Intel UHD Graphics 710 vs NVIDIA GeForce GT 745M Comparison

The NVIDIA GeForce GT 745M and Intel UHD Graphics 710 represent two very different approaches to integrated-class graphics, separated by nearly a decade of silicon evolution. The GT 745M is a 28 nm Kepler part from 2013 designed for discrete use in laptops, while the UHD 710 is a 10 nm Alder Lake iGPU from 2022. Benchmark data shows the older NVIDIA part maintains a lead in both shared tests, but the margin tells a more complex story about API efficiency and raw compute.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GT 745M scores 3953 on average, while the Intel UHD Graphics 710 scores 3792. The GT 745M sits at the 23rd percentile of all GPUs, with the UHD 710 just behind at the 22nd percentile.

Q: How large is the performance gap in the shared OpenCL test?

A: In Geekbench OpenCL, the GT 745M scores 3580 versus 3496 for the UHD 710, a 2.4% advantage for NVIDIA. This is a narrow margin, close enough that the two could be considered roughly equivalent in compute-heavy workloads.

Q: What about Vulkan performance?

A: The GT 745M wins decisively in Geekbench Vulkan, scoring 5502 compared to 4088 for the UHD 710. That is a 34.6% lead for NVIDIA, making it the single largest performance difference between the two in any measured test.

Q: How do their closest rivals compare?

A: The GT 745M's nearest rival is the AMD Radeon R5 M420 with an average score of 3956, a 0.1% difference. The UHD 710's closest competitor is the NVIDIA GeForce GTX 650 at 3823, which sits 0.8% higher than Intel's part.

Q: Do both GPUs support DirectX 12?

A: Yes, both support DirectX 12, but at different feature levels. The GT 745M supports DirectX 12 (11_0), while the UHD 710 supports DirectX 12 (12_1), indicating a more modern feature set on the Intel side.

Q: What is the memory configuration for each?

A: The GT 745M has 2 GB of dedicated GDDR5 memory on a 128-bit bus, providing 64.00 GB/s of bandwidth. The UHD 710 uses System Shared memory with System Dependent bandwidth, meaning it relies on the host system's RAM.

Architecture Differences

The GT 745M is built on NVIDIA's Kepler architecture using the GK107 chip, fabricated on a 28 nm process at TSMC. It packs 1,270 million transistors into a 118 mm² die, yielding a transistor density of 10.8M per mm². The UHD 710, by contrast, is based on Intel's Generation 12.2 architecture with an Alder Lake chip on Intel's 10 nm process. No transistor count or die size is listed for the Intel part, reflecting its integration into a larger processor package.

The compute configuration differs substantially. The GT 745M has 384 shading units, 32 texture mapping units, and 16 ROPs. The UHD 710 has only 128 shading units, 8 TMUs, and 8 ROPs. This 3:1 ratio in shader count favors NVIDIA heavily, but clock speeds compensate. The UHD 710 runs at a base of 300 MHz and boosts to 1300 MHz, while the GT 745M's core clocks are not listed in the data. The memory clock for NVIDIA is 1000 MHz (4 Gbps effective), while Intel's is system-shared.

The GT 745M has a 45 W TDP, three times the 15 W TDP of the UHD 710. Both are listed as IGP slot width and are end-of-life products. The NVIDIA part connects via PCIe 3.0 x16, while Intel uses a Ring Bus interface. Display outputs are Portable Device Dependent for NVIDIA and Motherboard Dependent for Intel, reflecting their intended deployment. The GT 745M supports Vulkan 1.2.175, while the UHD 710 supports Vulkan 1.4, a newer API revision.

Head-to-Head Benchmarks

The shared benchmark suite covers two tests: Geekbench OpenCL and Geekbench Vulkan. In OpenCL, the GT 745M scores 3580 against 3496 for the UHD 710. The 2.4% delta is small enough that run-to-run variation could plausibly flip the result in some scenarios, but the data consistently places NVIDIA ahead. The UHD 710's higher boost clock of 1300 MHz likely narrows what would otherwise be a wider gap given the shader count disparity.

Vulkan presents a very different picture. The GT 745M scores 5502, while the UHD 710 manages only 4088. The 34.6% delta is the dominant headline in this comparison. Kepler's older architecture appears to handle Vulkan's explicit command model more efficiently relative to its raw hardware resources, or the Intel driver stack for this iGPU is less optimized for the workload. Either way, the data shows a decisive NVIDIA advantage in this modern graphics API.

The GT 745M wins both head-to-head tests, giving it a 2-0 record. The UHD 710 has no victories in the measured benchmarks. However, the average scores tell a slightly different story: the GT 745M averages 3953 across all its benchmarks, while the UHD 710 averages 3792. The GT 745M's Vulkan score of 5502 pulls its average up significantly, whereas the UHD 710's average is dragged down by its weaker Vulkan showing.

Specification Differences

The two GPUs differ in nearly every measured specification. Process node: 28 nm for NVIDIA versus 10 nm for Intel. Transistors: 1,270 million for NVIDIA, none listed for Intel. Die size: 118 mm² for NVIDIA, none listed for Intel. Memory: 2 GB GDDR5 on a 128-bit bus with 64.00 GB/s bandwidth for NVIDIA; System Shared memory with System Dependent bandwidth for Intel.

Compute units: 384 shading units, 32 TMUs, and 16 ROPs for NVIDIA versus 128 shading units, 8 TMUs, and 8 ROPs for Intel. Pixel rate: 4.392 GPixel/s for NVIDIA versus 10.40 GPixel/s for Intel, a significant Intel advantage despite fewer ROPs. Texture rate: 17.57 GTexel/s for NVIDIA versus 10.40 GTexel/s for Intel, favoring NVIDIA.

FP32 performance: 421.6 GFLOPS for NVIDIA versus 332.8 GFLOPS for Intel. The UHD 710 also lists FP16 at 665.6 GFLOPS (2:1), while the GT 745M has no FP16 figure. TDP: 45 W for NVIDIA versus 15 W for Intel. Bus interface: PCIe 3.0 x16 versus Ring Bus. DirectX support: 12 (11_0) versus 12 (12_1). Vulkan: 1.2.175 versus 1.4. OpenGL is identical at 4.6 for both.

Release dates differ by roughly nine years: the GT 745M launched in early 2013, while the UHD 710 launched in early 2022. The GT 745M has listed predecessors and successors (GeForce 600M and GeForce 800M), while the UHD 710 has neither.

Where Each One Wins

The GT 745M wins in raw compute throughput. Its FP32 output of 421.6 GFLOPS exceeds the UHD 710's 332.8 GFLOPS by a meaningful margin. It also wins in texture rate, delivering 17.57 GTexel/s against 10.40 GTexel/s. The dedicated 64.00 GB/s of memory bandwidth is another clear advantage, particularly for workloads that exceed the UHD 710's system-shared memory capabilities.

The UHD 710 wins in pixel throughput, producing 10.40 GPixel/s versus 4.392 GPixel/s for NVIDIA. It also offers FP16 support at 665.6 GFLOPS, which the GT 745M lacks entirely. The 15 W TDP is one-third of NVIDIA's 45 W, making it far more power-efficient. The newer Vulkan 1.4 API support and DirectX 12 (12_1) feature level outclass the older NVIDIA part's API versions.

The UHD 710's higher boost clock of 1300 MHz, compared to no listed boost for the GT 745M, suggests it can ramp up quickly for short bursts. The Ring Bus interface reflects its integration into the processor, eliminating the need for a separate PCIe connection. In practical terms, the UHD 710 excels in scenarios where power draw is critical and memory is abundant in the host system.

The Verdict

The data points to a clear split: the GT 745M is the stronger performer in both measured benchmarks, with a 2.4% lead in OpenCL and a 34.6% lead in Vulkan. Its average benchmark score of 3953 is 4.2% higher than the UHD 710's 3792. For users prioritizing raw graphics performance, the GT 745M is the better choice, particularly in Vulkan-based applications where the NVIDIA part's 5502 score dwarfs Intel's 4088.

However, the UHD 710 occupies a different niche. Its 15 W TDP makes it suitable for power-constrained systems where the GT 745M's 45 W draw would be prohibitive. The pixel rate advantage of 10.40 GPixel/s versus 4.392 GPixel/s suggests it handles fill-rate-bound tasks more efficiently. The FP16 support at 665.6 GFLOPS could benefit compute workloads that leverage half-precision arithmetic, something the GT 745M cannot do.

The production status of both is end-of-life, so neither represents a forward-looking purchase. For legacy systems with a free PCIe slot and a 45 W thermal budget, the GT 745M offers better benchmark scores and dedicated memory. For modern Alder Lake platforms where the iGPU is already present, the UHD 710 provides adequate performance with zero additional power draw. The 23rd percentile ranking for the GT 745M and 22nd for the UHD 710 place both near the bottom of the GPU hierarchy, but within that low tier, the NVIDIA part is measurably faster in the tests that matter most.