Intel Arc Pro A60 vs Intel Arc Pro B65 Comparison

Head-to-Head Benchmarks

The recorded data presents an unusual comparison situation. The Intel Arc Pro A60 has two benchmark scores in the database, while the Intel Arc Pro B65 currently has no recorded benchmark results. This makes a direct head-to-head comparison of measured performance impossible at this time. The A60 delivers a Geekbench OpenCL score of 63485 and a Geekbench Vulkan score of 57166, producing an average benchmark score of 60326. The B65, by contrast, holds an average benchmark score of 0, with no entries in its benchmark list.

Examining the A60's position against its nearest rivals provides useful context for its performance tier. The A60's average score sits within a narrow band of comparable GPUs. The NVIDIA GeForce RTX 4090 posts an average score of 60347, which is essentially identical to the A60's 60326, a delta of 0 percent. The AMD Radeon Pro Vega 48 scores 60140, just 0.3 percent behind the A60. The AMD Radeon Pro W6600M leads the A60 by 2.5 percent with a score of 61896. The AMD Radeon PRO V710 trails by 2.8 percent, scoring 58657. This clustering indicates that the A60 occupies a well-defined performance band, with all four nearest rivals within roughly 3 percent of its average score.

The A60 also holds an 88th percentile ranking against all GPUs in the database. This places it firmly in the upper tier of recorded hardware. The B65's 50th percentile ranking and zero average score reflect the absence of measurement data rather than a determination of its capabilities. The database contains no head-to-head benchmark entries between these two cards, and neither card has a recorded win count in direct competition.

The Verdict

From the available data, the Intel Arc Pro A60 is the only card in this comparison with verified performance measurements. Its average benchmark score of 60326 places it alongside the NVIDIA GeForce RTX 4090 within a 0 percent delta, a notable positioning given the different market segments these products typically occupy. The A60's 88th percentile ranking across all GPUs confirms that its measured performance is competitive with high-end hardware.

The Intel Arc Pro B65 cannot be evaluated on performance terms from this database. Its zero average score and empty benchmark list mean that no empirical comparison is possible. Users seeking a card with confirmed, quantified benchmark results would find the A60 the only option with data supporting its performance claims. The B65 may offer architectural or specification advantages, which are discussed in later sections, but its actual benchmark behavior remains unmeasured.

The verdict from the data is straightforward: the A60 has proven, recorded performance. The B65 does not. Any decision between these two cards based on benchmark scores must favor the A60 solely because it has scores at all. The B65's 50th percentile ranking, contrasted with the A60's 88th, further illustrates the gap in available measurement data.

Architecture Differences

The two cards come from different Intel architectures. The A60 uses the DG2-256 chip built on the Xe-HPG architecture, belonging to the Alchemist (Pro Series) generation. The B65 uses the BMG-G21 chip built on the Xe2-HPG architecture, belonging to the Battlemage (Pro Series) generation. This generational shift brings substantial changes to the underlying design.

The manufacturing process differs between the two. The A60 is fabricated on a 6 nm process at TSMC, while the B65 uses a 5 nm process, also at TSMC. The transistor counts reflect this advancement: the A60 integrates 11,500 million transistors on a 269 mm² die, while the B65 packs 19,600 million transistors onto a 272 mm² die. The transistor density tells a stark story: the A60 achieves 42.8 million transistors per square millimeter, while the B65 reaches 72.1 million per square millimeter. Despite nearly identical die sizes, the B65 fits 70 percent more transistors into roughly the same area, a direct consequence of the denser 5 nm node.

The A60's architecture provides 2048 shading units, 128 texture mapping units, and 64 raster output units. It includes 16 ray tracing cores. The B65 expands on each of these: 2560 shading units, 160 TMUs, 80 ROPs, and 20 ray tracing cores. These increases scale with the larger transistor budget, giving the B65 a 25 percent advantage in shading units and ray tracing cores.

Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither card lists tensor cores in the database. The A60's memory operates at 2000 MHz with 16 Gbps effective transfer, while the B65's memory runs at 2375 MHz with 19 Gbps effective. The B65's higher memory clock aligns with its wider memory interface and larger capacity.

Specification Differences

Several specification fields differ between the two cards, and these differences are more concrete than benchmark comparisons. The A60 carries 12 GB of GDDR6 memory on a 192 bit bus, delivering 384.0 GB/s of bandwidth. The B65 carries 32 GB of GDDR6 memory on a 256 bit bus, delivering 608.0 GB/s of bandwidth. The B65 offers substantially more memory capacity and 58 percent more memory bandwidth.

Clock speeds differ significantly. The A60 has a base clock of 900 MHz and a boost clock of 2050 MHz. The B65 runs at a flat 2400 MHz for both base and boost, meaning it does not rely on dynamic boosting above its base frequency. The B65's sustained clock is 17 percent higher than the A60's boost clock.

Compute throughput follows the architectural and clock differences. The A60 delivers 8.397 TFLOPS of FP32 performance and 16.79 TFLOPS of FP16 performance. The B65 delivers 12.29 TFLOPS of FP32 and 24.58 TFLOPS of FP16, representing a 46 percent advantage in both precision formats. Pixel fill rates are 131.2 GPixel/s for the A60 versus 192.0 GPixel/s for the B65. Texture fill rates are 262.4 GTexel/s versus 384.0 GTexel/s.

Power requirements and physical specifications also diverge. The A60 has a 130 W TDP and requires a 300 W suggested power supply. It is a single-slot card. The B65 has a 200 W TDP and requires a 550 W suggested power supply. It is a dual-slot card with a single 8-pin power connector. The A60 lists no power connector in the database.

The A60 uses a PCIe 4.0 x16 bus interface, while the B65 uses PCIe 5.0 x16. Both cards provide four DisplayPort outputs, but the A60 features DisplayPort 2.0 while the B65 features DisplayPort 2.1. The A60 was released on June 5, 2023; the B65's release date is recorded as March 31, 2026. Both cards are listed as Active in production status. Neither card has a recorded launch MSRP.

FAQ

Q: Which card has higher benchmark scores?

A: The Intel Arc Pro A60 has recorded benchmark scores of 63485 in Geekbench OpenCL and 57166 in Geekbench Vulkan. The Intel Arc Pro B65 has no recorded benchmark scores in the database.

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

A: The A60's average score of 60326 is within 0 percent of the NVIDIA GeForce RTX 4090 (60347), 0.3 percent ahead of the AMD Radeon Pro Vega 48 (60140), 2.5 percent behind the AMD Radeon Pro W6600M (61896), and 2.8 percent ahead of the AMD Radeon PRO V710 (58657).

Q: What are the memory capacity differences?

A: The A60 has 12 GB of GDDR6 memory on a 192 bit bus with 384.0 GB/s bandwidth. The B65 has 32 GB of GDDR6 memory on a 256 bit bus with 608.0 GB/s bandwidth.

Q: Do the cards use the same architecture?

A: No. The A60 uses the DG2-256 chip on the Xe-HPG architecture from the Alchemist generation. The B65 uses the BMG-G21 chip on the Xe2-HPG architecture from the Battlemage generation.

Q: How do the power requirements differ?

A: The A60 has a 130 W TDP with a suggested 300 W power supply and a single-slot design. The B65 has a 200 W TDP with a suggested 550 W power supply, a dual-slot design, and one 8-pin power connector.

Q: Which card has higher compute throughput?

A: The B65 delivers 12.29 TFLOPS FP32 and 24.58 TFLOPS FP16, compared to the A60's 8.397 TFLOPS FP32 and 16.79 TFLOPS FP16. The B65 is 46 percent higher in both FP32 and FP16 throughput.

Where Each One Wins

The A60 wins in verified performance data. It is the only card of the two with benchmark scores in the database, and those scores place it at the 88th percentile of all GPUs. Its average score of 60326 puts it in the same performance class as the NVIDIA GeForce RTX 4090, with a 0 percent delta. For any workload where measured performance is the deciding factor, the A60 is the only option with supporting evidence.

The A60 also wins on power efficiency as expressed by TDP. It requires 130 W versus the B65's 200 W, and its suggested power supply is 300 W versus 550 W. The A60's single-slot design offers a smaller physical footprint than the B65's dual-slot configuration. These factors favor the A60 in space-constrained or power-limited environments.

The B65 wins on raw specification counts across nearly every hardware category. Its 32 GB memory capacity is nearly triple the A60's 12 GB. Its 608.0 GB/s bandwidth exceeds the A60's 384.0 GB/s by 58 percent. Its FP32 throughput of 12.29 TFLOPS is 46 percent higher than the A60's 8.397 TFLOPS. Its shading unit count of 2560 surpasses the A60's 2048, and its 20 ray tracing cores exceed the A60's 16. Its flat 2400 MHz clock exceeds the A60's 2050 MHz boost clock, and its 5 nm process node with 72.1M transistors per square millimeter represents a denser, newer manufacturing technology.

The B65 further wins on interface modernity. It uses PCIe 5.0 x16 versus the A60's PCIe 4.0 x16, and DisplayPort 2.1 versus DisplayPort 2.0. The B65's 200 W TDP is higher, but it also supports a larger feature set. The B65's release date of March 31, 2026, places it as a newer product than the A60's June 5, 2023 release.

The data presents a clear split. The A60 wins where measurement exists, efficiency matters, and physical size is a constraint. The B65 wins where raw specifications, memory capacity, bandwidth, and interface versions are the priority. Whether the B65's specification advantages translate into benchmark performance remains unmeasured in the database. The A60's verified scores give it an evidence-backed position, while the B65's potential rests on its architectural and specification sheet alone.