AMD Radeon RX 590 vs Intel Arc B580 Comparison

The AMD Radeon RX 590 and Intel Arc B580 are separated by six years of GPU architecture evolution, and the benchmark data reflects a decisive generational shift. In the only directly comparable head-to-head test, the Intel Arc B580 delivers a 3DMark Steel Nomad DX12 score of 3068, obliterating the RX 590’s 1081 by a 64.8% margin. While the RX 590 holds a slightly higher overall percentile ranking (70th vs. 68th), that statistic is buoyed by a legacy Geekbench Metal score of 48,407. The Arc B580’s average benchmark score of 23,021 sits just 1.9% below the RX 590’s 24,744, but the newer card’s raw performance ceiling in modern workloads is dramatically higher, as shown by its 13.67 TFLOPS FP32 throughput versus the RX 590’s 7.119 TFLOPS.

Head-to-Head Benchmarks

The sole direct comparison available is the 3DMark Steel Nomad DX12 test, and it is not close. The Intel Arc B580 scores 3068, while the AMD Radeon RX 590 trails at 1081. That translates to a 64.8% deficit for the older AMD card, meaning the Arc B580 achieves nearly three times the frame complexity in this DirectX 12 workload. This single result encapsulates the architectural gap: the RX 590’s GCN 4.0 design, with 2,304 shading units and a 256-bit memory bus, cannot compete with the Arc B580’s Xe2-HPG architecture, which pairs 2,560 shading units with a 192-bit bus but far faster GDDR6 memory.

Beyond the direct test, the aggregate benchmark scores tell a more nuanced story. The RX 590’s average score of 24,744 is actually 7.5% higher than the Arc B580’s 23,021, a counterintuitive result driven by the inclusion of the RX 590’s Geekbench Metal result. In that specific API test, the RX 590 scores 48,407, which is a strong showing for a 2018 card. However, the Arc B580 counters with a Geekbench Vulkan score of 109,672 and an OpenCL score of 92,821 — both figures that dwarf anything the RX 590 can produce in comparable compute workloads. The PassMark suite further illustrates the B580’s modern strengths: it scores 15,748 in G3D, 7,729 in GPU Compute, and 709 in G2D, while the RX 590 has no equivalent PassMark entries in the data. The B580’s PassMark DX9 score of 183 and DX11 score of 128 show it handles legacy APIs well, though its DX10 and DX12 scores drop to 76 apiece — a quirk of that benchmark’s methodology rather than a reflection of raw capability.

Where Each One Wins

The Intel Arc B580 wins in every scenario involving modern graphics APIs or compute-heavy tasks. Its 3DMark Steel Nomad DX12 victory is the headline, but the compute data is equally lopsided. The B580’s 13.67 TFLOPS FP32 and 27.34 TFLOPS FP16 (at a 2:1 ratio) performance are roughly double the RX 590’s 7.119 TFLOPS in both precision formats. The B580 also brings dedicated ray tracing hardware with 20 RT cores, a feature entirely absent from the RX 590. Memory bandwidth is another decisive win: the B580’s 456.0 GB/s over a 192-bit bus dwarfs the RX 590’s 256.0 GB/s over a 256-bit bus, thanks to GDDR6 memory running at 19 Gbps effective versus GDDR5 at 8 Gbps effective. The B580 also supports PCIe 4.0 x8, doubling the per-lane bandwidth of the RX 590’s PCIe 3.0 x16 connection.

The AMD Radeon RX 590 retains a few specific advantages, though they are narrow. Its 70th percentile ranking versus the B580’s 68th suggests that in the broader database of all GPUs, the RX 590 holds its own better than the B580 does — likely due to the Metal benchmark inclusion. The RX 590 also offers a 256-bit memory bus, which provides more physical lanes for data movement, though the B580’s higher effective bandwidth makes this a moot point in practice. In terms of nearest rivals, the RX 590’s closest competitor is the NVIDIA RTX A5000 Mobile (0.1% behind), while the B580’s closest is the AMD Radeon RX 580 2048SP (0.2% behind) — the B580 also edges out the NVIDIA GeForce RTX 2080 by 0.6% and the RTX 3080 by 0.7% in average score, a remarkable result for a card in this class. The RX 590, by contrast, only leads its nearest rivals by margins of 0.4% to 1.9%.

Architecture Differences

The architectural chasm between these two GPUs is vast. The RX 590 uses AMD’s GCN 4.0 architecture on a 12 nm GlobalFoundries process, packing 5,700 million transistors into a 232 mm² die with a transistor density of 24.6 million per mm². The Arc B580 uses Intel’s Xe2-HPG architecture on a 5 nm TSMC process, integrating 19,600 million transistors into a 272 mm² die — a density of 72.1 million per mm², nearly three times higher. This density advantage allows Intel to fit 2,560 shading units, 160 TMUs, and 80 ROPs, versus the RX 590’s 2,304 shading units, 144 TMUs, and 32 ROPs. The ROP count difference is particularly stark: 80 versus 32 explains the B580’s 213.6 GPixel/s pixel rate versus the RX 590’s 49.44 GPixel/s.

Clock speeds also diverge significantly. The RX 590 runs at a base of 1469 MHz and boosts to 1545 MHz, while the Arc B580 operates at a flat 2670 MHz for both base and boost — a 73% higher clock speed. The B580’s memory runs at 2375 MHz (19 Gbps effective) across a 192-bit bus, yielding 456.0 GB/s; the RX 590’s GDDR5 memory runs at 2000 MHz (8 Gbps effective) across a 256-bit bus, yielding 256.0 GB/s. The B580 supports DirectX 12 Ultimate (12_2), Vulkan 1.4, and has 20 RT cores; the RX 590 is limited to DirectX 12 (12_0), Vulkan 1.3, and has no ray tracing hardware. The B580 also has a 2:1 FP16 ratio, enabling 27.34 TFLOPS half-precision compute, while the RX 590 is locked at 1:1 (7.119 TFLOPS FP16). Display outputs favor the B580 with HDMI 2.1a and DisplayPort 2.1, versus the RX 590’s HDMI 2.0b and DisplayPort 1.4a. Power consumption is similar — 190 W for the B580 versus 175 W for the RX 590 — and both use a single 8-pin connector with a 450 W suggested PSU. The B580 is physically larger at 272 mm long, 115 mm tall, and 45 mm wide, versus the RX 590’s 241 mm length.

The Verdict

The data is unambiguous: the Intel Arc B580 is the superior GPU in nearly every measurable category. Its 64.8% lead in the 3DMark Steel Nomad DX12 test is the single most important benchmark result, as it represents a modern, demanding workload that reflects current gaming expectations. The B580 doubles the RX 590’s FP32 throughput (13.67 vs. 7.119 TFLOPS), quadruples its pixel rate (213.6 vs. 49.44 GPixel/s), and provides 78% more memory bandwidth (456.0 vs. 256.0 GB/s). It also adds ray tracing, supports DirectX 12 Ultimate, and offers a 12 GB frame buffer versus 8 GB — a 50% capacity increase that matters for high-resolution textures.

The AMD Radeon RX 590 should only be considered by users with legacy software dependencies or those specifically targeting Metal-based workloads, where its 48,407 Geekbench Metal score remains respectable. Its 70th percentile ranking and higher average benchmark score (24,744 vs. 23,021) are statistical artifacts of benchmark selection, not evidence of real-world superiority. The B580’s rivals list includes the NVIDIA GeForce RTX 2080 and RTX 3080, against which it trades blows within 0.7% — a stunning feat for a card that launched at a lower MSRP. The RX 590’s nearest rivals are mobile workstation parts and budget desktop cards, placing it in a fundamentally lower performance tier. For anyone building a new system or upgrading from a pre-2020 GPU, the Arc B580 is the clear choice. The RX 590 is end-of-life, while the B580 is an active product with forward-looking API support.

FAQ

Q: How much faster is the Intel Arc B580 than the AMD Radeon RX 590 in DirectX 12?

A: In the 3DMark Steel Nomad DX12 benchmark, the Arc B580 scores 3068 versus the RX 590’s 1081, giving the B580 a 64.8% advantage.

Q: Does the AMD Radeon RX 590 have a higher benchmark ranking than the Intel Arc B580?

A: Yes. The RX 590 ranks in the 70th percentile of all GPUs, while the Arc B580 ranks in the 68th percentile. However, the RX 590’s average benchmark score of 24,744 is higher than the B580’s 23,021.

Q: What memory advantages does the Intel Arc B580 have over the AMD Radeon RX 590?

A: The B580 has 12 GB of GDDR6 memory on a 192-bit bus with 456.0 GB/s bandwidth, while the RX 590 has 8 GB of GDDR5 on a 256-bit bus with 256.0 GB/s bandwidth. The B580’s memory runs at 19 Gbps effective versus 8 Gbps effective.

Q: Which GPU supports ray tracing?

A: The Intel Arc B580 has 20 dedicated RT cores and supports DirectX 12 Ultimate (12_2). The AMD Radeon RX 590 has no RT cores and is limited to DirectX 12 (12_0).

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

A: The Arc B580 delivers 13.67 TFLOPS FP32 and 27.34 TFLOPS FP16 (2:1 ratio). The RX 590 delivers 7.119 TFLOPS FP32 and 7.119 TFLOPS FP16 (1:1 ratio). The B580 also scores 92,821 in Geekbench OpenCL and 109,672 in Geekbench Vulkan.

Q: What are the nearest rivals for each card based on average benchmark score?

A: The RX 590’s closest rival is the NVIDIA RTX A5000 Mobile (0.1% behind), followed by the Intel Arc A350M (0.4% ahead). The Arc B580’s closest rival is the AMD Radeon RX 580 2048SP (0.2% ahead), with the NVIDIA GeForce RTX 2080 (0.6% behind) and RTX 3080 (0.7% ahead) nearby.