AMD Radeon Instinct MI60 vs Intel Arc A570M Comparison

Head-to-Head Benchmarks

The recorded data contains a single direct comparison between these two accelerators: the Geekbench OpenCL workload. In this test, the AMD Radeon Instinct MI60 posts a score of 92,488, while the Intel Arc A570M records 58,239. The AMD part wins this head-to-head by a substantial 58.8% margin.

This is the only benchmark where both products have overlapping measurements, so the verdict in this category is unambiguous. The MI60’s score places it at the 93rd percentile among all GPUs in the database, whereas the Arc A570M sits at the 88th percentile. That percentile gap, while narrower than the raw score gap, still reflects a meaningful tier difference in overall compute performance.

Looking at the MI60’s nearest rivals, it edges out the NVIDIA RTX A4500 by 0.9% (91,671 average) and the RTX A4500 Mobile by 1.5% (91,134). It trails the AMD Radeon Pro VII by 4.8% (97,131) and the Radeon RX 7900M by 5.2% (97,487). These are close margins, suggesting the MI60 sits in a competitive band of high-end workstation and datacenter cards.

The Arc A570M’s nearest rivals tell a different story. Its 58,239 average score is nearly identical to the AMD Radeon RX 6950 XT, which scores 58,392, a difference of only 0.3% in favor of the RX card. It also sits within 0.3% of the Radeon RX 5600 OEM (58,085) and within 0.5% of the NVIDIA P102-100 (58,528). The AMD Radeon PRO V710 leads it by 0.7% (58,657). The Arc A570M is essentially at parity with a cluster of mid-range desktop and mobile GPUs, making its 58.8% deficit to the MI60 a chasm in absolute performance terms.

In short, the data shows a clear hierarchy: the MI60 operates in a performance class roughly 58% above the Arc A570M in the one workload where both are measured. No other direct comparison exists in the database, so the head-to-head picture is defined entirely by this single OpenCL result.

Where Each One Wins

The AMD Radeon Instinct MI60 wins the only available head-to-head benchmark, the Geekbench OpenCL test, with a 58.8% advantage. That is the sole recorded win for either product in direct comparison, giving the MI60 a 1-0 win count.

Beyond the head-to-head, the MI60 also holds advantages in several specification-driven categories that are not directly benchmarked but are relevant to workload suitability. Its FP32 throughput of 14.75 TFLOPS versus the Arc A570M’s 5.325 TFLOPS indicates a 2.77x advantage in single-precision compute peak. The FP16 figures follow the same pattern: 29.49 TFLOPS for the MI60 versus 10.65 TFLOPS for the Arc, a 2.77x lead. Texture rate also favors the MI60 heavily: 460.8 GTexel/s versus 166.4 GTexel/s. Pixel rate is less lopsided but still favors AMD: 115.2 GPixel/s versus 83.20 GPixel/s.

The Intel Arc A570M, meanwhile, has no recorded benchmark wins against the MI60. Its strengths lie in efficiency and portability rather than raw performance. The 75 W TDP is one-quarter of the MI60’s 300 W, and the Arc is an IGP form factor, meaning it is designed for mobile integration, while the MI60 is a dual-slot add-in card. The Arc also supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the MI60 is limited to DirectX 12 (12_1) and Vulkan 1.3. For workloads that rely on the latest graphics API features, such as ray tracing (the Arc has 16 dedicated RT cores), Intel’s part is the more modern option despite its lower compute ceiling.

In terms of raw compute density per watt, the Arc A570M delivers 5.325 TFLOPS at 75 W, or roughly 71 GFLOPS per watt. The MI60 delivers 14.75 TFLOPS at 300 W, or roughly 49 GFLOPS per watt. The Intel part is more power-efficient in this arithmetic, though neither figure is directly benchmarked.

Architecture Differences

The AMD Radeon Instinct MI60 is built on the Vega 20 chip using the GCN 5.1 architecture, manufactured on a 7 nm process at TSMC. It packs 13,230 million transistors onto a 331 mm² die, yielding a transistor density of 40.0 million per mm². The Intel Arc A570M uses the DG2-256 chip with the Xe-HPG architecture, fabricated on a 6 nm process, also at TSMC. Its die contains 11,500 million transistors across 269 mm², for a density of 42.8 million per mm². Intel’s process node is denser by 2.8 million transistors per mm², despite having fewer total transistors.

Core configurations differ substantially. The MI60 has 4,096 shading units, 256 texture mapping units, and 64 ROPs. The Arc A570M has 2,048 shading units, 128 TMUs, and 64 ROPs. The AMD part has exactly double the shader and texture resources, while ROP counts match. The Arc adds 16 RT cores, which the MI60 lacks entirely. Neither product lists tensor cores.

Memory architecture is a major divergence. The MI60 uses 32 GB of HBM2 on a 4096-bit bus, delivering 1.02 TB/s of bandwidth. Memory clock is 1000 MHz, or 2 Gbps effective. The Arc A570M uses 8 GB of GDDR6 on a 128-bit bus, delivering 224.0 GB/s, with a memory clock of 1750 MHz (14 Gbps effective). The MI60 has 4x the capacity and 4.55x the bandwidth, a decisive advantage for memory-bound workloads.

Clock speeds differ significantly. The MI60 has a 1200 MHz base and 1800 MHz boost. The Arc A570M has a 900 MHz base and 1300 MHz boost. The AMD part runs 33% higher at base and 38% higher at boost. Despite the higher clocks, the MI60’s TDP is 300 W versus 75 W for the Arc.

Bus interfaces also differ: the MI60 uses PCIe 4.0 x16, while the Arc A570M uses PCIe 4.0 x8. The MI60 has a dual-slot form factor with 1x mini-DisplayPort 1.4a output, while the Arc is an IGP with portable-device-dependent outputs. The MI60 requires 1x 6-pin and 1x 8-pin power connectors and a 700 W suggested PSU; the Arc lists no power connectors or PSU recommendation.

API support favors Intel for DirectX: the Arc supports 12 Ultimate (12_2), while the MI60 is capped at 12 (12_1). Both support OpenGL 4.6. Vulkan support differs: the Arc lists 1.4, the MI60 lists 1.3.

FAQ

Q: Which GPU has higher raw compute performance?

A: The AMD Radeon Instinct MI60. Its FP32 throughput is 14.75 TFLOPS versus 5.325 TFLOPS for the Intel Arc A570M, a 2.77x advantage. In the Geekbench OpenCL test, the MI60 scores 92,488 versus 58,239 for the Arc, a 58.8% lead.

Q: Does the Intel Arc A570M support ray tracing?

A: Yes. The Arc A570M has 16 dedicated RT cores and supports DirectX 12 Ultimate (12_2). The MI60 has no RT cores and only supports DirectX 12 (12_1).

Q: How much memory does each GPU have, and what type?

A: The MI60 has 32 GB of HBM2 on a 4096-bit bus with 1.02 TB/s bandwidth. The Arc A570M has 8 GB of GDDR6 on a 128-bit bus with 224.0 GB/s bandwidth. The MI60 offers 4x the capacity and 4.55x the bandwidth.

Q: Which GPU is more power-efficient?

A: Based on the recorded specs, the Intel Arc A570M. It has a 75 W TDP and delivers 5.325 TFLOPS FP32, while the MI60 has a 300 W TDP for 14.75 TFLOPS. The Arc delivers roughly 71 GFLOPS per watt versus roughly 49 for the MI60.

Q: What are the production statuses of these two GPUs?

A: The AMD Radeon Instinct MI60 is end-of-life, released in November 2018. The Intel Arc A570M is active and was released in July 2023.

Q: How does the MI60 compare to its closest rivals in the database?

A: The MI60’s 92,466 average benchmark score is 0.9% higher than the NVIDIA RTX A4500 (91,671) and 1.5% higher than the RTX A4500 Mobile (91,134). It is 4.8% lower than the AMD Radeon Pro VII (97,131) and 5.2% lower than the AMD Radeon RX 7900M (97,487).

Specification Differences

| Specification | AMD Radeon Instinct MI60 | Intel Arc A570M |

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

| Architecture | GCN 5.1 | Xe-HPG |

| Process Node | 7 nm | 6 nm |

| Transistors | 13,230 million | 11,500 million |

| Die Size | 331 mm² | 269 mm² |

| Transistor Density | 40.0M / mm² | 42.8M / mm² |

| Base Clock | 1200 MHz | 900 MHz |

| Boost Clock | 1800 MHz | 1300 MHz |

| Memory Clock | 1000 MHz (2 Gbps effective) | 1750 MHz (14 Gbps effective) |

| Memory Size | 32 GB | 8 GB |

| Memory Type | HBM2 | GDDR6 |

| Memory Bus Width | 4096 bit | 128 bit |

| Memory Bandwidth | 1.02 TB/s | 224.0 GB/s |

| Shading Units | 4096 | 2048 |

| TMUs | 256 | 128 |

| ROPs | 64 | 64 |

| RT Cores | None | 16 |

| Pixel Rate | 115.2 GPixel/s | 83.20 GPixel/s |

| Texture Rate | 460.8 GTexel/s | 166.4 GTexel/s |

| FP32 | 14.75 TFLOPS | 5.325 TFLOPS |

| FP16 | 29.49 TFLOPS (2:1) | 10.65 TFLOPS (2:1) |

| TDP | 300 W | 75 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 1x 6-pin + 1x 8-pin | None |

| Suggested PSU | 700 W | None |

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

| Display Outputs | 1x mini-DisplayPort 1.4a | Portable Device Dependent |

| DirectX | 12 (12_1) | 12 Ultimate (12_2) |

| Vulkan | 1.3 | 1.4 |

| Dimensions | 267 mm (10.5 inches) long, 111 mm (4.4 inches) high | Not recorded |

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

| Release Date | November 2018 | July 2023 |

The Verdict

The AMD Radeon Instinct MI60 is the clear performance leader in every measured category. Its 58.8% OpenCL score advantage over the Arc A570M is decisive, and the underlying specifications reinforce this: double the shading units, double the TMUs, 2.77x the FP32 throughput, 4x the memory capacity, and 4.55x the memory bandwidth. The MI60 is also the only one of the two with HBM2 memory, a 4096-bit bus, and a PCIe 4.0 x16 interface. For compute-heavy datacenter or workstation workloads where raw throughput and memory bandwidth dominate, the MI60 is the correct choice, provided the 300 W power draw and dual-slot form factor are acceptable.

The Intel Arc A570M is the better option for scenarios where power efficiency, modern API support, and ray tracing matter. Its 75 W TDP is a fraction of the MI60’s 300 W, and it fits as an IGP in portable devices. The Arc supports DirectX 12 Ultimate and Vulkan 1.4, plus 16 RT cores, making it suitable for the latest graphics features on mobile platforms. Its performance, however, is firmly mid-range: it matches the AMD Radeon RX 6950 XT within 0.3% and the RX 5600 OEM within 0.3%, placing it far below the MI60’s class.

The MI60’s percentile ranking of 93 versus the Arc’s 88 reflects the performance gap, but the production status tells the broader story. The MI60 is end-of-life, while the Arc is active. The MI60 also requires a 700 W suggested PSU and external power connectors, while the Arc needs neither. For a user with an existing dual-slot PCIe 4.0 x16 workstation, the MI60 offers unmatched memory bandwidth and compute density. For a user building a low-power mobile system, the Arc is the only viable option between the two, though its benchmark data is limited to a single OpenCL score. The database records one head-to-head, and the MI60 wins it by a wide margin. That is the objective answer to which is faster. The Arc’s advantages are architectural and efficiency-based, not performance-based.