NVIDIA GeForce GTX 1660 vs NVIDIA GeForce MX350 Comparison

The NVIDIA GeForce GTX 1660 and the NVIDIA GeForce MX350 occupy opposite ends of the mobile and desktop graphics spectrum, yet both share the NVIDIA brand and target users who need more than integrated graphics. The GTX 1660 is a desktop-oriented Turing-based card with substantial compute resources, while the MX350 is a low-power Pascal-based mobile chip designed for thin-and-light laptops. Benchmark data shows a stark performance gap, with the GTX 1660 dominating every shared test by a wide margin. This analysis breaks down where each card wins, the architectural reasons behind the gap, and what the numbers mean for potential users.

Where Each One Wins

The data is unambiguous: the GTX 1660 wins both head-to-head benchmark comparisons, giving it a 2-0 record against the MX350. In the Geekbench OpenCL test, the GTX 1660 scores 47,850 versus the MX350’s 8,689, a 450.7% advantage. In Geekbench Vulkan, the GTX 1660 scores 50,137 against 13,077, a 283.4% lead. There is no benchmark in the shared dataset where the MX350 comes out ahead.

The MX350’s only claim to relevance is its efficiency and form factor. With a 20 W TDP and no power connectors, it is designed for portable devices where power draw and space are at a premium. The GTX 1660, at 120 W with a dual-slot cooler and an 8-pin connector, cannot fit in such environments. For users prioritizing battery life and thin laptop designs, the MX350 is the only option between the two; for anyone seeking raw graphics performance, the GTX 1660 wins every measurable test.

Architecture Differences

The two cards come from different architectural generations and process nodes. The GTX 1660 is built on the Turing architecture using TSMC’s 12 nm process, while the MX350 uses the older Pascal architecture on Samsung’s 14 nm node. This generational gap explains much of the performance difference. The GTX 1660’s TU116 chip contains 6,600 million transistors on a 284 mm² die, while the MX350’s GP107S chip has 3,300 million transistors on a 132 mm² die. Notably, the MX350 has a higher transistor density at 25.0M / mm² compared to 23.2M / mm² for the GTX 1660, but the GTX 1660’s larger die and newer architecture deliver far more compute capability.

The GTX 1660 features 1,408 shading units, 88 texture mapping units, and 48 raster operation pipelines. In contrast, the MX350 is equipped with 640 shading units, 32 TMUs, and 16 ROPs. This more than 2:1 ratio in shading units and 2.75:1 in TMUs directly translates to higher fill rates and pixel throughput. The GTX 1660 achieves 85.68 GPixel/s and 157.1 GTexel/s, while the MX350 manages 23.49 GPixel/s and 46.98 GTexel/s. Neither card has ray tracing or tensor cores, so both rely on traditional rasterization.

Memory architecture also diverges sharply. The GTX 1660 has 6 GB of GDDR5 on a 192-bit bus, yielding 192.1 GB/s of bandwidth. The MX350 has 2 GB of GDDR5 on a 64-bit bus, providing just 56.06 GB/s. Clock speeds differ as well: the GTX 1660 runs at 1530 MHz base and 1785 MHz boost, while the MX350 runs at 1354 MHz base and 1468 MHz boost. The GTX 1660’s memory clock is 2001 MHz (8 Gbps effective) versus 1752 MHz (7 Gbps effective) for the MX350. The GTX 1660 also offers FP32 performance of 5.027 TFLOPS and FP16 of 10.05 TFLOPS (2:1), whereas the MX350 delivers 1.879 TFLOPS FP32 and a negligible 29.36 GFLOPS FP16 (1:64). The MX350’s FP16 implementation is clearly not designed for compute workloads.

Head-to-Head Benchmarks

Only two benchmarks appear in both cards’ datasets, but they tell a consistent story. In Geekbench OpenCL, the GTX 1660 scores 47,850, which is 450.7% higher than the MX350’s 8,689. This is the largest relative gap between the two cards and reflects the massive difference in shading units, memory bandwidth, and compute throughput. OpenCL workloads often scale with raw shader count and memory bandwidth, both of which heavily favor the GTX 1660.

In Geekbench Vulkan, the GTX 1660 scores 50,137 versus 13,077 for the MX350, a 283.4% advantage. While still a landslide, the Vulkan gap is narrower than OpenCL, suggesting the MX350’s Pascal architecture handles Vulkan’s driver overhead relatively better than OpenCL. However, the absolute scores remain far apart, and the GTX 1660’s lead is decisive in both APIs.

Looking at average benchmark scores, the GTX 1660 posts 11,680 across its benchmark suite, while the MX350 averages 10,883. This parity in average scores is misleading because the two cards are tested on different suites; the GTX 1660 includes PassMark and 3DMark tests that the MX350 does not run. The head-to-head results are the only fair comparison, and they show the GTX 1660 leading by 283% to 451%. The GTX 1660 also holds a 51st percentile ranking among all GPUs, while the MX350 sits at the 49th percentile, indicating that both are mid-range performers in their respective contexts.

Specification Differences

The most obvious specification difference is memory size and bandwidth. The GTX 1660 has 6 GB of GDDR5 on a 192-bit bus, while the MX350 has 2 GB on a 64-bit bus. This 3x memory capacity and 3.4x bandwidth advantage means the GTX 1660 can handle larger textures and higher resolutions without stuttering. The bus interface also differs: the GTX 1660 uses PCIe 3.0 x16, while the MX350 uses PCIe 3.0 x4, which limits the MX350’s ability to transfer data quickly with the host system.

Power and physical specifications diverge completely. The GTX 1660 has a 120 W TDP, is dual-slot, requires a single 8-pin power connector, and needs a 300 W power supply. The MX350 has a 20 W TDP, no power connectors, and no specified power supply requirement, making it suitable for laptop motherboards. The GTX 1660 measures 229 mm in length, 111 mm in height, and 35 mm in width, while the MX350 has no listed dimensions because it is embedded in portable devices. Display outputs also differ: the GTX 1660 offers 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a, whereas the MX350’s outputs are described as "Portable Device Dependent."

Process node and foundry differ as well. The GTX 1660 uses a 12 nm TSMC process, while the MX350 uses a 14 nm Samsung process. The GTX 1660’s transistor count is double that of the MX350, and its die size is more than double. The GTX 1660 was released on 2019-03-13, while the MX350 came later on 2020-02-09. Both are end-of-life products. The GTX 1660 has a launch MSRP of 219 USD; the MX350 has no listed MSRP. Clock speeds, shading units, TMUs, ROPs, pixel rate, texture rate, FP32, and FP16 all differ substantially, with the GTX 1660 ahead in every compute metric.

FAQ

Q: Which card is faster in Geekbench OpenCL?

A: The NVIDIA GeForce GTX 1660 scores 47,850 versus 8,689 for the MX350, a 450.7% advantage.

Q: What is the memory bandwidth difference?

A: The GTX 1660 has 192.1 GB/s bandwidth from a 192-bit bus, while the MX350 has 56.06 GB/s from a 64-bit bus.

Q: Do either of these cards support ray tracing?

A: No. Both the GTX 1660 and MX350 have no ray tracing cores listed in their specifications.

Q: What is the TDP difference?

A: The GTX 1660 has a 120 W TDP, while the MX350 has a 20 W TDP.

Q: Which card has more shading units?

A: The GTX 1660 has 1,408 shading units, compared to 640 for the MX350.

Q: Are both cards still in production?

A: No. Both the GTX 1660 and MX350 are listed as end-of-life products.

The Verdict

The data supports a straightforward conclusion: the GTX 1660 is categorically superior to the MX350 in every benchmarked metric. In Geekbench OpenCL, it is 450.7% faster; in Geekbench Vulkan, it is 283.4% faster. It has more than double the shading units, nearly triple the TMUs, triple the ROPs, and over three times the memory bandwidth. For any task involving 3D rendering, compute, or gaming, the GTX 1660 is the clear choice.

The MX350’s only advantages are its 20 W TDP, lack of power connectors, and portable-device form factor. It is designed for laptops where the GTX 1660’s 120 W power draw, dual-slot cooler, and 8-pin connector would be impossible to accommodate. Users who need a discrete GPU in a thin laptop with no external power will find the MX350 functional, but they should expect performance levels roughly one-third to one-fifth of the GTX 1660 depending on the workload.

Given that the GTX 1660 holds a 51st percentile ranking versus the MX350’s 49th, both are mid-pack performers, but the GTX 1660’s absolute scores are far higher in shared tests. The GTX 1660 also has a 6 GB memory buffer versus 2 GB, making it viable for modern game textures, while the MX350 is limited to lighter workloads. For desktop users or anyone with access to a power supply, the GTX 1660 is the only rational pick. For ultra-portable laptop users who prioritize battery life over performance, the MX350 exists as a compromise, but the benchmark data shows it is not a competitor to the GTX 1660 in any performance sense.