AMD Radeon PRO W6600 vs Intel Arc Pro A60 Comparison

The Verdict

The benchmark data splits these two professional graphics cards into clear roles. The AMD Radeon PRO W6600 is the faster card overall, with an average benchmark score of 81,995 compared to the Intel Arc Pro A60’s 60,326. That is a 35.9% gap in average score. The AMD card wins both recorded head-to-head tests, with a 15.8% lead in OpenCL and a commanding 37.2% lead in Vulkan. For users who need maximum compute throughput in cross-platform workloads, the W6600 is the straightforward choice.

The Intel Arc Pro A60, however, is not without its own arguments. It carries 12 GB of memory versus 8 GB on the AMD card, paired with a 192-bit bus and 384.0 GB/s of bandwidth, which is 71.4% more bandwidth than the W6600’s 224.0 GB/s. It is also an active product, while the W6600 is end-of-life. For workloads that are memory-capacity bound or memory-bandwidth sensitive, the A60 offers structural advantages that raw compute scores do not capture. The A60 also uses a PCIe 4.0 x16 interface, double the lane width of the W6600’s PCIe 4.0 x8, which can matter for data transfer in certain systems.

The verdict, strictly from the data: pick the AMD Radeon PRO W6600 if raw compute performance in OpenCL and Vulkan is the priority, and if the 8 GB memory pool is sufficient for the target applications. Pick the Intel Arc Pro A60 if memory capacity, memory bandwidth, or a longer production lifecycle matters more than peak compute scores. The A60 sits at the 88th percentile of all GPUs, while the W6600 sits at the 92nd, so both are well above average, but the AMD card is the stronger performer in the recorded tests.

Architecture Differences

The two cards come from different architectural lineages. The AMD Radeon PRO W6600 uses the Navi 23 chip built on RDNA 2.0, fabricated on a 7 nm process at TSMC. The Intel Arc Pro A60 uses the DG2-256 chip built on Xe-HPG, fabricated on a 6 nm process, also at TSMC. Both are single-slot cards, but the underlying designs diverge significantly.

Transistor counts are nearly identical: 11,060 million on the AMD chip versus 11,500 million on the Intel chip, a difference of about 4%. The die sizes differ more noticeably, with AMD at 237 mm² and Intel at 269 mm², making the Intel die about 13.5% larger. Transistor density favors AMD: 46.7 million transistors per mm² versus 42.8 million on Intel, a 9.1% higher density.

Memory architecture is a major differentiator. The W6600 has 8 GB of GDDR6 on a 128-bit bus, yielding 224.0 GB/s of bandwidth. The A60 has 12 GB of GDDR6 on a 192-bit bus, yielding 384.0 GB/s, which is 71.4% more bandwidth. The memory clock rates also differ, with the AMD memory rated at 1750 MHz (14 Gbps effective) and the Intel memory at 2000 MHz (16 Gbps effective).

Compute resources are split differently. The Intel card has more shaders (2048 versus 1792, a 14.3% advantage), more texture mapping units (128 versus 112, a 14.3% advantage), and the same 64 ROPs. But the AMD card has more ray tracing cores: 28 versus 16, a 75% advantage. Clock speeds favor AMD heavily: the W6600 runs at a 2331 MHz base and 2580 MHz boost, while the A60 runs at 900 MHz base and 2050 MHz boost. The AMD boost clock is 25.9% higher.

The pixel and texture rates reflect these clock advantages. The W6600 achieves 165.1 GPixel/s and 289.0 GTexel/s, while the A60 achieves 131.2 GPixel/s and 262.4 GTexel/s. AMD leads by 25.8% in pixel rate and 10.1% in texture rate. FP32 compute is 9.247 TFLOPS on the AMD card versus 8.397 TFLOPS on the Intel card, a 10.1% lead. FP16 follows the same pattern: 18.49 TFLOPS versus 16.79 TFLOPS, both at 2:1 ratio.

Power draws differ: the W6600 has a TDP of 100 W with a single 6-pin connector, while the A60 has a TDP of 130 W and no power connector listed. Both suggest a 300 W power supply. Display outputs are modern on both, with 4x DisplayPort 1.4a on the AMD card and 4x DisplayPort 2.0 on the Intel card. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Where Each One Wins

The recorded benchmark wins go entirely to the AMD Radeon PRO W6600. In Geekbench OpenCL, the W6600 scores 73,514 against 63,485 for the A60, a 15.8% advantage. In Geekbench Vulkan, the W6600 scores 78,428 against 57,166, a 37.2% advantage. The Vulkan gap is particularly notable, suggesting the RDNA 2.0 architecture has a significant efficiency edge in that API on these workloads.

The average benchmark score reinforces the pattern: 81,995 for the W6600 versus 60,326 for the A60. The AMD card also has a Geekbench Metal score of 94,042, though the Intel card has no recorded Metal score for comparison. The W6600’s percentile ranking of 92 versus the A60’s 88 means the AMD card sits higher in the overall GPU distribution.

The Intel Arc Pro A60 wins on structural and capacity metrics rather than recorded compute benchmarks. Its 12 GB memory pool is 50% larger than the AMD card’s 8 GB. Its 384.0 GB/s bandwidth is 71.4% higher. Its PCIe 4.0 x16 interface doubles the lane count of the AMD card’s PCIe 4.0 x8. These factors can favor the A60 in memory-intensive professional workloads such as large dataset manipulation, high-resolution textures, or GPU-accelerated rendering that exceeds 8 GB.

The production status also favors Intel. The A60 is listed as active, while the W6600 is end-of-life. For organizations planning long procurement cycles, an active product has practical advantages. The AMD card’s launch MSRP was 649 USD, which can be stated once as a reference point, but the Intel card has no launch MSRP recorded in the database.

FAQ

Q: Which card is faster in Geekbench Vulkan?

A: The AMD Radeon PRO W6600 scores 78,428 in Vulkan, which is 37.2% ahead of the Intel Arc Pro A60’s 57,166.

Q: How much memory does each card have?

A: The AMD Radeon PRO W6600 has 8 GB of GDDR6 on a 128-bit bus, while the Intel Arc Pro A60 has 12 GB of GDDR6 on a 192-bit bus.

Q: Which card has higher memory bandwidth?

A: The Intel Arc Pro A60 has 384.0 GB/s of bandwidth, which is 71.4% higher than the AMD Radeon PRO W6600’s 224.0 GB/s.

Q: Are both cards single-slot designs?

A: Yes, both the AMD Radeon PRO W6600 and the Intel Arc Pro A60 are listed as single-slot cards.

Q: What is the production status of each card?

A: The AMD Radeon PRO W6600 is end-of-life, while the Intel Arc Pro A60 is active.

Q: Which card has more ray tracing cores?

A: The AMD Radeon PRO W6600 has 28 ray tracing cores, while the Intel Arc Pro A60 has 16, a 75% advantage for the AMD card.

Head-to-Head Benchmarks

The head-to-head data covers two tests: Geekbench OpenCL and Geekbench Vulkan. The AMD Radeon PRO W6600 wins both.

In Geekbench OpenCL, the W6600 records 73,514 points against the A60’s 63,485. The delta is 15.8% in favor of AMD. This is a solid but not overwhelming margin. It places the W6600 in a competitive position against its own nearest rivals: the AMD Radeon Pro Vega 64X scores 80,959 (1.3% behind the W6600), the NVIDIA GeForce RTX 5090 scores 79,842 (2.7% behind), the NVIDIA Tesla P100 PCIe 16 GB scores 79,605 (3.0% behind), and the NVIDIA Tesla P100 PCIe 12 GB scores 79,396 (3.3% behind). The W6600’s average score of 81,995 puts it ahead of all four nearest rivals by small margins.

The Intel Arc Pro A60’s nearest rivals tell a different story. The NVIDIA GeForce RTX 4090 scores 60,347, essentially tied with the A60’s 60,326 (0.0% delta). The AMD Radeon Pro Vega 48 scores 60,140, just 0.3% behind. The AMD Radeon Pro W6600M scores 61,896, which is 2.5% ahead of the A60. The AMD Radeon PRO V710 scores 58,657, which is 2.8% behind. The A60 sits in a crowded performance band where the differences between adjacent cards are within a few percentage points.

In Geekbench Vulkan, the gap widens dramatically. The W6600 scores 78,428, while the A60 scores 57,166. The delta is 37.2% in favor of AMD. This is the single largest margin in the head-to-head comparison. The Vulkan result suggests the AMD architecture extracts substantially more performance from the same class of workload in this API. The W6600’s Vulkan score is also higher than its OpenCL score, while the A60’s Vulkan score is lower than its OpenCL score, a reversal that highlights the architectural difference.

Looking at the broader benchmark picture, the W6600 has three recorded tests: Geekbench Metal at 94,042, Geekbench OpenCL at 73,514, and Geekbench Vulkan at 78,428. The average of these is 81,995. The A60 has two recorded tests: Geekbench OpenCL at 63,485 and Geekbench Vulkan at 57,166, averaging 60,326. The W6600’s Metal score is its strongest, and that test is not even available for the Intel card, which means the AMD card’s advantage could be understated in the head-to-head comparison.

The wins tally is 2 for the AMD Radeon PRO W6600 and 0 for the Intel Arc Pro A60 in the recorded head-to-head benchmarks. The overall percentile ranks reinforce this: the W6600 at the 92nd percentile versus the A60 at the 88th. Both cards are above the median GPU, but the AMD card sits four percentile points higher, a meaningful gap when selecting for peak compute performance.

The takeaway from the numbers is consistent. The AMD Radeon PRO W6600 dominates the Intel Arc Pro A60 in every recorded compute benchmark, with margins ranging from 15.8% to 37.2%. The Intel card counters with a larger memory pool, higher bandwidth, and active production status. Those are real advantages for specific use cases, but in the raw performance metrics recorded in the database, the AMD card is the clear winner.