Intel Arc B580 vs NVIDIA GeForce RTX 2080 Comparison

The Intel Arc B580 and NVIDIA GeForce RTX 2080 land at the same overall performance tier, with average benchmark scores of 23021 and 22895 respectively. That 0.6% gap in favor of the Arc B580 is statistically negligible, yet the two cards achieve parity through wildly different strengths. The data reveals a generational clash: Intel’s modern Battlemage architecture, built on a 5 nm process, squares off against NVIDIA’s older 12 nm Turing design. The benchmark breakdown shows a split personality, where the newer card dominates in raw compute-oriented tests while the veteran holds its ground in legacy and rasterization workloads.

Head-to-Head Benchmarks

The most dramatic divergence appears in the 3DMark Steel Nomad DX12 test. The Intel Arc B580 scores 3068, a full 75.1% higher than the RTX 2080’s 1752. This is not a marginal win; it is a generational stomping in a modern, API-heavy workload. The Arc B580’s advantage here suggests its architecture is far more efficient at extracting performance from contemporary DirectX 12 features, likely due to its newer design philosophy.

However, the RTX 2080 strikes back in the PassMark suite. In PassMark G3D, the NVIDIA card scores 18720 versus 15748 for the Arc B580, a 15.9% lead. The legacy API tests are even more lopsided. PassMark DirectX 10 shows the RTX 2080 winning by 44.1% (136 vs 76), and DirectX 9 shows a 17.9% edge (223 vs 183). The DirectX 11 result is closer but still favors NVIDIA, 158 vs 128, a 19% margin.

The compute-oriented tests tell a nuanced story. In Geekbench OpenCL, the Arc B580 edges ahead with 92821 vs 91313, a slim 1.7% gain. Geekbench Vulkan shows the identical 1.7% margin, with the Intel card scoring 109672 against 107797. PassMark GPU Compute flips the script slightly, with the RTX 2080 winning 7872 vs 7729, a 1.8% margin. The PassMark DirectX 12 result is a narrow Intel win, 76 vs 72, a 5.6% difference. The 2D test, PassMark G2D, goes decisively to NVIDIA, 907 vs 709, a 21.8% gap.

Overall, the Arc B580 wins 4 of the 10 head-to-head tests, while the RTX 2080 wins 6. But the magnitude of those wins matters more than the count. Intel’s victories in Steel Nomad and the Geekbench suite are relatively close except for the massive DX12 win, while NVIDIA’s wins in legacy DX10 and DX9 are enormous percentage swings. The data implies that the Arc B580 is built for the future, while the RTX 2080 remains a strong performer for older software stacks.

Architecture Differences

The architectural divide is stark. The Intel Arc B580 uses the BMG-G21 chip on a 5 nm TSMC process, packing 19,600 million transistors into a 272 mm² die. That yields a transistor density of 72.1M per mm². The NVIDIA RTX 2080, in contrast, uses the TU104 chip on a 12 nm process, with 13,600 million transistors spread across a much larger 545 mm² die, resulting in only 25.0M transistors per mm². The Arc B580 is more than twice as dense, a direct consequence of the newer process node.

Clock speeds reflect this efficiency gap. The Arc B580 runs at a flat 2670 MHz for both base and boost, while the RTX 2080 starts at 1515 MHz and boosts to 1710 MHz. The Intel card’s higher clocks contribute to its FP32 throughput of 13.67 TFLOPS, versus 10.07 TFLOPS for the RTX 2080. Yet the RTX 2080 fields more shading units, 2944 vs 2560, and more TMUs, 184 vs 160. The Intel card has more ROPs, 80 vs 64, and a higher pixel rate of 213.6 GPixel/s against 109.4 GPixel/s.

Ray tracing and tensor hardware differ fundamentally. The Arc B580 includes 20 RT cores, while the RTX 2080 has 46 RT cores and 368 tensor cores. The Intel card has no listed tensor core count. The RTX 2080’s tensor cores enable DLSS, a feature absent from the Arc B580’s spec sheet. Memory configurations are similar in bandwidth but differ in capacity: the Arc B580 offers 12 GB of GDDR6 on a 192-bit bus, delivering 456.0 GB/s, while the RTX 2080 has 8 GB on a 256-bit bus, achieving 448.0 GB/s. The Intel card’s extra 4 GB could be decisive in memory-heavy scenarios.

Power and connectivity also diverge. The Arc B580 is rated at 190 W TDP with a single 8-pin connector, while the RTX 2080 draws 215 W and requires both a 6-pin and an 8-pin connector. The Intel card uses PCIe 4.0 x8, whereas the RTX 2080 uses PCIe 3.0 x16. Display outputs favor Intel, with one HDMI 2.1a and three DisplayPort 2.1 ports, versus the RTX 2080’s single HDMI 2.0 and three DisplayPort 1.4a plus a USB Type-C port.

Where Each One Wins

The Intel Arc B580 is the clear winner for modern DirectX 12 gaming and compute workloads. The 75.1% lead in 3DMark Steel Nomad DX12 is the headline statistic. The Geekbench OpenCL and Vulkan scores, both 1.7% ahead, reinforce that the architecture handles current-generation APIs with ease. The 12 GB memory buffer provides headroom for high-resolution textures and large datasets, making it suitable for content creation and AI experimentation, despite lacking dedicated tensor cores.

The NVIDIA RTX 2080 dominates legacy and 2D workloads. The 44.1% margin in PassMark DirectX 10 and the 17.9% win in DirectX 9 show that older games and applications run significantly better on Turing. The 21.8% lead in PassMark G2D indicates superior 2D rendering performance, useful for desktop environments and certain productivity tasks. The 15.9% win in PassMark G3D suggests that the RTX 2080 remains competitive in general 3D rasterization, even if it falls behind in the most demanding modern scenarios.

The compute picture is nearly even. PassMark GPU Compute shows the RTX 2080 ahead by 1.8%, while Geekbench OpenCL shows the Arc B580 ahead by 1.7%. This suggests that for raw compute throughput, neither card holds a decisive edge, making the choice dependent on specific software optimizations. The DirectX 12 PassMark result, a 5.6% Intel win, is a minor data point that aligns with the Steel Nomad outcome.

The Verdict

The data points to a clear split based on workload, not a universal winner. For gamers playing current-generation titles with DirectX 12, the Intel Arc B580 is the superior choice. The 75.1% advantage in Steel Nomad DX12 is not a fluke; it reflects a fundamental architectural advantage in handling modern rendering techniques. The 12 GB memory capacity also future-proofs against increasing texture demands, a factor the RTX 2080’s 8 GB cannot match.

For users running older software, the NVIDIA RTX 2080 remains the better option. The 44.1% lead in DirectX 10 and the 17.9% win in DirectX 9 demonstrate that legacy API performance is significantly stronger on Turing. The 21.8% 2D advantage also matters for productivity-focused users. The RTX 2080’s tensor cores provide access to DLSS, a feature that could extend its lifespan in supported games, though the benchmark data does not quantify that advantage.

The overall benchmark parity, with the Arc B580’s average score of 23021 sitting just 0.6% above the RTX 2080’s 22895, means that neither card is objectively faster across the board. The Arc B580’s active production status and newer process node suggest better long-term driver support and efficiency, while the RTX 2080 is end-of-life. Ultimately, the choice hinges on whether the user prioritizes modern API performance or legacy compatibility.

FAQ

Q: How much faster is the Intel Arc B580 in 3DMark Steel Nomad DX12?

A: The Arc B580 scores 3068, which is 75.1% higher than the RTX 2080’s 1752 in that test.

Q: Which card wins in PassMark DirectX 10 and by how much?

A: The NVIDIA GeForce RTX 2080 wins with a score of 136 versus 76 for the Arc B580, a 44.1% margin.

Q: What are the average benchmark scores for both cards?

A: The Intel Arc B580 has an average score of 23021, while the NVIDIA RTX 2080 averages 22895, a 0.6% difference.

Q: Does the RTX 2080 have more memory than the Arc B580?

A: No, the Arc B580 has 12 GB of GDDR6, while the RTX 2080 has 8 GB, though the RTX 2080 uses a wider 256-bit bus.

Q: Which card has a higher transistor count?

A: The Intel Arc B580 has 19,600 million transistors, compared to 13,600 million for the RTX 2080, despite the Intel die being smaller at 272 mm² versus 545 mm².

Q: How do the power requirements compare?

A: The Arc B580 has a 190 W TDP with a single 8-pin connector, while the RTX 2080 has a 215 W TDP and requires both a 6-pin and an 8-pin connector.

Specification Differences

| Specification | Intel Arc B580 | NVIDIA GeForce RTX 2080 |

|---|---|---|

| Architecture | Xe2-HPG | Turing |

| Process Node | 5 nm | 12 nm |

| Transistors | 19,600 million | 13,600 million |

| Die Size | 272 mm² | 545 mm² |

| Transistor Density | 72.1M / mm² | 25.0M / mm² |

| Base Clock | 2670 MHz | 1515 MHz |

| Boost Clock | 2670 MHz | 1710 MHz |

| Memory Size | 12 GB | 8 GB |

| Memory Bus Width | 192 bit | 256 bit |

| Shading Units | 2560 | 2944 |

| TMUs | 160 | 184 |

| ROPs | 80 | 64 |

| RT Cores | 20 | 46 |

| Tensor Cores | Not specified | 368 |

| Pixel Rate | 213.6 GPixel/s | 109.4 GPixel/s |

| Texture Rate | 427.2 GTexel/s | 314.6 GTexel/s |

| FP32 | 13.67 TFLOPS | 10.07 TFLOPS |

| TDP | 190 W | 215 W |

| Power Connectors | 1x 8-pin | 1x 6-pin + 1x 8-pin |

| Suggested PSU | 450 W | 550 W |

| Bus Interface | PCIe 4.0 x8 | PCIe 3.0 x16 |

| Display Outputs | 1x HDMI 2.1a, 3x DisplayPort 2.1 | 1x HDMI 2.0, 3x DisplayPort 1.4a, 1x USB Type-C |

| Production Status | Active | End-of-life |

| Launch MSRP | 249 USD | 699 USD |