Intel HD Graphics 630 vs NVIDIA GeForce GT 735M Comparison

Head-to-Head Benchmarks

The recorded database contains a single direct comparison between these two parts: the Geekbench OpenCL test. In that test, the NVIDIA GeForce GT 735M scored 3616, while the Intel HD Graphics 630 scored 3587. That is a delta of negative 0.8 percent from the perspective of the Intel part, meaning the NVIDIA GPU is ahead by eight tenths of one percent. This is effectively a statistical tie, a margin so narrow that real-world application behavior will likely flip back and forth between the two depending on driver version and workload.

The Intel HD Graphics 630 has two additional benchmark entries that the GT 735M lacks. In Geekbench Metal, the Intel part scores 5099, which is substantially higher than its own OpenCL result. In Geekbench Vulkan, it scores 3540. The GT 735M has no recorded Metal or Vulkan scores, so cross-API comparisons are impossible from the data. The average benchmark score across all recorded tests tells a similar story: the Intel HD Graphics 630 averages 4075, while the GT 735M averages 3616. That is a gap of roughly 12.7 percent in favor of the Intel part, but this is driven entirely by the Metal score, which is an Apple-specific API that the NVIDIA part cannot even run.

Look at the nearest rival data for context. The Intel HD Graphics 630 sits at the 24th percentile of all GPUs in the database. Its closest rivals include the AMD Radeon RX 9060 XT 8 GB at an average score of 4093 (0.4 percent ahead), the NVIDIA GeForce GT 755M at 4033 (1 percent behind the Intel part), the AMD FirePro M4150 at 4013 (1.5 percent behind), and the NVIDIA Quadro K2100M at 4151 (1.8 percent ahead). The GT 735M sits at the 21st percentile, with rivals like the NVIDIA GeForce GTX 1050 at 3629 (0.3 percent ahead), the NVIDIA RTX 5000 Mobile Ada Generation at 3596 (0.6 percent behind), the NVIDIA GeForce GT 545 at 3594 (0.6 percent behind), and the AMD Radeon HD 6770 at 3649 (0.9 percent ahead). Both parts are clustered tightly with their respective peers, but the Intel part's peer group is operating at a higher absolute score level.

The one head-to-head result shows the GT 735M winning by a hair, but the broader benchmark picture shows the Intel part with a higher average due to its Metal performance. For OpenCL specifically, these two are interchangeable in practice.

Architecture Differences

The Intel HD Graphics 630 is built on the Kaby Lake GT2 chip, using Intel's Generation 9.5 architecture and a 14 nm++ process node from Intel's own foundry. The GT 735M uses the GK208 chip, which is Kepler 2.0 architecture, fabricated by TSMC on a 28 nm process. The process node difference is significant: Intel's 14 nm++ is two generations newer and denser, but the NVIDIA chip compensates with a larger, more specialized design.

The GT 735M is a discrete-class GPU with 1,020 million transistors on an 87 mm² die, giving a transistor density of 11.7 million per square millimeter. The Intel part has no recorded transistor count, die size, or density in the database. The NVIDIA chip has 384 shading units, 32 texture mapping units, and 8 render output units. The Intel part has 192 shading units, 24 TMUs, and only 3 ROPs. In raw compute resources, the NVIDIA part has double the shaders and more than double the ROPs.

Clock speeds tell the opposite story. The Intel part runs at a base of 350 MHz and boosts to 1000 MHz. The GT 735M runs at 575 MHz base and 628 MHz boost. The Intel part's boost clock is nearly 60 percent higher than the NVIDIA part's boost, which helps close the gap in raw throughput. The pixel rate for the Intel part is 3.000 GPixel/s versus 5.024 GPixel/s for the NVIDIA part. The texture rate is 24.00 GTexel/s for Intel versus 20.10 GTexel/s for NVIDIA. The FP32 compute is 384.0 GFLOPS for Intel versus 482.3 GFLOPS for NVIDIA. The Intel part also lists FP16 at 768.0 GFLOPS with a 2:1 ratio, while the NVIDIA part has no recorded FP16 capability.

Memory architecture is fundamentally different. The Intel HD Graphics 630 uses system shared memory, with the bus width also listed as system shared and bandwidth described as system dependent. The GT 735M has 2 GB of dedicated DDR3 memory on a 64-bit bus, with 14.40 GB/s of bandwidth and an effective memory clock of 1800 Mbps. The NVIDIA part's dedicated memory gives it predictable performance, while the Intel part's bandwidth depends entirely on the host system's RAM configuration.

The GT 735M has a TDP of 33 W, more than double the Intel part's 15 W. The GT 735M uses a PCIe 3.0 x8 bus interface and has no power connectors required. The Intel part uses a Ring Bus interface and is described as an IGP, meaning it lives on the CPU die. Both are end-of-life products. The GT 735M lists a predecessor in the GeForce 600M and a successor in the GeForce 800M; the Intel part has no recorded predecessor or successor.

Where Each One Wins

The GT 735M wins the only direct head-to-head test, the Geekbench OpenCL benchmark, by 0.8 percent. It also has a dedicated memory subsystem, which means its performance does not depend on system RAM speed or capacity. In any workload where consistent memory bandwidth matters, such as sustained compute tasks or texture-heavy rendering, the GT 735M has the structural advantage.

The Intel HD Graphics 630 wins in overall average benchmark score, 4075 versus 3616, driven by its strong Metal result of 5099. For users running Apple platforms or applications that leverage Metal, the Intel part is clearly superior. The Intel part also has a much higher boost clock at 1000 MHz versus 628 MHz, and a higher texture fill rate at 24.00 GTexel/s versus 20.10 GTexel/s. For tasks that depend on texture throughput rather than raw pixel output, the Intel part has the edge.

The Intel part also supports Vulkan 1.3, while the GT 735M supports Vulkan 1.2.175. The Intel part supports DirectX 12 (12_1), while the GT 735M only supports DirectX 12 (11_0), meaning the Intel part has the newer feature level. Both support OpenGL 4.6. For modern API compatibility, Intel has the advantage.

Power consumption is a clear win for Intel. At 15 W versus 33 W, the Intel part uses less than half the power of the NVIDIA part. This makes it the better choice for thin-and-light systems, fanless designs, or any chassis where thermal headroom is scarce.

Specification Differences

| Specification | Intel HD Graphics 630 | NVIDIA GeForce GT 735M |

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

| Process node | 14 nm++ | 28 nm |

| Foundry | Intel | TSMC |

| Transistors | Not recorded | 1,020 million |

| Die size | Not recorded | 87 mm² |

| Transistor density | Not recorded | 11.7M / mm² |

| Base clock | 350 MHz | 575 MHz |

| Boost clock | 1000 MHz | 628 MHz |

| Memory size | System Shared | 2 GB |

| Memory type | System Shared | DDR3 |

| Memory bus width | System Shared | 64 bit |

| Memory bandwidth | System Dependent | 14.40 GB/s |

| Shading units | 192 | 384 |

| TMUs | 24 | 32 |

| ROPs | 3 | 8 |

| Pixel rate | 3.000 GPixel/s | 5.024 GPixel/s |

| Texture rate | 24.00 GTexel/s | 20.10 GTexel/s |

| FP32 | 384.0 GFLOPS | 482.3 GFLOPS |

| FP16 | 768.0 GFLOPS (2:1) | Not recorded |

| TDP | 15 W | 33 W |

| Bus interface | Ring Bus | PCIe 3.0 x8 |

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

| Vulkan | 1.3 | 1.2.175 |

| Release date | 2016-08-29 | 2013-03-31 |

FAQ

Q: Which GPU is faster in the only direct benchmark comparison?

A: The NVIDIA GeForce GT 735M wins the Geekbench OpenCL test with a score of 3616 versus 3587 for the Intel HD Graphics 630, a delta of 0.8 percent.

Q: Why does the Intel HD Graphics 630 have a higher average benchmark score?

A: The Intel part has three recorded benchmark scores including a Geekbench Metal result of 5099, which lifts its average to 4075. The GT 735M only has the OpenCL score of 3616, so its average is that same value.

Q: Does the GT 735M have dedicated memory?

A: Yes, it has 2 GB of DDR3 on a 64-bit bus with 14.40 GB/s of bandwidth. The Intel HD Graphics 630 uses system shared memory with bandwidth described as system dependent.

Q: Which GPU supports the newer DirectX feature level?

A: The Intel HD Graphics 630 supports DirectX 12 (12_1), while the NVIDIA GeForce GT 735M supports DirectX 12 (11_0). The Intel part also supports Vulkan 1.3 versus Vulkan 1.2.175 on the NVIDIA part.

Q: How do power requirements compare?

A: The Intel HD Graphics 630 has a TDP of 15 W, while the NVIDIA GeForce GT 735M has a TDP of 33 W. The Intel part consumes less than half the power.

Q: What are the release dates?

A: The Intel HD Graphics 630 was released on 2016-08-29. The NVIDIA GeForce GT 735M was released on 2013-03-31. Both are listed as end-of-life products.

The Verdict

The data paints a clear picture for different use cases. If you need OpenCL compute performance in a system where the GPU is the only accelerator, the GT 735M is marginally ahead, but by a margin that is within noise. The 0.8 percent delta is not something you will feel in any real workload. The GT 735M's dedicated 2 GB DDR3 memory and 14.40 GB/s bandwidth give it a stability advantage, and its higher FP32 throughput of 482.3 GFLOPS versus 384.0 GFLOPS means it will handle raw compute tasks with a bit more headroom.

However, the Intel HD Graphics 630 is the newer part by more than three years, and it shows in API support. DirectX 12 (12_1) versus 12 (11_0), Vulkan 1.3 versus 1.2.175, and the presence of FP16 support all point toward better longevity in modern software. The Metal score of 5099 is the standout number in the entire dataset, and if you are on an Apple platform, the Intel part is the only one that even has a recorded Metal result.

For power-constrained systems, the decision is obvious. The Intel part draws 15 W versus 33 W for the NVIDIA part. That is a difference that matters in a laptop chassis, and the Intel part's performance is close enough that the power savings come at almost no cost in the only direct benchmark.

The GT 735M makes sense if you have a system with an available PCIe 3.0 x8 slot and need dedicated memory for consistency. The Intel part makes sense if you want the better API feature set, lower power draw, and the ability to run Metal workloads. Given that both are end-of-life products, the practical choice comes down to what your system supports. If it is a modern Intel platform with an integrated GPU, the HD Graphics 630 is the better bet. If you are upgrading an older laptop with a free mini-PCIe or MXM slot, the GT 735M is a serviceable, if dated, option. The benchmark data shows these two are close enough that system compatibility should be the deciding factor, not raw performance.