Intel Arc 140T Mobile vs Intel Arc B370 Comparison

The Verdict

The recorded data presents two distinct Intel integrated graphics solutions aimed at different segments of the mobile market. The Intel Arc 140T Mobile, based on the Arrow Lake-H chip with Xe-LPG+ architecture, is positioned as the higher-power option with a 35 W TDP, while the Intel Arc B370, built on the Panther Lake chip with Xe3-LPG architecture, operates at a lower 25 W TDP. Both are integrated graphics processors (IGP) with system-shared memory, but their architectural generations and performance profiles differ substantially.

The Intel Arc B370 has a single recorded benchmark result: a 3DMark Steel Nomad DX12 score of 1184. This places it at the 5th percentile of all GPUs in the database. Its nearest rivals are all older discrete graphics cards, including the ATI Mobility Radeon HD 5570 (scoring 1186, only 0.2% faster), the ATI Radeon HD 5770 (scoring 1190, 0.5% faster), and the AMD Radeon HD 7650M (scoring 1192, 0.7% faster). It also edges out the AMD FirePro M2000, which scores 1168, by 1.4%. This cluster of results indicates the Arc B370 delivers performance comparable to legacy discrete mobile GPUs from over a decade ago.

The Intel Arc 140T Mobile has no recorded benchmark scores in the database, no average score, and no nearest rivals. Its percentile rank stands at 50, which is a neutral midpoint pending actual measurements. Consequently, the data cannot confirm a performance advantage for either part through direct comparison. However, the architectural specifications reveal a clear divide: the Arc 140T offers higher pixel throughput at 75.20 GPixel/s versus 48.00 GPixel/s for the B370, and higher texture fill at 150.4 GTexel/s versus 96.00 GTexel/s. The B370 counters with higher raw FP32 compute at 6.144 TFLOPS versus 4.813 TFLOPS for the 140T.

For system integrators, the choice depends on whether the workload favors texture and pixel throughput or raw shader compute. The Arc 140T appears oriented toward scenarios where rasterization throughput matters, while the Arc B370 emphasizes general compute density. The B370 also draws 10 W less power, a meaningful difference in thermally constrained portable devices. Until direct head-to-head measurements are recorded, the data supports selecting the Arc 140T for fill-rate-bound tasks and the Arc B370 for compute-bound tasks within their respective power envelopes.

Architecture Differences

The two GPUs belong to different architectural generations despite sharing the Intel Arc Graphics-M family lineage. The Intel Arc 140T Mobile uses the Xe-LPG+ architecture on the Arrow Lake-H chip, fabricated on a 5 nm process by TSMC. The Intel Arc B370 uses the newer Xe3-LPG architecture on the Panther Lake chip, fabricated on a 3 nm process by Intel. This process difference is significant: the B370 packs more shading units into a presumably smaller die, though transistor counts and die sizes are not recorded.

The shading unit counts differ notably. The Arc 140T contains 1024 shading units, while the Arc B370 contains 1280, a 25% increase. However, the texture mapping unit (TMU) and render output unit (ROP) counts are inverted. The Arc 140T has 64 TMUs and 32 ROPs, whereas the Arc B370 has only 40 TMUs and 20 ROPs. This configuration makes the 140T better suited for texture-heavy and pixel-heavy workloads, while the B370's larger shader array favors compute-heavy tasks. The ray tracing core counts follow the shader trend: 8 RT cores for the 140T versus 10 for the B370.

Clock speeds are close but not identical. Both GPUs share a 300 MHz base clock. The boost clock reaches 2350 MHz on the Arc 140T and 2400 MHz on the Arc B370, a 50 MHz difference that slightly favors the B370 in sustained peak performance. Neither part has a dedicated game clock. Memory subsystems are identical in configuration: both use system-shared memory with system-shared bus width and system-dependent bandwidth. Neither GPU has its own VRAM, so performance will scale with the host system's memory capabilities.

The power envelopes differ by 10 W. The Arc 140T has a 35 W TDP, while the Arc B370 operates at 25 W. The B370 is listed as having no power connectors, consistent with its lower power draw and IGP form factor. The 140T has no power connector data recorded. Both are IGP slot-width designs with portable-device-dependent display outputs. The API support is identical: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither GPU lists tensor cores.

Production status is active for both. The Arc 140T was released on January 12, 2025, with the HD Graphics-M listed as its predecessor. The Arc B370 was released on January 26, 2026, a little over a year later, and has no predecessor listed. Neither part has a recorded successor or launch MSRP.

Where Each One Wins

The benchmark data currently favors the Arc B370 only because it has a recorded score. The 3DMark Steel Nomad DX12 result of 1184, despite the low 5th percentile ranking, provides a concrete reference point. The Arc 140T has no recorded measurements, so any claims about its wins must be inferred from architectural specifications rather than observed results.

For pixel throughput, the Arc 140T delivers 75.20 GPixel/s, which is 56.7% higher than the B370's 48.00 GPixel/s. This suggests the 140T has an advantage in fill-rate-bound scenarios such as high-resolution rendering with simple shaders or heavy overdraw situations. The texture rate follows a similar pattern: 150.4 GTexel/s for the 140T versus 96.00 GTexel/s for the B370, a 56.7% advantage in texture fetch and filtering throughput. Games or applications that rely on detailed texture sampling would benefit from the 140T's configuration.

The Arc B370 wins on raw compute. Its FP32 throughput of 6.144 TFLOPS exceeds the 140T's 4.813 TFLOPS by 27.7%. The FP16 rates show the same proportional gap: 12.29 TFLOPS versus 9.626 TFLOPS, both at 2:1 ratios. This makes the B370 the stronger choice for shader-heavy compute workloads, including general-purpose GPU tasks and effects that scale with shader core count. The B370 also has 2 more ray tracing cores, which may benefit workloads that use ray tracing, though no RT-specific benchmark is recorded.

The power efficiency comparison is straightforward from the data. The B370 achieves higher FP32 compute at 25 W, while the 140T achieves higher fill rates at 35 W. For thermally limited portable devices, the B370's lower TDP combined with higher shader throughput is a meaningful advantage. The 140T's higher TDP may be justified in devices with adequate cooling where fill-rate performance is prioritized.

Neither GPU has a clear win in memory performance since both rely on system-shared memory with system-dependent bandwidth. The recorded data shows no wins for either part in head-to-head benchmarks, as the database lists zero wins for both. The Arc 140T's 50th percentile ranking is not derived from actual scores and should not be interpreted as a measured performance level.

FAQ

Q: Which GPU has a higher benchmark score in the database?

A: The Intel Arc B370 has a recorded 3DMark Steel Nomad DX12 score of 1184. The Intel Arc 140T Mobile has no recorded benchmark scores in the database, so no comparative score exists for it.

Q: How does the Arc B370 compare to its nearest rivals?

A: The Arc B370 scores 1184, which is 0.2% below the ATI Mobility Radeon HD 5570 (1186), 0.5% below the ATI Radeon HD 5770 (1190), and 0.7% below the AMD Radeon HD 7650M (1192). It is 1.4% above the AMD FirePro M2000 (1168).

Q: Which GPU has more shading units?

A: The Intel Arc B370 has 1280 shading units, while the Intel Arc 140T Mobile has 1024 shading units. The B370 also has 10 ray tracing cores versus 8 for the 140T.

Q: Which GPU has higher texture and pixel fill rates?

A: The Intel Arc 140T Mobile has higher fill rates: 75.20 GPixel/s pixel rate and 150.4 GTexel/s texture rate. The Arc B370 records 48.00 GPixel/s and 96.00 GTexel/s respectively.

Q: What are the power consumption levels?

A: The Intel Arc 140T Mobile has a TDP of 35 W. The Intel Arc B370 has a TDP of 25 W, which is 10 W lower.

Q: Do these GPUs use dedicated video memory?

A: No. Both the Intel Arc 140T Mobile and the Intel Arc B370 use system-shared memory with system-shared type and bus width, and their bandwidth is system dependent.

Head-to-Head Benchmarks

The database currently lists no head-to-head benchmark entries for the Intel Arc 140T Mobile versus the Intel Arc B370. The wins counter shows 0 for both parts. This means the only quantitative comparison available comes from the single 3DMark Steel Nomad DX12 score recorded for the Arc B370 and the architectural specifications of both GPUs.

Starting with the Arc B370's recorded result, a score of 1184 in 3DMark Steel Nomad DX12 places it at the 5th percentile of all GPUs in the database. The nearest rivals provide context: the ATI Mobility Radeon HD 5570 scores 1186, which is 0.2% higher; the ATI Radeon HD 5770 scores 1190, 0.5% higher; the AMD Radeon HD 7650M scores 1192, 0.7% higher; and the AMD FirePro M2000 scores 1168, which is 1.4% lower. The B370 therefore sits in a tight performance band spanning roughly 2 percentage points among these four legacy discrete GPUs. Its average benchmark score equals its single recorded score at 1184.

Turning to compute throughput, the Arc B370's FP32 rate of 6.144 TFLOPS is 27.7% higher than the Arc 140T's 4.813 TFLOPS. In FP16, the B370 delivers 12.29 TFLOPS versus 9.626 TFLOPS for the 140T, a 27.7% advantage as well. These figures indicate that for shader compute, the B370 holds a clear lead despite its lower TDP and lower pixel rate.

The fill-rate comparison reverses the outcome. The Arc 140T's pixel rate of 75.20 GPixel/s is 56.7% higher than the B370's 48.00 GPixel/s. Its texture rate of 150.4 GTexel/s is likewise 56.7% higher than the B370's 96.00 GTexel/s. These ratios derive directly from the TMU and ROP counts: the 140T has 64 TMUs and 32 ROPs, while the B370 has 40 TMUs and 20 ROPs.

Clock behavior is nearly identical. Both GPUs boost to similar frequencies, with the B370 reaching 2400 MHz and the 140T reaching 2350 MHz. The B370's 50 MHz boost advantage is small relative to the 27.7% FP32 gap, suggesting the shader count difference, not the clock, drives its compute lead. The 140T's fill-rate advantage is entirely structural, stemming from its higher TMU and ROP counts.

The power data adds another layer. The B370 achieves its higher FP32 throughput at 25 W, while the 140T requires 35 W for its fill-rate advantage. Per watt, the B370 delivers roughly 0.246 TFLOPS/W in FP32, while the 140T delivers roughly 0.138 TFLOPS/W, assuming TDP scales linearly with power draw. These are derived figures from the recorded TDPs and FP32 rates. The B370's efficiency advantage in compute is substantial, while the 140T's fill-rate-per-watt is 2.15 GPixel/s/W versus 1.92 GPixel/s/W for the B370, a narrower edge.

Neither GPU has a recorded memory bandwidth figure, as both depend on the host system's shared memory configuration. The absence of head-to-head benchmark results means the 3DMark Steel Nomad score for the B370 cannot be directly compared to any 140T measurement. The database shows zero wins for each part, and the 140T has no average benchmark score, only a 50th percentile ranking that reflects no measured data. Until direct comparisons are recorded, the architectural analysis above represents the full extent of the quantitative record.