AMD Radeon AI PRO 9600D vs Intel Arc Pro B65 Comparison

Where Each One Wins

The recorded data for the AMD Radeon AI PRO 9600D and the Intel Arc Pro B65 shows a split based on compute and rendering workloads. The AMD Radeon AI PRO 9600D holds the advantage in raw floating-point throughput. Its FP32 performance of 24.82 TFLOPS doubles the Intel Arc Pro B65's 12.29 TFLOPS, which indicates a significant lead in single-precision compute tasks such as simulation, scientific computing, and general GPU compute.

The Intel Arc Pro B65 counters with a higher FP16 rate of 24.58 TFLOPS, achieved through a 2:1 ratio, while the AMD card delivers 24.82 TFLOPS at a 1:1 ratio. This means the Intel card can match the AMD card in half-precision work despite having fewer physical shading units, making it competitive in AI inference or machine learning workloads that rely on FP16 arithmetic.

In memory bandwidth, the Intel Arc Pro B65 edges out the AMD part with 608.0 GB/s compared to 576.0 GB/s. This gives Intel a slight advantage in bandwidth-sensitive tasks like texture-heavy rendering or large dataset streaming. Pixel rate and texture rate are nearly identical: the AMD card posts 193.9 GPixel/s and 387.8 GTexel/s, while the Intel card posts 192.0 GPixel/s and 384.0 GTexel/s. These differences are minimal and unlikely to produce a decisive win in standard rasterization.

The AMD card wins on transistor density and die complexity, with 53,900 million transistors on a 357 mm² die, whereas the Intel card uses 19,600 million transistors on a 272 mm² die. The AMD architecture also includes 48 ray tracing cores versus Intel's 20, which suggests stronger ray tracing throughput per clock. Power consumption favors AMD as well, with a 150 W TDP versus Intel's 200 W, and the AMD card fits in a single-slot form factor while the Intel card requires dual-slot spacing.

Architecture Differences

The AMD Radeon AI PRO 9600D uses the Navi 48 chip built on RDNA 4.0 architecture, fabricated on a 4 nm process at TSMC. The Intel Arc Pro B65 uses the BMG-G21 chip built on Xe2-HPG architecture, fabricated on a 5 nm process at TSMC. The process node difference gives AMD a density advantage: 151.0M transistors per mm² versus Intel's 72.1M transistors per mm².

AMD's chip contains 3,072 shading units, 192 texture mapping units, and 96 render output units. Intel's chip contains 2,560 shading units, 160 TMUs, and 80 ROPs. Both cards feature 32 GB of GDDR6 memory on a 256-bit bus, but the AMD card runs memory at 2250 MHz (18 Gbps effective) while the Intel card runs at 2375 MHz (19 Gbps effective), which explains the bandwidth difference.

Ray tracing hardware differs substantially. The AMD card has 48 RT cores, while the Intel card has 20. Neither card lists tensor cores in the database. The AMD card achieves FP16 at a 1:1 ratio with FP32, while the Intel card achieves FP16 at a 2:1 ratio, meaning Intel packs more half-precision operations per clock.

Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both use PCIe 5.0 x16 interfaces. The AMD card provides a single DisplayPort 2.1a output, while the Intel card provides four DisplayPort 2.1 outputs. The AMD card requires a single 16-pin power connector and a 450 W suggested PSU, while the Intel card uses a single 8-pin connector and a 550 W suggested PSU.

Physical dimensions are recorded for AMD only: 241 mm length, 111 mm height, and 19 mm width. The Intel card's dimensions are not recorded in the database. The AMD card is single-slot and the Intel card is dual-slot.

Release timing differs: the AMD card has a release date of 2025-12-10, while the Intel card has a release date of 2026-03-31. The AMD card lists a predecessor (Radeon Pro Vega) while the Intel card has no predecessor recorded. Both cards are marked as Active in production status.

The Verdict

The database shows the AMD Radeon AI PRO 9600D as the stronger choice for FP32 compute workloads. Its 24.82 TFLOPS is more than double the Intel Arc Pro B65's 12.29 TFLOPS, and it delivers this at a lower TDP of 150 W versus 200 W. The AMD card also carries more RT cores (48 versus 20) and a higher transistor count (53,900 million versus 19,600 million), which points to a more complex and capable compute engine.

The Intel Arc Pro B65 is the better fit for FP16-heavy tasks. Its 24.58 TFLOPS at a 2:1 ratio nearly matches AMD's 24.82 TFLOPS at 1:1, and its 608.0 GB/s bandwidth exceeds AMD's 576.0 GB/s. The Intel card also offers four display outputs versus AMD's single output, which matters for multi-monitor professional setups.

Neither card has an average benchmark score or nearest rivals recorded, so direct performance percentages are unavailable. The percentile versus all GPUs is 50 for both cards, indicating median positioning in the database. The choice reduces to workload type: AMD for FP32 and ray tracing, Intel for FP16 and multi-display configurations.

FAQ

Q: Which card has higher FP32 performance?

A: The AMD Radeon AI PRO 9600D delivers 24.82 TFLOPS FP32, which is more than double the Intel Arc Pro B65's 12.29 TFLOPS.

Q: Do both cards have the same memory capacity?

A: Yes, both the AMD Radeon AI PRO 9600D and the Intel Arc Pro B65 have 32 GB of GDDR6 memory on a 256-bit bus.

Q: Which card provides more display outputs?

A: The Intel Arc Pro B65 provides four DisplayPort 2.1 outputs, while the AMD Radeon AI PRO 9600D provides a single DisplayPort 2.1a output.

Q: How do the power requirements compare?

A: The AMD Radeon AI PRO 9600D has a 150 W TDP and a suggested PSU of 450 W, while the Intel Arc Pro B65 has a 200 W TDP and a suggested PSU of 550 W.

Q: What is the difference in ray tracing hardware?

A: The AMD Radeon AI PRO 9600D has 48 ray tracing cores, while the Intel Arc Pro B65 has 20 ray tracing cores.

Q: Which card has higher memory bandwidth?

A: The Intel Arc Pro B65 has 608.0 GB/s bandwidth, exceeding the AMD Radeon AI PRO 9600D's 576.0 GB/s.

Head-to-Head Benchmarks

The database records no head-to-head benchmark results for these two cards, and neither card has individual benchmark entries. The comparison therefore relies on specification-derived performance indicators. The largest win for the AMD Radeon AI PRO 9600D is in FP32 throughput: 24.82 TFLOPS versus 12.29 TFLOPS, a gap of 12.53 TFLOPS. This is a 101.9% advantage, meaning the AMD card processes more than twice the single-precision operations per second.

The largest win for the Intel Arc Pro B65 is in memory bandwidth: 608.0 GB/s versus 576.0 GB/s, a difference of 32 GB/s or 5.6%. This is a modest edge but can matter in memory-bound scenarios. The Intel card also leads in FP16 throughput when factoring the 2:1 ratio: 24.58 TFLOPS versus AMD's 24.82 TFLOPS at 1:1. While AMD still holds a slight numerical lead of 0.24 TFLOPS, the Intel card achieves this with fewer shading units (2,560 versus 3,072), indicating higher efficiency per unit.

Pixel rate shows near parity: AMD at 193.9 GPixel/s and Intel at 192.0 GPixel/s, a difference of 1.9 GPixel/s in AMD's favor. Texture rate follows the same pattern: AMD at 387.8 GTexel/s and Intel at 384.0 GTexel/s, a difference of 3.8 GTexel/s. These margins are within 1%, suggesting rasterization performance would be nearly indistinguishable in practice.

Clock speeds favor Intel on paper: the Intel card has a base and boost clock of 2400 MHz, while the AMD card has a base clock of 1080 MHz and a boost clock of 2020 MHz. However, the AMD card compensates with more shading units and TMUs, which explains the comparable pixel and texture rates despite lower clocks.

Specification Differences

The two cards differ across several specification fields. The process node is 4 nm for AMD and 5 nm for Intel. Transistor count is 53,900 million for AMD and 19,600 million for Intel. Die size is 357 mm² for AMD and 272 mm² for Intel. Transistor density is 151.0M per mm² for AMD and 72.1M per mm² for Intel.

Clocks differ: AMD has a base clock of 1080 MHz and boost of 2020 MHz, while Intel has both base and boost at 2400 MHz. Memory clocks are 2250 MHz (18 Gbps effective) for AMD and 2375 MHz (19 Gbps effective) for Intel. Bandwidth is 576.0 GB/s for AMD and 608.0 GB/s for Intel.

Compute units differ: AMD has 3,072 shading units, 192 TMUs, 96 ROPs, and 48 RT cores. Intel has 2,560 shading units, 160 TMUs, 80 ROPs, and 20 RT cores. FP32 is 24.82 TFLOPS for AMD and 12.29 TFLOPS for Intel. FP16 is 24.82 TFLOPS at 1:1 for AMD and 24.58 TFLOPS at 2:1 for Intel.

TDP is 150 W for AMD and 200 W for Intel. Slot width is single-slot for AMD and dual-slot for Intel. Power connectors are 1x 16-pin for AMD and 1x 8-pin for Intel. Suggested PSU is 450 W for AMD and 550 W for Intel.

Display outputs are 1x DisplayPort 2.1a for AMD and 4x DisplayPort 2.1 for Intel. Dimensions are recorded only for AMD: 241 mm length, 111 mm height, 19 mm width. Release dates are 2025-12-10 for AMD and 2026-03-31 for Intel. The AMD card has a predecessor (Radeon Pro Vega), while the Intel card has none. Both share the same bus interface (PCIe 5.0 x16) and API support (DirectX 12 Ultimate, OpenGL 4.6, Vulkan 1.4).