AMD Instinct MI455X vs Intel Arc Pro A60 Comparison

Head-to-Head Benchmarks

The recorded data presents an unusual comparison: the AMD Instinct MI455X has no benchmark scores in the database, while the Intel Arc Pro A60 has two recorded results. The Intel Arc Pro A60 achieves an average benchmark score of 60,326 across its tested workloads. In the Geekbench OpenCL test, it scores 63,485, and in the Geekbench Vulkan test, it scores 57,166. The AMD Instinct MI455X, by contrast, shows an average benchmark score of 0, meaning no comparable measurements exist for this part in the database, and its percentile ranking sits at 50 among all GPUs, while the Intel card ranks at 88.

Because the head-to-head benchmark array is empty and the AMD card has no scores, a direct numerical comparison of measured performance is not possible. What can be stated is the relative position of the Intel Arc Pro A60 against its closest rivals, which provides context for where this card lands. The Arc Pro A60 sits within 0.0% of the NVIDIA GeForce RTX 4090 in average score, a delta of exactly zero, placing it effectively at parity with that card in the database's recorded measurements. It is 0.3% ahead of the AMD Radeon Pro Vega 48, 2.5% behind the AMD Radeon Pro W6600M, and 2.8% ahead of the AMD Radeon PRO V710. These deltas are small, indicating that the Arc Pro A60 competes in a tight cluster of mid-range workstation cards.

The FP32 compute figures, however, are dramatically different. The AMD Instinct MI455X delivers 157.3 TFLOPS of FP32 performance, while the Intel Arc Pro A60 delivers 8.397 TFLOPS. That is a gap of roughly 18.7 times in raw single-precision throughput. FP16 performance tells a similar story: the AMD card hits 157.3 TFLOPS with a 1:1 ratio to FP32, while the Intel card reaches 16.79 TFLOPS with a 2:1 ratio. The AMD part's memory bandwidth, measured at 23.3 TB/s, dwarfs the Intel card's 384.0 GB/s, a roughly 60-fold difference. These figures indicate that the MI455X is designed for a completely different performance class, even though the database lacks direct benchmark scores for it.

Where Each One Wins

The AMD Instinct MI455X wins decisively in raw compute and memory throughput. Its FP32 rating of 157.3 TFLOPS is nearly nineteen times that of the Intel Arc Pro A60, and its FP16 rating matches that FP32 figure, whereas the Intel card halves its FP16 throughput relative to FP32. The memory subsystem on the AMD card is in another tier: 432 GB of HBM4 memory on a 24,576-bit bus produces 23.3 TB/s of bandwidth, compared with 12 GB of GDDR6 on a 192-bit bus at 384.0 GB/s for the Intel part. Texture rate also favors AMD, with 2,457.6 GTexel/s versus 262.4 GTexel/s. The AMD card uses a 2 nm process node from TSMC, packs 320,000 million transistors on a 2,990 mm² die, and supports PCIe 6.0 x16.

The Intel Arc Pro A60 wins in areas relevant to workstation graphics output and API support. It has 4 DisplayPort 2.0 outputs, while the AMD card lists no display outputs. The Intel card supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, while the AMD card reports N/A for all three APIs. The Intel card also has dedicated ray tracing hardware with 16 RT cores, whereas the AMD card reports no RT core count. Pixel rate tells the same story: the Intel card produces 131.2 GPixel/s, while the AMD card is rated at 0 MPixel/s, reflecting its compute-only design. The Intel part fits in a single slot and draws 130 W, while the AMD module uses an EAM form factor and consumes 2,300 W.

The database's benchmark results only cover the Intel card, so the measured wins belong entirely to it. The AMD card's wins are theoretical, based on specifications rather than recorded scores. The Intel Arc Pro A60's average score of 60,326 places it in the 88th percentile among all GPUs, indicating strong measured performance in its class. The AMD Instinct MI455X, with no recorded benchmarks, holds a 50th percentile ranking by default, but its specification sheet suggests a far higher compute ceiling.

Architecture Differences

The two cards come from different architectural families and target different workloads. The AMD Instinct MI455X uses the CDNA 5.0 architecture with a chip design labeled MI450 256CU. It is built on a 2 nm process at TSMC, with 320,000 million transistors on a 2,990 mm² die, resulting in a transistor density of 107.0 million transistors per square millimeter. The Intel Arc Pro A60 uses the Xe-HPG architecture with a DG2-256 chip, built on a 6 nm process at TSMC, with 11,500 million transistors on a 269 mm² die, for a density of 42.8 million transistors per square millimeter. The AMD chip is about 11 times larger in die area and holds nearly 28 times more transistors.

Shader resources differ enormously. The AMD card has 32,768 shading units and 1,024 texture mapping units, with no ROPs listed. The Intel card has 2,048 shading units, 128 TMUs, and 64 ROPs. The AMD card's texture rate is 2,457.6 GTexel/s, while the Intel card manages 262.4 GTexel/s. The Intel card includes 16 RT cores for hardware ray tracing, a feature absent from the AMD card's specification list. Neither card lists tensor cores or dedicated AI accelerator counts.

Memory architecture is fundamentally different. The AMD card uses HBM4 memory totaling 432 GB on a 24,576-bit bus, with a 1,900 MHz memory clock running at 7.6 Gbps effective and 23.3 TB/s bandwidth. The Intel card uses GDDR6 memory totaling 12 GB on a 192-bit bus, with a 2,000 MHz memory clock running at 16 Gbps effective and 384.0 GB/s bandwidth. Clock speeds tell a mixed story: the AMD card has a 1,000 MHz base clock and 2,400 MHz boost clock, while the Intel card has a 900 MHz base clock and 2,050 MHz boost clock. The AMD card runs at higher clocks in both states, but its power envelope of 2,300 W dwarfs the Intel card's 130 W.

The bus interfaces also differ. The AMD card uses PCIe 6.0 x16, while the Intel card uses PCIe 4.0 x16. The AMD card requires a suggested PSU of 2,700 W and uses no power connectors, as it is an EAM module. The Intel card requires a 300 W suggested PSU and uses a single-slot form factor. The Intel card's production status is listed as Active, while the AMD card's production status is not specified. Release dates place the Intel card in June 2023 and the AMD card in July 2026, a gap of more than three years that explains the architectural differences.

FAQ

Q: How much faster is the AMD Instinct MI455X in FP32 compute than the Intel Arc Pro A60?

A: The AMD card delivers 157.3 TFLOPS of FP32 performance, while the Intel card delivers 8.397 TFLOPS, a difference of roughly 18.7 times.

Q: Which card has more memory bandwidth?

A: The AMD Instinct MI455X has 23.3 TB/s of bandwidth from 432 GB of HBM4 memory on a 24,576-bit bus. The Intel Arc Pro A60 has 384.0 GB/s from 12 GB of GDDR6 on a 192-bit bus.

Q: Does the Intel Arc Pro A60 support display outputs?

A: Yes, it has 4 DisplayPort 2.0 outputs. The AMD Instinct MI455X lists no display outputs.

Q: What is the average benchmark score of the Intel Arc Pro A60?

A: Its average benchmark score is 60,326, based on a Geekbench OpenCL score of 63,485 and a Geekbench Vulkan score of 57,166.

Q: Does the AMD Instinct MI455X have any recorded benchmark scores?

A: No, the database lists an average benchmark score of 0 for the AMD card, with no individual benchmark entries.

Q: How does the Intel Arc Pro A60 compare to the NVIDIA GeForce RTX 4090 in the database?

A: The Intel card's average score is within 0.0% of the RTX 4090's average score of 60,347, indicating parity in recorded measurements.

Specification Differences

| Specification | AMD Instinct MI455X | Intel Arc Pro A60 |

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

| Architecture | CDNA 5.0 | Xe-HPG |

| Process Node | 2 nm | 6 nm |

| Transistors | 320,000 million | 11,500 million |

| Die Size | 2,990 mm² | 269 mm² |

| Transistor Density | 107.0M / mm² | 42.8M / mm² |

| Base Clock | 1,000 MHz | 900 MHz |

| Boost Clock | 2,400 MHz | 2,050 MHz |

| Memory Size | 432 GB | 12 GB |

| Memory Type | HBM4 | GDDR6 |

| Memory Bus Width | 24,576 bit | 192 bit |

| Memory Bandwidth | 23.3 TB/s | 384.0 GB/s |

| Shading Units | 32,768 | 2,048 |

| TMUs | 1,024 | 128 |

| ROPs | 0 | 64 |

| RT Cores | Not listed | 16 |

| Pixel Rate | 0 MPixel/s | 131.2 GPixel/s |

| Texture Rate | 2,457.6 GTexel/s | 262.4 GTexel/s |

| FP32 Performance | 157.3 TFLOPS | 8.397 TFLOPS |

| FP16 Performance | 157.3 TFLOPS (1:1) | 16.79 TFLOPS (2:1) |

| TDP | 2,300 W | 130 W |

| Slot Width | EAM Module | Single-slot |

| Suggested PSU | 2,700 W | 300 W |

| Bus Interface | PCIe 6.0 x16 | PCIe 4.0 x16 |

| Display Outputs | No outputs | 4x DisplayPort 2.0 |

| DirectX Support | N/A | 12 Ultimate (12_2) |

| OpenGL Support | N/A | 4.6 |

| Vulkan Support | N/A | 1.4 |

| Release Date | July 2026 | June 2023 |

| Production Status | Not specified | Active |

| Percentile vs All GPUs | 50 | 88 |

| Average Benchmark Score | 0 | 60,326 |