Intel Arc A580 vs Intel Arc Pro A60 Comparison

Intel Arc Pro A60 and Intel Arc A580 are two distinct Alchemist-generation GPUs from Intel, built on the same Xe-HPG architecture and 6 nm TSMC process, yet they target very different roles. The benchmark data shows a clear performance hierarchy: the Arc A580 wins both shared tests decisively, while the Pro A60 counters with a workstation-oriented feature set, lower power demands, and a professional display configuration. The A580 posts a Geekbench OpenCL score of 91,657 against the A60’s 63,485 (a 30.7% deficit for the Pro card), and a Vulkan score of 79,381 versus 57,166 (a 28% gap). However, the A60’s 88th percentile ranking across all GPUs slightly edges the A580’s 87th, indicating that raw compute averages don’t tell the whole story—the Pro card’s average benchmark score of 60,326 actually exceeds the A580’s 57,756, despite losing every head-to-head test. This paradox stems from different benchmark weighting and the A60’s proximity to higher-tier rivals like the RTX 4090 (0% delta) and Radeon Pro W6600M (-2.5%), while the A580 sits near mid-range gaming cards such as the RX 5600 OEM (-0.6%) and RX 6950 XT (-1.1%).

Head-to-Head Benchmarks

The Geekbench OpenCL test delivers the largest single-score gap between these two cards. The Intel Arc A580 scores 91,657, which is 30.7% higher than the Arc Pro A60’s 63,485. This is not a marginal difference; it represents a substantial compute throughput advantage for the A580, driven by its larger silicon and higher shader count. In practical terms, any OpenCL-accelerated workload—rendering, physics simulation, or data processing—will favor the A580 by roughly a third, which is a decisive margin for most users. The A580’s nearest rivals include the AMD Radeon RX 5600 OEM (average score 58,085, delta -0.6%) and the RX 9070 GRE (57,367, delta 0.7%), placing it in a competitive mid-range bracket where its OpenCL lead over the A60 is unusually large.

The Vulkan test narrows the gap slightly but still favors the A580 by a wide margin. Scores are 79,381 for the A580 versus 57,166 for the A60, a 28% difference. Vulkan is a lower-level API that often scales with raw hardware resources, and the A580’s additional execution units and memory bandwidth show here. The A580’s 512.0 GB/s bandwidth versus the A60’s 384.0 GB/s provides a theoretical foundation for this win, though the benchmark score itself is the authoritative metric. Notably, the A60’s Vulkan score of 57,166 is closer to its OpenCL result (63,485) than the A580’s scores are to each other, suggesting the Pro card has more balanced performance across APIs, while the A580 excels specifically in compute-heavy Vulkan workloads.

There are no tests where the A60 wins, as the head-to-head dataset records 0 wins for the Pro card and 2 for the A580. However, the aggregate picture is nuanced: the A60’s average benchmark score of 60,326 is 4.4% higher than the A580’s 57,756. This counterintuitive result occurs because the A60’s nearest rivals include the NVIDIA RTX 4090 (average 60,347, delta 0%) and AMD Radeon Pro Vega 48 (60,140, delta 0.3%), while the A580 competes against lower-scoring cards like the RX 5600 OEM (58,085). The percentile rankings reinforce this: A60 sits at the 88th percentile versus A580’s 87th, meaning the Pro card outperforms a slightly larger fraction of the GPU population despite losing to the A580 in direct comparison.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The Intel Arc Pro A60 has a higher average benchmark score of 60,326, compared to the Intel Arc A580’s 57,756. This is despite the A580 winning both direct head-to-head tests, highlighting how different benchmark suites weight workloads.

Q: What is the performance difference in OpenCL compute?

A: The Arc A580 leads by 30.7% in Geekbench OpenCL, scoring 91,657 versus the A60’s 63,485. This is the largest margin in any shared test.

Q: How do the two cards compare in Vulkan performance?

A: The A580 wins Vulkan by 28%, with a score of 79,381 against the A60’s 57,166. The gap is slightly narrower than OpenCL but remains substantial.

Q: Which card has a higher transistor count?

A: The Arc A580 uses the DG2-512 chip with 21,700 million transistors, while the Arc Pro A60 uses the DG2-256 chip with 11,500 million transistors. The A580’s die is also larger at 406 mm² versus 269 mm².

Q: Are these cards from the same architecture generation?

A: Yes, both are based on Xe-HPG architecture and belong to the Alchemist generation. The A60 is in the Pro Series, while the A580 is in the Arc 5 series.

Q: What is the memory configuration difference?

A: The A60 has 12 GB of GDDR6 on a 192-bit bus with 384.0 GB/s bandwidth, while the A580 has 8 GB on a 256-bit bus with 512.0 GB/s bandwidth. The A580 offers higher bandwidth but less capacity.

Where Each One Wins

The Intel Arc A580 is the clear winner for raw compute performance and gaming-style workloads. Its OpenCL and Vulkan dominance, combined with higher FP32 throughput (12.29 TFLOPS versus 8.397 TFLOPS), makes it the better choice for tasks that leverage massive parallel processing. The A580’s 24 ray tracing cores and 192 texture mapping units provide additional headroom for modern graphics features, and its dual-slot design with two 8-pin power connectors suggests it is built for sustained high-load operation. Users prioritizing maximum frame rates or compute-heavy rendering should select the A580 without hesitation.

The Intel Arc Pro A60 wins in professional deployment scenarios. Its single-slot form factor is a critical advantage for dense multi-GPU workstations, and its four DisplayPort 2.0 outputs (versus the A580’s one HDMI 2.1 and three DisplayPort 2.0) offer more flexible multi-monitor setups without adapter needs. The A60’s lower 130 W TDP versus the A580’s 175 W reduces thermal and power infrastructure requirements, and its 300 W suggested PSU versus 450 W lowers system integration costs. For users who need professional validation, the Pro Series branding and 12 GB memory capacity (50% more than the A580) provide advantages in large dataset handling, even if raw speed is lower.

The average benchmark score tells a third story: the A60’s 60,326 average exceeds the A580’s 57,756, suggesting that in a broader range of mixed workloads—not just the two direct tests—the Pro card holds its own. This is likely due to its higher percentile ranking (88th versus 87th) and closer alignment with professional-grade rivals like the Radeon Pro W6600M (-2.5% delta) and Radeon PRO V710 (2.8% delta), whereas the A580’s rivals are primarily consumer gaming cards. Choose the A60 for workstation stability and balanced professional performance; choose the A580 for uncompromising compute speed.

Specification Differences

The core silicon differs fundamentally: the A60 uses the DG2-256 chip with 11,500 million transistors on a 269 mm² die, while the A580 uses the DG2-512 chip with 21,700 million transistors on a 406 mm² die. Transistor density also varies, with the A580 at 53.4M per mm² versus the A60’s 42.8M per mm². Clock speeds diverge significantly: the A60 has a 900 MHz base clock and 2050 MHz boost, while the A580 runs at 1700 MHz base and 2000 MHz boost. The A580’s higher base clock is notable, but the A60’s boost clock is 50 MHz faster.

Memory configurations are nearly opposite: the A60 offers 12 GB GDDR6 on a 192-bit bus with 384.0 GB/s bandwidth, while the A580 has 8 GB on a 256-bit bus with 512.0 GB/s bandwidth. Memory clock is identical at 2000 MHz (16 Gbps effective). Compute resources favor the A580: 3072 shading units versus 2048, 192 TMUs versus 128, 96 ROPs versus 64, and 24 RT cores versus 16. The A580’s pixel rate is 192.0 GPixel/s versus 131.2 GPixel/s, and its texture rate is 384.0 GTexel/s versus 262.4 GTexel/s. FP32 performance is 12.29 TFLOPS versus 8.397 TFLOPS, with FP16 at 24.58 TFLOPS versus 16.79 TFLOPS (both 2:1).

Power and physical specs differ: the A60 has a 130 W TDP with a 300 W suggested PSU, while the A580 has a 175 W TDP with a 450 W suggested PSU. The A60 is single-slot with no listed power connectors, whereas the A580 is dual-slot with two 8-pin connectors. Display outputs differ: the A60 has four DisplayPort 2.0, while the A580 has one HDMI 2.1 and three DisplayPort 2.0. Both use PCIe 4.0 x16 and support DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4.

Architecture Differences

Both GPUs share the Xe-HPG architecture and are manufactured on TSMC’s 6 nm process, but they implement it at different scales. The A580’s DG2-512 chip is a full-fat implementation with 21,700 million transistors, representing the maximum configuration of the Alchemist design. The A60’s DG2-256 chip is a cut-down variant with 11,500 million transistors, likely designed for lower power and thermal envelopes. This explains the A580’s higher core counts across every unit type—shading units, TMUs, ROPs, and RT cores—while the A60 compensates with a higher boost clock (2050 MHz versus 2000 MHz).

The generation labels differ: the A60 is listed as “Alchemist (Pro Series)” while the A580 is “Alchemist (Arc 5),” indicating Intel’s segmentation of the same architecture into professional and consumer lines. The A580 has a predecessor (Xe Graphics) and successor (Battlemage), while the A60 has neither, suggesting it is a standalone professional product. The A60’s 12 GB memory capacity is unusual for a 192-bit bus, indicating a specific configuration for workstation workloads, whereas the A580’s 8 GB on a 256-bit bus prioritizes bandwidth for gaming. Both support identical API feature sets (DirectX 12 Ultimate, OpenGL 4.6, Vulkan 1.4), but the A580’s larger RT core count (24 versus 16) provides more ray tracing throughput. The A60’s four DisplayPort outputs versus the A580’s mixed HDMI/DisplayPort setup reflects different multi-monitor strategies: the Pro card targets display walls, while the consumer card accommodates typical TV and monitor combinations.