Intel UHD Graphics 750 vs NVIDIA GeForce GTX 1660 Ti Comparison

The NVIDIA GeForce GTX 1660 Ti and the Intel UHD Graphics 750 occupy opposite ends of the graphics spectrum, and the benchmark data reflects that chasm. The GTX 1660 Ti is a discrete, Turing-based GPU with dedicated GDDR6 memory, while the UHD 750 is an integrated processor graphics solution from Intel's Rocket Lake generation. In the two available head-to-head tests, each card wins one: the GTX 1660 Ti dominates OpenCL with a score of 51,029 against 6,743, a 656.8% advantage, while the UHD 750 takes Vulkan with 8,138 versus 5,827, a 28.4% edge. The verdict is straightforward: the GTX 1660 Ti is for anyone needing real graphics compute or gaming performance, while the UHD 750 is strictly for basic display output and light workloads where a discrete card is absent.

The Verdict

The data splits cleanly along usage lines. For anyone running GPU-accelerated compute, the NVIDIA GeForce GTX 1660 Ti is the only rational choice. Its Geekbench OpenCL score of 51,029 is nearly 7.6 times higher than the Intel UHD Graphics 750's 6,743, a delta of 656.8%. This is not a marginal gap; it is an order-of-magnitude difference in raw compute throughput. The GTX 1660 Ti also carries a Passmark G3D score of 12,871 and an average benchmark score of 7,723, placing it at the 41st percentile of all GPUs. Its nearest rival, the AMD Radeon 540, sits just 0.7% away at 7,673, meaning the GTX 1660 Ti is competitive with low-end discrete cards from the same era.

Conversely, the Intel UHD Graphics 750 is not a gaming or compute part. Its average benchmark score is 7,441, a mere 3.7% below the GTX 1660 Ti's average, but that figure is misleading because the UHD 750's benchmark pool is far smaller. Its 40th percentile ranking places it just one percentile point below the GTX 1660 Ti, yet its best head-to-head result is a Vulkan win at 8,138, which still trails the GTX 1660 Ti's Passmark G3D by 36.7%. The data suggests the UHD 750 is for systems where a discrete GPU is unnecessary—office desktops, HTPCs, or basic productivity machines. The GTX 1660 Ti is for anyone who needs to render, encode, or play modern titles, even if its 41st percentile rank indicates it is a mid-range card, not a high-end one.

Architecture Differences

The two GPUs share no common architectural lineage. The NVIDIA GeForce GTX 1660 Ti uses the TU116 chip built on Turing architecture, fabricated by TSMC on a 12 nm process. It packs 6,600 million transistors into a 284 mm² die, yielding a transistor density of 23.2 million per square millimeter. The Intel UHD Graphics 750, by contrast, is built on Generation 12.1 architecture with a Rocket Lake chip, fabricated by Intel on a 14 nm+++ process. Intel does not list transistor count or die size for this part, indicating it is a small, integrated block within a larger processor.

Core counts differ drastically. The GTX 1660 Ti has 1,536 shading units, 96 texture mapping units, and 48 raster output pipelines. The UHD 750 has 256 shading units, 16 TMUs, and 8 ROPs—exactly one-sixth, one-sixth, and one-sixth of the NVIDIA part's counts, respectively. Clock speeds tell a different story: the GTX 1660 Ti runs at a 1500 MHz base and 1770 MHz boost, while the UHD 750 starts at 300 MHz and boosts to 1300 MHz. The NVIDIA card's higher core count and clocks translate to a pixel rate of 84.96 GPixel/s and a texture rate of 169.9 GTexel/s, versus the Intel part's 10.40 GPixel/s and 20.80 GTexel/s. FP32 throughput is 5.437 TFLOPS for the GTX 1660 Ti versus 665.6 GFLOPS for the UHD 750, a 8.2-to-1 ratio.

Memory architecture is a fundamental split. The GTX 1660 Ti has 6 GB of GDDR6 on a 192-bit bus, delivering 288.0 GB/s of bandwidth. The UHD 750 uses system shared memory with a system-dependent bus width and bandwidth, meaning its performance scales with the host system's RAM speed. The NVIDIA card also has a 120 W TDP, requires a dual-slot cooler with a 1x 8-pin power connector, and a 300 W suggested PSU. The UHD 750 is an IGP with a 15 W TDP, no power connectors, and no suggested PSU. Both support DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, but the GTX 1660 Ti offers dedicated display outputs (1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a) while the UHD 750's outputs are motherboard dependent.

Head-to-Head Benchmarks

Only two benchmark tests appear in the head-to-head data, and they produce opposite winners. In Geekbench OpenCL, the NVIDIA GeForce GTX 1660 Ti scores 51,029 against the Intel UHD Graphics 750's 6,743. The delta is 656.8% in NVIDIA's favor, representing the single largest performance gap in the dataset. This test measures raw compute throughput across multiple workloads, and the GTX 1660 Ti's dedicated GDDR6 memory and 5.437 TFLOPS FP32 performance simply overwhelm the UHD 750's 665.6 GFLOPS and system-shared memory.

In Geekbench Vulkan, the result flips. The Intel UHD Graphics 750 scores 8,138, beating the GTX 1660 Ti's 5,827 by 28.4%. This is a notable anomaly given the NVIDIA card's hardware advantage. Vulkan is a low-level API that can favor architectures with specific driver optimizations, and the data suggests Intel's Generation 12.1 implementation handles this particular Vulkan workload more efficiently. However, the absolute scores are low for both cards—the GTX 1660 Ti's Vulkan result is far below its own OpenCL score, indicating the test may be driver-bound rather than hardware-bound.

Outside the head-to-head, the GTX 1660 Ti has a Passmark G3D score of 12,871 and a Passmark GPU Compute score of 5,226. The UHD 750 has no Passmark entries in the data, so its compute performance outside Geekbench is unknown. The GTX 1660 Ti also shows strong legacy DirectX performance with Passmark scores of 190 in DirectX 9, 103 in DirectX 11, and 66 in DirectX 10, though its DirectX 12 score drops to 51. The UHD 750's absence from these tests means the GTX 1660 Ti's win count of 1 against 1 in the head-to-head is a partial picture; the broader benchmark suite heavily favors NVIDIA.

Specification Differences

The two cards differ in nearly every measurable specification. Process node: 12 nm (TSMC) for the GTX 1660 Ti versus 14 nm+++ (Intel) for the UHD 750. Transistor count: 6,600 million versus not listed. Die size: 284 mm² versus not listed. Transistor density: 23.2M per mm² versus not listed. Base clock: 1500 MHz versus 300 MHz. Boost clock: 1770 MHz versus 1300 MHz. Memory clock: 1500 MHz (12 Gbps effective) versus system shared. Memory size: 6 GB GDDR6 versus system shared. Bus width: 192-bit versus system shared. Bandwidth: 288.0 GB/s versus system dependent. Shading units: 1536 versus 256. TMUs: 96 versus 16. ROPs: 48 versus 8. Pixel rate: 84.96 GPixel/s versus 10.40 GPixel/s. Texture rate: 169.9 GTexel/s versus 20.80 GTexel/s. FP32: 5.437 TFLOPS versus 665.6 GFLOPS. FP16: 10.87 TFLOPS versus 1,331.2 GFLOPS. TDP: 120 W versus 15 W. Slot width: dual-slot versus IGP. Power connectors: 1x 8-pin versus none. Suggested PSU: 300 W versus none. Bus interface: PCIe 3.0 x16 versus Ring Bus. Display outputs: 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a versus motherboard dependent. The GTX 1660 Ti has a launch MSRP of 279 USD; the UHD 750 has no launch MSRP. Release dates: 2019-02-21 versus 2021-03-29.

FAQ

Q: Which GPU has better raw compute performance?

A: The NVIDIA GeForce GTX 1660 Ti, by a wide margin. Its Geekbench OpenCL score of 51,029 is 656.8% higher than the Intel UHD Graphics 750's 6,743. The GTX 1660 Ti also delivers 5.437 TFLOPS FP32 versus the UHD 750's 665.6 GFLOPS.

Q: Why does the Intel UHD Graphics 750 win in Vulkan?

A: The UHD 750 scores 8,138 in Geekbench Vulkan versus the GTX 1660 Ti's 5,827, a 28.4% advantage. The data does not explain the cause, but it suggests that Intel's Generation 12.1 architecture and drivers handle this specific Vulkan workload more efficiently than NVIDIA's Turing.

Q: Is the GTX 1660 Ti worth its 120 W TDP?

A: The data shows a clear trade-off. The GTX 1660 Ti consumes 105 W more than the UHD 750 (120 W versus 15 W) but delivers roughly 8 times the FP32 throughput (5.437 TFLOPS versus 665.6 GFLOPS) and 8.2 times the pixel rate (84.96 GPixel/s versus 10.40 GPixel/s). The performance-per-watt is not directly calculable from the data, but the raw output is far higher.

Q: Can the UHD 750 handle gaming?

A: The data does not include any gaming benchmarks. Its best head-to-head score is 8,138 in Vulkan, which is below the GTX 1660 Ti's Passmark G3D score of 12,871. The UHD 750's 256 shading units and 8 ROPs suggest very limited rasterization capability, but no game-specific scores exist.

Q: How do these cards compare to their nearest rivals?

A: The GTX 1660 Ti's average score of 7,723 places it 0.7% ahead of the AMD Radeon 540 (7,673) and 2.3% ahead of the Intel Arc A310 (7,550). The UHD 750's average of 7,441 is 0.1% behind the AMD Radeon HD 8850M (7,447) and 1.4% behind the Intel Arc A310 (7,550). Both cards sit in a tight cluster of low-to-mid-range GPUs.

Q: What is the memory bandwidth difference?

A: The GTX 1660 Ti has 288.0 GB/s of dedicated GDDR6 bandwidth on a 192-bit bus. The UHD 750 uses system shared memory with system-dependent bandwidth, meaning its effective bandwidth varies with the host system's RAM configuration. The data does not provide a fixed number for the UHD 750.

Where Each One Wins

The NVIDIA GeForce GTX 1660 Ti wins in every compute-heavy scenario. Its OpenCL score of 51,029 versus 6,743 is the definitive metric for GPU compute workloads like rendering, video encoding, or scientific calculations. It also has the Passmark G3D score of 12,871, which indicates strong rasterization for gaming, though no specific game benchmarks are in the data. Its 6 GB of GDDR6 memory and 288.0 GB/s bandwidth make it suitable for textures and buffers that would overwhelm an IGP. The GTX 1660 Ti's launch MSRP of 279 USD reflects its position as a mid-range discrete card, and its 120 W TDP is a reasonable power envelope for a desktop system with a 300 W PSU.

The Intel UHD Graphics 750 wins in efficiency and integration. Its 15 W TDP and lack of power connectors make it ideal for compact or low-power systems where a discrete card is impractical. Its Vulkan score of 8,138 shows it can outperform the GTX 1660 Ti in specific low-level API workloads, though the absolute performance is modest. The UHD 750 is also the only option in this comparison that requires no additional hardware—its memory, bus, and display outputs are all system dependent. For a user building a basic office PC or a media center, the UHD 750's 40th percentile ranking is sufficient for 2D workloads and light 3D acceleration. The GTX 1660 Ti, at the 41st percentile, is not dramatically higher in percentile terms, but its benchmark scores are 3.7% higher on average (7,723 versus 7,441), and its compute advantage in OpenCL is overwhelming. The data clearly positions the GTX 1660 Ti as the performance pick and the UHD 750 as the integrated, low-power alternative.