Intel HD Graphics 630 vs NVIDIA GeForce MX110 Comparison

Intel HD Graphics 630 and NVIDIA GeForce MX110 represent two fundamentally different approaches to integrated and entry-level discrete graphics, yet their benchmark averages place them within striking distance of each other. The Intel part, built on a 14 nm++ process as part of the Kaby Lake GT2 die, delivers an average benchmark score of 4075, while the MX110, a 28 nm Maxwell-based discrete chip from TSMC, averages 3834. This 6.3% gap in favor of the Intel solution is notable, but the individual workload results tell a more nuanced story: the MX110 wins Geekbench OpenCL by 15.7%, while the HD 630 counters with a 3.7% victory in Geekbench Vulkan. Both products are end-of-life, with the Intel part launching in August 2016 and the NVIDIA part arriving in November 2017.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The Intel HD Graphics 630, with an average benchmark score of 4075, outperforms the NVIDIA GeForce MX110’s 3834 average. This places the Intel part at the 24th percentile of all GPUs, versus the 23rd percentile for the MX110.

Q: How do the two compare in Geekbench OpenCL performance?

A: The NVIDIA GeForce MX110 is decisively ahead, scoring 4255 versus the Intel HD Graphics 630’s 3587. That is a 15.7% advantage for the discrete NVIDIA chip in this compute-oriented test.

Q: Is there any test where the Intel integrated graphics wins?

A: Yes, in Geekbench Vulkan the Intel HD Graphics 630 scores 3540 against the MX110’s 3413, a 3.7% lead. This suggests the Intel driver stack handles Vulkan workloads efficiently relative to its OpenCL showing.

Q: What are the closest rivals to each GPU based on average score?

A: For the Intel HD Graphics 630, the nearest rival is the AMD Radeon RX 9060 XT 8 GB (average 4093, just 0.4% higher), followed by the NVIDIA Quadro K2100M (4151, 1.8% higher). The MX110’s closest rival is the NVIDIA GeForce GTX 650 (3823, 0.3% lower), with the AMD Radeon R5 Graphics (3883) sitting 1.2% above.

Q: What are the memory configurations for each product?

A: The Intel HD Graphics 630 uses system-shared memory, with bandwidth described as "System Dependent." The NVIDIA GeForce MX110 has 2 GB of dedicated GDDR5 memory on a 64-bit bus, providing 40.10 GB/s of bandwidth.

Q: What is the difference in shading unit count?

A: The NVIDIA GeForce MX110 has 256 shading units, while the Intel HD Graphics 630 has 192. However, the Intel part has more texture mapping units (24 versus 16), though the NVIDIA chip has more ROPs (8 versus 3).

Architecture Differences

The architectural divide here is stark, stemming from different process nodes and design philosophies. Intel’s HD Graphics 630 is built on a 14 nm++ process at Intel’s own foundry, using the Generation 9.5 architecture (Kaby Lake GT2). NVIDIA’s GeForce MX110 is a 28 nm TSMC part based on the older Maxwell architecture, using the GM108S chip. The process advantage for Intel means the integrated GPU can operate at a lower thermal envelope — 15 W TDP versus 30 W for the MX110 — despite having a smaller raw transistor budget in terms of dedicated graphics resources.

The Intel chip integrates 192 shading units, 24 TMUs, and just 3 ROPs, reflecting its role as a shared-resource IGP. It relies entirely on system memory, with no dedicated VRAM, and its clock speeds are modest: a 350 MHz base and 1000 MHz boost. In contrast, the MX110 packs 256 shading units, 16 TMUs, and 8 ROPs, running at a 978 MHz base and 1006 MHz boost. The NVIDIA part also has dedicated GDDR5 memory (2 GB, 64-bit, 40.10 GB/s), which eliminates the bandwidth contention that plagues shared-memory solutions.

Compute capabilities differ significantly. The Intel HD Graphics 630 delivers 384.0 GFLOPS of FP32 performance and 768.0 GFLOPS of FP16 (at a 2:1 ratio), while the MX110 provides 515.1 GFLOPS of FP32 and has no listed FP16 support. The MX110’s pixel rate of 8.048 GPixel/s dwarfs the Intel part’s 3.000 GPixel/s, but the Intel chip has a higher texture rate (24.00 GTexel/s versus 16.10 GTexel/s) due to its extra TMUs. API support also diverges: the Intel part supports DirectX 12 (12_1) and Vulkan 1.3, while the MX110 is limited to DirectX 12 (11_0) but offers Vulkan 1.4.

Head-to-Head Benchmarks

The two GPUs split their two common benchmark tests, and the margin of victory tells us about their respective strengths. In Geekbench OpenCL, the NVIDIA GeForce MX110 posts a score of 4255 against the Intel HD Graphics 630’s 3587. That 15.7% delta is the largest gap in either direction, and it aligns with the MX110’s higher FP32 throughput (515.1 GFLOPS versus 384.0 GFLOPS) and dedicated memory bandwidth. OpenCL workloads that scale with raw shader count and memory access benefit from the MX110’s 256 shading units and 40.10 GB/s of dedicated GDDR5 bandwidth, versus the Intel part’s shared-memory bottleneck.

Conversely, in Geekbench Vulkan, the Intel HD Graphics 630 wins with 3540 against the MX110’s 3413. The 3.7% margin is much narrower, but it suggests that Intel’s Vulkan driver implementation is more efficient than its OpenCL path. Despite having fewer shading units and lower raw FP32, the Intel chip’s newer architecture generation (Generation 9.5 versus Maxwell) likely helps with API-level overhead. The MX110’s Vulkan score of 3413 is actually lower than its OpenCL score, whereas the Intel part’s Vulkan score (3540) is only 1.3% below its OpenCL score (3587), indicating more consistent performance across compute APIs for the integrated solution.

Looking at the broader benchmark context, the Intel HD Graphics 630’s average of 4075 is propped up by its Geekbench Metal result of 5099, which is not available for the MX110. That Metal score is 41.9% higher than the Intel part’s OpenCL score, showing Apple’s Metal API plays favorably to this Intel GPU. The MX110’s two available scores (OpenCL 4255, Vulkan 3413) average to 3834, which is dragged down by its weaker Vulkan showing. If the MX110 had a Metal score, the comparison might shift, but as measured, the Intel part’s cross-API consistency gives it the overall edge.

Specification Differences

| Specification | Intel HD Graphics 630 | NVIDIA GeForce MX110 |

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

| Process Node | 14 nm++ | 28 nm |

| Transistors | Not listed | 1,020 million |

| Die Size | Not listed | 77 mm² |

| Transistor Density | Not listed | 13.2M / mm² |

| Base Clock | 350 MHz | 978 MHz |

| Boost Clock | 1000 MHz | 1006 MHz |

| Memory Size | System Shared | 2 GB GDDR5 |

| Memory Bus Width | System Shared | 64 bit |

| Memory Bandwidth | System Dependent | 40.10 GB/s |

| Shading Units | 192 | 256 |

| TMUs | 24 | 16 |

| ROPs | 3 | 8 |

| Pixel Rate | 3.000 GPixel/s | 8.048 GPixel/s |

| Texture Rate | 24.00 GTexel/s | 16.10 GTexel/s |

| FP32 Performance | 384.0 GFLOPS | 515.1 GFLOPS |

| FP16 Performance | 768.0 GFLOPS (2:1) | Not listed |

| TDP | 15 W | 30 W |

| Bus Interface | Ring Bus | PCIe 3.0 x4 |

| DirectX Version | 12 (12_1) | 12 (11_0) |

| Vulkan Version | 1.3 | 1.4 |

| Release Date | 2016-08-29 | 2017-11-16 |

The table above highlights that the MX110 is a more conventional discrete GPU with higher clocks, dedicated memory, and more shading units, while the HD 630 compensates with a newer process, higher TMU count, and broader DirectX feature level. The MX110’s 30 W TDP is double the Intel part’s 15 W, reflecting the cost of dedicated VRAM and higher shader counts. Notably, the Intel part has no power connectors (being an IGP), and the MX110 also has none, fitting its low-power mobile niche.

Where Each One Wins

The Intel HD Graphics 630 wins in scenarios where API efficiency and low power consumption matter. Its Vulkan score of 3540 exceeds the MX110’s 3413, making it the better choice for Vulkan-based compute or gaming workloads on systems where the CPU’s integrated graphics are already present. The 15 W TDP is half that of the MX110, making it far more suitable for ultra-portable laptops where battery life and thermals are paramount. Its higher texture rate (24.00 GTexel/s) also suggests better performance in texture-bound tasks, despite the lower pixel throughput. The Metal score of 5099, while not directly comparable, indicates strong macOS compatibility, a platform where the MX110 is rarely found.

The NVIDIA GeForce MX110 wins decisively in OpenCL-heavy applications, where its 15.7% lead (4255 versus 3587) is substantial. The dedicated 2 GB GDDR5 memory at 40.10 GB/s eliminates the system-memory contention that penalizes the Intel part, and the higher FP32 throughput (515.1 GFLOPS versus 384.0 GFLOPS) makes it better suited for GPGPU workloads like video encoding or scientific compute. The MX110’s 8 ROPs also provide a 2.7x advantage in pixel rate (8.048 GPixel/s versus 3.000 GPixel/s), which helps in fill-rate-limited rendering. For users who need a discrete GPU for light gaming or compute tasks without upgrading to a larger discrete card, the MX110’s dedicated resources offer a clear edge.

In terms of overall standing, the Intel HD Graphics 630’s average score of 4075 puts it 0.4% below the AMD Radeon RX 9060 XT 8 GB and 1.8% below the NVIDIA Quadro K2100M, while the MX110’s 3834 average sits within 1.6% of the NVIDIA Quadro 2000 (3898) and 1.1% above the Intel UHD Graphics 710 (3792). The Intel part’s percentile rank (24) edges out the MX110 (23), reinforcing that despite the MX110’s discrete nature, the HD 630’s newer architecture and driver maturity give it a slight overall advantage in these benchmark aggregates.