NVIDIA GeForce GTX 960 vs NVIDIA GeForce MX350 Comparison

Head-to-Head Benchmarks

The recorded data shows a split decision between these two NVIDIA parts, with each claiming a victory in one of the two shared tests. The most dramatic difference appears in Geekbench OpenCL, where the GTX 960 posts a score of 18,925 against the MX350’s 8,689. That translates to a 54.1% deficit for the MX350, meaning the older desktop card delivers more than double the raw compute throughput in this particular workload. The margin is substantial and reflects the GTX 960’s larger hardware footprint, something that becomes clear when examining the architecture section below.

However, the story flips entirely in Geekbench Vulkan. Here, the MX350 scores 13,077 while the GTX 960 trails at 9,231. The MX350’s advantage is 41.7%, a commanding lead that suggests its driver stack or hardware design handles the Vulkan API more efficiently. This is not a narrow win; it is a decisive gap that flips the overall impression of which card is faster depending on the API being tested. The average benchmark score across all tests in the database also favors the MX350, at 10,883 versus 9,273 for the GTX 960, though that figure includes different test suites for each card, so direct comparison requires caution.

Looking at percentile placement, the MX350 ranks at the 49th percentile among all GPUs, while the GTX 960 sits at the 45th. This puts the MX350 slightly ahead in overall database standing, but the nearest rivals tell a more nuanced tale. For the MX350, the closest competitors are the AMD Radeon Pro 450 (average score 10,804, just 0.7% behind) and the NVIDIA Quadro K2200 (10,761, 1.1% behind). On the other side, the GeForce GTX 1650 SUPER (11,047) leads the MX350 by 1.5%, and the AMD Radeon RX 550 (11,075) is 1.7% ahead. For the GTX 960, the surrounding field is tightly packed: the GeForce GTX 465 (9,294) is only 0.2% faster, the GTX 850M (9,302) is 0.3% ahead, and the AMD Radeon R7 M380 (9,313) leads by 0.4%. The AMD Radeon Vega 8 (9,221) trails the GTX 960 by 0.6%. These deltas indicate that the MX350 sits in a slightly more competitive neighborhood, while the GTX 960 is essentially tied with a group of older or lower-tier parts.

Architecture Differences

The two GPUs come from different architectural generations and manufacturing processes, which explains much of the benchmark behavior. The MX350 uses the GP107S chip built on a 14 nm process at Samsung, featuring 3,300 million transistors on a 132 mm² die. The transistor density works out to 25.0M per mm², a figure that reflects the more modern fabrication node. The GTX 960, in contrast, relies on the GM206 chip manufactured on TSMC’s 28 nm process, with 2,940 million transistors spread across a much larger 228 mm² die. Its transistor density is only 12.9M per mm². The MX350 packs more transistors into a smaller area, a hallmark of the newer Pascal architecture versus the older Maxwell 2.0 design.

Clock speeds also differ. The MX350 runs at a base of 1354 MHz and boosts to 1468 MHz. The GTX 960 starts lower at 1127 MHz base and 1178 MHz boost. Despite the lower clocks, the GTX 960 has far more execution resources: 1024 shading units, 64 texture mapping units, and 32 raster output processors, compared to the MX350’s 640 shading units, 32 TMUs, and 16 ROPs. This doubling of key units in the GTX 960 gives it higher theoretical peak rates: 75.39 GTexel/s texture fill and 37.70 GPixel/s pixel throughput, versus 46.98 GTexel/s and 23.49 GPixel/s for the MX350. Floating-point performance follows the same pattern, with the GTX 960 at 2.413 TFLOPS FP32 against the MX350’s 1.879 TFLOPS.

Memory subsystems diverge sharply. The MX350 uses 2 GB of GDDR5 on a 64-bit bus, yielding 56.06 GB/s of bandwidth. The GTX 960 also has 2 GB of GDDR5 but on a 128-bit bus, doubling bandwidth to 112.2 GB/s. Both run memory at effectively 7 Gbps, but the wider bus is the differentiator. The GTX 960 also requires a 1x 6-pin power connector and lists a suggested PSU of 300 W, while the MX350 draws power solely from the slot and has no external connectors. The MX350’s TDP is 20 W, a stark contrast to the GTX 960’s 120 W. The GTX 960 is a dual-slot card measuring 241 mm (9.5 inches), whereas the MX350 is portable-device dependent, meaning its dimensions vary by laptop implementation. The GTX 960 offers fixed display outputs (1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.2), while the MX350 relies on the host device’s connectors.

Both cards support DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, so API compatibility is not a differentiator. The MX350 lists FP16 performance at 29.36 GFLOPS (1:64 ratio), while the GTX 960 has no recorded FP16 figure. The MX350 connects via PCIe 3.0 x4, a narrower lane configuration than the GTX 960’s PCIe 3.0 x16.

Where Each One Wins

The benchmark data points to clear use-case separation. The GTX 960 dominates in OpenCL workloads, where its 54.1% lead over the MX350 indicates a major advantage in compute-heavy tasks that leverage this API. This could include general-purpose GPU computing, certain rendering tasks, or physics simulations. The GTX 960’s higher texture rate (75.39 GTexel/s) and pixel rate (37.70 GPixel/s) also suggest it handles traditional rasterization workloads, such as older games or non-Vulkan titles, more effectively. Its double memory bandwidth (112.2 GB/s) helps in scenarios where large textures or framebuffers are in play.

The MX350, conversely, wins decisively in Vulkan, with a 41.7% advantage. This makes it the stronger choice for applications built around modern low-overhead APIs, including many current game engines and emulators. Its higher boost clock (1468 MHz versus 1178 MHz) and smaller die with denser transistors may contribute to better per-clock efficiency in this API. The MX350’s lower TDP (20 W versus 120 W) makes it the only viable option for thin-and-light laptops, as it requires no external power and produces far less heat. The GTX 960’s dual-slot cooler and 300 W suggested PSU rule it out for such portable scenarios.

In terms of overall database standing, the MX350’s 49th percentile slightly edges the GTX 960’s 45th, and its average benchmark score (10,883) is higher. However, that average includes the Vulkan result, which heavily favors the MX350. If a user’s software stack relies primarily on OpenCL, the GTX 960’s superior score in that specific test would matter more than the aggregate. The nearest rival data reinforces this: the MX350 sits close to mid-range parts like the GTX 1650 SUPER and RX 550, while the GTX 960 trades blows with much older hardware like the GTX 465 and GTX 850M.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The MX350 averages 10,883 across its recorded tests, while the GTX 960 averages 9,273. The MX350 also ranks at the 49th percentile among all GPUs, versus the 45th percentile for the GTX 960.

Q: How large is the performance gap in OpenCL?

A: The GTX 960 scores 18,925 in Geekbench OpenCL, which is 54.1% higher than the MX350’s 8,689. This is the largest single-test margin in the head-to-head data.

Q: Does the MX350 win any benchmark outright?

A: Yes, in Geekbench Vulkan the MX350 scores 13,077 against the GTX 960’s 9,231, giving the MX350 a 41.7% lead in that test.

Q: What are the closest rivals for each card?

A: For the MX350, the AMD Radeon Pro 450 is nearest (0.7% behind), followed by the NVIDIA Quadro K2200 (1.1% behind). For the GTX 960, the NVIDIA GeForce GTX 465 is closest (0.2% ahead), then the GTX 850M (0.3% ahead).

Q: Do both cards support the same modern APIs?

A: Yes, both support DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The MX350 additionally records FP16 performance at 29.36 GFLOPS, while the GTX 960 has no listed FP16 figure.

Q: Which card has higher memory bandwidth?

A: The GTX 960 offers 112.2 GB/s over a 128-bit bus, exactly double the MX350’s 56.06 GB/s from a 64-bit bus. Both use 2 GB of GDDR5 at 7 Gbps effective.

Specification Differences

The two cards differ across nearly every hardware specification. The process node is 14 nm for the MX350 versus 28 nm for the GTX 960, with the foundry being Samsung versus TSMC. Transistor count is 3,300 million on the MX350 versus 2,940 million, while die size is 132 mm² versus 228 mm², giving transistor densities of 25.0M/mm² and 12.9M/mm², respectively. Base clocks are 1354 MHz versus 1127 MHz, and boost clocks are 1468 MHz versus 1178 MHz. Memory bandwidth is 56.06 GB/s versus 112.2 GB/s, with bus widths of 64-bit versus 128-bit.

Shading units count 640 versus 1024, TMUs 32 versus 64, and ROPs 16 versus 32. Pixel rate is 23.49 GPixel/s versus 37.70 GPixel/s, texture rate is 46.98 GTexel/s versus 75.39 GTexel/s, and FP32 performance is 1.879 TFLOPS versus 2.413 TFLOPS. The MX350 lists FP16 at 29.36 GFLOPS; the GTX 960 has no FP16 entry. TDP is 20 W versus 120 W. The GTX 960 is dual-slot with a 1x 6-pin power connector and a 300 W suggested PSU; the MX350 has no power connectors and is portable-device dependent. Bus interface is PCIe 3.0 x4 for the MX350 versus PCIe 3.0 x16 for the GTX 960. Display outputs are portable-device dependent versus 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.2. The GTX 960 measures 241 mm (9.5 inches) long; the MX350 has no recorded dimensions. Release dates are February 2020 for the MX350 and January 2015 for the GTX 960, with both marked end-of-life. The GTX 960 lists a launch MSRP of 199 USD; the MX350 has no MSRP recorded.

The Verdict

The data suggests two very different products aimed at different scenarios. For users running OpenCL-heavy applications, the GTX 960 is the clear pick: its 54.1% lead in that benchmark, combined with higher pixel and texture rates, points to stronger raw compute and rasterization throughput. It also offers double the memory bandwidth, which benefits texture-heavy workloads. The GTX 960’s nearest rivals are all within 0.4% of its average score, indicating it sits in a stable but crowded performance tier.

For those prioritizing Vulkan performance, the MX350 is the winner, with a 41.7% advantage in that test. Its higher boost clock and newer Pascal architecture appear to translate into better API efficiency. The MX350 also holds a higher overall percentile (49 versus 45) and a higher average benchmark score (10,883 versus 9,273). Its 20 W TDP and lack of power connectors make it the only practical choice for portable devices, while the GTX 960’s 120 W draw and dual-slot design require a desktop chassis with adequate cooling and a 300 W PSU.

The choice hinges on the workload. If the target applications use Vulkan or require low power consumption, the MX350’s data justifies selecting it. If the priority is OpenCL compute or maximum memory bandwidth, the GTX 960’s recorded performance is superior despite its older architecture. Both cards are end-of-life, so availability in the used market would be the deciding factor, but the benchmark evidence points to a genuine split: the MX350 for modern API efficiency and mobility, the GTX 960 for classic compute-heavy tasks.