Intel Arc B580 vs NVIDIA GeForce RTX 2080 SUPER Comparison

The NVIDIA GeForce RTX 2080 SUPER and the Intel Arc B580 represent two very different eras of GPU design, yet their benchmark data places them in a surprisingly close overall standing. The RTX 2080 SUPER, a Turing-generation part from 2019, holds a narrow lead in average benchmark score at 24,170, while the Arc B580, a Battlemage-generation part from late 2024, trails slightly at 23,021. The two cards sit within 4.8% of each other in that aggregate, and their percentile rankings are nearly identical at 69 and 68, respectively. The head-to-head results, however, reveal a clear split: the older NVIDIA card wins 8 of 10 tests, while the Intel card takes 2, including a decisive victory in the most modern DirectX 12 workload.

Head-to-Head Benchmarks

The most striking result in the comparison is the 3DMark Steel Nomad DX12 test. The Intel Arc B580 scores 3,068, a massive 38.7% advantage over the RTX 2080 SUPER’s 1,882. This is not a marginal win; it is a generational leap in raw throughput for a modern, heavy API workload. The data indicates that the Arc B580 is substantially better equipped to handle the demands of contemporary game engines that leverage DirectX 12’s advanced features. The RTX 2080 SUPER’s only other loss comes in the Passmark DirectX 12 test, where it scores 75 against 76, a negligible 1.3% deficit. These two results suggest that the Intel card’s architecture, Xe2-HPG, is fundamentally more efficient at processing the low-level, draw-call-heavy workloads that define modern titles.

Outside of those two DirectX 12 tests, the NVIDIA GeForce RTX 2080 SUPER is the clear winner, often by significant margins. The largest victory comes in Passmark DirectX 10, where the RTX 2080 SUPER scores 140 versus 76, a staggering 84.2% advantage. This is a legacy API test, but the difference is so large that it indicates a fundamental strength in the Turing architecture’s handling of older rendering paths. Similarly, in Passmark DirectX 11, the NVIDIA card posts 165 against 128, a 28.9% win, and in DirectX 9 it scores 227 versus 183, a 24% lead. The pattern is consistent: the older the API, the more dominant the RTX 2080 SUPER becomes.

The compute and general-purpose performance also favors NVIDIA, though by smaller margins. In Geekbench OpenCL, the RTX 2080 SUPER scores 99,226 to 92,821, a 6.9% advantage. Geekbench Vulkan is closer, with the NVIDIA card at 111,284 versus 109,672, just a 1.5% lead. The Passmark G3D test, which measures overall 3D graphics performance, shows the RTX 2080 SUPER at 19,490, a 23.8% advantage over the Intel card’s 15,748. Passmark GPU Compute follows a similar trend, with NVIDIA at 8,290 and Intel at 7,729, a 7.3% lead. The final win for NVIDIA is in Passmark G2D, a 2D graphics test, where it scores 920 versus 709, a 29.8% margin. The data across these tests shows that the RTX 2080 SUPER is a more balanced performer in mixed workloads, while the Arc B580 is a specialist in modern API efficiency.

Where Each One Wins

The Intel Arc B580’s wins are concentrated in the most forward-looking benchmark: 3DMark Steel Nomad DX12. Its 3,068 score is not just a win; it is a 38.7% blowout that dwarfs any other margin in the entire comparison. This suggests that for users primarily interested in the latest DirectX 12 Ultimate games, the Arc B580 offers significantly more headroom. Its secondary win in Passmark DirectX 12, though narrow, reinforces this strength. The architecture’s design, with its 20 ray tracing cores and 5 nm process node, appears optimized for the parallel, compute-heavy tasks that modern engines generate.

The NVIDIA GeForce RTX 2080 SUPER wins everywhere else, making it the more versatile card for a broader software library. Its dominance in legacy APIs is the most notable feature. The 84.2% lead in DirectX 10 and the 24% lead in DirectX 9 indicate that it will handle older games and applications with far greater ease. The 28.9% win in DirectX 11 is also substantial, covering a vast catalog of games released in the last decade. Its wins in OpenCL and Vulkan, though smaller, show that it is also competitive in modern compute and cross-platform graphics. The 23.8% lead in the Passmark G3D overall score suggests that for general 3D performance across a mix of workloads, the RTX 2080 SUPER is the safer choice. The 29.8% win in G2D is a niche but clear advantage for desktop and 2D rendering tasks.

The Verdict

The benchmark data presents a clear trade-off. For users whose primary focus is playing the newest DirectX 12 games, the Intel Arc B580 is the stronger performer. Its 38.7% lead in 3DMark Steel Nomad DX12 is the single most significant data point in the comparison. It is not a close margin; it is a generational gap in modern rendering capability. The 12 GB memory size and higher boost clock of 2670 MHz contribute to this edge, though the primary differentiator appears to be the efficiency of the Xe2-HPG architecture.

For every other scenario, the NVIDIA GeForce RTX 2080 SUPER is the superior card. It wins 8 out of 10 benchmarks, including all the legacy API tests, where it shows advantages ranging from 1.5% to 84.2%. Its higher average benchmark score of 24,170 also indicates better overall consistency. Users with a backlog of older games, or those who rely on OpenCL or Vulkan compute tasks, will find the NVIDIA card to be the more reliable performer. The data suggests that the RTX 2080 SUPER, despite its age, offers a higher level of general-purpose performance, while the Arc B580 is a specialized tool for a specific, modern workload. The choice depends entirely on the user’s software priority.

FAQ

Q: Which card has the higher average benchmark score?

A: The NVIDIA GeForce RTX 2080 SUPER has a higher average benchmark score of 24,170, compared to the Intel Arc B580’s 23,021.

Q: How large is the Intel Arc B580’s win in the 3DMark Steel Nomad DX12 test?

A: The Intel Arc B580 scores 3,068 in 3DMark Steel Nomad DX12, which is a 38.7% advantage over the NVIDIA GeForce RTX 2080 SUPER’s score of 1,882.

Q: In which test does the NVIDIA GeForce RTX 2080 SUPER have its largest margin of victory?

A: The NVIDIA card’s largest margin is in Passmark DirectX 10, where it scores 140 versus the Intel card’s 76, a 84.2% difference.

Q: What is the difference in their DirectX 11 performance?

A: In Passmark DirectX 11, the NVIDIA GeForce RTX 2080 SUPER scores 165, which is 28.9% higher than the Intel Arc B580’s score of 128.

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

A: The NVIDIA GeForce RTX 2080 SUPER leads in Geekbench Vulkan with a score of 111,284, a 1.5% advantage over the Intel Arc B580’s 109,672.

Q: Which card wins in the Passmark DirectX 12 test?

A: The Intel Arc B580 wins the Passmark DirectX 12 test, scoring 76 to the NVIDIA GeForce RTX 2080 SUPER’s 75, a margin of 1.3%.

Architecture Differences

The underlying architectures of these two GPUs are fundamentally different. The NVIDIA GeForce RTX 2080 SUPER is built on the Turing architecture with the TU104 chip, manufactured on a 12 nm process at TSMC. This is a monolithic design with 13,600 million transistors on a 545 mm² die, yielding a transistor density of 25.0M / mm². In contrast, the Intel Arc B580 uses the Xe2-HPG architecture with the BMG-G21 chip, built on a more advanced 5 nm process, also at TSMC. This process allows Intel to pack 19,600 million transistors into a significantly smaller 272 mm² die, achieving a much higher transistor density of 72.1M / mm². The Arc B580’s process advantage is clear, but it does not translate into universal performance superiority.

The compute configurations also differ. The RTX 2080 SUPER has 3,072 shading units, 192 texture mapping units, and 64 raster operation pipelines. It also includes 48 ray tracing cores and 384 tensor cores, a feature set that the Intel card lacks, as its tensor core count is null. The Arc B580, meanwhile, has fewer shading units at 2,560 but more ROPs at 80, and it includes 20 ray tracing cores. The architectural focus is different: NVIDIA’s Turing was designed with a heavy emphasis on tensor cores for AI and deep learning, while Intel’s Xe2-HPG appears to concentrate on raw throughput and ray tracing, with its 20 RT cores. The APIs supported are identical, with both cards offering DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Specification Differences

The specifications reveal a clear generational divide in memory and power characteristics. The NVIDIA GeForce RTX 2080 SUPER has 8 GB of GDDR6 memory on a 256-bit bus, providing a bandwidth of 495.9 GB/s. The Intel Arc B580 offers more capacity with 12 GB of GDDR6 memory, but on a narrower 192-bit bus, resulting in slightly lower bandwidth of 456.0 GB/s. This trade-off between capacity and bandwidth is a key differentiator. The Intel card’s memory runs at 2375 MHz (19 Gbps effective), while the NVIDIA card’s memory runs at 1937 MHz (15.5 Gbps effective).

Clock speeds and power draw also differ substantially. The RTX 2080 SUPER has a base clock of 1650 MHz and a boost clock of 1815 MHz, with a TDP of 250 W. The Arc B580 has a much higher base and boost clock, both at 2670 MHz, yet a lower TDP of 190 W. This indicates that the 5 nm process is significantly more power-efficient. The power connectors reflect this: the NVIDIA card requires a 1x 6-pin and 1x 8-pin setup, while the Intel card only needs a single 8-pin connector. The suggested PSU also differs, with NVIDIA recommending a 600 W unit and Intel recommending a 450 W unit. The bus interface is another point of difference, with the RTX 2080 SUPER using PCIe 3.0 x16 and the Arc B580 using the newer PCIe 4.0 x8.

The physical dimensions and display outputs differ as well. The RTX 2080 SUPER is 267 mm long, 116 mm high, and 35 mm wide, while the Arc B580 is slightly longer at 272 mm, marginally shorter at 115 mm, and notably thicker at 45 mm. Both are dual-slot cards. For display outputs, the NVIDIA card offers 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C, while the Intel card provides 1x HDMI 2.1a and 3x DisplayPort 2.1. The production status also contrasts, with the RTX 2080 SUPER being end-of-life and the Arc B580 being an active product.