AMD Instinct MI300 vs Intel Arc Pro A60 Comparison

AMD Instinct MI300 and Intel Arc Pro A60 represent two very different approaches to GPU hardware. The MI300 is a data center accelerator built for massive compute workloads, while the Arc Pro A60 is a professional workstation graphics card. The recorded data for these two parts shows a complete separation in purpose, performance profile, and physical design. This analysis walks through the benchmark results, architectural details, and specification differences as captured in the database.

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark runs between the AMD Instinct MI300 and the Intel Arc Pro A60. The MI300 has no recorded benchmark scores in the database, and its average benchmark score is listed as zero. Its percentile ranking against all GPUs is 50, which places it in the middle of the distribution, but this is based on an empty benchmark set. The absence of scores means no direct comparison can be made from the available measurements.

The Intel Arc Pro A60, by contrast, has two recorded benchmark results. In Geekbench OpenCL, it scores 63485. In Geekbench Vulkan, it scores 57166. Its average benchmark score across these runs is 60326. This average places the Arc Pro A60 in the 88th percentile among all GPUs in the database, a strong showing for a workstation card.

When comparing the Arc Pro A60 to its nearest rivals, the data indicates a tight cluster. The NVIDIA GeForce RTX 4090 has an average score of 60347, which is essentially identical to the Arc Pro A60’s 60326, with a delta of 0 percent. The AMD Radeon Pro Vega 48 scores 60140, a delta of 0.3 percent behind. The AMD Radeon Pro W6600M scores 61896, which is 2.5 percent ahead of the Arc Pro A60. The AMD Radeon PRO V710 scores 58657, putting it 2.8 percent behind the Arc Pro A60. These deltas are small, indicating the Arc Pro A60 performs within a few percentage points of several competing professional and workstation GPUs.

For the MI300, no such rival comparisons exist in the database. Its raw compute capabilities, when expressed in the recorded specifications, dwarf the Arc Pro A60, but those capabilities are not measured by the benchmark suite used here. The data shows the MI300’s FP32 throughput is 47.87 TFLOPS, while the Arc Pro A60 delivers 8.397 TFLOPS. That is a 5.7x difference in raw floating-point performance. The MI300’s texture rate is 1,496.0 GTexel/s versus 262.4 GTexel/s for the Arc Pro A60, a 5.7x difference as well. The MI300 has no pixel rate (0 MPixel/s) because it lacks ROPs entirely, while the Arc Pro A60 outputs 131.2 GPixel/s.

The Verdict

The data points to two entirely separate use cases. The AMD Instinct MI300 is a compute accelerator with no display outputs, no graphics API support, and no pixel processing capability. Its 128 GB of HBM3 memory and 5.32 TB/s bandwidth are designed for large-scale data processing, AI training, and scientific simulation. The Intel Arc Pro A60 is a professional graphics card with 4x DisplayPort 2.0 outputs, full DirectX 12 Ultimate support, OpenGL 4.6, and Vulkan 1.4. It is meant for rendering, CAD, and visual workloads where output to a screen is required.

From the benchmark data alone, the Arc Pro A60 is the only one of the two with measured performance. Its average score of 60326 places it in the 88th percentile, and its nearest rivals are within a few percentage points. The MI300 has no scores, so any performance claim about it must come from the specification sheet, not from the database’s benchmark results.

For a user who needs a GPU for graphics work, the Arc Pro A60 is the clear choice from the recorded data. For a user who needs raw compute throughput without graphics, the MI300’s specifications suggest it is built for that role, but the database does not verify this with benchmarks. The verdict from the data is simple: the Arc Pro A60 is a measured performer in workstation graphics, while the MI300 is an unmeasured compute accelerator with far higher theoretical limits.

Architecture Differences

The AMD Instinct MI300 uses the CDNA 3.0 architecture, specifically designed for compute and AI workloads. Its chip is code-named Aqua Vanjaram. The Intel Arc Pro A60 uses the Xe-HPG architecture, part of the Alchemist generation for the Pro Series. Its chip is DG2-256.

The MI300 is manufactured on a 5 nm process at TSMC, with 153,000 million transistors on a die size of 1017 mm². That yields a transistor density of 150.4 million per square millimeter. The Arc Pro A60 is on a 6 nm process at TSMC, with 11,500 million transistors on a 269 mm² die, giving a density of 42.8 million per square millimeter. The MI300 has over 13 times the transistor count and nearly four times the die area.

The MI300 has 14,080 shading units and 880 texture mapping units. It has no ROPs, no ray tracing cores, and no tensor cores listed. Its FP16 performance is listed as 47.87 TFLOPS with a 1:1 ratio to FP32, meaning it does not double-rate half-precision work. The Arc Pro A60 has 2,048 shading units, 128 TMUs, and 64 ROPs. It includes 16 ray tracing cores. Its FP16 throughput is 16.79 TFLOPS with a 2:1 ratio, meaning it processes half-precision at twice the rate of FP32.

The MI300’s architecture is compute-focused, lacking any graphics pipeline elements. The Arc Pro A60’s architecture includes a full graphics pipeline with ray tracing support. The process node difference is small (5 nm vs 6 nm), but the transistor budget is vastly different, reflecting the MI300’s role as a high-end accelerator.

Specification Differences

The memory subsystems differ completely. The MI300 has 128 GB of HBM3 memory on an 8192-bit bus, delivering 5.32 TB/s of bandwidth. The Arc Pro A60 has 12 GB of GDDR6 memory on a 192-bit bus, delivering 384.0 GB/s. The MI300’s memory clock is 1300 MHz, effectively 5.2 Gbps, while the Arc Pro A60 runs at 2000 MHz, effectively 16 Gbps per pin.

Clock speeds show the MI300 with a base of 1000 MHz and a boost of 1700 MHz. The Arc Pro A60 has a base of 900 MHz and a boost of 2050 MHz. Despite the higher boost clock on the Arc Pro A60, the MI300’s much larger execution unit count gives it far higher throughput.

Power requirements differ sharply. The MI300 has a TDP of 600 W and requires a 1000 W power supply. It uses two 8-pin power connectors. The Arc Pro A60 has a TDP of 130 W and a suggested power supply of 300 W, with no additional power connectors listed. The MI300’s physical dimensions are 267 mm in length and 111 mm in height. The Arc Pro A60 is single-slot, but its length and height are not recorded.

The bus interfaces differ as well. The MI300 uses PCIe 5.0 x16, while the Arc Pro A60 uses PCIe 4.0 x16. Display outputs are absent on the MI300, while the Arc Pro A60 provides four DisplayPort 2.0 connections.

The MI300’s API support is listed as N/A for DirectX, OpenGL, and Vulkan. The Arc Pro A60 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The release dates show the MI300 launched on January 3, 2023, and the Arc Pro A60 on June 5, 2023. The MI300’s predecessor is listed as Radeon Instinct, while the Arc Pro A60 has no predecessor. The Arc Pro A60 has an active production status; the MI300’s status is not recorded.

FAQ

Q: Does the AMD Instinct MI300 have any benchmark scores in the database?

A: No. The MI300 has an empty benchmark list, an average benchmark score of 0, and a percentile ranking of 50 among all GPUs.

Q: How does the Intel Arc Pro A60 compare to its nearest rivals?

A: The Arc Pro A60’s average score is 60326. The NVIDIA GeForce RTX 4090 scores 60347 with a 0 percent delta, the AMD Radeon Pro Vega 48 scores 60140 with a 0.3 percent delta, the AMD Radeon Pro W6600M scores 61896 with a -2.5 percent delta, and the AMD Radeon PRO V710 scores 58657 with a 2.8 percent delta.

Q: Which card has higher FP32 performance?

A: The MI300, at 47.87 TFLOPS, is 5.7 times higher than the Arc Pro A60’s 8.397 TFLOPS.

Q: Can the MI300 output video to a display?

A: No. The MI300 has no display outputs and its API support is listed as N/A for DirectX, OpenGL, and Vulkan.

Q: What is the memory bandwidth of each card?

A: The MI300 has 5.32 TB/s from 128 GB of HBM3 on an 8192-bit bus. The Arc Pro A60 has 384.0 GB/s from 12 GB of GDDR6 on a 192-bit bus.

Q: Does the Arc Pro A60 support ray tracing?

A: Yes. It has 16 ray tracing cores and supports DirectX 12 Ultimate, which includes ray tracing features.

Where Each One Wins

The AMD Instinct MI300 wins in raw compute throughput. Its FP32 figure of 47.87 TFLOPS is nearly six times higher than the Arc Pro A60’s 8.397 TFLOPS. Its texture rate of 1,496.0 GTexel/s far exceeds the 262.4 GTexel/s of the Arc Pro A60. The MI300’s memory bandwidth of 5.32 TB/s is over 13 times the Arc Pro A60’s 384.0 GB/s, and its 128 GB capacity dwarfs the 12 GB available on the Intel card. For workloads that need massive memory pools and extreme bandwidth, such as large model inference or scientific data processing, the MI300’s specifications give it a clear advantage. It also has over 13 times the transistor count, which suggests a much larger compute fabric.

The Intel Arc Pro A60 wins in graphics and display functionality. It has 4x DisplayPort 2.0 outputs, supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. It has 64 ROPs, producing a pixel rate of 131.2 GPixel/s, while the MI300 has zero ROPs and zero pixel rate. The Arc Pro A60 also includes 16 ray tracing cores, which the MI300 lacks entirely. Its FP16 performance doubles to 16.79 TFLOPS, while the MI300’s FP16 is locked at 1:1 with FP32. The Arc Pro A60 has a higher boost clock (2050 MHz vs 1700 MHz) and a much lower power draw (130 W vs 600 W), making it practical for workstation environments where a display output is required.

The benchmark data places the Arc Pro A60 in the 88th percentile with a measured average score of 60326, while the MI300 has no measured scores. In the database’s recorded results, only the Arc Pro A60 demonstrates real-world performance, and that performance is competitive with the NVIDIA GeForce RTX 4090 and other professional cards. The MI300’s wins are purely theoretical, based on its specification sheet. The data shows that the MI300 is built for compute density and the Arc Pro A60 for graphics output, and neither card can substitute for the other in its intended role.