AMD Radeon 860M vs Intel Arc A570M Comparison

Where Each One Wins

The recorded data shows a single head-to-head comparison between the AMD Radeon 860M and the Intel Arc A570M, and the outcome is unambiguous. In the Geekbench OpenCL test, the Intel Arc A570M claims the only win, outperforming the AMD Radeon 860M by 60.9%. The Radeon 860M has zero wins in this matchup.

This one-sided result extends beyond the direct comparison. The Intel Arc A570M sits at the 88th percentile among all GPUs in the database, while the AMD Radeon 860M is at the 72nd percentile. That 16-percentile gap indicates the A570M occupies a distinctly higher performance tier in the overall distribution of recorded scores.

The AMD Radeon 860M does have a second benchmark entry, a Geekbench Vulkan score of 30043, but no corresponding Vulkan result exists for the Intel Arc A570M in the database. Without a matching test, that Vulkan figure cannot be used for direct comparison. The only common metric is OpenCL, and there the Intel part is decisively ahead.

Architecture Differences

The two GPUs come from different manufacturers and use fundamentally different architectures. The AMD Radeon 860M is built on RDNA 3.5, belongs to the Navi III IGP generation, and uses the Krackan Point chip. The Intel Arc A570M is built on Xe-HPG, belongs to the Alchemist generation under the Arc 5 Mobile family, and uses the DG2-256 chip.

Manufacturing processes differ substantially. The AMD part is produced on a 4 nm node at TSMC, while the Intel part uses a 6 nm node, also at TSMC. The Intel chip is physically larger, with a die size of 269 mm² and 11,500 million transistors, resulting in a transistor density of 42.8M per mm². The AMD part has unknown transistor count and die size in the database, so no direct density comparison is possible.

Compute resource allocation differs sharply. The Intel Arc A570M carries 2048 shading units, 128 texture mapping units, and 64 raster operation units. The AMD Radeon 860M has 512 shading units, 32 TMUs, and 16 ROPs. In every one of these categories, the Intel part has exactly four times the hardware. Ray tracing cores follow the same pattern: 16 on the Intel side versus 8 on the AMD side.

Clock behavior inverts the resource ratio. The AMD Radeon 860M has a base clock of 600 MHz and a boost clock of 3000 MHz. The Intel Arc A570M has a base clock of 900 MHz and a boost clock of 1300 MHz. The AMD part therefore has a much higher peak clock, but the Intel part compensates with its larger execution width.

Memory architecture is a fundamental separation point. The AMD Radeon 860M uses system shared memory, with type, bus width, and bandwidth all listed as system dependent. The Intel Arc A570M has dedicated 8 GB of GDDR6 memory on a 128-bit bus, delivering 224.0 GB/s of bandwidth. The Intel memory clock is 1750 MHz, with 14 Gbps effective data rate.

Power envelopes differ considerably. The AMD Radeon 860M is rated at 15 W TDP, while the Intel Arc A570M is rated at 75 W TDP. Both are listed as IGP slot width, meaning they are integrated into a host device rather than discrete add-in cards. Both use PCIe 4.0 x8 as the bus interface.

Head-to-Head Benchmarks

The only directly comparable benchmark is Geekbench OpenCL. The Intel Arc A570M scores 58239, while the AMD Radeon 860M scores 22759. The delta is -60.9% from the AMD side, meaning the Radeon 860M delivers less than 40% of the Intel part's OpenCL score. To frame it another way, the Intel Arc A570M is roughly 2.56 times faster in this single test.

That margin is substantial enough to place the two parts in different performance classes. The Intel Arc A570M's average benchmark score of 58239 places it alongside the AMD Radeon RX 6950 XT, which has an average score of 58392, a 0.3% difference in Intel's favor. The closest rival on the Intel side, the AMD Radeon RX 5600 OEM, scores 58085, which is 0.3% behind the A570M. The NVIDIA P102-100 and AMD Radeon PRO V710 trail by 0.5% and 0.7%, respectively.

The AMD Radeon 860M's average benchmark score of 26401 puts it in a much lower performance tier. Its nearest rivals cluster tightly around it: the NVIDIA GeForce MX550 scores 26421, which is 0.1% ahead of the 860M. The AMD Radeon RX 5700 XT 50th Anniversary scores 26553, 0.6% ahead. The NVIDIA RTX A4000 scores 26683, 1.1% ahead. The NVIDIA GeForce RTX 5060 scores 26331, which is 0.3% behind the 860M. These deltas are all within roughly one percent, indicating the Radeon 860M sits in a crowded band of near-identical average scores.

The gap between the two parts in this comparison is therefore not a marginal edge but a tier-level separation. The Intel Arc A570M's average score is more than double the AMD Radeon 860M's average score.

The Verdict

The database indicates a clear performance hierarchy. The Intel Arc A570M is the faster GPU in the one benchmark where both have recorded results. Its OpenCL score of 58239 versus 22759 for the AMD Radeon 860M is a decisive margin, and its 88th percentile standing versus the 72nd percentile for the AMD part reinforces that ordering.

The AMD Radeon 860M does have advantages in efficiency and integration. Its 15 W TDP is one-fifth of the Intel part's 75 W TDP. It also uses system shared memory, which simplifies the host design by eliminating the need for dedicated VRAM. Its 4 nm process node is more advanced than the 6 nm node used by the Intel chip. For a portable device where power draw and physical footprint are primary constraints, the Radeon 860M is the more economical choice.

The Intel Arc A570M wins on raw compute. It has four times the shading units, TMUs, and ROPs, plus twice the ray tracing cores. Its dedicated 8 GB GDDR6 memory with 224.0 GB/s bandwidth removes the dependency on system memory performance. Its pixel rate of 83.20 GPixel/s and texture rate of 166.4 GTexel/s are well above the 48.00 GPixel/s and 96.00 GTexel/s of the AMD part.

Users who need maximum compute throughput in OpenCL workloads should select the Intel Arc A570M. Users who prioritize low power consumption and a simpler memory setup may prefer the AMD Radeon 860M, accepting its lower performance ceiling.

FAQ

Q: Which GPU is faster in OpenCL benchmarks?

A: The Intel Arc A570M scores 58239 in Geekbench OpenCL, while the AMD Radeon 860M scores 22759. The Intel part is 60.9% ahead.

Q: How do their power requirements compare?

A: The AMD Radeon 860M has a 15 W TDP, while the Intel Arc A570M has a 75 W TDP. The AMD part draws considerably less power.

Q: What memory configurations do they use?

A: The AMD Radeon 860M uses system shared memory with system dependent bandwidth. The Intel Arc A570M has 8 GB of GDDR6 on a 128-bit bus with 224.0 GB/s bandwidth.

Q: How do their compute resources differ?

A: The Intel Arc A570M has 2048 shading units, 128 TMUs, 64 ROPs, and 16 ray tracing cores. The AMD Radeon 860M has 512 shading units, 32 TMUs, 16 ROPs, and 8 ray tracing cores.

Q: What process nodes are used?

A: The AMD Radeon 860M is fabricated on a 4 nm node at TSMC. The Intel Arc A570M is fabricated on a 6 nm node, also at TSMC.

Q: Which GPU has a higher percentile ranking?

A: The Intel Arc A570M ranks at the 88th percentile among all GPUs, while the AMD Radeon 860M ranks at the 72nd percentile.

Specification Differences

| Specification | AMD Radeon 860M | Intel Arc A570M |

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

| Architecture | RDNA 3.5 | Xe-HPG |

| Generation | Navi III IGP (Strix Point Mobile) | Alchemist (Arc 5 Mobile) |

| Process Node | 4 nm | 6 nm |

| Transistors | Unknown | 11,500 million |

| Die Size | Unknown | 269 mm² |

| Transistor Density | N/A | 42.8M / mm² |

| Base Clock | 600 MHz | 900 MHz |

| Boost Clock | 3000 MHz | 1300 MHz |

| Memory Size | System Shared | 8 GB |

| Memory Type | System Shared | GDDR6 |

| Memory Bus Width | System Shared | 128 bit |

| Memory Bandwidth | System Dependent | 224.0 GB/s |

| Shading Units | 512 | 2048 |

| TMUs | 32 | 128 |

| ROPs | 16 | 64 |

| Ray Tracing Cores | 8 | 16 |

| Pixel Rate | 48.00 GPixel/s | 83.20 GPixel/s |

| Texture Rate | 96.00 GTexel/s | 166.4 GTexel/s |

| FP32 Performance | 3.072 TFLOPS | 5.325 TFLOPS |

| FP16 Performance | 3.072 TFLOPS (1:1) | 10.65 TFLOPS (2:1) |

| TDP | 15 W | 75 W |

| Release Date | 2025-02-28 | 2023-07-31 |