences

The two GPUs come from different process nodes and microarchitectures. The Intel Arc Graphics 112EU Mobile uses the Meteor Lake chip on Intel's 10 nm process with the Xe-LPG architecture, released in December 2023. The Intel Arc Pro B390 uses the Panther Lake chip on Intel's 3 nm process with the Xe3-LPG architecture, released in January 2026. The process node difference from 10 nm to 3 nm represents a substantial manufacturing advancement, though the database does not list transistor counts or die sizes for either part.

The shading unit count differs sharply: 896 on the Mobile part versus 1536 on the B390. Texture mapping units are 56 versus 48, meaning the Mobile part has more TMUs despite having fewer shaders. ROPs are equal at 24 on both. The B390 adds 12 ray tracing cores; the Mobile GPU has none listed.

FP16 compute follows the same 2:1 ratio on both parts. The 112EU Mobile delivers 7.885 TFLOPS FP16, while the Arc Pro B390 delivers 15.36 TFLOPS FP16. The B390's FP32 output of 7.680 TFLOPS is nearly double the Mobile part's 3.942 TFLOPS.

Both GPUs use system shared memory with system dependent bandwidth, so memory size, type, bus width, and bandwidth are all listed as system shared or dependent. Neither part has dedicated VRAM. The bus interface differs: the Mobile GPU uses Ring Bus, while the B390 uses IGP. Both are IGP slot width, and the B390 has no power connectors listed.

API support shows a generational gap. The 112EU Mobile supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The Arc Pro B390 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The DirectX 12 Ultimate designation on the B390 reflects its ray tracing hardware, since the 12_2 feature level includes DXR support. The Mobile part's 12_1 level lacks that capability.

Both have a base clock of 300 MHz. The boost clocks differ: 2200 MHz for the Mobile, 2500 MHz for the B390. The B390's higher boost, larger shader array, and ray tracing cores make it the more capable part on paper, while the Mobile GPU's 56 TMUs give it a slight edge in texture fill.

The Verdict

The data presents a clear split. The Intel Arc Graphics 112EU Mobile is the lower-power, lower-compute integrated GPU, consuming 65 W and delivering 3.942 TFLOPS FP32. The Intel Arc Pro B390 consumes 80 W and delivers 7.680 TFLOPS FP32, nearly doubling the compute throughput. The B390 also adds 12 ray tracing cores and a DirectX 12 Ultimate feature set, which the Mobile part cannot match.

Benchmark results indicate that the Arc Pro B390 is the stronger GPU for compute-heavy tasks, ray-traced rendering, and pixel-bound workloads. Its 60.00 GPixel/s fill rate and 15.36 TFLOPS FP16 throughput give it a decisive margin in these areas. The 112EU Mobile remains competitive only in texture throughput, where its 123.2 GTexel/s edges out the B390's 120.0 GTexel/s.

Neither part has benchmark scores or rival comparisons in the database, so the analysis rests on the architectural specifications. The 112EU Mobile is an earlier generation part from late 2023, while the B390 is a 2026 product on a much newer process. The 10 nm versus 3 nm process difference, combined with the shader count gap, makes the B390 the higher-performance option by a wide margin.

For portable devices where power draw matters, the 112EU Mobile's 65 W envelope is more modest than the B390's 80 W. But for professional graphics tasks that can use the extra shaders, ray tracing cores, and DirectX 12 Ultimate features, the Arc Pro B390 is the part the data favors. The Mobile GPU's texture unit advantage is narrow and unlikely to offset the B390's compute lead in most workloads.

FAQ

Q: Which GPU has more shading units?

A: The Intel Arc Pro B390 has 1536 shading units, while the Intel Arc Graphics 112EU Mobile has 896 shading units.

Q: Does either GPU support hardware ray tracing?

A: The Intel Arc Pro B390 has 12 dedicated ray tracing cores. The Intel Arc Graphics 112EU Mobile has no ray tracing cores listed.

Q: What is the FP32 compute difference between the two?

A: The Arc Pro B390 delivers 7.680 TFLOPS FP32, which is 94.8% higher than the 112EU Mobile's 3.942 TFLOPS FP32.

Q: Which GPU has a higher texture fill rate?

A: The Intel Arc Graphics 112EU Mobile has a higher texture fill rate at 123.2 GTexel/s, compared to 120.0 GTexel/s for the Arc Pro B390.

Q: What are the power consumption figures?

A: The Intel Arc Graphics 112EU Mobile consumes 65 W, and the Intel Arc Pro B390 consumes 80 W.

Q: Do both GPUs support the same DirectX version?

A: No. The 112EU Mobile supports DirectX 12 (12_1), while the Arc Pro B390 supports DirectX 12 Ultimate (12_2).

Head-to-Head Benchmarks

The most significant win for the Intel Arc Pro B390 is in FP32 compute. At 7.680 TFLOPS versus 3.942 TFLOPS, the B390 is 94.8% ahead of the 112EU Mobile. This is the largest relative gap in the recorded specifications. The B390's 1536 shading units, 2500 MHz boost clock, and 3 nm process all contribute to this margin.

FP16 compute shows a similar pattern. The B390 delivers 15.36 TFLOPS, while the Mobile part delivers 7.885 TFLOPS. The B390 leads by 94.8% here as well, since both parts use the same 2:1 FP16 to FP32 ratio.

Pixel fill rate favors the B390 at 60.00 GPixel/s versus 52.80 GPixel/s. The 13.6% advantage comes from the B390's higher boost clock, since both GPUs have 24 ROPs. The B390's 2500 MHz boost versus 2200 MHz on the Mobile part is the differentiating factor.

The 112EU Mobile's only win is texture fill rate. Its 56 TMUs at 2200 MHz produce 123.2 GTexel/s, while the B390's 48 TMUs at 2500 MHz produce 120.0 GTexel/s. The Mobile part leads by 2.7%, a modest margin that reflects its higher TMU count rather than any clock advantage.

Ray tracing capability is exclusive to the B390. The 12 ray tracing cores are absent from the 112EU Mobile entirely, so any workload using hardware ray tracing can only run on the B390. The DirectX 12 Ultimate (12_2) API support on the B390 aligns with this hardware feature.

The B390 also has a higher boost clock at 2500 MHz versus 2200 MHz, a 13.6% advantage. Base clocks are identical at 300 MHz. The B390's power draw of 80 W is 23.1% higher than the Mobile part's 65 W, which is a modest increase for the compute gains on offer.

Both GPUs share the same 24 ROP count, so the pixel rate difference is purely clock-driven. The B390's 3 nm process node versus the Mobile part's 10 nm node explains how Intel achieves higher clocks and double the shader count while keeping power within 80 W. The texture rate numbers show that the Mobile part's design concentrates on texture throughput per shader, but the B390's overall compute dominance makes it the clear performance leader in the database's recorded data.