AMD Radeon Instinct MI60 vs Intel Arc A580 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon Instinct MI60 records an average benchmark score of 92,466, placing it in the 93rd percentile of all GPUs. The Intel Arc A580 averages 57,756, which lands in the 87th percentile.

Q: How do the two compare in Geekbench OpenCL?

A: The MI60 scores 92,488 against the Arc A580's 91,657. The AMD part wins by 0.9%, a narrow margin that puts it just ahead of the Intel card in compute workloads.

Q: What is the largest performance gap between them?

A: In Geekbench Vulkan, the MI60 scores 92,444 while the Arc A580 manages 79,381. The AMD card leads by 16.5%, the biggest spread recorded in the head-to-head data.

Q: Which GPU has more shading units?

A: The AMD Radeon Instinct MI60 packs 4,096 shading units, compared to 3,072 on the Intel Arc A580. The MI60 also has 256 texture mapping units versus 192 on the Arc A580.

Q: Do both cards support the same DirectX version?

A: No. The Intel Arc A580 supports DirectX 12 Ultimate (12_2), while the AMD Radeon Instinct MI60 is limited to DirectX 12 (12_1). Intel also lists Vulkan 1.4 support, one version ahead of AMD's Vulkan 1.3.

Q: What are the memory configurations?

A: The MI60 uses 32 GB of HBM2 on a 4096-bit bus, delivering 1.02 TB/s of bandwidth. The Arc A580 has 8 GB of GDDR6 on a 256-bit bus, providing 512.0 GB/s.

Architecture Differences

The two cards come from entirely different design philosophies. The AMD Radeon Instinct MI60 uses the Vega 20 chip built on GCN 5.1 architecture, manufactured on a 7 nm process at TSMC. The Intel Arc A580 uses the DG2-512 die with Xe-HPG architecture, produced on a 6 nm node, also at TSMC. Both foundries are the same, but the process generations differ.

Transistor counts reveal the scale of each design. The MI60 integrates 13,230 million transistors on a 331 mm² die, giving a density of 40.0 million transistors per square millimeter. The Arc A580 packs 21,700 million transistors into 406 mm², achieving a higher density of 53.4 million per square millimeter. Intel's chip is physically larger and more densely packed.

The MI60's compute layout favors raw throughput with 4,096 shading units, 256 TMUs, and 64 ROPs. The Arc A580 counters with 3,072 shading units, 192 TMUs, and 96 ROPs. Notably, the Arc A580 includes 24 ray tracing cores, a feature the MI60 lacks entirely. Neither card has tensor cores listed in the database.

Memory architecture splits them sharply. The MI60 uses 32 GB of HBM2 across a 4096-bit bus, reaching 1.02 TB/s of bandwidth. The Arc A580 uses 8 GB of GDDR6 on a 256-bit bus, capping at 512.0 GB/s, exactly half the AMD card's bandwidth. Clock speeds also differ: the MI60 runs at 1200 MHz base and 1800 MHz boost, while the Arc A580 starts at 1700 MHz base and reaches 2000 MHz boost.

The MI60's FP32 throughput is 14.75 TFLOPS, ahead of the Arc A580's 12.29 TFLOPS. FP16 figures follow the same pattern: 29.49 TFLOPS on the MI60 versus 24.58 TFLOPS on the Arc A580, both with a 2:1 ratio relative to FP32. Pixel and texture rates tell a mixed story. The Arc A580 posts 192.0 GPixel/s, higher than the MI60's 115.2 GPixel/s, but the MI60 leads in texture rate at 460.8 GTexel/s against 384.0 GTexel/s.

Power and connectivity differ as well. The MI60 has a 300 W TDP with a suggested 700 W PSU and uses 1x 6-pin plus 1x 8-pin connectors. The Arc A580 draws 175 W, needs a 450 W PSU, and uses 2x 8-pin connectors. Both are dual-slot cards on PCIe 4.0 x16.

Head-to-Head Benchmarks

The database records two direct benchmark comparisons. The results show a clear sweep for the AMD Radeon Instinct MI60, which wins both tests.

In Geekbench OpenCL, the MI60 scores 92,488 against the Arc A580's 91,657. The delta is 0.9%, indicating a near tie. This small gap suggests that in compute-heavy OpenCL workloads, the two cards perform almost identically, with the AMD part holding a slight edge. The MI60's higher shading unit count and memory bandwidth contribute to this lead, but the Arc A580's higher clock speeds narrow the distance.

Geekbench Vulkan presents a different picture. The MI60 records 92,444, while the Arc A580 drops to 79,381. The 16.5% delta is substantial. This result shows the MI60 maintaining its OpenCL-level performance in Vulkan, whereas the Arc A580 loses roughly 13% of its OpenCL score when switching to the Vulkan API. The Arc A580's 24 ray tracing cores do not translate into a Vulkan advantage in this benchmark; the MI60's older architecture simply performs more consistently across APIs.

The overall benchmark averages reinforce the head-to-head pattern. The MI60's average of 92,466 sits far above the Arc A580's 57,756. That average includes the Arc A580's additional 3DMark Steel Nomad DX12 score of 2,229, which drags its aggregate down. The MI60's two benchmark scores are tightly clustered around 92,500, showing stability across different test types.

The nearest rival data provides context for each card's standing. The MI60's closest competitors are the NVIDIA RTX A4500 at 91,671 (0.9% behind), the RTX A4500 Mobile at 91,134 (1.5% behind), the AMD Radeon Pro VII at 97,131 (4.8% ahead), and the AMD Radeon RX 7900M at 97,487 (5.2% ahead). The Arc A580's rivals include the AMD Radeon RX 5600 OEM at 58,085 (0.6% behind), the RX 9070 GRE at 57,367 (0.7% ahead), the Intel Arc A570M at 58,239 (0.8% behind), and the RX 6950 XT at 58,392 (1.1% behind). The MI60 competes in a higher performance tier, while the Arc A580 sits among mid-range parts.

The Verdict

The data directs different buyers to different cards. The AMD Radeon Instinct MI60 wins both head-to-head benchmarks, leading by 0.9% in OpenCL and 16.5% in Vulkan. Its average benchmark score of 92,466 places it in the 93rd percentile of all GPUs, a full six points above the Arc A580's 87th percentile. For compute tasks that rely on OpenCL or Vulkan, the MI60 is the stronger choice.

The Intel Arc A580, however, is not without merit in specific contexts. Its 192.0 GPixel/s pixel rate exceeds the MI60's 115.2 GPixel/s, and its 96 ROPs outnumber the MI60's 64. The Arc A580 also supports DirectX 12 Ultimate and Vulkan 1.4, newer API versions than the MI60's DirectX 12 (12_1) and Vulkan 1.3. For gaming or applications that use ray tracing, the Arc A580's 24 RT cores give it a feature the MI60 cannot match. The Arc A580's lower TDP of 175 W also makes it easier to integrate into systems with modest power budgets.

The MI60 targets data center and compute environments. Its 32 GB of HBM2 memory and 1.02 TB/s bandwidth suit large datasets, and its end-of-life status indicates it has been superseded. The Arc A580, released in October 2023, remains active and has a successor, Battlemage, listed in the database.

The verdict comes down to workload. For raw compute throughput, the MI60 wins. For modern API support, ray tracing, and pixel-heavy rendering, the Arc A580 holds the advantage. The 16.5% Vulkan gap shows the MI60's dominance in that API, but the Arc A580's feature set covers newer use cases.

Specification Differences

| Specification | AMD Radeon Instinct MI60 | Intel Arc A580 |

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

| Architecture | GCN 5.1 | Xe-HPG |

| Process Node | 7 nm | 6 nm |

| Transistors | 13,230 million | 21,700 million |

| Die Size | 331 mm² | 406 mm² |

| Transistor Density | 40.0M / mm² | 53.4M / mm² |

| Base Clock | 1200 MHz | 1700 MHz |

| Boost Clock | 1800 MHz | 2000 MHz |

| Memory Size | 32 GB | 8 GB |

| Memory Type | HBM2 | GDDR6 |

| Memory Bus Width | 4096 bit | 256 bit |

| Memory Bandwidth | 1.02 TB/s | 512.0 GB/s |

| Shading Units | 4096 | 3072 |

| TMUs | 256 | 192 |

| ROPs | 64 | 96 |

| Ray Tracing Cores | None | 24 |

| Pixel Rate | 115.2 GPixel/s | 192.0 GPixel/s |

| Texture Rate | 460.8 GTexel/s | 384.0 GTexel/s |

| FP32 | 14.75 TFLOPS | 12.29 TFLOPS |

| FP16 | 29.49 TFLOPS (2:1) | 24.58 TFLOPS (2:1) |

| TDP | 300 W | 175 W |

| Power Connectors | 1x 6-pin + 1x 8-pin | 2x 8-pin |

| Suggested PSU | 700 W | 450 W |

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

| Vulkan | 1.3 | 1.4 |

| Display Outputs | 1x mini-DisplayPort 1.4a | 1x HDMI 2.1, 3x DisplayPort 2.0 |

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

| Release Date | 2018-11-17 | 2023-10-09 |

Where Each One Wins

The AMD Radeon Instinct MI60 wins in compute throughput. Its FP32 output of 14.75 TFLOPS and FP16 output of 29.49 TFLOPS beat the Arc A580's 12.29 and 24.58 TFLOPS respectively. The MI60's texture rate of 460.8 GTexel/s also surpasses the Arc A580's 384.0 GTexel/s. Memory bandwidth is a decisive factor: 1.02 TB/s versus 512.0 GB/s means the MI60 can feed data to its compute units at twice the rate. The MI60 wins both recorded benchmark tests, with a notable 16.5% lead in Vulkan. Its 93rd percentile ranking places it among the top GPUs in the database.

The Intel Arc A580 wins in rendering throughput. Its pixel rate of 192.0 GPixel/s exceeds the MI60's 115.2 GPixel/s by a wide margin, and its 96 ROPs outnumber the MI60's 64. The Arc A580 supports DirectX 12 Ultimate and Vulkan 1.4, newer API versions that enable features the MI60 cannot use. Its 24 ray tracing cores provide hardware acceleration for RT workloads, a capability entirely absent on the MI60. The Arc A580 also runs cooler from a power perspective, with a 175 W TDP versus 300 W, and requires only a 450 W PSU.

The release timeline favors the Arc A580 for longevity. The MI60 launched in November 2018 and is end-of-life, while the Arc A580 launched in October 2023 and remains active with a successor, Battlemage, already listed. The Arc A580's display outputs include multiple modern ports (1x HDMI 2.1, 3x DisplayPort 2.0), whereas the MI60 offers only a single mini-DisplayPort 1.4a.

Use-case guidance follows the data. Choose the MI60 for compute-heavy workloads like data processing or scientific simulation where FP32 throughput and memory bandwidth dominate. Choose the Arc A580 for gaming, ray tracing, or any workload that relies on modern DirectX features, where its pixel rate and API support provide tangible benefits.