Intel Arc Graphics 64EU Mobile vs Intel Arc Pro B65 Comparison

Head-to-Head Benchmarks

The recorded data contains no direct head-to-head benchmark results between the Intel Arc Graphics 64EU Mobile and the Intel Arc Pro B65. Both entries have empty benchmark arrays, zero average benchmark scores, and no nearest rival comparisons. The database shows a percentile rank of 50 for both GPUs, indicating they sit at the median of all recorded graphics processors, but this is a positional figure rather than a performance measurement.

What the data does permit is a comparison of theoretical compute ceilings derived from each part's specified clock rates and shader configuration. The Arc Pro B65 reaches 12.29 TFLOPS of FP32 throughput, while the Arc Graphics 64EU Mobile delivers 1.792 TFLOPS. This is a 6.86x gap in raw single-precision compute. In FP16, the B65 records 24.58 TFLOPS versus 3.584 TFLOPS on the mobile part, again a 6.86x difference, since both architectures use a 2:1 ratio for FP16 to FP32.

Pixel throughput tells a similar story. The Arc Pro B65 sustains 192.0 GPixel/s, while the mobile chip manages 28.00 GPixel/s, a 6.86x difference. Texture rate follows the same multiplier: 384.0 GTexel/s against 56.00 GTexel/s. These ratios are consistent because the B65 scales every relevant unit (shading units, TMUs, ROPs) by roughly the same factor relative to the 64EU part.

Clock behavior differentiates the two. The Arc Pro B65 runs at a fixed 2400 MHz for both base and boost, meaning no thermal or power headroom variability in the specification. The mobile chip has a 300 MHz base and a 1750 MHz boost, so its actual operating point depends heavily on the host system's thermal envelope. The B65's constant clock suggests a workstation-oriented stability profile, while the mobile part's wide clock range indicates adaptive performance.

FAQ

Q: Which GPU has a higher FP32 compute throughput?

A: The Intel Arc Pro B65 records 12.29 TFLOPS of FP32, which is 6.86 times the 1.792 TFLOPS delivered by the Intel Arc Graphics 64EU Mobile.

Q: What memory configuration does each GPU use?

A: The Arc Pro B65 uses 32 GB of GDDR6 on a 256-bit bus with 608.0 GB/s of bandwidth. The Arc Graphics 64EU Mobile uses system shared memory, with bandwidth listed as system dependent and memory clock tied to the host's shared memory controller.

Q: How do the shading unit counts compare?

A: The Arc Pro B65 has 2560 shading units, while the Arc Graphics 64EU Mobile has 512. The B65 also has 160 texture mapping units versus 32, and 80 ROPs versus 16.

Q: Do both GPUs support the same DirectX version?

A: No. The Arc Pro B65 supports DirectX 12 Ultimate (12_2), while the Arc Graphics 64EU Mobile supports DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4.

Q: What is the power draw specification for each?

A: The Arc Pro B65 has a TDP of 200 W and requires a 550 W suggested power supply with a single 8-pin connector. The Arc Graphics 64EU Mobile has a TDP of 65 W and uses the system's integrated power delivery, with no separate power connector listed.

Q: Which GPU has a larger process node?

A: The Arc Graphics 64EU Mobile uses Intel's 10 nm process, while the Arc Pro B65 uses a 5 nm process fabricated by TSMC. The B65's die measures 272 mm² and contains 19,600 million transistors, giving a density of 72.1M transistors per mm².

The Verdict

The Arc Pro B65 is the clear performance leader across every measurable compute metric in the database. Its FP32 output of 12.29 TFLOPS dwarfs the mobile part's 1.792 TFLOPS, and its fixed 2400 MHz clock eliminates boost variability. The B65 also brings dedicated memory: 32 GB of GDDR6 with 608.0 GB/s bandwidth, whereas the mobile chip shares system memory with bandwidth dependent on the host platform.

The Arc Graphics 64EU Mobile is not a performance competitor; it is a 65 W integrated solution for portable devices. Its 512 shading units and 28.00 GPixel/s pixel rate suit basic graphics workloads, not the professional rendering tasks the B65 targets. The B65's 20 ray tracing cores provide hardware RT support that the mobile chip lacks entirely, a decisive feature gap for modern 3D workloads.

For a system builder needing sustained compute, the B65's dual-slot form factor, 200 W TDP, and PCIe 5.0 x16 interface position it as a dedicated workstation accelerator. The mobile chip's ring bus interface and IGP slot class indicate it belongs in laptops or compact systems where discrete graphics are impractical. The data supports only one conclusion: choose the B65 for compute-heavy professional work, choose the 64EU Mobile for integrated graphics in portable devices.

Specification Differences

| Field | Intel Arc Graphics 64EU Mobile | Intel Arc Pro B65 |

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

| Architecture | Xe-LPG | Xe2-HPG |

| Process node | 10 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | Not specified | 19,600 million |

| Die size | Not specified | 272 mm² |

| Transistor density | Not specified | 72.1M / mm² |

| Base clock | 300 MHz | 2400 MHz |

| Boost clock | 1750 MHz | 2400 MHz |

| Memory size | System Shared | 32 GB 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 |

| RT cores | Not specified | 20 |

| Pixel rate | 28.00 GPixel/s | 192.0 GPixel/s |

| Texture rate | 56.00 GTexel/s | 384.0 GTexel/s |

| FP32 | 1.792 TFLOPS | 12.29 TFLOPS |

| FP16 | 3.584 TFLOPS (2:1) | 24.58 TFLOPS (2:1) |

| TDP | 65 W | 200 W |

| Slot width | IGP | Dual-slot |

| Power connectors | Not specified | 1x 8-pin |

| Suggested PSU | Not specified | 550 W |

| Bus interface | Ring Bus | PCIe 5.0 x16 |

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

| DirectX | 12 (12_1) | 12 Ultimate (12_2) |

| Release date | 2023-12-13 | 2026-03-31 |

Architecture Differences

The two GPUs come from different Intel generations and design philosophies. The Arc Graphics 64EU Mobile uses the Xe-LPG architecture on Meteor Lake, built on Intel's 10 nm process. The Arc Pro B65 uses Xe2-HPG on the Battlemage Pro Series die (BMG-G21), fabricated by TSMC on a 5 nm process. This node transition accounts for the B65's ability to pack 19,600 million transistors into a 272 mm² die.

Shader configuration diverges sharply. The B65 has 2560 shading units, 160 TMUs, and 80 ROPs; the mobile part has 512, 32, and 16 respectively. The B65 also adds 20 dedicated ray tracing cores, a feature absent from the mobile chip's specification. Neither part lists tensor cores, so AI acceleration through dedicated tensor hardware is not recorded for either.

Memory architecture is fundamentally different. The mobile chip uses system shared memory, meaning its bus width and bandwidth are inherited from the host CPU's memory controller. The B65 carries its own 32 GB GDDR6 pool on a 256-bit interface, achieving 608.0 GB/s regardless of host memory. The mobile part's memory clock is listed as system shared, while the B65 runs at 2375 MHz with 19 Gbps effective transfer.

Clock behavior reflects their use cases. The mobile chip's 300 MHz base and 1750 MHz boost allow it to idle at low power and ramp under load, constrained by the 65 W TDP. The B65 holds a constant 2400 MHz base and boost, prioritizing consistent throughput over power savings. The B65's 200 W TDP and dual-slot cooler confirm it is designed for sustained load, not mobile operation.

API support overlaps partially. Both support OpenGL 4.6 and Vulkan 1.4, but the B65 reaches DirectX 12 Ultimate (12_2), while the mobile part stops at DirectX 12 (12_1). Display output also differs: the B65 provides 4x DisplayPort 2.1, while the mobile chip's outputs depend entirely on the portable device it is embedded in.

Where Each One Wins

The Arc Pro B65 wins in every compute-heavy scenario the database records. Its 12.29 TFLOPS FP32 throughput handles double-precision-adjacent workloads, AI inference at 24.58 TFLOPS FP16, and high-resolution rendering with a 192.0 GPixel/s pixel rate. The 20 ray tracing cores give it hardware-accelerated RT for CAD visualization, architectural previews, and media production. The 32 GB GDDR6 frame buffer with 608.0 GB/s bandwidth supports large scenes and multi-display output through 4x DisplayPort 2.1. The fixed 2400 MHz clock ensures predictable performance in long-running batch renders or simulation loops.

The Arc Graphics 64EU Mobile wins in power envelope and integration. Its 65 W TDP is a fraction of the B65's 200 W, making it viable for thin laptops and compact systems where a dual-slot card cannot fit. The IGP form factor and ring bus interface require no additional power connector or PSU upgrade. Its system shared memory model simplifies design, removing the cost and space of dedicated VRAM. For lightweight 2D workloads, video playback, and basic 3D acceleration, the mobile chip's 1.792 TFLOPS is sufficient. Its 512 shading units and 16 ROPs align with entry-level productivity rather than professional rendering.

The release timeline also separates their relevance. The mobile part launched in 2023, the B65 in 2026, so the B65 represents a later architecture generation with the Xe2-HPG design. The mobile chip's predecessor is listed as HD Graphics-M, while the B65 has no recorded predecessor, indicating a new product line entry. Both remain in active production status.

In practice, the B65 suits fixed workstations with 550 W power supplies, PCIe 5.0 x16 slots, and a need for dedicated memory bandwidth. The mobile chip suits portable devices prioritizing battery life, thermal headroom, and physical size. The data does not support using the mobile part for any workload that stresses FP32, FP16, pixel fill, or texture sampling, as the B65 outperforms it by a factor of 6.86 in each of these metrics.