Intel Arc A580 vs Intel Arc A770 Comparison

Head-to-Head Benchmarks

The data is unambiguous: the Intel Arc A770 outperforms the Intel Arc A580 in every recorded benchmark. Across three head-to-head tests, the A770 claims all three wins, with margins ranging from just under 19% to over 33%. The largest gap appears in the 3DMark Steel Nomad DX12 test, where the A770 scores 2969 against the A580’s 2229, a 33.2% advantage. That is a decisive lead for a synthetic DX12 workload, and it suggests the A770’s additional execution resources translate directly into higher rasterization throughput.

In Geekbench OpenCL, the A770 posts 109175 points, while the A580 reaches 91657, a 19.1% difference. This test stresses raw compute across a wide range of workloads, and the A770’s lead here is consistent with its larger shader array and higher clock speeds. The Vulkan result is nearly identical in margin: the A770 scores 94284 versus the A580’s 79381, a 18.8% gap. Vulkan tends to scale well with additional cores and memory bandwidth, so the A770’s consistent advantage across both OpenCL and Vulkan points to a well-balanced hardware design rather than a single-test anomaly.

The average benchmark scores reinforce this pattern. The A770’s average is 68809, placing it in the 90th percentile of all GPUs in the database. The A580’s average is 57756, which lands in the 87th percentile. That 3-percentile difference may sound modest, but the raw average score gap of 11053 points, roughly 19.1% lower for the A580, shows that the A770 occupies a distinctly higher performance tier. Looking at the nearest rivals for each card clarifies the competitive landscape. The A770 sits within 2% of the NVIDIA Quadro P6000 (69986, A770 is 1.7% behind) and the AMD Radeon Pro WX 8200 (69870, 1.5% behind). The A580, by contrast, trades blows with the AMD Radeon RX 5600 OEM (58085, A580 is 0.6% behind) and the AMD Radeon RX 9070 GRE (57367, A580 is 0.7% ahead). In short, the A770 competes against workstation-class cards from the previous generation, while the A580 aligns with mid-range gaming GPUs.

Where Each One Wins

The A770 wins everywhere in this comparison, but the magnitude of its wins varies by workload type. The 33.2% lead in 3DMark Steel Nomad is the standout, and it indicates that the A770 is substantially stronger in modern DX12 game workloads, particularly those that rely on heavy geometry and compute. The A580 still delivers playable performance, but the delta is too large to ignore for anyone prioritizing frame rates in demanding titles.

In compute-oriented tasks, the A770’s lead narrows to roughly 19% in both OpenCL and Vulkan. This suggests that while the A580 is a capable compute card, the A770’s extra 1024 shading units, 64 TMUs, and 32 ROPs provide a meaningful but not overwhelming advantage in general-purpose workloads. The A770 also benefits from higher clock speeds, with a 2100 MHz base and 2400 MHz boost versus the A580’s 1700 MHz base and 2000 MHz boost. That 400 MHz boost difference alone explains a significant portion of the gap, but the core count difference matters more at scale.

The A580 does have one clear advantage in this matchup: power consumption. Its TDP is 175 W, compared to the A770’s 225 W, and its suggested PSU is 450 W versus 550 W. For a system builder targeting lower overall power draw or working with a smaller power supply, the A580 is the more accommodating choice, though it comes with a performance penalty. The A580 also uses the same 8 GB memory bus width (256 bit) and identical bandwidth (512.0 GB/s), but only half the capacity: 8 GB versus 16 GB. In memory-heavy workloads, the A770’s larger frame buffer can prevent capacity-related stutters, though the theoretical bandwidth is identical.

The A580’s production status is Active, while the A770 is End-of-life. For buyers in 2025, this means the A580 remains in production, while the A770 may be harder to source new. The A770’s launch MSRP is 329 USD, which appears once in the database, but no pricing comparison is possible since the A580 has no recorded launch MSRP. The A770 also holds a percentile advantage, sitting at 90 versus 87, which confirms it outperforms a larger share of the GPU population.

The Verdict

The Intel Arc A770 is the clear winner for anyone who wants maximum performance from this generation of Intel discrete GPUs. Its 33.2% lead in 3DMark Steel Nomad and 19% leads in both OpenCL and Vulkan make it the obvious choice for gaming, content creation, and compute workloads. The 16 GB frame buffer is a meaningful upgrade over the A580’s 8 GB, and the higher core counts and clock speeds translate into measurable gains across every recorded test. If the choice is between these two cards and performance is the only criterion, the A770 wins without qualification.

The Intel Arc A580 is not without merit, but its appeal is narrower. It delivers around 81% of the A770’s OpenCL and Vulkan scores and roughly 75% of its 3DMark Steel Nomad score, while drawing 50 W less power and requiring a 450 W PSU instead of 550 W. For a compact or low-power build, or for a user who already owns a 450 W power supply, the A580 is a capable option that still lands in the 87th percentile of all GPUs. It is also the only one of the two that remains in active production, which matters for warranty availability and long-term supply.

The data does not support choosing the A580 for raw performance. The A770 wins 3 out of 3 head-to-head tests, and its nearest rivals include the NVIDIA Quadro P6000 and AMD Radeon Pro WX 8200, both of which are workstation-oriented cards with higher average scores. The A580’s nearest rivals, meanwhile, are mid-range gaming cards like the AMD Radeon RX 5600 OEM and RX 9070 GRE. That positioning difference is the clearest summary: the A770 is a top-tier performer in its generation, while the A580 is a solid mid-range option. For gamers and creators who want the best Intel Alchemist experience, the A770 is the definitive choice. For budget-conscious builders with power constraints, the A580 remains a viable alternative, but it is not a performance competitor to the A770.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The Intel Arc A770 has an average benchmark score of 68809, compared to the Intel Arc A580’s 57756. The A770 also sits in the 90th percentile of all GPUs, while the A580 sits in the 87th percentile.

Q: How much faster is the A770 in 3DMark Steel Nomad?

A: The A770 scores 2969 in 3DMark Steel Nomad DX12, while the A580 scores 2229. This represents a 33.2% advantage for the A770.

Q: Do both cards use the same memory bandwidth?

A: Yes, both cards have a 256 bit memory bus and 512.0 GB/s bandwidth. However, the A770 has 16 GB of GDDR6 memory, while the A580 has 8 GB.

Q: What is the power consumption difference?

A: The A770 has a TDP of 225 W and requires a 550 W PSU, while the A580 has a TDP of 175 W and requires a 450 W PSU. The A580 uses two 8-pin power connectors, while the A770 uses one 6-pin and one 8-pin connector.

Q: Are both cards based on the same silicon?

A: Yes, both use the DG2-512 chip on TSMC’s 6 nm process, with 21,700 million transistors and a 406 mm² die size. They share the Xe-HPG architecture and belong to the Alchemist generation, though the A770 is Arc 7 and the A580 is Arc 5.

Q: Which card has more shading units and ray tracing cores?

A: The A770 has 4096 shading units, 256 TMUs, 128 ROPs, and 32 ray tracing cores. The A580 has 3072 shading units, 192 TMUs, 96 ROPs, and 24 ray tracing cores.

Architecture Differences

Both cards are built on Intel’s Xe-HPG architecture and use the same DG2-512 chip, fabricated by TSMC on a 6 nm process. The transistor count is identical at 21,700 million, and the die size is the same 406 mm², yielding a transistor density of 53.4 million per square millimeter. This means the A580 is not a fundamentally different design; it is a cut-down version of the same silicon with fewer enabled execution resources.

The most significant architectural difference lies in the number of active compute units. The A770 enables 4096 shading units, while the A580 enables 3072, a reduction of exactly 25%. This pattern repeats across the rest of the pipeline: the A770 has 256 texture mapping units versus 192 on the A580, and 128 ROPs versus 96. Ray tracing hardware follows the same ratio, with 32 ray tracing cores on the A770 and 24 on the A580. The result is that the A770’s pixel rate is 307.2 GPixel/s versus 192.0 GPixel/s on the A580, and its texture rate is 614.4 GTexel/s versus 384.0 GTexel/s. These are not minor differences; they reflect a 60% advantage in pixel throughput and a 60% advantage in texture throughput for the A770.

Clock speeds add another layer of separation. The A770 runs at a 2100 MHz base clock and 2400 MHz boost, while the A580 runs at 1700 MHz base and 2000 MHz boost. That 400 MHz boost gap translates into higher raw compute: the A770 delivers 19.66 TFLOPS of FP32 performance, while the A580 delivers 12.29 TFLOPS. FP16 performance follows the same 2:1 ratio, with 39.32 TFLOPS on the A770 and 24.58 TFLOPS on the A580. The memory subsystem is architecturally identical in bus width and bandwidth, but the A770 carries 16 GB of GDDR6, double the A580’s 8 GB. Both run at 2000 MHz with 16 Gbps effective speed. Because the chips share the same process node and die, the A580’s lower clocks likely stem from binning, where chips that do not meet the A770’s frequency targets are sold as A580 parts.

Specification Differences

The specification sheets for these two cards are nearly identical except for the core configuration and power envelope. The process node is the same (6 nm TSMC), the chip is the same (DG2-512), and the architecture is the same (Xe-HPG). Both use PCIe 4.0 x16 and both provide 1x HDMI 2.1 and 3x DisplayPort 2.0 outputs. The API support is also identical: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither card has a recorded game clock, tensor cores, or dimensions, and both are dual-slot designs.

The differences begin with clocks: the A770 has a 2100 MHz base and 2400 MHz boost, while the A580 has a 1700 MHz base and 2000 MHz boost. Memory capacity is 16 GB on the A770 versus 8 GB on the A580, though both use GDDR6 with a 256 bit bus and 512.0 GB/s bandwidth. The compute configuration differs as listed above: 4096 vs 3072 shading units, 256 vs 192 TMUs, 128 vs 96 ROPs, and 32 vs 24 ray tracing cores. Pixel rate is 307.2 GPixel/s on the A770 versus 192.0 GPixel/s on the A580, and texture rate is 614.4 GTexel/s versus 384.0 GTexel/s. FP32 compute is 19.66 TFLOPS versus 12.29 TFLOPS, and FP16 is 39.32 TFLOPS versus 24.58 TFLOPS.

The power delivery is another clear differentiator. The A770 has a TDP of 225 W, a suggested PSU of 550 W, and uses one 6-pin plus one 8-pin connector. The A580 has a TDP of 175 W, a suggested PSU of 450 W, and uses two 8-pin connectors. Production status differs as well: the A770 is End-of-life, while the A580 is Active. The release dates are separated by roughly a year: the A770 launched on October 11, 2022, and the A580 followed on October 9, 2023. The A770 has a recorded launch MSRP of 329 USD, while the A580 has no launch MSRP in the database. Both cards list Xe Graphics as their predecessor and Battlemage as their successor, and both belong to the Alchemist generation, with the A770 in the Arc 7 tier and the A580 in the Arc 5 tier.