GPU Comparison

The NVIDIA RTX A5000 Mobile and Intel Arc B580 occupy very different corners of the GPU market, yet benchmark results show them trading blows across a range of workloads. The A5000 Mobile, an end-of-life Ampere-generation part aimed at professional mobile workstations, averages a score of 24763, placing it in the 70th percentile of all GPUs. The Arc B580, an active desktop Battlemage part, averages 23021, sitting in the 68th percentile. The data reveals a split decision: the A5000 Mobile wins 4 of the 9 head-to-head benchmarks, while the Arc B580 wins 5. The margin of victory matters more than the count, however, as some of these wins are decisive while others are near-statistical ties.

Head-to-Head Benchmarks

The most lopsided result in this comparison is in the Passmark DirectX 10 test, where the NVIDIA RTX A5000 Mobile scores 115 against the Arc B580's 76, a 51.3% advantage. This is a stark generational gap, suggesting the older Ampere architecture retains a significant edge in legacy DirectX 10 workloads. The A5000 Mobile also leads in Geekbench OpenCL, scoring 110877 versus 92821, a 19.5% margin that reflects its higher raw FP32 throughput of 19.35 TFLOPS compared to the Arc B580's 13.67 TFLOPS. In the Passmark DirectX 11 test, the A5000 Mobile wins again, 133 to 128, though the 3.9% delta is much narrower and within the range of typical run-to-run variance.

The Intel Arc B580 strikes back decisively in the Geekbench Vulkan test, posting a score of 109672 against the A5000 Mobile's 88144. That is a 19.6% margin in Intel's favor, a strong indicator that the newer Xe2-HPG architecture handles modern Vulkan APIs far more efficiently than the Ampere part. The B580 also wins the Passmark GPU Compute test, 7729 versus 6945, a 10.1% advantage that points to better raw compute throughput in OpenCL-style workloads. In the Passmark DirectX 9 test, the B580 takes a 183 to 169 win, a 7.7% margin, showing it handles legacy DirectX 9 titles better than the NVIDIA part. The Passmark G2D test goes to the B580 as well, 709 to 629, an 11.3% lead that suggests faster 2D rendering and window composition.

Two benchmarks are essentially ties. In Passmark DirectX 12, the B580 scores 76 against the A5000 Mobile's 72, a 5.3% delta that is close but still a win for Intel. The Passmark G3D test is the closest of all: the A5000 Mobile scores 15779 against the B580's 15748, a margin of just 0.2%, effectively a dead heat. The A5000 Mobile's average benchmark score of 24763 is 7.6% higher than the B580's 23021, but that gap is heavily influenced by the A5000's massive OpenCL lead. For overall gaming performance as measured by G3D, these two cards are indistinguishable.

Where Each One Wins

The NVIDIA RTX A5000 Mobile is the clear choice for compute-heavy professional workloads that lean on OpenCL. Its 19.5% lead over the B580 in Geekbench OpenCL, combined with its 51.3% dominance in DirectX 10, makes it the stronger part for legacy software and compute-accelerated applications that are not optimized for the latest graphics APIs. The A5000 Mobile also has a slight but measurable edge in DirectX 11, which remains the baseline for many older CAD and 3D modeling tools. Its architecture, with 6144 shading units and 192 tensor cores, is built for sustained professional workloads, and the data shows it delivers there.

The Intel Arc B580 wins in modern API scenarios. Its 19.6% lead in Geekbench Vulkan is the headline result, indicating superior performance in Vulkan-based games and applications. The B580 also wins the compute test by 10.1%, which is notable because it achieves this with fewer shading units (2560 versus 6144) and lower FP32 throughput (13.67 TFLOPS versus 19.35 TFLOPS). This suggests the Xe2-HPG architecture is significantly more efficient per shader. The B580's wins in DirectX 9 and G2D further position it as the better part for legacy DirectX 9 games and general 2D desktop work. For DirectX 12, the B580 ekes out a 5.3% win, which is modest but consistent with its newer architecture being better tuned for modern graphics pipelines.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA RTX A5000 Mobile has an average benchmark score of 24763, which is 7.6% higher than the Intel Arc B580's 23021. However, the A5000 Mobile's nearest rival by average score is the AMD Radeon RX 590 (24744, 0.1% delta), while the B580's closest rival is the AMD Radeon RX 580 2048SP (23061, -0.2% delta).

Q: How do they compare in DirectX 12 performance?

A: The Intel Arc B580 wins the Passmark DirectX 12 test with a score of 76, against the NVIDIA RTX A5000 Mobile's 72, a 5.3% margin. Both GPUs support DirectX 12 Ultimate (12_2), so feature parity is not an issue.

Q: Which card is better for Vulkan workloads?

A: The Intel Arc B580 is significantly better in Vulkan. It scores 109672 in Geekbench Vulkan, which is 19.6% higher than the NVIDIA RTX A5000 Mobile's 88144.

Q: What about legacy DirectX 9 and DirectX 10 games?

A: The results are split. The Intel Arc B580 wins DirectX 9 (183 versus 169, a 7.7% lead), but the NVIDIA RTX A5000 Mobile dominates DirectX 10 (115 versus 76, a 51.3% lead).

Q: Is the NVIDIA card's higher memory bandwidth an advantage in benchmarks?

A: The A5000 Mobile has a 448.0 GB/s bandwidth on a 256-bit bus, while the Arc B580 has 456.0 GB/s on a 192-bit bus. Despite having slightly less bandwidth, the B580 wins the Passmark GPU Compute test by 10.1% (7729 versus 6945), indicating bandwidth alone does not determine compute performance.

Q: How does the Arc B580 compare to an NVIDIA GeForce RTX 2080 in average score?

A: The Intel Arc B580's average benchmark score of 23021 is 0.6% higher than the NVIDIA GeForce RTX 2080's 22895. This places the B580 in the company of last-generation high-end desktop GPUs.

Specification Differences

The two GPUs differ in nearly every core specification. The NVIDIA RTX A5000 Mobile uses 6144 shading units, 192 TMUs, and 96 ROPs, while the Intel Arc B580 uses 2560 shading units, 160 TMUs, and 80 ROPs. The A5000 Mobile also has 48 ray tracing cores and 192 tensor cores, whereas the B580 has 20 ray tracing cores and no listed tensor cores. Clock speeds diverge sharply: the A5000 Mobile runs at a 900 MHz base and 1575 MHz boost, while the B580 runs at a flat 2670 MHz for both base and boost. Memory configurations differ as well, the A5000 Mobile has 16 GB of GDDR6 on a 256-bit bus, while the B580 has 12 GB of GDDR6 on a 192-bit bus. The B580's memory runs at 2375 MHz (19 Gbps effective) versus the A5000's 1750 MHz (14 Gbps effective).

Power and physical specs also differ. The A5000 Mobile has a 150 W TDP and uses no power connectors, as it is designed for mobile integration. The B580 has a 190 W TDP, uses a single 8-pin power connector, and requires a 450 W suggested PSU. The B580 is a dual-slot card measuring 272 mm in length, 115 mm in height, and 45 mm in width, while the A5000 Mobile has no listed dimensions. The B580 offers 1x HDMI 2.1a and 3x DisplayPort 2.1 outputs, whereas the A5000 Mobile's display outputs are listed as "Portable Device Dependent." Both use a PCIe 4.0 interface, but the A5000 Mobile uses x16 while the B580 uses x8.

Architecture Differences

The NVIDIA RTX A5000 Mobile is built on the GA104 chip using the Ampere architecture, fabricated on Samsung's 8 nm process. The die size is 392 mm² with 17,400 million transistors, giving a transistor density of 44.4M per mm². The Intel Arc B580 uses the BMG-G21 chip with the Xe2-HPG architecture (Battlemage generation), fabricated on TSMC's 5 nm process. Its die is 272 mm² with 19,600 million transistors, achieving a much higher transistor density of 72.1M per mm².

The B580's 5 nm node allows for a significantly denser design, but the A5000 Mobile compensates with a much larger shader array. FP32 performance favors the A5000 Mobile at 19.35 TFLOPS versus the B580's 13.67 TFLOPS. However, FP16 performance tells a different story: the A5000 Mobile achieves 19.35 TFLOPS (1:1 ratio), while the B580 achieves 27.34 TFLOPS (2:1 ratio). Pixel and texture rates also favor the B580, which hits 213.6 GPixel/s and 427.2 GTexel/s, versus the A5000 Mobile's 151.2 GPixel/s and 302.4 GTexel/s. The B580's higher clocks and superior fill rates help explain its wins in rasterization-heavy benchmarks despite lower raw FP32 compute. The A5000 Mobile is end-of-life, released in April 2021, while the B580 is active, released in December 2024.

The Verdict

The data supports a clear split based on workload. For professional compute, legacy API support, and OpenCL-heavy applications, the NVIDIA RTX A5000 Mobile is the stronger card. Its 19.5% OpenCL lead and 51.3% DirectX 10 advantage are decisive, and its 16 GB memory capacity is double the B580's 12 GB. It also holds the 70th percentile rank versus the B580's 68th, and its average score is 7.6% higher.

For modern gaming, Vulkan-based titles, and general 2D performance, the Intel Arc B580 is the better pick. Its 19.6% Vulkan lead and 10.1% compute win show that the newer architecture is more efficient per transistor. It also wins DirectX 9, DirectX 12, and G2D tests. The B580's 12 GB of memory is less, but its 456.0 GB/s bandwidth is slightly higher than the A5000's 448.0 GB/s. The B580 is an active product with a launch MSRP of 249 USD, while the A5000 Mobile is end-of-life. Choose the A5000 Mobile for legacy compute and professional OpenCL workloads; choose the Arc B580 for modern APIs and Vulkan gaming. In DirectX 11 and G3D overall, the two are effectively tied, so those workloads will not break the tie.