Intel Arc A580 vs Intel Arc Pro B70 Comparison

Where Each One Wins

The Intel Arc A580 and Intel Arc Pro B70 occupy different positions in the database, and the recorded measurements reflect that separation. The Arc A580 is an active, shipping product with three completed benchmark runs, while the Arc Pro B70 has no recorded benchmark scores and a production status that is not listed. The data therefore shows a clear split: the Arc A580 wins in measured performance evidence, and the Arc Pro B70 wins in raw hardware specifications and memory capacity.

The Arc A580 delivers an average benchmark score of 57,756 across its test suite, placing it in the 87th percentile of all GPUs in the database. That percentile rank puts it ahead of roughly 87 percent of recorded graphics cards, with a 3DMark Steel Nomad DX12 score of 2,229, a Geekbench OpenCL score of 91,657, and a Geekbench Vulkan score of 79,381. The Arc Pro B70 has no benchmark entries, so it cannot claim a single measured victory in any test. Its percentile ranking sits at 50, and its average score is recorded as zero, meaning the database has no performance data to compare.

Where the Arc Pro B70 wins is in specification headroom. It offers four times the memory capacity at 32 GB versus 8 GB, a higher memory bandwidth at 608.0 GB/s versus 512.0 GB/s, more shading units at 4,096 versus 3,072, more texture mapping units at 256 versus 192, more raster operations units at 128 versus 96, and more ray tracing cores at 32 versus 24. Its FP32 compute rating of 22.94 TFLOPS nearly doubles the Arc A580's 12.29 TFLOPS. The Arc Pro B70 also boosts to 2800 MHz versus 2000 MHz for the Arc A580, and it runs on the newer Xe2-HPG architecture versus the older Xe-HPG design.

For users who need measured evidence of performance, the Arc A580 is the only one of the two with any. For users who need the maximum memory allocation and compute throughput on paper, the Arc Pro B70 has the clear specification advantage.

FAQ

Q: Which GPU has a higher benchmark score?

A: The Intel Arc A580 has an average benchmark score of 57,756, while the Intel Arc Pro B70 has no recorded benchmark scores and an average score of zero.

Q: How do the memory capacities compare?

A: The Arc Pro B70 has 32 GB of GDDR6 memory, while the Arc A580 has 8 GB of GDDR6 memory. The Arc Pro B70 also has a higher memory bandwidth at 608.0 GB/s versus 512.0 GB/s.

Q: Which GPU has a higher boost clock?

A: The Arc Pro B70 boosts to 2800 MHz, compared to the Arc A580's boost clock of 2000 MHz. The Arc Pro B70 also has a higher base clock at 2280 MHz versus 1700 MHz.

Q: What is the architecture difference between the two?

A: The Arc A580 uses the Xe-HPG architecture with the DG2-512 chip, while the Arc Pro B70 uses the Xe2-HPG architecture with the BMG-G31 chip. The Arc A580 is from the Alchemist (Arc 5) generation, and the Arc Pro B70 is from the Battlemage (Pro Series) generation.

Q: Which GPU has more shading units?

A: The Arc Pro B70 has 4,096 shading units, compared to 3,072 for the Arc A580. The Arc Pro B70 also has more texture mapping units (256 versus 192), more ROPs (128 versus 96), and more ray tracing cores (32 versus 24).

Q: What is the process node for each GPU?

A: The Arc A580 is built on a 6 nm process at TSMC, while the Arc Pro B70 is built on a 5 nm process at TSMC. The Arc A580 has a die size of 406 mm², and the Arc Pro B70 has a die size of 368 mm².

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark results between the Intel Arc A580 and the Intel Arc Pro B70. The head-to-head benchmark list is empty, and neither card has a recorded win in a direct comparison. The only measurable performance data comes from the Arc A580's three standalone benchmark runs.

The Arc A580's 3DMark Steel Nomad DX12 score of 2,229 represents its strongest single-test result in the modern DirectX 12 workload. Its Geekbench OpenCL score of 91,657 shows compute performance in a general-purpose OpenCL context, and its Geekbench Vulkan score of 79,381 demonstrates graphics performance through the Vulkan API. These three results combine to produce an average benchmark score of 57,756.

The Arc A580's nearest rivals in the database provide context for these numbers. The AMD Radeon RX 5600 OEM has an average score of 58,085, which is 0.6 percent higher than the Arc A580. The Intel Arc A570M has an average score of 58,239, 0.8 percent higher. The AMD Radeon RX 6950 XT has an average score of 58,392, 1.1 percent higher. The AMD Radeon RX 9070 GRE has an average score of 57,367, which is 0.7 percent lower than the Arc A580. The Arc A580 sits within roughly one percentage point of all four rivals, indicating a tight cluster of similar overall performance.

The Arc Pro B70 has no benchmark scores, no nearest rivals, and no wins recorded. Its percentile ranking of 50 is the median default, and its average score of zero means it cannot be placed in any performance comparison. The data indicates that the Arc Pro B70 is a specification-defined product rather than a benchmark-verified one.

Specification Differences

The two GPUs differ across nearly every major specification field. The Arc A580 uses the DG2-512 chip, while the Arc Pro B70 uses the BMG-G31 chip. The Arc A580 is from the Alchemist (Arc 5) generation, and the Arc Pro B70 is from the Battlemage (Pro Series) generation.

The process node differs: the Arc A580 is on 6 nm, the Arc Pro B70 is on 5 nm, both from TSMC. The Arc A580 has a die size of 406 mm², the Arc Pro B70 has a die size of 368 mm². The Arc A580 has 21,700 million transistors, while the Arc Pro B70's transistor count is listed as unknown. The Arc A580's transistor density is 53.4M per mm², while the Arc Pro B70 has no recorded density.

Clock speeds differ substantially. The Arc A580 has a base clock of 1700 MHz and a boost clock of 2000 MHz. The Arc Pro B70 has a base clock of 2280 MHz and a boost clock of 2800 MHz. Memory clocks also differ: the Arc A580 runs at 2000 MHz with 16 Gbps effective, while the Arc Pro B70 runs at 2375 MHz with 19 Gbps effective.

Memory configuration is a major split. The Arc A580 has 8 GB of GDDR6 on a 256-bit bus with 512.0 GB/s bandwidth. The Arc Pro B70 has 32 GB of GDDR6 on a 256-bit bus with 608.0 GB/s bandwidth.

Compute units scale up on the Arc Pro B70. It has 4,096 shading units, 256 TMUs, 128 ROPs, and 32 ray tracing cores. The Arc A580 has 3,072 shading units, 192 TMUs, 96 ROPs, and 24 ray tracing cores. Pixel rate is 358.4 GPixel/s for the Arc Pro B70 versus 192.0 GPixel/s for the Arc A580. Texture rate is 716.8 GTexel/s versus 384.0 GTexel/s. FP32 compute is 22.94 TFLOPS versus 12.29 TFLOPS, and FP16 compute is 45.88 TFLOPS versus 24.58 TFLOPS, both at a 2:1 ratio.

Power and connectivity differ. The Arc A580 has a TDP of 175 W with dual 8-pin power connectors and a suggested PSU of 450 W. The Arc Pro B70 has a TDP of 230 W with a single 8-pin power connector and a suggested PSU of 550 W. Both are dual-slot cards. The Arc A580 uses PCIe 4.0 x16, while the Arc Pro B70 uses PCIe 5.0 x16.

Display outputs differ slightly. The Arc A580 has 1x HDMI 2.1 and 3x DisplayPort 2.0. The Arc Pro B70 has 1x HDMI 2.1a and 3x DisplayPort 2.1. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Arc Pro B70 has recorded dimensions of 267 mm length, 110 mm height, and 39 mm width, while the Arc A580 has no recorded dimensions.

Architecture Differences

The architecture generation is the defining difference between these two GPUs. The Arc A580 uses the Xe-HPG architecture, which is the foundation of Intel's Alchemist generation. The Arc Pro B70 uses the Xe2-HPG architecture, which is the foundation of Intel's Battlemage generation. This means the Arc Pro B70 represents a newer architectural design with the second generation of Intel's high-performance graphics architecture.

The chip designs reflect this generational shift. The Arc A580 uses the DG2-512 chip, a large die at 406 mm² built on TSMC's 6 nm process. The Arc Pro B70 uses the BMG-G31 chip, a slightly smaller die at 368 mm² built on TSMC's 5 nm process. The smaller die on a more advanced process node allows the Arc Pro B70 to pack more shading units, TMUs, ROPs, and ray tracing cores into a smaller physical area.

The transistor counts tell a partial story. The Arc A580 has 21,700 million transistors on its 406 mm² die, giving a density of 53.4M per mm². The Arc Pro B70's transistor count is unknown, so a direct density comparison is not possible from the recorded data. However, the 5 nm process node is listed as the manufacturing technology for the Arc Pro B70, which is a more advanced node than the 6 nm process used for the Arc A580.

Memory architecture also differs. While both cards use GDDR6 memory on a 256-bit bus, the Arc Pro B70 runs its memory at 2375 MHz with 19 Gbps effective speed, yielding 608.0 GB/s of bandwidth. The Arc A580 runs at 2000 MHz with 16 Gbps effective speed, yielding 512.0 GB/s. The Arc Pro B70's 32 GB capacity is four times the Arc A580's 8 GB.

The rendering pipelines scale accordingly. The Arc Pro B70's pixel rate of 358.4 GPixel/s is nearly double the Arc A580's 192.0 GPixel/s. Its texture rate of 716.8 GTexel/s is also nearly double the Arc A580's 384.0 GTexel/s. FP32 compute scales from 12.29 TFLOPS on the Arc A580 to 22.94 TFLOPS on the Arc Pro B70, a roughly 87 percent increase. FP16 compute scales from 24.58 TFLOPS to 45.88 TFLOPS, also at a 2:1 ratio for both.

The bus interface advances from PCIe 4.0 x16 on the Arc A580 to PCIe 5.0 x16 on the Arc Pro B70. Display outputs also move from DisplayPort 2.0 on the Arc A580 to DisplayPort 2.1 on the Arc Pro B70, with the HDMI version advancing from 2.1 to 2.1a.

The Verdict

The Intel Arc A580 and Intel Arc Pro B70 serve different roles in the database, and the recorded data supports a clear separation of use cases.

For measured performance evidence, the Arc A580 is the only option. It has three benchmark scores, an average score of 57,756, and an 87th percentile ranking among all GPUs. Its nearest rivals cluster within about one percentage point, confirming that it performs in a tight competitive band. The Arc Pro B70 has no benchmark data, an average score of zero, and a 50th percentile default ranking. There is no performance evidence to support a claim that the Arc Pro B70 outperforms the Arc A580 in any test.

For raw specification capacity, the Arc Pro B70 is the stronger card on paper. It offers 32 GB of memory versus 8 GB, higher bandwidth at 608.0 GB/s versus 512.0 GB/s, more shading units, more TMUs, more ROPs, more ray tracing cores, higher clocks, and more than double the FP32 compute throughput. Its 5 nm process node and Xe2-HPG architecture represent a newer design generation than the 6 nm Xe-HPG architecture in the Arc A580.

The Arc A580 is an active product with a release date of October 2023 and a successor listed as Battlemage. The Arc Pro B70 has a release date of March 2026, no production status, and no predecessor or successor recorded. The Arc Pro B70 also has a launch MSRP of 949 USD, which the database records as the only pricing information for either card.

The data suggests that the Arc A580 is a benchmark-verified mid-range graphics card, while the Arc Pro B70 is a specification-defined professional card with a much larger memory pool and higher compute ceiling. Users who require measured performance data should look to the Arc A580. Users who require maximum memory capacity and compute throughput on paper should look to the Arc Pro B70. The two cards do not compete on the same evidence basis in the database.