Intel Arc A370M vs NVIDIA P104-100 Comparison

Head-to-Head Benchmarks

The recorded data shows a clear and decisive advantage for the NVIDIA P104-100 across the two shared benchmark tests. In Geekbench OpenCL, the NVIDIA P104-100 scores 52,368 points, while the Intel Arc A370M manages 29,676 points. This translates to a 76.5% lead for the NVIDIA card, a substantial margin that indicates a major gap in raw compute throughput under this API.

The gap narrows somewhat in the Vulkan test, but the outcome remains the same. The NVIDIA P104-100 posts 45,165 points against the Intel Arc A370M's 28,673 points. The delta here is 57.5% in favor of the NVIDIA part. While the percentage difference is smaller than in OpenCL, it still represents a commanding victory for the Pascal-based card.

Across the head-to-head benchmarks, the NVIDIA P104-100 secures 2 wins, while the Intel Arc A370M records 0 wins. The average benchmark score in the database reinforces this trend: the NVIDIA P104-100 sits at 32,982, whereas the Intel Arc A370M trails at 29,175. The NVIDIA card also holds a higher percentile ranking among all GPUs, placing in the 77th percentile compared to the Intel part's 74th percentile. The data indicates that in every measurable comparison available, the NVIDIA P104-100 outperforms the Intel Arc A370M, often by a wide margin.

Where Each One Wins

Given that the NVIDIA P104-100 wins both head-to-head benchmark tests, the use-case split is heavily skewed toward the older mining card. The Geekbench OpenCL result is particularly telling, as the 76.5% advantage suggests that workloads leveraging OpenCL compute will see significantly better performance on the NVIDIA P104-100. This is relevant for general-purpose GPU compute tasks, which often rely on OpenCL as a cross-platform API.

In Vulkan-based applications, the NVIDIA P104-100 still holds a 57.5% edge. Vulkan is commonly used in modern game engines and high-performance graphics applications, so the NVIDIA card would be the stronger choice for any workload that depends on this API. The Intel Arc A370M, despite being a newer architecture, does not close the gap enough to claim a single victory in these tests.

The Intel Arc A370M does have its own advantages, but they do not manifest in the benchmark scores recorded. Its architectural features, such as support for DirectX 12 Ultimate and 8 ray tracing cores, suggest capabilities that the NVIDIA P104-100 lacks, but the database shows no test where these features translate into a performance win. For users prioritizing the two benchmark results available, the NVIDIA P104-100 is the only logical pick.

The Verdict

The data is unambiguous. The NVIDIA P104-100 is the superior performer in every benchmark where both cards were measured. Its 76.5% lead in OpenCL and 57.5% lead in Vulkan make it the default choice for any workload represented by these tests. The NVIDIA card also achieves a higher average benchmark score of 32,982 versus 29,175 for the Intel Arc A370M, and it ranks higher in the overall GPU percentile distribution.

However, the Intel Arc A370M should not be dismissed entirely. It is a mobile-oriented part with a 35 W TDP and an IGP form factor, making it suitable for thin-and-light portable devices. The NVIDIA P104-100, by contrast, is a dual-slot card with a 1x 8-pin power connector and a suggested 200 W PSU, and it has no display outputs. This makes the NVIDIA card unsuitable for direct display tasks, while the Intel part is designed for portable device integration.

For users who need raw compute performance and have the chassis and power budget to accommodate a dual-slot card, the NVIDIA P104-100 is the clear winner. For users who require an integrated solution for a portable device, the Intel Arc A370M is the only viable option, despite its lower scores. The decision hinges on form factor and power constraints rather than benchmark performance, since the benchmark data overwhelmingly favors the NVIDIA card.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA P104-100 has an average benchmark score of 32,982, while the Intel Arc A370M scores 29,175.

Q: How much faster is the NVIDIA P104-100 in Geekbench OpenCL?

A: The NVIDIA P104-100 scores 52,368 points compared to 29,676 for the Intel Arc A370M, a difference of 76.5%.

Q: Does the Intel Arc A370M win any head-to-head benchmark?

A: No, the Intel Arc A370M does not win any of the recorded head-to-head benchmarks. The NVIDIA P104-100 wins both tests.

Q: What is the percentile ranking for each GPU?

A: The NVIDIA P104-100 is in the 77th percentile of all GPUs, and the Intel Arc A370M is in the 74th percentile.

Q: What are the memory specifications for each card?

A: The NVIDIA P104-100 has 4 GB of GDDR5X on a 256-bit bus with 320.3 GB/s bandwidth. The Intel Arc A370M has 4 GB of GDDR6 on a 64-bit bus with 112.0 GB/s bandwidth.

Q: Which card has ray tracing cores?

A: The Intel Arc A370M has 8 ray tracing cores, while the NVIDIA P104-100 has none.

Architecture Differences

The two GPUs come from different architectural generations and process nodes. The NVIDIA P104-100 is built on the Pascal architecture using TSMC's 16 nm process, with a die size of 314 mm². The Intel Arc A370M uses the Xe-HPG architecture (Alchemist generation) on TSMC's 6 nm node, with a much smaller die size of 157 mm². Both chips pack the same 7,200 million transistors, but the Intel part achieves a significantly higher transistor density of 45.9M per mm² compared to 22.9M per mm² for the NVIDIA chip.

The compute configurations differ notably. The NVIDIA P104-100 features 1,920 shading units, 120 texture mapping units, and 64 ROPs. The Intel Arc A370M has 1,024 shading units, 64 TMUs, and 32 ROPs. These numbers explain the NVIDIA card's higher pixel rate of 110.9 GPixel/s and texture rate of 208.0 GTexel/s, versus 65.60 GPixel/s and 131.2 GTexel/s for the Intel part. In raw FP32 throughput, the NVIDIA card delivers 6.655 TFLOPS, while the Intel card manages 4.198 TFLOPS.

Memory architecture is another major divergence. The NVIDIA P104-100 uses 4 GB of GDDR5X on a 256-bit bus, yielding 320.3 GB/s of bandwidth. The Intel Arc A370M uses 4 GB of GDDR6 on a 64-bit bus, providing only 112.0 GB/s. Clock speeds also favor the NVIDIA card: it has a base clock of 1607 MHz and a boost of 1733 MHz, while the Intel part runs at 1550 MHz base and 2050 MHz boost. The higher boost clock for Intel does not compensate for its narrower memory bus and fewer shaders.

Feature support shows the generational difference. The Intel Arc A370M supports DirectX 12 Ultimate (12_2), includes 8 ray tracing cores, and has FP16 throughput of 8.397 TFLOPS (2:1 ratio). The NVIDIA P104-100 is limited to DirectX 12 (12_1), has no ray tracing cores, and its FP16 performance is a negligible 104.0 GFLOPS (1:64 ratio). Both cards support OpenGL 4.6 and Vulkan 1.4.

Power and physical characteristics are starkly different. The Intel Arc A370M has a 35 W TDP and is classified as an IGP, with no power connectors and display outputs described as portable device dependent. The NVIDIA P104-100 has no listed TDP but requires a dual-slot cooler, a single 8-pin power connector, and a suggested 200 W power supply. It also has no display outputs, reflecting its mining-oriented design. The bus interfaces differ as well: the NVIDIA card uses PCIe 1.0 x4, while the Intel card uses PCIe 4.0 x8. The NVIDIA P104-100 measures 267 mm in length, while the Intel part has no listed dimensions.

Release timing also separates the pair. The NVIDIA P104-100 launched on 2017-12-11, and the Intel Arc A370M followed on 2022-03-29. Both are now end-of-life products. The NVIDIA card belongs to the Mining GPUs generation, while the Intel part is from the Alchemist (Arc 3 Mobile) generation. These architectural and physical differences explain why the benchmark results favor the NVIDIA card so heavily, despite the Intel part being several years newer.