GPU Comparison

The Intel Arc B580 and AMD Radeon RX 6700S occupy different corners of the GPU landscape: one is a desktop-focused discrete card built on a new architecture, the other a mobile part from a previous generation. The benchmark data shows a clear overall winner in raw performance, but the RX 6700S holds its own in specific legacy and 2D workloads. This analysis breaks down the head-to-head results, architectural differences, and practical implications based solely on the provided metrics.

Head-to-Head Benchmarks

The most decisive victory for the Intel Arc B580 comes in the 3DMark Steel Nomad DX12 test, where it scores 3068 against the RX 6700S’s 1451. That is a massive 111.4% lead, more than doubling the AMD part’s output. This is the single largest delta in the entire comparison and signals a generational gap in modern DirectX 12 rendering performance.

The Arc B580 also dominates in compute-oriented workloads. In Geekbench OpenCL, it posts 92,821 points versus 70,965 for the RX 6700S, a 30.8% advantage. The gap widens further in Geekbench Vulkan, where the Intel card scores 109,672 compared to 76,150, a 44% lead. These results indicate that the B580’s architecture is substantially better equipped for heterogeneous compute and modern API workloads.

Passmark results tell a more nuanced story. In Passmark G3D (the overall 3D graphics score), the Arc B580 wins with 15,748 points against 15,066 for the RX 6700S, but the margin is only 4.5%. The Intel card also takes Passmark DirectX 12 with 76 versus 58, a 31% win, and Passmark GPU Compute with 7,729 versus 5,990, a 29% advantage.

However, the AMD Radeon RX 6700S wins four of the ten head-to-head tests. It takes Passmark DirectX 10 with 87 versus 76, a 12.6% lead, and Passmark DirectX 11 with 134 versus 128, a 4.5% margin. It also wins Passmark DirectX 9 (195 vs 183, a 6.2% edge) and Passmark G2D (739 vs 709, a 4.1% lead). These wins are concentrated in older API paths and 2D rasterization, suggesting the RX 6700S retains some compatibility and driver efficiency in legacy scenarios.

The average benchmark scores reinforce the overall picture: the Arc B580 averages 23,021 across all tests, while the RX 6700S averages 22,154. That is a 3.9% difference in favor of Intel, despite the B580’s massive wins in modern tests being offset by the AMD card’s narrower victories in older ones. The Arc B580’s percentile rank among all GPUs is 68, versus 67 for the RX 6700S, placing both cards in the same general performance tier despite their architectural differences.

Where Each One Wins

The Intel Arc B580 is the clear choice for modern, API-heavy workloads. Its 111.4% lead in 3DMark Steel Nomad DX12 is the headline result, indicating that games and applications built around DirectX 12 will see dramatically higher frame rates. The 44% lead in Geekbench Vulkan similarly points to strong performance in Vulkan-based titles and compute tasks. For users running current AAA games, ray-traced content, or GPU-accelerated compute, the B580’s 30.8% OpenCL and 29% Passmark Compute leads make it the superior option.

The AMD Radeon RX 6700S wins in legacy DirectX paths. Its 12.6% lead in DirectX 10 and 6.2% lead in DirectX 9 suggest that older games, particularly those from the early 2010s or before, may run better on the AMD part. The 4.5% win in DirectX 11 is notable, as many games still ship with DX11 renderers. The RX 6700S also edges out the B580 in Passmark G2D (4.1% lead), which measures 2D desktop and windowed performance. This makes the AMD card a reasonable choice for users with large libraries of older titles or who prioritize 2D application responsiveness.

The Passmark G3D result, where the B580 wins by only 4.5%, shows that the two cards are closer in general 3D performance than the modern API tests suggest. However, the trend is clear: the B580 wins where it matters most for future-proofing, while the RX 6700S holds onto its advantages in older software stacks.

Architecture Differences

The Intel Arc B580 is built on the Xe2-HPG architecture, part of the Battlemage generation (Arc 5). It uses a 5 nm process at TSMC, packing 19,600 million transistors into a 272 mm² die, for a transistor density of 72.1M per mm². The chip, codenamed BMG-G21, has a base and boost clock of 2670 MHz for both. Memory is 12 GB of GDDR6 on a 192-bit bus, yielding 456.0 GB/s of bandwidth.

The AMD Radeon RX 6700S belongs to the RDNA 2.0 architecture, part of the Navi Mobile generation (RX 6000M series). It uses a 7 nm process at TSMC, with 11,060 million transistors on a 237 mm² die, for a density of 46.7M per mm². The Navi 23 chip has a base clock of 1700 MHz, a boost of 2000 MHz, and a game clock of 1890 MHz. Memory is 8 GB of GDDR6 on a 128-bit bus, providing 224.0 GB/s of bandwidth.

The B580 has a significant resource advantage: 2560 shading units, 160 TMUs, and 80 ROPs, versus 1792 shading units, 112 TMUs, and 64 ROPs on the RX 6700S. The Intel card also has 20 ray tracing cores, while the AMD part has 28, though the architectural implementation differs. The B580’s pixel rate is 213.6 GPixel/s and texture rate is 427.2 GTexel/s, compared to 128.0 GPixel/s and 224.0 GTexel/s for the RX 6700S. FP32 compute is 13.67 TFLOPS for Intel versus 7.168 TFLOPS for AMD, and FP16 is 27.34 TFLOPS versus 14.34 TFLOPS, respectively.

Power and form factor differences are stark. The B580 is a dual-slot card with a 190 W TDP and a single 8-pin power connector, requiring a 450 W PSU. The RX 6700S is an IGP (integrated for mobile) with an 80 W TDP and no power connectors, making it a laptop-oriented part. The B580 uses PCIe 4.0 x8, same as the RX 6700S, but offers 1x HDMI 2.1a and 3x DisplayPort 2.1 outputs, while the AMD card’s outputs are "Portable Device Dependent."

FAQ

Q: Which GPU is faster in 3DMark Steel Nomad DX12?

A: The Intel Arc B580 scores 3068 versus 1451 for the AMD Radeon RX 6700S, giving the Intel card a 111.4% lead.

Q: Does the AMD Radeon RX 6700S win any benchmarks?

A: Yes, it wins four tests: Passmark DirectX 10 (87 vs 76), DirectX 11 (134 vs 128), DirectX 9 (195 vs 183), and Passmark G2D (739 vs 709).

Q: What is the memory capacity difference?

A: The Intel Arc B580 has 12 GB of GDDR6, while the AMD Radeon RX 6700S has 8 GB of GDDR6. The B580’s memory bandwidth is 456.0 GB/s versus 224.0 GB/s.

Q: How do the power requirements compare?

A: The Intel Arc B580 has a 190 W TDP and requires a 450 W PSU with a single 8-pin connector. The AMD Radeon RX 6700S has an 80 W TDP and uses no power connectors.

Q: What are the average benchmark scores for each?

A: The Intel Arc B580 averages 23,021 across all tests, and the AMD Radeon RX 6700S averages 22,154.

Q: Which GPU has a higher transistor count?

A: The Intel Arc B580 has 19,600 million transistors, compared to 11,060 million in the AMD Radeon RX 6700S.

The Verdict

The data points to the Intel Arc B580 as the superior performer for modern workloads. Its 111.4% lead in 3DMark Steel Nomad DX12 and 44% lead in Geekbench Vulkan are decisive, and its 30.8% OpenCL and 29% Passmark Compute leads make it the better choice for compute tasks. The B580 also offers more memory (12 GB vs 8 GB), higher bandwidth (456.0 GB/s vs 224.0 GB/s), and more shading units (2560 vs 1792). Its 68th percentile ranking versus 67th for the RX 6700S confirms its edge, even if the average score gap is only 3.9%.

The AMD Radeon RX 6700S is a mobile part with an 80 W TDP, making it suitable for laptops where power efficiency is paramount. Its wins in DirectX 9, 10, and 11, plus G2D, show that it retains some utility for older games and 2D applications. However, its 8 GB memory and 224.0 GB/s bandwidth are limiting factors, and its end-of-life production status means it is not a forward-looking choice.

Users building a desktop system with a 450 W PSU and looking for current-generation performance should pick the Intel Arc B580. Its launch MSRP is 249 USD. Laptop users constrained to an 80 W power envelope, or those with a heavy library of legacy DirectX titles, may find the RX 6700S acceptable, but the benchmark evidence shows the B580 is the stronger all-around GPU.