Intel UHD Graphics 750 vs NVIDIA GeForce GTX 675MX Comparison

Where Each One Wins

The recorded benchmark data splits cleanly between these two graphics solutions, but only one of them emerges with a win in the head-to-head comparison. The NVIDIA GeForce GTX 675MX takes the single direct comparison available in the database, the Geekbench OpenCL test, with a score of 10,723 against the Intel UHD Graphics 750's 6,743. That is a 59% advantage, and it represents the only direct head-to-head measurement recorded for this pairing.

The Intel UHD Graphics 750 does not win any of the direct comparisons, but it does have its own strengths in the broader benchmark context. Its Geekbench Vulkan score of 8,138 shows a different API profile, while its OpenCL result trails significantly. For compute workloads measured through OpenCL, the NVIDIA part is clearly dominant. For Vulkan-based tasks, the Intel part has a recorded score that is higher than its OpenCL score, suggesting the architecture responds differently depending on the API used.

Looking at average benchmark scores across the full database, the NVIDIA GeForce GTX 675MX sits at 8,427, which places it in the 43rd percentile of all GPUs. The Intel UHD Graphics 750 averages 7,441, placing it in the 40th percentile. The gap between their averages is 986 points, roughly 13% in favor of the NVIDIA part. These percentile positions indicate both parts sit in the lower-middle range of the overall GPU landscape, but the NVIDIA part consistently records higher compute throughput in the tests available.

The use-case split is therefore straightforward: the NVIDIA GeForce GTX 675MX wins in raw compute performance as measured by OpenCL, and the Intel UHD Graphics 750's case rests on its Vulkan performance being respectable for an integrated solution, even if it cannot match the discrete NVIDIA part in the head-to-head test.

Architecture Differences

The two GPUs come from fundamentally different design philosophies separated by nearly a decade of silicon evolution. The NVIDIA GeForce GTX 675MX uses the GK104 chip built on Kepler architecture, built on TSMC's 28 nm process. It packs 3,540 million transistors into a 294 mm² die, yielding a transistor density of 12.0 million per square millimeter. The Intel UHD Graphics 750 uses the Rocket Lake chip with Generation 12.1 architecture, built on Intel's 14 nm+++ process. Its transistor count and die size are not recorded in the database, but the process node difference alone signals a substantial manufacturing gap.

The compute resources differ dramatically. The NVIDIA part has 960 shading units, 80 texture mapping units, and 32 raster output units. The Intel part has 256 shading units, 16 TMUs, and 8 ROPs. In every unit category, the NVIDIA part has at least three times the resources. This explains the compute throughput differences: the NVIDIA part records 1,255.7 GFLOPS of FP32 performance, while the Intel part manages 665.6 GFLOPS. The Intel part does have an FP16 capability of 1,331.2 GFLOPS at a 2:1 ratio, a feature the NVIDIA part does not record, but for conventional FP32 workloads the NVIDIA part is roughly 89% faster.

Memory configurations could not be more different. The NVIDIA GeForce GTX 675MX uses 2 GB of GDDR5 on a 256-bit bus, delivering 115.2 GB/s of bandwidth. The Intel UHD Graphics 750 uses system shared memory, with bus width listed as system shared and bandwidth described as system dependent. Memory clocks on the Intel part are also system shared. The NVIDIA part has a memory clock of 900 MHz, which translates to 3.6 Gbps effective, a figure that has no meaning for the Intel part since it relies on shared system memory.

Pixel and texture rates follow the resource counts. The NVIDIA part records 13.08 GPixel/s pixel rate and 52.32 GTexel/s texture rate. The Intel part records 10.40 GPixel/s and 20.80 GTexel/s. The pixel rate gap is modest at about 26%, but the texture rate gap is far larger at roughly 152%. Clock behavior also differs: the Intel part has a base clock of 300 MHz and a boost of 1300 MHz, while the NVIDIA part has no base or boost clock recorded, only its memory clock.

Power draw shows the fundamental design gap. The NVIDIA part is rated at 100 W and comes as an MXM Module with no power connectors, meaning it draws power through the MXM interface. The Intel part is rated at 15 W and is an IGP, integrated into the processor package. API support also differs: the NVIDIA part supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The Intel part supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4.

Head-to-Head Benchmarks

The only direct head-to-head benchmark recorded is Geekbench OpenCL, and the result is decisive. The NVIDIA GeForce GTX 675MX scores 10,723, while the Intel UHD Graphics 750 scores 6,743. The delta is 59% in favor of the NVIDIA part. This is not a narrow margin; it is a commanding lead that reflects the underlying hardware differences in shading units, memory bandwidth, and dedicated graphics design.

The Intel UHD Graphics 750 has no direct head-to-head win in the database. Its best recorded result is the Vulkan score of 8,138, which is still below the NVIDIA part's OpenCL score of 10,723, though the two tests are not directly comparable due to different APIs. The Intel part's OpenCL score of 6,743 is its weaker of the two recorded benchmarks, and it falls 3,980 points short of the NVIDIA part in the same test.

For context, the NVIDIA part's average benchmark score of 8,427 puts it just behind the AMD Radeon 880M at 8,436, a delta of -0.1%, and behind the NVIDIA GeForce MX330 at 8,458, a delta of -0.4%. It sits ahead of the AMD Radeon R9 M375X at 8,325, a delta of 1.2%. The Intel part's average of 7,441 puts it just behind the AMD Radeon HD 8850M at 7,447, a delta of -0.1%, and behind the NVIDIA GeForce GTX 1650 at 7,472, a delta of -0.4%. It sits ahead of the AMD Radeon R7 350 at 7,425, a delta of 0.2%.

The relative positions in the broader database reinforce the head-to-head result. The NVIDIA part's average is 986 points higher, and its nearest rivals all cluster around the 8,300 to 8,500 range. The Intel part's nearest rivals cluster around 7,400 to 7,550. The two parts occupy different performance tiers, with the NVIDIA part roughly one tier above the Intel part in aggregate database scores.

The Verdict

The data points to a clear choice for anyone comparing these two directly. The NVIDIA GeForce GTX 675MX wins the only head-to-head benchmark recorded, and it does so by a wide 59% margin in OpenCL. Its average benchmark score is 13% higher than the Intel UHD Graphics 750, and it ranks in the 43rd percentile versus the Intel part's 40th. For compute-heavy workloads measured through OpenCL, the NVIDIA part is the stronger option without qualification.

The Intel UHD Graphics 750 is not without merit. Its Vulkan score of 8,138 exceeds its own OpenCL score, and for an integrated part with a 15 W TDP, it delivers a level of performance that is respectable for its class. Its nearest rivals include the AMD Radeon R7 350 and the AMD Radeon HD 8850M, which places it in a specific performance bracket. But against the NVIDIA GeForce GTX 675MX, it loses the direct comparison in every recorded metric that overlaps.

The NVIDIA part's 100 W TDP and MXM form factor make it suitable for systems designed to accommodate a discrete mobile GPU. The Intel part's 15 W TDP and IGP form factor make it suitable for systems where power efficiency and integration matter more than raw compute throughput. The choice depends on whether the user needs the compute performance of a discrete GPU or the integration of an iGPU.

From the data alone, the NVIDIA GeForce GTX 675MX is the faster part. The Intel UHD Graphics 750 is the more efficient part. Users who prioritize compute performance should select the NVIDIA part. Users who prioritize low power draw and integrated simplicity should select the Intel part, but they should expect significantly lower compute throughput.

FAQ

Q: Which GPU wins the only direct head-to-head benchmark?

A: The NVIDIA GeForce GTX 675MX wins the Geekbench OpenCL test with a score of 10,723 against the Intel UHD Graphics 750's 6,743, a 59% advantage.

Q: What is the average benchmark score difference between the two?

A: The NVIDIA GeForce GTX 675MX averages 8,427, while the Intel UHD Graphics 750 averages 7,441, a difference of 986 points in favor of the NVIDIA part.

Q: Does the Intel UHD Graphics 750 have any benchmark where it scores higher than the NVIDIA part?

A: No. The Intel part's best recorded score is 8,138 in Geekbench Vulkan, which is still below the NVIDIA part's 10,723 in Geekbench OpenCL, though the tests use different APIs.

Q: How do the shading unit counts compare?

A: The NVIDIA GeForce GTX 675MX has 960 shading units, while the Intel UHD Graphics 750 has 256 shading units.

Q: What are the TDP values for each GPU?

A: The NVIDIA GeForce GTX 675MX is rated at 100 W, while the Intel UHD Graphics 750 is rated at 15 W.

Q: What memory configurations do the two GPUs use?

A: The NVIDIA GeForce GTX 675MX uses 2 GB of GDDR5 on a 256-bit bus with 115.2 GB/s bandwidth. The Intel UHD Graphics 750 uses system shared memory with system dependent bandwidth.

Specification Differences

| Specification | NVIDIA GeForce GTX 675MX | Intel UHD Graphics 750 |

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

| Architecture | Kepler | Generation 12.1 |

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

| Foundry | TSMC | Intel |

| Transistors | 3,540 million | Not recorded |

| Die Size | 294 mm² | Not recorded |

| Transistor Density | 12.0M / mm² | Not recorded |

| Base Clock | Not recorded | 300 MHz |

| Boost Clock | Not recorded | 1300 MHz |

| Memory Clock | 900 MHz (3.6 Gbps effective) | System Shared |

| Memory Size | 2 GB | System Shared |

| Memory Type | GDDR5 | System Shared |

| Memory Bus Width | 256 bit | System Shared |

| Memory Bandwidth | 115.2 GB/s | System Dependent |

| Shading Units | 960 | 256 |

| TMUs | 80 | 16 |

| ROPs | 32 | 8 |

| Pixel Rate | 13.08 GPixel/s | 10.40 GPixel/s |

| Texture Rate | 52.32 GTexel/s | 20.80 GTexel/s |

| FP32 Performance | 1,255.7 GFLOPS | 665.6 GFLOPS |

| FP16 Performance | Not recorded | 1,331.2 GFLOPS (2:1) |

| TDP | 100 W | 15 W |

| Slot Width | MXM Module | IGP |

| Power Connectors | None | Not recorded |

| Bus Interface | MXM-B (3.0) | Ring Bus |

| Display Outputs | Portable Device Dependent | Motherboard Dependent |

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

| OpenGL Support | 4.6 | 4.6 |

| Vulkan Support | 1.2.175 | 1.4 |