Intel Arc Pro A60M vs NVIDIA Rubin GPU Comparison

Intel Arc Pro A60M and NVIDIA Rubin GPU occupy opposite ends of the hardware spectrum. One is a mobile professional graphics part aimed at portable workstations; the other is a massive server accelerator designed for data center compute. The database records no direct benchmark comparisons between them, so the analysis below relies strictly on their recorded specifications and architectural characteristics.

Where Each One Wins

The Intel Arc Pro A60M is built for on-device graphics rendering and display output. Its specification sheet shows a 128-bit memory bus with 8 GB of GDDR6 memory delivering 256.0 GB/s of bandwidth. That configuration suits tasks like CAD viewport rendering, video encoding, and lightweight 3D modeling on a laptop. The part carries a 95 W TDP, making it suitable for integrated graphics packages (the slot width is listed as IGP). It also supports PCIe 4.0 x16, which is a standard interface for mobile workstations. The presence of 64 ROPs and 128 texture mapping units gives it a balanced pixel and texture throughput for conventional rasterization workloads.

The NVIDIA Rubin GPU, by contrast, is a server-class compute engine. It uses 288 GB of HBM4 memory across a 16384-bit bus, yielding 22.1 TB/s of memory bandwidth. That is the highest bandwidth figure in the database, and it dwarfs the Intel part’s 256.0 GB/s. The Rubin GPU also has 28672 shading units and 896 tensor cores, which point toward massive parallel compute for AI training, scientific simulation, and large-scale data processing. Its 896 texture mapping units provide a 2,031.2 GTexel/s texture rate, which is over 12 times the Intel part’s 166.4 GTexel/s. The Rubin GPU has no display outputs, confirming its role as a compute accelerator rather than a graphics card.

In terms of raw computational throughput, the Rubin GPU wins decisively. Its FP32 performance is 130.0 TFLOPS versus 5.325 TFLOPS for the Intel Arc Pro A60M. That is roughly a 24.4 times difference based on the recorded numbers. The Rubin GPU also delivers 260.0 TFLOPS for FP16 (2:1 ratio), while the Intel part manages 10.65 TFLOPS for FP16 (2:1). For FP16-heavy workloads like deep learning inference, the Rubin GPU provides about 24.4 times the throughput.

The Intel part wins in areas related to graphics output and power efficiency. It supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, making it compatible with modern graphics APIs. The Rubin GPU lists N/A for all graphics APIs, so it cannot run traditional gaming or CAD applications that rely on those interfaces. The Intel Arc Pro A60M also has a much lower TDP: 95 W versus 2300 W. That difference allows the Intel part to operate in portable devices, while the Rubin GPU requires a suggested PSU of 2700 W and uses an SXM Module slot.

Architecture Differences

The two GPUs come from different architecture families. The Intel Arc Pro A60M uses the Xe-HPG architecture, built on the DG2-256 chip, and belongs to the Alchemist generation for Pro-Series Mobile. The process node is 6 nm from TSMC, with 11,500 million transistors on a 269 mm² die. That yields a transistor density of 42.8M per mm². The NVIDIA Rubin GPU uses the Rubin architecture on the GR100 chip, from the Server Rubin (Rxx) generation. It is fabricated on a 3 nm process from TSMC, with 336,000 million transistors on a 1456 mm² die. That gives a transistor density of 230.8M per mm², which is more than five times higher than the Intel part.

The die size difference is substantial: 1456 mm² versus 269 mm². That makes the Rubin GPU one of the largest chips in the database, while the Intel part is a modest mobile chip. The transistor count difference is even more pronounced: 336,000 million versus 11,500 million. The Rubin GPU packs nearly 30 times more transistors into a die that is about 5.4 times larger.

Memory architecture also diverges sharply. The Intel Arc Pro A60M uses GDDR6 memory with a 128-bit bus. The NVIDIA Rubin GPU uses HBM4 with a 16384-bit bus. HBM4 is a stacked memory type that provides extremely high bandwidth per pin, and the bus width of 16384 bits is enormous compared to the Intel part’s 128 bits. The memory clock speeds differ as well: the Intel part runs at 2000 MHz with 16 Gbps effective data rate, while the Rubin GPU operates at 2695 MHz with 10.8 Gbps effective. Despite the lower effective data rate per pin, the Rubin GPU’s much wider bus gives it a massive bandwidth advantage.

The compute resources are structured differently. The Intel Arc Pro A60M has 2048 shading units, 128 TMUs, 64 ROPs, and 16 ray tracing cores. The NVIDIA Rubin GPU has 28672 shading units, 896 TMUs, 24 ROPs, and 896 tensor cores. The Rubin GPU has no listed ray tracing cores, but its tensor core count of 896 is a key differentiator. Tensor cores are specialized for matrix operations used in AI and machine learning. The Intel part has no tensor cores listed. The ROP count is unusual: the Rubin GPU has only 24 ROPs versus 64 for the Intel part. That suggests the Rubin GPU is not optimized for pixel output, consistent with its lack of display outputs.

Clock speeds also differ. The Intel Arc Pro A60M has a base clock of 900 MHz and a boost clock of 1300 MHz. The NVIDIA Rubin GPU has a base clock of 700 MHz and a boost clock of 2267 MHz. The Rubin GPU’s boost clock is significantly higher, which helps it reach 130.0 TFLOPS despite having a lower base clock.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark results for these two GPUs. The wins and losses fields show zero for both parts. That means no direct performance comparison is available from recorded measurements. However, the specification data allows for indirect comparisons based on theoretical throughput.

The most striking difference is in FP32 compute. The NVIDIA Rubin GPU records 130.0 TFLOPS, while the Intel Arc Pro A60M records 5.325 TFLOPS. That puts the Rubin GPU at roughly 24.4 times the FP32 throughput. For FP16, the Rubin GPU achieves 260.0 TFLOPS versus 10.65 TFLOPS for the Intel part, again about a 24.4 times difference. These ratios are consistent because both parts list their FP16 performance as 2:1 relative to FP32.

Texture rate shows a similar pattern. The Rubin GPU delivers 2,031.2 GTexel/s, while the Intel part delivers 166.4 GTexel/s. That is about 12.2 times higher. The pixel rate, however, tells a different story. The Intel Arc Pro A60M records 83.20 GPixel/s, while the NVIDIA Rubin GPU records 54.41 GPixel/s. The Intel part is about 1.5 times faster in pixel rate, which aligns with its higher ROP count (64 versus 24). This indicates that for pure rasterization output, the Intel part is more capable, despite its overall lower compute power.

Memory bandwidth is where the Rubin GPU dominates most clearly. Its 22.1 TB/s is about 86.3 times the Intel part’s 256.0 GB/s. That bandwidth is essential for feeding the Rubin GPU’s massive compute array and its large memory pool. The Intel part’s 8 GB memory capacity is tiny compared to the Rubin GPU’s 288 GB, a 36 times difference.

The pixel rate result is the only major specification where the Intel part leads. This is likely because the Intel part has dedicated ROPs and a graphics-oriented pipeline, while the Rubin GPU is designed for compute and uses a minimal number of ROPs. The data shows that raw pixel throughput does not correlate with overall compute performance in these two parts.

FAQ

Q: What is the memory capacity difference between the two GPUs?

A: The Intel Arc Pro A60M has 8 GB of GDDR6 memory, while the NVIDIA Rubin GPU has 288 GB of HBM4 memory. The Rubin GPU offers 36 times more memory capacity.

Q: Which GPU has higher FP32 compute performance?

A: The NVIDIA Rubin GPU records 130.0 TFLOPS FP32, compared to 5.325 TFLOPS for the Intel Arc Pro A60M. That makes the Rubin GPU about 24.4 times faster in FP32.

Q: Can the NVIDIA Rubin GPU be used for traditional graphics rendering?

A: No. The Rubin GPU lists N/A for DirectX, OpenGL, and Vulkan support, and it has no display outputs. The Intel Arc Pro A60M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What is the TDP of each GPU?

A: The Intel Arc Pro A60M has a TDP of 95 W, while the NVIDIA Rubin GPU has a TDP of 2300 W. The Rubin GPU also requires a suggested PSU of 2700 W.

Q: Which GPU has a higher pixel rate?

A: The Intel Arc Pro A60M has a pixel rate of 83.20 GPixel/s, which is higher than the NVIDIA Rubin GPU’s 54.41 GPixel/s. This is due to the Intel part having 64 ROPs versus 24 on the Rubin GPU.

Q: What are the production statuses and release dates?

A: Both parts are listed as Active in production. The Intel Arc Pro A60M was released on 2023-06-05, while the NVIDIA Rubin GPU has a release date of 2025-12-31.

Specification Differences

The two GPUs differ in nearly every major specification field. The table below highlights the key differences:

| Specification | Intel Arc Pro A60M | NVIDIA Rubin GPU |

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

| Chip | DG2-256 | GR100 |

| Architecture | Xe-HPG | Rubin |

| Generation | Alchemist (Pro-Series Mobile) | Server Rubin (Rxx) |

| Process Node | 6 nm | 3 nm |

| Transistors | 11,500 million | 336,000 million |

| Die Size | 269 mm² | 1456 mm² |

| Transistor Density | 42.8M / mm² | 230.8M / mm² |

| Base Clock | 900 MHz | 700 MHz |

| Boost Clock | 1300 MHz | 2267 MHz |

| Memory Size | 8 GB | 288 GB |

| Memory Type | GDDR6 | HBM4 |

| Memory Bus | 128 bit | 16384 bit |

| Memory Bandwidth | 256.0 GB/s | 22.1 TB/s |

| Shading Units | 2048 | 28672 |

| TMUs | 128 | 896 |

| ROPs | 64 | 24 |

| Tensor Cores | N/A | 896 |

| Pixel Rate | 83.20 GPixel/s | 54.41 GPixel/s |

| Texture Rate | 166.4 GTexel/s | 2,031.2 GTexel/s |

| FP32 | 5.325 TFLOPS | 130.0 TFLOPS |

| FP16 | 10.65 TFLOPS (2:1) | 260.0 TFLOPS (2:1) |

| TDP | 95 W | 2300 W |

| Slot Width | IGP | SXM Module |

| Suggested PSU | N/A | 2700 W |

| Bus Interface | PCIe 4.0 x16 | PCIe 6.0 x16 |

| Display Outputs | Portable Device Dependent | No outputs |

| DirectX | 12 Ultimate (12_2) | N/A |

| OpenGL | 4.6 | N/A |

| Vulkan | 1.4 | N/A |

| Release Date | 2023-06-05 | 2025-12-31 |

| Predecessor | N/A | Server Blackwell |

The Intel part is a mobile graphics solution with a lower transistor count, smaller die, and a graphics-oriented feature set. The NVIDIA part is a server accelerator with an enormous transistor budget, high-bandwidth memory, and compute-focused resources. The two parts share only their manufacturer status (both are from different manufacturers, Intel and NVIDIA) and their production status (both Active). The data confirms they serve entirely different market segments with no overlap in typical use cases.