Intel UHD Graphics P750 vs NVIDIA GeForce GTX 880M Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 880M has an average benchmark score of 8040, while the Intel UHD Graphics P750 scores 6554. This places the GTX 880M in the 42nd percentile of all GPUs, compared to the 38th percentile for the Intel part.

Q: How do the two compare in the only shared benchmark test?

A: In Geekbench OpenCL, the Intel UHD Graphics P750 scores 6554 versus 5622 for the NVIDIA GeForce GTX 880M, giving Intel a 14.2% advantage in that specific workload.

Q: What are the closest rivals for each GPU?

A: The GTX 880M's nearest rival is the NVIDIA Quadro P5000 with a score of 8039, a 0% delta. The Intel UHD Graphics P750's closest competitor is the AMD Radeon HD 7730M at 6581, which is 0.4% ahead.

Q: Which GPU has more shading units?

A: The NVIDIA GeForce GTX 880M has 1536 shading units, compared to 256 for the Intel UHD Graphics P750. The NVIDIA part also has 128 texture mapping units versus 64 for Intel.

Q: What are the process node differences?

A: The NVIDIA GeForce GTX 880M uses a 28 nm process at TSMC, while the Intel UHD Graphics P750 is built on Intel's 14 nm+++ process. The NVIDIA chip contains 3,540 million transistors on a 294 mm² die.

Q: Which GPU has a higher pixel fill rate?

A: The Intel UHD Graphics P750 has a pixel rate of 41.60 GPixel/s, which is higher than the GTX 880M's 31.78 GPixel/s, despite the NVIDIA part having far higher FP32 compute.

Architecture Differences

The NVIDIA GeForce GTX 880M is built on the Kepler architecture using the GK104 chip, a design focused on high compute throughput in a mobile MXM form factor. It integrates 3,540 million transistors on a 294 mm² die manufactured at TSMC's 28 nm node, yielding a transistor density of 12.0 million per mm². The Intel UHD Graphics P750, by contrast, uses the Generation 12.1 architecture on the Rocket Lake chip, fabricated on Intel's 14 nm+++ process, though its transistor count and die size are not recorded in the database.

The GPU core configurations differ substantially. The GTX 880M deploys 1536 shading units, 128 TMUs, and 32 ROPs, giving it a texture rate of 127.1 GTexel/s and FP32 throughput of 3.050 TFLOPS. The Intel part has 256 shading units, 64 TMUs, and 32 ROPs, with a texture rate of 83.20 GTexel/s and FP32 of 665.6 GFLOPS. Interestingly, the Intel GPU supports FP16 at 1,331.2 GFLOPS (2:1 ratio), a capability the GTX 880M lacks, while the NVIDIA part has no recorded FP16 data.

Memory architecture is another fundamental divider. The GTX 880M uses 8 GB of GDDR5 on a 256-bit bus, delivering 160.0 GB/s of bandwidth. The Intel UHD Graphics P750 relies on system shared memory, with its bus width, type, and bandwidth all system dependent. Clock behavior also differs: the NVIDIA GPU runs at 954 MHz base and 993 MHz boost with memory at 1250 MHz (5 Gbps effective), while Intel's integrated GPU starts at 350 MHz base and boosts to 1300 MHz.

API support shows generational differences. The GTX 880M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The Intel UHD Graphics P750 supports DirectX 12 (12_1), OpenGL 4.6, and the newer Vulkan 1.4. Both are end-of-life products, but the NVIDIA part has a release date of March 2014 and sits between the GeForce 700M and 900M generations, while Intel's is in the HD Graphics-W (Rocket Lake) line.

The Verdict

The data presents a clear split: the NVIDIA GeForce GTX 880M is the stronger compute GPU overall, with an average benchmark score of 8040 that places it 22.7% above the Intel UHD Graphics P750's 6554 average. Its nearest rivals, including the Quadro P5000 and GTX 650 Ti family, all cluster within 0.5% of its score, indicating consistent performance in its class. The Intel part sits in a lower tier, with rivals like the GTX 670M and GT 555M within 0.9% of its result.

However, in the single head-to-head Geekbench OpenCL test, the Intel GPU wins by 14.2%, scoring 6554 against 5622. This suggests that for OpenCL-specific workloads, the Intel integrated solution can outperform the older discrete GPU. The GTX 880M compensates with far higher FP32 throughput (3.050 TFLOPS versus 665.6 GFLOPS), more shading units, and dedicated GDDR5 memory with 160.0 GB/s bandwidth.

For users seeking maximum raw compute and dedicated graphics memory, the GTX 880M is the pick from this data. Its 42nd percentile ranking and proximity to the Quadro P5000 indicate solid mid-range standing. For systems where integrated graphics suffice and OpenCL performance matters most, the Intel UHD Graphics P750 holds a measurable edge in that specific benchmark, despite lower overall compute specifications. The 15 W TDP of the Intel part versus 122 W for the NVIDIA also matters for power-constrained environments, though the database does not record performance per watt directly.

Specification Differences

| Specification | NVIDIA GeForce GTX 880M | Intel UHD Graphics P750 |

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

| Architecture | Kepler | Generation 12.1 |

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

| Shading Units | 1536 | 256 |

| TMUs | 128 | 64 |

| ROPs | 32 | 32 |

| Base Clock | 954 MHz | 350 MHz |

| Boost Clock | 993 MHz | 1300 MHz |

| FP32 | 3.050 TFLOPS | 665.6 GFLOPS |

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

| Pixel Rate | 31.78 GPixel/s | 41.60 GPixel/s |

| Texture Rate | 127.1 GTexel/s | 83.20 GTexel/s |

| Memory Size | 8 GB GDDR5 | System Shared |

| Memory Bus | 256 bit | System Shared |

| Memory Bandwidth | 160.0 GB/s | System Dependent |

| TDP | 122 W | 15 W |

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

| Display Outputs | Portable Device Dependent | Motherboard Dependent |

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

| Vulkan | 1.2.175 | 1.4 |

Head-to-Head Benchmarks

The only direct benchmark comparison in the database is Geekbench OpenCL, where the Intel UHD Graphics P750 posts 6554 against the GTX 880M's 5622. This 14.2% margin in favor of Intel is notable given the NVIDIA part's superior hardware specifications. The GTX 880M's 3.050 TFLOPS FP32 compute is over 4.5 times the Intel's 665.6 GFLOPS, yet the OpenCL workload favors the newer architecture, suggesting that driver optimization or workload characteristics, rather than raw throughput, drive this result.

Looking at broader benchmark data, the GTX 880M achieves 10458 in Geekbench Metal, a test the Intel GPU does not appear in. Its OpenCL score of 5622 contributes to an average of 8040 across both tests. The Intel part's single OpenCL score of 6554 equals its average, meaning it has no additional workload data to offset the comparison.

The nearest rival clusters reinforce the performance tiers. The GTX 880M sits within 0.5% of the Quadro P5000 (8039), GTX 650 Ti (8053), GTX 650 Ti Boost (8067), and GRID K2 (8080), showing stable performance around the 8040 mark. The Intel UHD Graphics P750's rivals, including the HD 7730M (6581, 0.4% ahead), GTX 670M (6513, 0.6% behind), R7 M460 (6612, 0.9% ahead), and GT 555M (6493, 0.9% behind), all fall within a narrow 1.8% band, indicating consistent positioning.

In terms of fill rates, the Intel GPU's 41.60 GPixel/s pixel rate exceeds the GTX 880M's 31.78 GPixel/s, an advantage that likely helps in certain rendering tasks despite the NVIDIA part's higher texture rate of 127.1 GTexel/s versus 83.20 GTexel/s. The GTX 880M compensates with 8 GB of dedicated GDDR5 at 160.0 GB/s, whereas the Intel GPU's bandwidth is system dependent and cannot be quantified in isolation.

The data indicates that the GTX 880M wins the overall compute comparison through its average score and higher FP32 throughput, while the Intel UHD Graphics P750 wins the sole head-to-head OpenCL test and offers a newer API profile with Vulkan 1.4 and DirectX 12 (12_1). Users prioritizing OpenCL performance with minimal power draw (15 W versus 122 W) may prefer the Intel part, while those needing dedicated memory bandwidth and higher raw compute should select the NVIDIA GPU.