Intel Arc Pro A30M vs NVIDIA GRID M60-1Q Comparison

Intel Arc Pro A30M and NVIDIA GRID M60-1Q occupy the same performance percentile but represent opposite ends of the hardware spectrum. The Arc Pro A30M is a modern, power-efficient mobile solution built on TSMC's 6 nm process, while the GRID M60-1Q is a legacy server-oriented card from 2015 built on 28 nm. Both sit at the 76th percentile against all GPUs, yet their benchmark scores are measured through different APIs, making direct comparisons dependent on the specific test. The data shows a clear split: the GRID M60-1Q leads in raw compute throughput per the Vulkan test, while the Arc Pro A30M matches it in overall percentile ranking with a different architectural approach. There are no shared head-to-head benchmark results, so the analysis relies on their individual scores, nearest rival comparisons, and specification sheets.

The Verdict

From the data, the Intel Arc Pro A30M is the pick for anyone needing a modern feature set with minimal power draw and a compact footprint. Its 50 W TDP, PCIe 4.0 x8 interface, and support for DirectX 12 Ultimate (12_2) position it for current-generation workloads that leverage ray tracing and advanced shading features. The GRID M60-1Q, with its 225 W TDP, dual-slot design, and 1x 8-pin power connector, is a different beast entirely—it is built for datacenter virtualization, not desktop or mobile use. Its lack of display outputs confirms this server-only role.

For raw compute per the available benchmarks, the GRID M60-1Q posts a Geekbench Vulkan score of 31220, while the Arc Pro A30M posts a Geekbench OpenCL score of 31894. These are different tests, so the 674-point gap is not a direct apples-to-apples win. The GRID M60-1Q’s nearest rivals include the NVIDIA Quadro M5000 (31206, 0% delta) and the NVIDIA GeForce RTX 4070 Ti SUPER (31087, 0.4% delta), showing it holds its own against much newer cards. The Arc Pro A30M’s nearest rivals include the NVIDIA TITAN RTX (31676, 0.7% delta) and the AMD Radeon Pro 570X (32176, -0.9% delta), indicating it trades blows with high-end parts from other generations.

Strictly from the data, the Arc Pro A30M is the better choice for a workstation requiring modern API support, low power consumption, and mobile form-factor flexibility. The GRID M60-1Q is the better choice for a vGPU deployment where compute density per card and server integration matter more than efficiency. Neither card wins outright; the decision hinges on the use case. The Arc Pro A30M is end-of-life but still offers contemporary features, while the GRID M60-1Q is also end-of-life but represents older technology. For new projects, the Arc Pro A30M’s architectural advantages—6 nm process, 7,200 million transistors on a 157 mm² die, and 8 ray tracing cores—make it the more future-proof option. For legacy virtualized environments, the GRID M60-1Q’s 2048 shading units and 128 texture mapping units provide a higher theoretical ceiling in certain workloads.

FAQ

Q: Which card has a higher raw benchmark score?

A: The Intel Arc Pro A30M scores 31894 in Geekbench OpenCL, while the NVIDIA GRID M60-1Q scores 31220 in Geekbench Vulkan. These are different API tests, so the comparison is indirect; the Arc leads by 674 points in its test, but the GRID’s score comes from a separate benchmark.

Q: How do these cards compare to their nearest rivals?

A: The Arc Pro A30M is 0.7% ahead of the NVIDIA TITAN RTX (31676) and 1.1% ahead of the NVIDIA RTX PRO 4500 Blackwell (31532), but 0.9% behind the AMD Radeon Pro 570X (32176). The GRID M60-1Q is 0.4% ahead of the GeForce RTX 4070 Ti SUPER (31087) and 0% against the Quadro M5000 (31206), but 1% behind the RTX PRO 4500 Blackwell (31532).

Q: What are the power requirements for each card?

A: The Arc Pro A30M has a 50 W TDP and requires no power connectors. The GRID M60-1Q has a 225 W TDP, uses a 1x 8-pin power connector, and the suggested power supply is 550 W.

Q: Do these cards support modern graphics APIs?

A: The Arc Pro A30M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The GRID M60-1Q supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. Both support OpenGL and Vulkan equally, but the Arc has a higher DirectX feature level.

Q: What is the memory configuration difference?

A: The Arc Pro A30M has 4 GB of GDDR6 on a 64-bit bus with 128.0 GB/s bandwidth. The GRID M60-1Q has 1024 MB (1 GB) of GDDR5 on a 256-bit bus with 160.4 GB/s bandwidth. The GRID has a wider bus and higher bandwidth, but the Arc has four times the capacity and faster memory type.

Q: Which card has more shading units and texture units?

A: The GRID M60-1Q has 2048 shading units and 128 TMUs, compared to the Arc Pro A30M’s 1024 shading units and 64 TMUs. The GRID has exactly double the shading units and double the TMUs.

Architecture Differences

The architectural gap between these two GPUs is generational. The Intel Arc Pro A30M is built on the Xe-HPG architecture, specifically the DG2-128 chip, which belongs to the Alchemist generation for Pro-Series Mobile. It uses a 6 nm process at TSMC, packing 7,200 million transistors into a 157 mm² die, achieving a transistor density of 45.9 million per square millimeter. The architecture includes 8 dedicated ray tracing cores, a feature entirely absent from the GRID M60-1Q. The Arc also supports DirectX 12 Ultimate (12_2), which enables hardware ray tracing and variable rate shading.

The NVIDIA GRID M60-1Q is built on the Maxwell 2.0 architecture, using the GM204 chip from the GRID (Mx) generation. It uses a 28 nm process at TSMC, with 5,200 million transistors on a much larger 398 mm² die, resulting in a transistor density of just 13.1 million per square millimeter. Maxwell 2.0 lacks ray tracing cores and tensor cores, and its DirectX support stops at 12 (12_1), which omits the DXR ray tracing tier. The GRID’s architecture is optimized for virtualized graphics workloads in datacenters, not for client-side rendering with advanced features.

The FP16 capability further separates them. The Arc Pro A30M delivers 8.192 TFLOPS FP16 with a 2:1 ratio against its FP32 output of 4.096 TFLOPS, making it suitable for AI inference and compute tasks that leverage half-precision. The GRID M60-1Q has no listed FP16 performance, indicating it likely operates at FP32 only, where it achieves 4.825 TFLOPS. This makes the Arc more versatile for mixed-precision workloads, while the GRID is a pure FP32 machine.

Specification Differences

The specification sheets diverge significantly across nearly every field. The Arc Pro A30M uses a 64-bit memory bus with 4 GB of GDDR6 at 16 Gbps effective, delivering 128.0 GB/s bandwidth. The GRID M60-1Q uses a 256-bit bus with 1024 MB of GDDR5 at 5 Gbps effective, delivering 160.4 GB/s. The GRID has 25% more bandwidth despite older memory, but the Arc has 4x the capacity.

Clock speeds tell a story of efficiency versus brute force. The Arc runs at a 1500 MHz base and 2000 MHz boost, while the GRID runs at a 557 MHz base and 1178 MHz boost. The Arc’s boost clock is nearly double the GRID’s base clock, reflecting the process node advantage. Pixel and texture rates follow: the Arc achieves 64.00 GPixel/s and 128.0 GTexel/s, while the GRID achieves 75.39 GPixel/s and 150.8 GTexel/s. The GRID wins on both raw throughput metrics due to its higher ROPS (64 vs 32) and TMUs (128 vs 64).

Power and physical design are polar opposites. The Arc is 50 W with no power connectors and a PCIe 4.0 x8 interface, designed for mobile integration. The GRID is 225 W, dual-slot, requires a 1x 8-pin connector, a 550 W suggested power supply, and measures 267 mm (10.5 inches) in length. The GRID uses PCIe 3.0 x16. Display outputs also differ: the Arc is portable-device dependent, while the GRID has no outputs at all. The Arc’s transistor density is 3.5 times higher (45.9M/mm² vs 13.1M/mm²), and its release date is seven years later.

Head-to-Head Benchmarks

There are no direct head-to-head benchmark results in the data, so the comparison relies on each card’s individual Geekbench scores and their nearest rival deltas. The Arc Pro A30M scores 31894 in Geekbench OpenCL. This score places it 0.7% ahead of the NVIDIA TITAN RTX (31676) and 1.1% ahead of the NVIDIA RTX PRO 4500 Blackwell (31532). It trails the AMD Radeon Pro 570X (32176) by 0.9% and the AMD FirePro S10000 (32388) by 1.5%. These deltas show the Arc is competitive with high-end workstation and enthusiast GPUs, despite its low power envelope.

The GRID M60-1Q scores 31220 in Geekbench Vulkan. This is 0% against the NVIDIA Quadro M5000 (31206), effectively a tie, and 0.4% ahead of the GeForce RTX 4070 Ti SUPER (31087). It trails the RTX PRO 4500 Blackwell (31532) by 1% and the TITAN RTX (31676) by 1.4%. The GRID’s scores are remarkably close to the Arc’s, with a 674-point difference, but the different APIs prevent a direct winner. The Arc’s OpenCL score is higher in absolute terms, but the GRID’s Vulkan score is also strong, suggesting both are capable of similar levels of compute performance.

The largest wins are contextual. The Arc wins on memory capacity (4 GB vs 1 GB), clock speed (2000 MHz boost vs 1178 MHz boost), and process efficiency (6 nm vs 28 nm). The GRID wins on shading units (2048 vs 1024), TMUs (128 vs 64), ROPS (64 vs 32), texture rate (150.8 GTexel/s vs 128.0 GTexel/s), and pixel rate (75.39 GPixel/s vs 64.00 GPixel/s). In terms of FP32 compute, the GRID’s 4.825 TFLOPS edges out the Arc’s 4.096 TFLOPS by 17.8%.

Where Each One Wins

The Intel Arc Pro A30M wins in scenarios that demand modern features and efficiency. Its 8 ray tracing cores and DirectX 12 Ultimate support make it the only option for hardware-accelerated ray tracing in this comparison. The 4 GB GDDR6 memory is essential for modern workloads that exceed the 1 GB capacity of the GRID, such as larger textures in 3D rendering or bigger datasets in compute. The 50 W TDP means it can be deployed in thin-and-light mobile workstations without exotic cooling. The 2000 MHz boost clock and 16 Gbps effective memory speed indicate strong sustained performance in bursty workloads. The Arc’s transistor density of 45.9M/mm² suggests a highly optimized design that extracts more performance per watt, a critical factor in battery-powered devices.

The NVIDIA GRID M60-1Q wins in scenarios that prioritize raw throughput and server virtualization. Its 2048 shading units and 128 TMUs provide double the parallel execution resources of the Arc, making it better suited for highly parallel compute tasks that are not memory-capacity-limited. The 160.4 GB/s memory bandwidth, delivered over a 256-bit bus, exceeds the Arc’s 128.0 GB/s, which is advantageous for memory-bandwidth-bound operations. The 225 W TDP and dual-slot design are acceptable in a datacenter rack where power and space are less constrained than in a laptop. The lack of display outputs is a feature for vGPU deployments, where the card serves multiple virtual machines headlessly. The PCIe 3.0 x16 interface provides more lanes than the Arc’s PCIe 4.0 x8, though the latter has higher per-lane bandwidth. The GRID’s FP32 output of 4.825 TFLOPS is the highest raw compute number in this comparison, making it the choice for FP32-only compute workloads where the newer architecture’s features are not required.