AMD Radeon 860M vs Intel Arc Pro B65 Comparison

AMD Radeon 860M and Intel Arc Pro B65 occupy opposite ends of the graphics spectrum. The 860M is an integrated GPU embedded in AMD's Krackan Point chip, built for power-sipping portable devices. The Arc Pro B65 is a discrete, dual-slot professional card with 32 GB of GDDR6 memory, designed for sustained workstation workloads. The recorded data shows two products with fundamentally different missions, and the benchmark results, where available, reflect that divide.

Where Each One Wins

The AMD Radeon 860M wins in the integrated graphics arena by default, as it is the only one of the two that functions as an IGP. Its 15 W TDP and PCIe 4.0 x8 interface make it suitable for thin-and-light laptops where power draw is the primary constraint. The 860M's benchmark scores, with a Geekbench OpenCL score of 22759 and a Geekbench Vulkan score of 30043, give it an average benchmark score of 26401, placing it in the 72nd percentile of all GPUs. That percentile ranking is notable for an integrated part, and the data shows it sits within 1.1% of discrete mobile GPUs like the NVIDIA GeForce MX550 and desktop cards like the AMD Radeon RX 5700 XT 50th Anniversary. In scenarios where space and power are limited, the 860M delivers competitive performance without needing a separate power connector or a dual-slot cooler.

The Intel Arc Pro B65 wins in raw throughput and memory capacity. It uses 2560 shading units, 160 texture mapping units, and 80 render output units, all operating at a fixed 2400 MHz clock. The B65's FP32 compute is rated at 12.29 TFLOPS, which is four times the 860M's 3.072 TFLOPS. Its memory subsystem is a dedicated 32 GB of GDDR6 on a 256-bit bus, yielding 608.0 GB/s of bandwidth, a figure that dwarfs the 860M's system-shared memory. The B65 also maintains a boost clock equal to its base clock, 2400 MHz, indicating a steady-state design intended for prolonged rendering or compute tasks. For professional applications that require large datasets resident in VRAM, the B65 is the clear choice, provided the host system can supply its 200 W TDP and 550 W suggested PSU.

Architecture Differences

The two GPUs come from different architectural lineages. The AMD Radeon 860M uses RDNA 3.5, part of the Navi III IGP generation under the Krackan Point chip. It is fabricated on a 4 nm process at TSMC. The architecture supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The 860M has 512 shading units, 32 TMUs, 16 ROPs, and 8 ray tracing cores. Its FP16 performance matches FP32 at 3.072 TFLOPS, indicating a 1:1 ratio, which is typical for consumer-oriented graphics where full-rate FP16 is not prioritized. The chip's base clock is 600 MHz, boosting to 3000 MHz, and memory is entirely system-shared with bandwidth described as system dependent.

The Intel Arc Pro B65 uses the Xe2-HPG architecture, part of the Battlemage (Pro Series) generation, built on the BMG-G21 chip. It too is fabricated at TSMC, but on a 5 nm process. The die measures 272 mm² and contains 19,600 million transistors, yielding a transistor density of 72.1 million per square millimeter. The B65 has 2560 shading units, 160 TMUs, 80 ROPs, and 20 ray tracing cores. Unlike the 860M, the B65's FP16 throughput is double its FP32 rate, at 24.58 TFLOPS versus 12.29 TFLOPS, reflecting a 2:1 ratio that accelerates compute workloads. The B65's memory runs at 2375 MHz with 19 Gbps effective speed. It also supports the same API set as the 860M: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The B65 uses a PCIe 5.0 x16 interface and provides four DisplayPort 2.1 outputs, whereas the 860M's display outputs are portable device dependent.

Head-to-Head Benchmarks

Direct head-to-head benchmark data is absent from the database, as no shared test scores exist for both products. The 860M has recorded Geekbench OpenCL and Vulkan results, while the B65 has no benchmark entries, yielding an average benchmark score of zero and a 50th percentile rank. The comparison therefore relies on the 860M's nearest rivals and the B65's architectural specifications.

The 860M's average score of 26401 places it 0.1% below the NVIDIA GeForce MX550, 0.3% above the NVIDIA GeForce RTX 5060, 0.6% below the AMD Radeon RX 5700 XT 50th Anniversary, and 1.1% below the NVIDIA RTX A4000. These deltas are all within 1.1 percentage points, meaning the 860M's integrated design competes with a range of discrete GPUs from previous generations. Its Vulkan score of 30043 is significantly higher than its OpenCL score of 22759, indicating that the Vulkan API extracts more from the RDNA 3.5 architecture.

The B65, by contrast, has no measured scores, so its performance must be inferred from its compute rates and memory bandwidth. Its pixel rate is 192.0 GPixel/s, four times the 860M's 48.00 GPixel/s. Its texture rate is 384.0 GTexel/s, also four times the 860M's 96.00 GTexel/s. These figures suggest that in fill-rate-bound workloads, the B65 would deliver roughly four times the throughput. The B65's FP32 output of 12.29 TFLOPS is exactly four times the 860M's 3.072 TFLOPS, and the FP16 output of 24.58 TFLOPS is eight times the 860M's 3.072 TFLOPS. The absence of measured data prevents a direct percentage comparison, but the specification sheet points to a consistent 4x advantage in core compute and a larger gap in memory bandwidth.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The AMD Radeon 860M has an average benchmark score of 26401, while the Intel Arc Pro B65 has no recorded benchmark scores, leaving its average at zero.

Q: What is the memory configuration difference?

A: The 860M uses system-shared memory with system-dependent bandwidth, while the B65 has 32 GB of dedicated GDDR6 memory on a 256-bit bus with 608.0 GB/s bandwidth.

Q: How do the compute capabilities compare?

A: The 860M delivers 3.072 TFLOPS FP32 and 3.072 TFLOPS FP16. The B65 delivers 12.29 TFLOPS FP32 and 24.58 TFLOPS FP16, representing a 4x advantage in FP32 and an 8x advantage in FP16.

Q: Which GPU has a higher clock speed?

A: The Intel Arc Pro B65 runs at a fixed 2400 MHz for both base and boost clocks. The AMD Radeon 860M has a 600 MHz base clock and a 3000 MHz boost clock.

Q: Are both GPUs compatible with the same APIs?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What are the power requirements?

A: The 860M has a 15 W TDP and no power connectors. The B65 has a 200 W TDP, requires a single 8-pin power connector, and has a suggested PSU of 550 W.

The Verdict

The data indicates two distinct purchasing decisions. The AMD Radeon 860M is for portable systems where an integrated GPU must deliver acceptable performance within a 15 W envelope. Its benchmark scores place it in the 72nd percentile of all GPUs, and its nearest rivals include discrete GPUs like the NVIDIA GeForce MX550 and the AMD Radeon RX 5700 XT 50th Anniversary, all within 1.1% of its average score. For users who cannot install a discrete card, the 860M provides a capable baseline for general graphics and light gaming, with a Vulkan score of 30043 that suggests strong API efficiency.

The Intel Arc Pro B65 is for workstation builds that require substantial VRAM and sustained compute. Its 32 GB GDDR6 memory and 608.0 GB/s bandwidth enable large model and texture datasets that the 860M's system-shared memory cannot match. The B65's 4x advantage in pixel rate, texture rate, and FP32 throughput, along with an 8x advantage in FP16, positions it for professional rendering and compute tasks. The 200 W TDP and dual-slot cooler are acceptable trade-offs for that capability. The B65's lack of recorded benchmark scores means its real-world performance is not yet quantified in the database, but its specifications indicate a clear performance tier above the 860M.

Specification Differences

| Specification | AMD Radeon 860M | Intel Arc Pro B65 |

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

| Architecture | RDNA 3.5 | Xe2-HPG |

| Generation | Navi III IGP (Strix Point Mobile) | Battlemage (Pro Series) |

| Process Node | 4 nm | 5 nm |

| Transistors | unknown | 19,600 million |

| Die Size | unknown | 272 mm² |

| Transistor Density | not specified | 72.1M / mm² |

| Base Clock | 600 MHz | 2400 MHz |

| Boost Clock | 3000 MHz | 2400 MHz |

| Memory Size | System Shared | 32 GB |

| Memory Type | System Shared | GDDR6 |

| Memory Bus Width | System Shared | 256 bit |

| Memory Bandwidth | System Dependent | 608.0 GB/s |

| Shading Units | 512 | 2560 |

| TMUs | 32 | 160 |

| ROPs | 16 | 80 |

| Ray Tracing Cores | 8 | 20 |

| Pixel Rate | 48.00 GPixel/s | 192.0 GPixel/s |

| Texture Rate | 96.00 GTexel/s | 384.0 GTexel/s |

| FP32 | 3.072 TFLOPS | 12.29 TFLOPS |

| FP16 | 3.072 TFLOPS (1:1) | 24.58 TFLOPS (2:1) |

| TDP | 15 W | 200 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | 1x 8-pin |

| Suggested PSU | not specified | 550 W |

| Bus Interface | PCIe 4.0 x8 | PCIe 5.0 x16 |

| Display Outputs | Portable Device Dependent | 4x DisplayPort 2.1 |

| Release Date | 2025-02-28 | 2026-03-31 |

| Percentile vs All GPUs | 72 | 50 |

| Average Benchmark Score | 26401 | 0 |