Intel Arc Pro B370 vs NVIDIA GeForce RTX 5070 Comparison

Where Each One Wins

The benchmark database splits these two GPUs into entirely different performance classes. The NVIDIA GeForce RTX 5070 holds the 82nd percentile among all GPUs, while the Intel Arc Pro B370 sits at the 50th percentile. That gap is reflected in the recorded measurements: the RTX 5070's average benchmark score is 40,377, whereas the Intel part has no recorded average score, as the database lists no completed benchmark runs for it.

The RTX 5070 wins on every measurable axis. Its PassMark G3D score of 29,137 places it far ahead of the Intel's unrecorded results, and its compute-oriented PassMark GPU Compute score of 15,787 reinforces its lead in general-purpose throughput. The RTX 5070's Geekbench scores are also substantial: 172,660 in OpenCL and 178,923 in Vulkan. The Intel Arc Pro B370, by contrast, has zero entries in the benchmark suite, meaning the database cannot attribute any single win to it across the tested workloads.

The use-case split is therefore one-sided. For any task that appears in the database, the RTX 5070 is the only part with verified performance data. The Intel part's role is limited to its specifications, which suggest it targets integrated graphics duty rather than discrete performance. Its 25 W TDP and IGP slot width indicate a power-constrained design, whereas the RTX 5070's 250 W TDP and dual-slot footprint show a dedicated add-in board. In practical terms, the RTX 5070 handles DirectX 12, DirectX 11, DirectX 10, and DirectX 9 workloads with measurable scores (108, 277, 180, and 320 on PassMark respectively), while the Intel part offers no comparable numbers.

Architecture Differences

The two chips come from different manufacturing and design philosophies. The Intel Arc Pro B370 uses the Panther Lake chip with an Xe3-LPG architecture, built on Intel's own 3 nm process. It is part of the Arc Graphics-WM (Panther Lake) generation. The RTX 5070 uses the GB205 chip with a Blackwell 2.0 architecture, manufactured by TSMC on a 5 nm process. The RTX 5070's transistor count is recorded as 31,100 million, with a die size of 263 mm² and a transistor density of 118.3M per mm². The Intel part's transistor count and die size are listed as unknown, so no density comparison is possible.

The RTX 5070 is a discrete card with a PCIe 5.0 x16 bus interface, while the Intel Arc Pro B370 is an integrated graphics processor (IGP) with no bus interface beyond the IGP itself. The RTX 5070's memory subsystem is fully specified: 12 GB of GDDR7 on a 192-bit bus, delivering 672.0 GB/s of bandwidth. The Intel part shares system memory, with its bus width and bandwidth described as system dependent. The RTX 5070's memory clock is 1750 MHz (28 Gbps effective), while the Intel part's memory clock is simply "System Shared."

Core counts differ sharply. The RTX 5070 has 6,144 shading units, 192 texture mapping units, 80 raster output units, 48 ray tracing cores, and 192 tensor cores. The Intel Arc Pro B370 has 1,280 shading units, 40 TMUs, 20 ROPs, and 10 ray tracing cores, with no tensor core count listed. The RTX 5070's pixel rate is 201.0 GPixel/s and its texture rate is 482.3 GTexel/s, compared to the Intel part's 48.00 GPixel/s and 96.00 GTexel/s. FP32 throughput is 30.87 TFLOPS for the RTX 5070 versus 6.144 TFLOPS for the Intel part. FP16 rates are 30.87 TFLOPS (1:1) for NVIDIA and 12.29 TFLOPS (2:1) for Intel.

Clock behavior also differs. The RTX 5070 runs at a 2325 MHz base and 2512 MHz boost. The Intel part's base clock is 300 MHz with a 2400 MHz boost. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RTX 5070's display outputs are 1x HDMI 2.1b and 3x DisplayPort 2.1b, while the Intel part's outputs are listed as portable device dependent. Power delivery is distinct: the RTX 5070 uses a 1x 16-pin connector with a suggested PSU of 600 W, while the Intel part has no power connectors and a 25 W TDP versus the RTX 5070's 250 W TDP.

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark entries between these two products, and the Intel Arc Pro B370 has no individual benchmark scores recorded. As a result, the comparison relies entirely on the RTX 5070's measured results and the Intel part's absence from the test suite.

The RTX 5070's strongest recorded result is its PassMark G3D score of 29,137. Its PassMark GPU Compute score of 15,787 is roughly half of that, indicating that graphics rendering tasks scale well on this architecture. The Geekbench scores are higher in magnitude: 172,660 in OpenCL and 178,923 in Vulkan, suggesting strong compute API performance in both OpenCL and Vulkan environments. The PassMark DirectX 12 score of 108 is the lowest among the DirectX tests, while DirectX 9 scores 320, DirectX 11 scores 277, and DirectX 10 scores 180. The PassMark G2D score of 1,305 reflects 2D graphics throughput.

The RTX 5070's nearest rivals in the database are all AMD professional parts. It leads the AMD Radeon Pro 580 (average score 40,318) by 0.1%, and the AMD Radeon Pro WX 7100 (average score 40,063) by 0.8%. It trails the AMD Radeon Pro 5300 (average score 40,870) by 1.2%, and leads the NVIDIA RTX A500 Mobile (average score 39,568) by 2.0%. These margins are narrow, placing the RTX 5070 in a tightly contested cluster around the 40,000 average score mark.

The Intel Arc Pro B370 has no nearest rivals listed, no benchmark scores, and no wins attributed to it. The database shows zero wins for item A and zero wins for item B in head-to-head tests, but that absence is due to a lack of recorded test data for the Intel part, not to any measured parity. The RTX 5070's average benchmark score of 40,377 is the only aggregate figure available for comparison, and it stands alone as the sole data point in this matchup.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce RTX 5070 has an average benchmark score of 40,377. The Intel Arc Pro B370 has an average benchmark score of 0, as the database records no benchmark runs for it.

Q: What are the memory specifications for each GPU?

A: The RTX 5070 has 12 GB of GDDR7 on a 192-bit bus with 672.0 GB/s bandwidth. The Intel Arc Pro B370 uses system shared memory, with type, bus width, and bandwidth all listed as system dependent.

Q: How do the shading unit counts compare?

A: The RTX 5070 has 6,144 shading units, while the Intel Arc Pro B370 has 1,280 shading units. The RTX 5070 also has 192 TMUs and 80 ROPs, versus 40 TMUs and 20 ROPs on the Intel part.

Q: What is the power draw difference?

A: The RTX 5070 has a TDP of 250 W and uses a 1x 16-pin power connector with a suggested PSU of 600 W. The Intel Arc Pro B370 has a TDP of 25 W and uses no power connectors.

Q: Which GPU supports ray tracing?

A: Both support ray tracing. The RTX 5070 has 48 ray tracing cores, while the Intel Arc Pro B370 has 10 ray tracing cores.

Q: What is the process node for each chip?

A: The Intel Arc Pro B370 uses a 3 nm process from Intel. The NVIDIA GeForce RTX 5070 uses a 5 nm process from TSMC.

The Verdict

The data presents a clear hierarchy. The NVIDIA GeForce RTX 5070 is the only one of the two with recorded performance, and its numbers place it in the 82nd percentile of all GPUs. Its 30.87 TFLOPS FP32 throughput, 672.0 GB/s memory bandwidth, and 6,144 shading units describe a high-end discrete solution. Its nearest rivals, all within 2% of its average score, confirm it sits at the top of a competitive tier.

The Intel Arc Pro B370, with no benchmark scores and no percentile rank beyond the 50th, cannot be assessed for performance from the database. Its specifications describe an integrated processor with 25 W TDP, system shared memory, and 1,280 shading units. It is a different class of product, aimed at portable devices where the RTX 5070's dual-slot, 250 W design would not fit.

For anyone selecting a GPU based on the recorded data, the RTX 5070 is the only viable choice. It delivers measurable wins across DirectX 9, 10, 11, and 12 workloads, plus strong OpenCL and Vulkan results. The Intel part offers no such evidence. The RTX 5070's launch MSRP is 549 USD, and its production status is active. The Intel part is also active, but it lacks any database measurements to justify a performance-based selection. The verdict is straightforward: the RTX 5070 wins every recorded comparison, and the Intel Arc Pro B370 remains a specification-only entry.