Intel Arc A570M vs NVIDIA GeForce RTX 5090 Comparison

The NVIDIA GeForce RTX 5090 and Intel Arc A570M occupy opposite ends of the graphics spectrum, and the benchmark data reflects that starkly. In the only directly comparable test, Geekbench OpenCL, the RTX 5090 scores 334,370 against the Arc A570M’s 58,239, a lead of 474.1%. This is not a close contest; it is a generational and architectural chasm. The RTX 5090 sits at the 92nd percentile of all GPUs in the database, while the Arc A570M sits at the 88th percentile, yet the raw score gap shows that percentile ranking alone masks the sheer magnitude of performance difference. The database records one head-to-head benchmark, and the RTX 5090 wins it outright. No benchmark in the recorded data shows the Arc A570M ahead.

Head-to-Head Benchmarks

The single shared benchmark between these two products is Geekbench OpenCL. The RTX 5090 produces a score of 334,370, while the Arc A570M produces 58,239. The delta is 474.1%, meaning the NVIDIA part delivers nearly five times the compute throughput in this workload. To contextualize this within the database’s rival groupings, the RTX 5090’s average benchmark score is 79,842, placing it 0.3% ahead of the NVIDIA Tesla P100 PCIe 16 GB (79,605) and 0.6% ahead of the Tesla P100 PCIe 12 GB (79,396). It is 1.1% ahead of the AMD Radeon RX 6850M XT (78,940) and 1.4% behind the AMD Radeon Pro Vega 64X (80,959). These are narrow margins against high-end data center and workstation parts, which indicates the RTX 5090 is competing at an elite tier where small percentage differences separate the top performers.

The Arc A570M, by contrast, has an average benchmark score of 58,239, which places it 0.3% behind the AMD Radeon RX 6950 XT (58,392) and 0.3% ahead of the AMD Radeon RX 5600 OEM (58,085). It is 0.5% behind the NVIDIA P102-100 (58,528) and 0.7% behind the AMD Radeon PRO V710 (58,657). These deltas are all within a single percentage point, showing that the Arc A570M is clustered with mid-range desktop and mobile parts from the previous generation. The performance gap between the two products in question is not incremental; it is a multiple of five in raw OpenCL output.

Looking at the RTX 5090’s broader benchmark suite, the data shows consistent strength across legacy and modern APIs. In Passmark DirectX 10 it scores 226, in DirectX 11 it scores 341, and in DirectX 12 it scores 185. The DirectX 9 score is 395, which is the highest of the Passmark DirectX tests. The Passmark G2D score is 1,413, and the G3D score is 39,650. The GPU compute score is 26,756. These figures are not directly comparable to the Arc A570M because the database does not list corresponding Arc scores for those tests, but they establish the RTX 5090 as a complete performer across multiple workloads. The Arc A570M only has a Geekbench OpenCL result in the database, so any comparison beyond that single test relies on the average score and rival deltas.

The 3DMark Steel Nomad DX12 score for the RTX 5090 is 18,355, a strong result for a modern DirectX 12 Ultimate workload, though again no equivalent Arc figure exists in the recorded data. The RTX 5090 also records Geekbench Vulkan at 376,728, which is higher than its OpenCL score, suggesting particular strength in Vulkan compute and rendering paths. The Arc A570M has no Vulkan score in the database, so this remains an RTX-only data point.

The Verdict

The data points to one conclusion: the NVIDIA GeForce RTX 5090 is the superior product by every measurable benchmark in the database. It wins the sole head-to-head test by 474.1%, and its average benchmark score of 79,842 is 37.1% higher than the Arc A570M’s 58,239, though that percentage is not directly listed and must be read from the raw scores. The RTX 5090’s nearest rivals are all high-end data center or workstation GPUs, with deltas under 1.5%, meaning it is firmly in the top tier of the database’s rankings. The Arc A570M’s nearest rivals are mid-range parts, and its deltas are also under 1%, but at a far lower absolute score level.

Who should pick which? From the data alone, the RTX 5090 is for anyone who needs maximum compute throughput, whether for OpenCL, Vulkan, or DirectX workloads. Its 92nd percentile ranking and its ability to outscore the Tesla P100 variants and the Radeon RX 6850M XT make it the choice for high-end desktop builds where performance is the only criterion. The Arc A570M, with its 88th percentile ranking and scores clustered around the RX 5600 OEM and RX 6950 XT, is a different class of product entirely. It might suit a user whose workload is light enough that a 58,239 OpenCL score is sufficient, but the benchmark data gives no reason to prefer it over the RTX 5090 if both are available and performance is the goal.

The production status of both is listed as Active, so neither is a discontinued or legacy part. The RTX 5090 launched on 2025-01-29, while the Arc A570M launched on 2023-07-31, a difference of roughly a year and a half in release timing. The RTX 5090 has a launch MSRP of 1,999 USD, which the database records; the Arc A570M has no recorded launch MSRP. No further pricing discussion is warranted.

Architecture Differences

The architectural gap between these two GPUs is as wide as the benchmark gap. The RTX 5090 uses the GB202 chip built on TSMC’s 5 nm process, with 92,200 million transistors on a 750 mm² die, yielding a transistor density of 122.9 million per square millimeter. The Arc A570M uses the DG2-256 chip on TSMC’s 6 nm process, with 11,500 million transistors on a 269 mm² die, for a density of 42.8 million per square millimeter. The RTX 5090’s die is nearly three times larger in area and holds eight times the transistor count, which explains the massive compute advantage.

The RTX 5090 is built on the Blackwell 2.0 architecture, part of the GeForce 50 series, while the Arc A570M uses Xe-HPG architecture from the Alchemist generation, specifically the Arc 5 Mobile family. The RTX 5090 has 21,760 shading units, 680 texture mapping units, and 176 ROPs. It also carries 170 ray tracing cores and 680 tensor cores. The Arc A570M has 2,048 shading units, 128 TMUs, and 64 ROPs, with 16 ray tracing cores and no tensor cores listed. The RTX 5090’s shading unit count is over ten times that of the Arc A570M.

Clock speeds differ substantially. The RTX 5090 runs at a base of 2017 MHz and a boost of 2407 MHz, while the Arc A570M runs at 900 MHz base and 1300 MHz boost. The RTX 5090’s boost clock is nearly double the Arc’s. Memory clocks also differ: the RTX 5090 uses 1750 MHz with 28 Gbps effective data rate, while the Arc A570M uses 1750 MHz with 14 Gbps effective. The memory type is GDDR7 for the RTX 5090 and GDDR6 for the Arc A570M.

The memory subsystem is another major divider. The RTX 5090 has 32 GB of GDDR7 on a 512-bit bus, delivering 1.79 TB/s of bandwidth. The Arc A570M has 8 GB of GDDR6 on a 128-bit bus, delivering 224.0 GB/s. The RTX 5090’s bandwidth is exactly eight times the Arc’s, which is consistent with the bus width and memory speed differences. Pixel rate for the RTX 5090 is 423.6 GPixel/s versus 83.20 GPixel/s for the Arc, and texture rate is 1,636.8 GTexel/s versus 166.4 GTexel/s.

Compute throughput follows the same pattern. The RTX 5090 delivers 104.8 TFLOPS of FP32 and 104.8 TFLOPS of FP16 at a 1:1 ratio. The Arc A570M delivers 5.325 TFLOPS of FP32 and 10.65 TFLOPS of FP16 at a 2:1 ratio. The RTX 5090’s FP32 output is roughly 19.7 times the Arc’s, and even its FP16 output, despite the Arc’s 2:1 ratio advantage, is nearly ten times higher.

Power and physical specifications differ as much as compute. The RTX 5090 has a TDP of 575 W, a dual-slot form factor, a single 16-pin power connector, and a suggested PSU of 950 W. It uses a PCIe 5.0 x16 bus interface. The Arc A570M has a TDP of 75 W, is listed as an IGP (integrated graphics processor), has no power connector listed, no suggested PSU, and uses a PCIe 4.0 x8 bus interface. The RTX 5090 measures 304 mm in length, 137 mm in height, and 40 mm in width. The Arc A570M has no recorded dimensions, consistent with its mobile or integrated positioning.

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The NVIDIA GeForce RTX 5090 scores 334,370, which is 474.1% higher than the Intel Arc A570M’s 58,239.

Q: What is the average benchmark score difference between the two?

A: The RTX 5090 has an average benchmark score of 79,842, while the Arc A570M has 58,239. The RTX 5090 also sits at the 92nd percentile of all GPUs, compared to the Arc A570M’s 88th percentile.

Q: How does the RTX 5090 compare to its nearest rivals?

A: The RTX 5090 is 0.3% ahead of the NVIDIA Tesla P100 PCIe 16 GB, 0.6% ahead of the Tesla P100 PCIe 12 GB, 1.1% ahead of the AMD Radeon RX 6850M XT, and 1.4% behind the AMD Radeon Pro Vega 64X.

Q: How does the Arc A570M compare to its nearest rivals?

A: The Arc A570M is 0.3% behind the AMD Radeon RX 6950 XT, 0.3% ahead of the AMD Radeon RX 5600 OEM, 0.5% behind the NVIDIA P102-100, and 0.7% behind the AMD Radeon PRO V710.

Q: What memory configurations do the two GPUs use?

A: The RTX 5090 has 32 GB of GDDR7 on a 512-bit bus with 1.79 TB/s bandwidth. The Arc A570M has 8 GB of GDDR6 on a 128-bit bus with 224.0 GB/s bandwidth.

Q: What are the power requirements for each?

A: The RTX 5090 has a TDP of 575 W and a suggested PSU of 950 W, with a single 16-pin power connector. The Arc A570M has a TDP of 75 W, no power connector listed, and no suggested PSU.

Where Each One Wins

The RTX 5090 wins every category where data exists. In the single head-to-head benchmark, Geekbench OpenCL, it wins by 474.1%. Its Passmark suite shows strength across DirectX 9, 10, 11, and 12, with scores of 395, 226, 341, and 185 respectively, plus a G3D score of 39,650 and a GPU compute score of 26,756. Its 3DMark Steel Nomad DX12 score is 18,355, and its Geekbench Vulkan score is 376,728. These results position it as a top-tier part for compute-heavy and graphics-heavy workloads alike.

The Arc A570M has no recorded wins in any benchmark against the RTX 5090. Its only recorded benchmark is Geekbench OpenCL at 58,239, and its nearest rival deltas show it is competitive with parts like the RX 5600 OEM and RX 6950 XT, but those are far below the RTX 5090’s tier. Where the Arc A570M could be considered preferable is in power envelope: its 75 W TDP is drastically lower than the RTX 5090’s 575 W, and its IGP form factor means no discrete power connector or large chassis requirement. For a user constrained by power or physical space, the Arc A570M is the only option between the two, but that is a system design consideration, not a performance one.

The RTX 5090 also wins on memory capacity, bandwidth, shading units, TMUs, ROPs, ray tracing cores, tensor cores, pixel rate, texture rate, and FP32 and FP16 throughput. It is the larger die, the newer architecture, the faster clocked part, and the higher bandwidth part. The Arc A570M wins on power draw and form factor integration, with 75 W versus 575 W and an IGP designation versus a dual-slot card. Those are the only two fields where the Arc is not behind.

Specification Differences

The two GPUs differ in nearly every recorded specification. The RTX 5090 uses the GB202 chip with Blackwell 2.0 architecture, while the Arc A570M uses DG2-256 with Xe-HPG architecture. Process nodes are 5 nm for the RTX 5090 and 6 nm for the Arc A570M, both from TSMC. Transistor counts are 92,200 million versus 11,500 million, and die sizes are 750 mm² versus 269 mm². Transistor density is 122.9 million per mm² versus 42.8 million per mm².

Base clocks are 2017 MHz versus 900 MHz, and boost clocks are 2407 MHz versus 1300 MHz. Memory effective data rates are 28 Gbps versus 14 Gbps. Memory size is 32 GB versus 8 GB, type is GDDR7 versus GDDR6, bus width is 512-bit versus 128-bit, and bandwidth is 1.79 TB/s versus 224.0 GB/s. Shading units are 21,760 versus 2,048, TMUs are 680 versus 128, ROPs are 176 versus 64, ray tracing cores are 170 versus 16, and tensor cores are 680 versus none listed.

Pixel rate is 423.6 GPixel/s versus 83.20 GPixel/s, texture rate is 1,636.8 GTexel/s versus 166.4 GTexel/s, FP32 is 104.8 TFLOPS versus 5.325 TFLOPS, and FP16 is 104.8 TFLOPS at 1:1 versus 10.65 TFLOPS at 2:1. TDP is 575 W versus 75 W, slot width is dual-slot versus IGP, power connectors are 1x 16-pin versus none, suggested PSU is 950 W versus none, and bus interface is PCIe 5.0 x16 versus PCIe 4.0 x8. Display outputs are 1x HDMI 2.1b and 3x DisplayPort 2.1b for the RTX 5090 versus portable device dependent for the Arc A570M. The RTX 5090 has recorded dimensions of 304 mm by 137 mm by 40 mm; the Arc A570M has none. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RTX 5090 has a launch MSRP of 1,999 USD; the Arc A570M has none recorded.