GPU Comparison

The AMD Radeon RX 590 and Intel Arc A350M occupy the same performance tier in the aggregate benchmark data, with average scores of 24,744 and 24,647 respectively, a difference of just 0.4% in favor of the RX 590. Both cards sit at the 70th percentile among all GPUs, placing them in a cluster that also includes the NVIDIA RTX A5000 Mobile (24,763) and the AMD Radeon RX 6600 XT (24,442). The data shows two fundamentally different designs, a 2018 desktop card built for raw rasterization and a 2022 mobile chip with modern API support, converging on nearly identical overall performance.

Where Each One Wins

The RX 590 wins in raw compute throughput and traditional rendering workloads. Its FP32 performance of 7.119 TFLOPS is more than double the Arc A350M's 3.379 TFLOPS, and its texture rate of 222.5 GTexel/s dwarfs the Intel part's 105.6 GTexel/s. The AMD card also holds a decisive memory bandwidth advantage: 256.0 GB/s over a 256-bit bus versus 112.0 GB/s over a 64-bit bus. These figures point to a clear edge in fill-rate-bound scenarios and high-resolution texture streaming. The RX 590's 8 GB of GDDR5 memory provides twice the capacity of the Arc A350M's 4 GB GDDR6, which matters for modern game assets and large framebuffers.

The Arc A350M wins in architectural efficiency and feature completeness. Its 25 W TDP is a fraction of the RX 590's 175 W, making it suitable for thin-and-light laptops where the AMD card's dual-slot cooler and 1x 8-pin power connector would be impractical. The Intel chip supports DirectX 12 Ultimate (12_2), which includes hardware ray tracing via its 6 RT cores, a feature entirely absent from the RX 590's DirectX 12 (12_0) feature set. The Arc A350M also leads in pixel rate at 52.80 GPixel/s versus 49.44 GPixel/s, and its boost clock of 2200 MHz is substantially higher than the RX 590's 1545 MHz, suggesting better per-clock efficiency.

Architecture Differences

The two GPUs come from different foundries and process nodes. The RX 590 uses GlobalFoundries' 12 nm process with Polaris 30 silicon, measuring 232 mm² with 5,700 million transistors for a density of 24.6M transistors per mm². The Arc A350M uses TSMC's 6 nm process with DG2-128 silicon, measuring 157 mm² with 7,200 million transistors for a density of 45.9M per mm². The Intel chip crams more transistors into a smaller die, reflecting a newer design methodology.

The compute architectures diverge sharply. The RX 590 employs GCN 4.0 with 2,304 shading units, 144 TMUs, and 32 ROPs. The Arc A350M uses Xe-HPG with only 768 shading units, 48 TMUs, and 24 ROPs, but compensates with 6 dedicated RT cores and a 2:1 FP16 ratio (6.758 TFLOPS) versus the AMD card's 1:1 FP16 (7.119 TFLOPS). Memory subsystems also differ: the RX 590 uses GDDR5 at 2000 MHz (8 Gbps effective), while the Arc A350M uses GDDR6 at 1750 MHz (14 Gbps effective). The AMD card's wider 256-bit bus gives it over twice the bandwidth, but the Intel part's newer memory type operates at higher effective speeds per pin.

The interface and connectivity reflect their different markets. The RX 590 is a desktop card with PCIe 3.0 x16, 1x HDMI 2.0b and 3x DisplayPort 1.4a outputs, and a 241 mm dual-slot length. The Arc A350M is an integrated mobile GPU with PCIe 4.0 x8 and "Portable Device Dependent" display outputs, meaning the laptop manufacturer determines the actual ports. The RX 590 supports Vulkan 1.3, while the Arc A350M supports Vulkan 1.4, indicating a newer driver stack on the Intel side.

Head-to-Head Benchmarks

The aggregate benchmark scores show near-identical overall performance. The RX 590's average score of 24,744 places it 0.4% ahead of the Arc A350M's 24,647. In the nearest rival comparison, the RX 590 is 0.4% ahead of the Arc A350M, while the Arc A350M is 0.4% behind the RX 590, the same delta viewed from opposite directions. Both cards trail the NVIDIA RTX A5000 Mobile by only 0.1% to 0.5%, and both lead the AMD Radeon RX 6600 XT by 0.8% to 1.2% and the NVIDIA GeForce GTX 1630 by 1.5% to 1.9%.

Individual benchmark results reveal different strengths. The RX 590 scores 1,081 in 3DMark Steel Nomad DX12, a modern rasterization test that stresses raw shader throughput. The Arc A350M has no score in this test, but its Geekbench OpenCL score of 24,546 and Vulkan score of 24,747 demonstrate strong compute and API-level performance. The RX 590's Geekbench Metal score of 48,407 shows its capability in Apple's Metal API, though the Arc A350M lacks a comparable Metal result. The data suggests the RX 590 excels in classic DX12 rasterization, while the Arc A350M holds its own in cross-platform compute workloads.

The largest numerical gap in the data is in FP32 throughput: the RX 590's 7.119 TFLOPS is 110.6% higher than the Arc A350M's 3.379 TFLOPS. Texture rate shows a similar story at 222.5 GTexel/s versus 105.6 GTexel/s. However, the Arc A350M counters with a 2200 MHz boost clock versus 1545 MHz, a 52.80 GPixel/s pixel rate versus 49.44 GPixel/s, and a 6 RT core count versus zero. These architectural advantages do not translate into a higher average benchmark score, indicating that the RX 590's brute-force compute wins where it matters most for overall performance.

FAQ

Q: Which GPU has higher average benchmark scores?

A: The AMD Radeon RX 590 has an average benchmark score of 24,744, which is 0.4% higher than the Intel Arc A350M's 24,647.

Q: Does the Intel Arc A350M support ray tracing?

A: Yes, the Arc A350M includes 6 RT cores and supports DirectX 12 Ultimate (12_2), whereas the RX 590 has no RT cores and only supports DirectX 12 (12_0).

Q: How do the memory bandwidths compare?

A: The RX 590 has 256.0 GB/s bandwidth over a 256-bit bus with 8 GB GDDR5, while the Arc A350M has 112.0 GB/s over a 64-bit bus with 4 GB GDDR6. The AMD card has over twice the bandwidth and double the capacity.

Q: What is the power consumption difference?

A: The RX 590 is rated at 175 W TDP with a suggested 450 W PSU, while the Arc A350M is rated at 25 W TDP with no PSU requirement listed, reflecting its mobile integrated design.

Q: Which GPU has a higher transistor density?

A: The Arc A350M has a density of 45.9M transistors per mm² on TSMC's 6 nm process, compared to the RX 590's 24.6M per mm² on GlobalFoundries' 12 nm process.

Q: Are both GPUs still in production?

A: No, both are listed as end-of-life. The RX 590 was released on 2018-11-14, and the Arc A350M on 2022-03-29.

Specification Differences

| Field | AMD Radeon RX 590 | Intel Arc A350M |

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

| Chip | Polaris 30 | DG2-128 |

| Architecture | GCN 4.0 | Xe-HPG |

| Process Node | 12 nm | 6 nm |

| Foundry | GlobalFoundries | TSMC |

| Transistors | 5,700 million | 7,200 million |

| Die Size | 232 mm² | 157 mm² |

| Transistor Density | 24.6M / mm² | 45.9M / mm² |

| Base Clock | 1469 MHz | 1150 MHz |

| Boost Clock | 1545 MHz | 2200 MHz |

| Memory Clock | 2000 MHz (8 Gbps effective) | 1750 MHz (14 Gbps effective) |

| Memory Size | 8 GB | 4 GB |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bus | 256 bit | 64 bit |

| Memory Bandwidth | 256.0 GB/s | 112.0 GB/s |

| Shading Units | 2304 | 768 |

| TMUs | 144 | 48 |

| ROPs | 32 | 24 |

| RT Cores | None | 6 |

| Pixel Rate | 49.44 GPixel/s | 52.80 GPixel/s |

| Texture Rate | 222.5 GTexel/s | 105.6 GTexel/s |

| FP32 | 7.119 TFLOPS | 3.379 TFLOPS |

| FP16 | 7.119 TFLOPS (1:1) | 6.758 TFLOPS (2:1) |

| TDP | 175 W | 25 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 1x 8-pin | None |

| Suggested PSU | 450 W | None |

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

| Display Outputs | 1x HDMI 2.0b, 3x DisplayPort 1.4a | Portable Device Dependent |

| DirectX | 12 (12_0) | 12 Ultimate (12_2) |

| Vulkan | 1.3 | 1.4 |

| Release Date | 2018-11-14 | 2022-03-29 |

| Launch MSRP | 279 USD | None |

The Verdict

The data supports a clear split by use case. Choose the AMD Radeon RX 590 if you need maximum raw compute and memory bandwidth in a desktop system. Its 7.119 TFLOPS FP32, 256.0 GB/s bandwidth, and 8 GB VRAM make it the stronger choice for traditional rasterization workloads and texture-heavy applications. The 3DMark Steel Nomad DX12 score of 1,081 demonstrates its capability in modern DX12 titles, and its 0.4% aggregate lead over the Arc A350M confirms this edge. The 279 USD launch MSRP also positions it as a conventional desktop add-in card with standard display outputs.

Choose the Intel Arc A350M if efficiency and modern features matter more than raw throughput. Its 25 W TDP enables mobile integration where the RX 590's 175 W dual-slot design cannot fit. The DirectX 12 Ultimate support, 6 RT cores, and Vulkan 1.4 compatibility provide a forward-looking feature set that the RX 590 lacks. The 2200 MHz boost clock and 52.80 GPixel/s pixel rate indicate strong per-clock efficiency, and its 45.9M transistors per mm² density reflects a more advanced 6 nm manufacturing process. The near-identical average benchmark score of 24,647 means the feature and efficiency advantages come at no significant performance cost in aggregate.

The verdict from the data is that neither GPU decisively outperforms the other. The 0.4% difference in average scores is within the margin of error, and both cards rank at the same 70th percentile. The RX 590 is the pick for desktop builders prioritizing compute and memory bandwidth. The Arc A350M is the pick for portable systems prioritizing power efficiency and modern API support. Both are end-of-life products, so the decision rests on platform compatibility and workload requirements rather than future upgrade paths.