GPU Comparison

The NVIDIA GRID M60-1Q and Intel Arc A370M are two very different graphics cards aimed at entirely different use cases, yet their benchmark scores put them in a similar tier. The GRID M60-1Q is a dual-slot, 225 W server-side accelerator from 2015, while the Arc A370M is a 35 W mobile integrated GPU from 2022. One is designed for datacenter virtualization; the other for thin-and-light laptops. In the single head-to-head benchmark available, the older NVIDIA part wins, but the data reveals a more nuanced picture when you look at their architectures, efficiency, and feature sets.

Where Each One Wins

The GRID M60-1Q wins the only direct comparison available, a Geekbench Vulkan test, scoring 31220 against the Arc A370M's 28673. That is an 8.9% lead for NVIDIA. This matters because Vulkan is a low-level API used heavily in modern games and compute workloads. For any task that relies on Vulkan performance, the GRID M60-1Q is the faster card. Its 76th percentile rank among all GPUs (versus 74th for Intel) also confirms it sits slightly higher in the overall performance distribution.

However, the Arc A370M wins in areas that the benchmark table does not fully capture. It has a second benchmark result (Geekbench OpenCL) of 29676, which is higher than its Vulkan score, suggesting it is more competitive in OpenCL compute workloads. The GRID M60-1Q has no OpenCL score listed, so if you are running OpenCL code, the Intel part has proven capability while the NVIDIA part's performance is unknown. The Arc A370M also supports DirectX 12 Ultimate (12_2) versus the GRID's DirectX 12 (12_1), meaning it is ready for hardware ray tracing and other modern graphics features. The GRID M60-1Q has no RT cores; the Arc A370M has 8.

The practical split is clear: the GRID M60-1Q is the better raw performer for Vulkan-based tasks and sits in a higher overall percentile, while the Arc A370M offers a broader feature set, better efficiency, and a second benchmark data point showing solid OpenCL performance.

Architecture Differences

The GRID M60-1Q uses NVIDIA's GM204 chip on the Maxwell 2.0 architecture, built on a 28 nm process at TSMC. It packs 5,200 million transistors into a 398 mm² die, giving a transistor density of 13.1M per mm². The chip runs at a base clock of 557 MHz and boosts to 1178 MHz. Memory is 1024 MB of GDDR5 on a 256-bit bus, delivering 160.4 GB/s of bandwidth. The card has 2048 shading units, 128 TMUs, and 64 ROPs. It produces a pixel rate of 75.39 GPixel/s, a texture rate of 150.8 GTexel/s, and 4.825 TFLOPS of FP32 compute. It is a dual-slot card measuring 267 mm (10.5 inches) long, requires a 550 W power supply, and draws 225 W.

The Intel Arc A370M uses the DG2-128 chip on the Xe-HPG architecture, built on a 6 nm process, also at TSMC. It has 7,200 million transistors on a much smaller 157 mm² die, yielding a far higher density of 45.9M per mm². Base clock is 1550 MHz, boosting to 2050 MHz. Memory is 4 GB of GDDR6 on a 64-bit bus, with 112.0 GB/s of bandwidth. It has 1024 shading units, 64 TMUs, 32 ROPs, and 8 RT cores. Pixel rate is 65.60 GPixel/s, texture rate is 131.2 GTexel/s, and FP32 compute is 4.198 TFLOPS. It also offers FP16 at 8.397 TFLOPS (2:1), which the GRID does not list. The Arc is an IGP (integrated GPU) with a 35 W TDP, uses PCIe 4.0 x8, and its display outputs are portable-device dependent.

The architectural gap is stark: 28 nm vs 6 nm, 5.2 billion vs 7.2 billion transistors, and a 398 mm² die vs a 157 mm² die. Intel crams more transistors into less space, which is why it achieves comparable compute with a fraction of the power. The GRID's massive die and higher memory bandwidth (160.4 vs 112.0 GB/s) are its advantages. The Arc's 4 GB VRAM is four times the GRID's 1 GB, which is crucial for modern workloads.

Head-to-Head Benchmarks

The only direct benchmark is Geekbench Vulkan. The GRID M60-1Q scores 31220, and the Arc A370M scores 28673. The GRID wins by 8.9%. That is a meaningful margin, roughly equivalent to the gap between the GRID and its nearest rival, the NVIDIA Quadro M5000, which scores 31206 (0% delta). The Arc's nearest rival is the AMD Radeon RX Vega M GH at 29197, a 0.1% difference against Intel (the Arc is slightly behind). So the GRID is performing near the level of a Quadro M5000, while the Arc is essentially tied with a Vega M GH.

Looking at the rivals, the GRID's deltaPct values show it is 0.4% ahead of a GeForce RTX 4070 Ti SUPER (31087), 1% behind an RTX PRO 4500 Blackwell (31532), and 1.4% behind a TITAN RTX (31676). This is remarkable for a 2015 card. The Arc A370M, by contrast, is 0.1% behind a Vega M GH, 0.1% behind a FirePro W8000 (29211), 0.6% ahead of a Radeon RX 470 (28996), and 1% ahead of a Radeon RX 6800M (28874). The Arc's relative position shows it is a solid mid-range mobile part, but it does not punch into the high-end desktop territory that the GRID occupies.

In the Vulkan test, the GRID also has a higher pixel rate (75.39 vs 65.60 GPixel/s) and texture rate (150.8 vs 131.2 GTexel/s), which likely contributes to its win. The Arc's higher clocks (2050 MHz boost vs 1178 MHz) do not overcome the GRID's wider memory bus and higher raw throughput. The GRID's 8.9% victory is consistent with its greater shading units (2048 vs 1024) and higher FP32 (4.825 vs 4.198 TFLOPS).

FAQ

Q: Which card has more VRAM?

A: The Intel Arc A370M has 4 GB of GDDR6, while the NVIDIA GRID M60-1Q has 1 GB of GDDR5. The Arc has four times the memory capacity.

Q: Is the NVIDIA card faster in Vulkan?

A: Yes. The GRID M60-1Q scores 31220 in Geekbench Vulkan, which is 8.9% higher than the Arc A370M's 28673.

Q: Which GPU is more power-efficient?

A: The Arc A370M has a 35 W TDP, while the GRID M60-1Q has a 225 W TDP. The Arc delivers comparable compute with roughly six times less power draw.

Q: Does the Arc support ray tracing?

A: Yes, it has 8 RT cores and supports DirectX 12 Ultimate (12_2). The GRID M60-1Q has no RT cores and only supports DirectX 12 (12_1).

Q: How does the GRID compare to a TITAN RTX?

A: The GRID M60-1Q is 1.4% behind the TITAN RTX in average benchmark score (31220 vs 31676).

Q: What is the Arc's performance relative to a Radeon RX 6800M?

A: The Arc A370M is 1% ahead of the RX 6800M in average score (29175 vs 28874).

The Verdict

Pick the NVIDIA GRID M60-1Q if you need raw Vulkan performance above all else. It wins the head-to-head by 8.9%, sits at the 76th percentile, and is within 1.4% of a TITAN RTX. It also has a wider memory bus (256-bit vs 64-bit) and higher bandwidth (160.4 vs 112.0 GB/s), which matters for certain compute tasks. This is a card for server-side rendering or virtualization where power consumption is not a primary concern.

Pick the Intel Arc A370M if you need modern features, efficiency, or more memory. It has 4 GB of VRAM (vs 1 GB), supports DirectX 12 Ultimate with 8 RT cores, and has a 35 W TDP versus 225 W. Its OpenCL score of 29676 shows it is competitive in that API, and its 74th percentile rank is only two points behind the GRID. The Arc is the practical choice for a laptop or a low-power system where the GRID's dual-slot, 267 mm, 550 W PSU requirement is impossible to accommodate.

The data does not support a universal winner. In the one benchmark they share, NVIDIA is faster. In every metric of efficiency, modernity, and memory capacity, Intel is ahead. Choose based on your workload, not on brand loyalty.

Specification Differences

| Specification | NVIDIA GRID M60-1Q | Intel Arc A370M |

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

| Architecture | Maxwell 2.0 | Xe-HPG |

| Process Node | 28 nm | 6 nm |

| Transistors | 5,200 million | 7,200 million |

| Die Size | 398 mm² | 157 mm² |

| Transistor Density | 13.1M / mm² | 45.9M / mm² |

| Base Clock | 557 MHz | 1550 MHz |

| Boost Clock | 1178 MHz | 2050 MHz |

| Memory Size | 1024 MB | 4 GB |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bus Width | 256 bit | 64 bit |

| Memory Bandwidth | 160.4 GB/s | 112.0 GB/s |

| Shading Units | 2048 | 1024 |

| TMUs | 128 | 64 |

| ROPs | 64 | 32 |

| RT Cores | None | 8 |

| Pixel Rate | 75.39 GPixel/s | 65.60 GPixel/s |

| Texture Rate | 150.8 GTexel/s | 131.2 GTexel/s |

| FP32 Compute | 4.825 TFLOPS | 4.198 TFLOPS |

| FP16 Compute | Not listed | 8.397 TFLOPS (2:1) |

| TDP | 225 W | 35 W |

| Slot Width | Dual-slot | IGP |

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

| Suggested PSU | 550 W | Not listed |

| Bus Interface | PCIe 3.0 x16 | PCIe 4.0 x8 |

| Display Outputs | No outputs | Portable Device Dependent |

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

| Length | 267 mm (10.5 inches) | Not listed |

| Release Date | 2015-08-29 | 2022-03-29 |

| Production Status | End-of-life | End-of-life |