Intel Arc 140T Mobile vs Intel Arc Pro B390 Comparison

Intel Arc 140T Mobile and Intel Arc Pro B390 are both integrated graphics solutions from Intel, but they target different performance envelopes and are built on distinct architectures. The database records no direct head-to-head benchmark matches between them, yet their specifications and compute capabilities allow a clear analytical comparison. The Arc 140T Mobile relies on the Xe-LPG+ architecture on a 5 nm TSMC process, while the Arc Pro B390 uses the newer Xe3-LPG architecture on Intel’s 3 nm node. Both sit at the 50th percentile among all GPUs in the database, indicating they are mid-pack performers rather than outliers. The data reveals a substantial difference in raw compute, memory configuration, and thermal design, which translates into distinct use-case advantages.

Where Each One Wins

For raw compute throughput, the Intel Arc Pro B390 dominates. Its FP32 performance of 7.680 TFLOPS is roughly 60% higher than the Arc 140T Mobile’s 4.813 TFLOPS. FP16 performance follows the same pattern, with the B390 delivering 15.36 TFLOPS versus 9.626 TFLOPS for the 140T. This makes the B390 the clear choice for workloads that scale with shader count, such as rendering, scientific simulation, or compute-heavy content creation. The B390 also has more shading units at 1536 versus 1024, and more ray tracing cores at 12 versus 8. For any application that leverages ray tracing, the B390 holds a structural advantage.

The Intel Arc 140T Mobile, however, wins on efficiency and throughput per watt. Its 35 W TDP is less than half of the B390’s 80 W TDP, yet it delivers more than half of the FP32 performance. The 140T also achieves a higher pixel rate of 75.20 GPixel/s compared to 60.00 GPixel/s for the B390, and a higher texture rate of 150.4 GTexel/s versus 120.0 GTexel/s. This is counterintuitive given the B390’s larger shader array, but the 140T’s higher TMU count of 64 versus 48 and higher ROP count of 32 versus 24 explain the advantage in fill-rate-bound tasks. The 140T is the better performer for lightweight graphics workloads, older game titles, or any scenario where power draw and heat generation must stay low.

The B390 wins in compute density and future-proofing. Its Xe3-LPG architecture is a newer generation, and its 3 nm process node indicates a more advanced manufacturing technology. The 140T’s Xe-LPG+ is a previous-generation design on 5 nm. For users prioritizing peak performance regardless of power, the B390 is the superior part. For users prioritizing portability, battery life, and adequate graphics performance in a thin-and-light device, the 140T is the more sensible choice.

FAQ

Q: Which GPU has higher FP32 performance?

A: The Intel Arc Pro B390 has an FP32 rating of 7.680 TFLOPS, which is significantly higher than the Intel Arc 140T Mobile’s 4.813 TFLOPS.

Q: Are both GPUs integrated or discrete?

A: Both are integrated graphics processors. The Arc 140T Mobile uses an IGP slot width and the Arc Pro B390 also uses an IGP slot width, with no power connectors required for the B390.

Q: What is the TDP difference between the two?

A: The Arc 140T Mobile has a TDP of 35 W, while the Arc Pro B390 has a TDP of 80 W. The 140T consumes less than half the power of the B390.

Q: Do they share the same API support?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. There is no API difference between the two in the recorded data.

Q: Which GPU has more ray tracing cores?

A: The Arc Pro B390 has 12 ray tracing cores, while the Arc 140T Mobile has 8. The B390 provides 50% more ray tracing hardware.

Q: What are the boost clock speeds?

A: The Arc 140T Mobile boosts to 2350 MHz, while the Arc Pro B390 boosts to 2500 MHz. The B390 has a higher boost clock by 150 MHz.

Head-to-Head Benchmarks

No head-to-head benchmark results exist in the database, but the specification-derived performance metrics provide a quantitative comparison. In FP32 compute, the B390’s 7.680 TFLOPS represents a 59.5% advantage over the 140T’s 4.813 TFLOPS. This is the largest single-metric gap in the comparison. FP16 compute shows a similar ratio: 15.36 TFLOPS versus 9.626 TFLOPS, a 59.6% lead for the B390. These figures suggest that any workload heavily dependent on shader arithmetic will see a near-linear scaling benefit from the B390’s extra shading units.

The texture rate and pixel rate metrics flip the script. The 140T’s 150.4 GTexel/s outperforms the B390’s 120.0 GTexel/s by 25.3%. Its pixel rate of 75.20 GPixel/s beats the B390’s 60.00 GPixel/s by 25.3% as well. This is a consistent pattern: the 140T has more texture mapping units (64 versus 48) and more render output units (32 versus 24), which directly boosts fill-rate performance. In games or applications that are fill-rate limited, such as those with heavy alpha blending or high-resolution texture sampling, the 140T can outperform the B390 despite lower raw compute.

Clock speeds also favor the B390, but not decisively. The B390’s boost clock of 2500 MHz is 6.4% higher than the 140T’s 2350 MHz. Both share the same 300 MHz base clock. The B390’s higher boost clock partially explains its compute advantage, but the larger shader count is the dominant factor. The 140T’s lower clock and lower shader count combine to reduce its peak throughput, yet its fill-rate advantage persists due to the higher TMU and ROP counts.

Ray tracing capacity is another clear B390 win. With 12 RT cores versus 8, the B390 has 50% more ray tracing hardware. This does not directly translate to a 50% performance gain in all ray-traced scenes, as memory bandwidth and driver efficiency also play roles, but the hardware headroom is present. For any ray-traced workload, the B390 is the stronger candidate based on the recorded specifications.

Power efficiency is where the 140T reclaims ground. At 35 W, the 140T delivers 4.813 TFLOPS, which equates to roughly 137.5 GFLOPS per watt. The B390 at 80 W delivers 7.680 TFLOPS, which is about 96 GFLOPS per watt. The 140T is approximately 43% more power-efficient in FP32 compute per watt. This is a significant differentiator for mobile devices where thermal and battery constraints are paramount.

Specification Differences

The most obvious difference is the architecture generation. The Arc 140T Mobile uses Xe-LPG+, while the Arc Pro B390 uses Xe3-LPG. This is not a minor revision; it represents a full architectural generation jump. The process node also differs substantially: the 140T is fabricated on a 5 nm TSMC process, while the B390 uses a 3 nm Intel process. The B390’s process node is more advanced, likely enabling higher transistor density and better power characteristics at the same clock speed, though the recorded data shows the B390 consumes more power overall.

Shading units differ significantly: 1024 for the 140T versus 1536 for the B390. This is a 50% increase in shader count for the B390. Texture mapping units, however, favor the 140T at 64 versus 48. Render output units also favor the 140T at 32 versus 24. Ray tracing cores favor the B390 at 12 versus 8. The B390 has no tensor cores recorded for either part, so no comparison is possible there.

Clock speeds differ in boost: 2350 MHz for the 140T versus 2500 MHz for the B390. Base clocks are identical at 300 MHz. The TDP difference is stark: 35 W for the 140T versus 80 W for the B390. Both are integrated graphics with IGP bus interfaces and no power connectors for the B390; the 140T has no power connector data recorded. Both use system-shared memory with system-dependent bandwidth, so no memory capacity or bandwidth figures can be compared.

Pixel rate and texture rate both favor the 140T, as previously noted. The B390’s FP32 and FP16 rates are higher. Both support the same API set: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both are listed as Active in production status. The 140T was released earlier, with a release date of January 2025, while the B390 has a later release date of January 2026. Neither has a recorded launch MSRP.

Architecture Differences

The architectural split is clean. The Arc 140T Mobile is built on Xe-LPG+, which is a refinement of Intel’s Xe-LPG architecture. It is part of the Arc Graphics-M (Arrow Lake) generation, indicating it is designed for mobile integration with Arrow Lake-H chips. The B390 uses Xe3-LPG, a third-generation low-power architecture, and belongs to the Arc Graphics-WM (Panther Lake) generation. This places the B390 in a newer product family with a different design philosophy.

The process node difference is significant. The 140T uses TSMC’s 5 nm process, while the B390 uses Intel’s 3 nm process. The foundry also differs: TSMC for the 140T, Intel for the B390. While the database does not list transistor counts or die sizes for either, the process node alone suggests the B390 has a more modern manufacturing foundation. The 3 nm node typically allows for higher clock speeds or lower power at the same performance, but the B390’s 80 W TDP shows Intel chose to use that efficiency for higher performance rather than lower power.

The shader configuration differs in a telling way. The B390 has more shading units and more RT cores but fewer TMUs and ROPs. This indicates the B390 is optimized for compute and ray tracing workloads, sacrificing fill-rate hardware for shader throughput. The 140T, with more TMUs and ROPs, is balanced more toward traditional rasterization and fill-rate-heavy tasks. This is visible in the metrics: the B390 leads in FP32 and FP16, while the 140T leads in pixel rate and texture rate.

Memory architecture is identical in the recorded data. Both use system-shared memory with system-dependent bandwidth. No dedicated VRAM is present on either. This means both are limited by the host system’s memory performance, which can vary significantly between platforms. The B390’s higher compute throughput may be partially constrained by shared memory bandwidth in real-world scenarios, though the database does not provide specific bandwidth figures.

The display outputs are listed as portable device dependent for both, meaning the actual connectors depend on the host device. Neither has a dedicated slot width beyond IGP, and neither has dimensions recorded. Both are active products with no successors listed. The 140T’s predecessor is HD Graphics-M, while the B390’s predecessor is HD Graphics-WM, indicating they descend from different integrated graphics lineages.

The Verdict

The data points to two different buyers. The Intel Arc Pro B390 is the performance pick. Its 7.680 TFLOPS FP32, 15.36 TFLOPS FP16, and 12 ray tracing cores make it the stronger choice for compute-heavy tasks, ray-traced rendering, and any workload that uses the full shader array. The 80 W TDP is high for an integrated GPU, but the performance headroom justifies it for users who prioritize throughput over power draw. The newer Xe3-LPG architecture and 3 nm process also suggest better long-term driver optimization potential, though the database does not include driver-specific data.

The Intel Arc 140T Mobile is the efficiency pick. Its 35 W TDP, higher pixel rate of 75.20 GPixel/s, and higher texture rate of 150.4 GTexel/s make it suitable for fill-rate-bound scenarios and power-constrained environments. It is the better option for thin-and-light laptops where battery life and thermals are critical. Its FP32 performance of 4.813 TFLOPS is still respectable for integrated graphics, and its FP16 performance of 9.626 TFLOPS covers most lightweight compute tasks.

Benchmark results indicate a clean trade-off. Users who need maximum compute, ray tracing, and shader throughput should choose the Arc Pro B390. Users who need a balanced integrated GPU with lower power consumption and superior fill-rate metrics should choose the Arc 140T Mobile. Neither part is universally better; the choice depends entirely on whether the workload favors raw compute or efficiency and fill-rate performance. The 50th percentile ranking for both in the overall GPU database confirms they are mainstream performers, not specialized outliers.