NVIDIA GeForce GTX 960M vs NVIDIA GeForce MX350 Comparison

Head-to-Head Benchmarks

The benchmark data splits cleanly between the two GPUs, with each claiming one decisive victory in the recorded tests. The NVIDIA GeForce GTX 960M dominates the Geekbench OpenCL workload, scoring 11,045 against the MX350's 8,689, a commanding 21.3% lead for the older Maxwell part. This is not a marginal gap; it reflects a fundamental difference in how the two chips handle compute-heavy tasks.

Conversely, the Vulkan results flip the script entirely. The MX350 posts 13,077, while the GTX 960M manages only 8,245. That gives the Pascal-based MX350 a massive 58.6% advantage, one of the largest swings in either direction. The delta here is almost three times the magnitude of the GTX 960M's OpenCL win.

Looking at the broader database percentiles, the MX350 sits at the 49th percentile of all GPUs, which places it just above the midpoint of the entire field. The GTX 960M trails at the 46th percentile. Their average benchmark scores tell a similar story: the MX350 averages 10,883, while the GTX 960M averages 9,645. The MX350's average is 12.8% higher than its rival's, driven almost entirely by the Vulkan result.

The nearest rival data adds context. The MX350's closest competitor, the AMD Radeon Pro 450, scores 10,804, a mere 0.7% behind, while the NVIDIA Quadro K2200 (10,761) sits 1.1% lower. The MX350 also trades blows with stronger parts: the GeForce GTX 1650 SUPER (11,047) beats it by 1.5%, and the Radeon RX 550 (11,075) by 1.7%. For the GTX 960M, the fight is even tighter: the Quadro K5000 (9,637) is only 0.1% behind, the Radeon Pro WX 2100 (9,653) is 0.1% ahead, the Quadro P4000 (9,665) is 0.2% ahead, and the Tesla C2070 (9,716) is 0.7% ahead.

What this means in practice: if the workload is OpenCL-centric, the GTX 960M is the clear pick. If Vulkan matters, the MX350 is in a different league entirely. The recorded data shows no middle ground in these tests, each GPU wins its preferred API by a wide margin.

FAQ

Q: Which GPU wins in OpenCL performance?

A: The NVIDIA GeForce GTX 960M wins decisively, scoring 11,045 versus the MX350's 8,689, a 21.3% advantage.

Q: Which GPU wins in Vulkan performance?

A: The NVIDIA GeForce MX350 wins by a wide margin, scoring 13,077 against the GTX 960M's 8,245, a 58.6% lead.

Q: How do the two GPUs rank against all other GPUs in the database?

A: The MX350 sits at the 49th percentile, while the GTX 960M sits at the 46th percentile. The MX350's average benchmark score is 10,883, compared to 9,645 for the GTX 960M.

Q: What are the closest rivals to each GPU?

A: For the MX350, the AMD Radeon Pro 450 is closest at 0.7% behind. For the GTX 960M, the NVIDIA Quadro K5000 is closest at 0.1% behind.

Q: Do both GPUs support Vulkan?

A: Yes, both list Vulkan 1.4 in their API support.

Q: Which GPU has the higher texture fill rate?

A: The GTX 960M has a slightly higher texture rate at 47.04 GTexel/s, versus 46.98 GTexel/s for the MX350, though the MX350 has a higher pixel rate at 23.49 GPixel/s versus 18.82 GPixel/s.

Architecture Differences

The two GPUs represent different NVIDIA architectures from different eras. The MX350 is built on the Pascal architecture, utilizing the GP107S chip, fabricated on a 14 nm process at Samsung. The GTX 960M uses the older Maxwell architecture with the GM107 chip, produced on a 28 nm process at TSMC. This process node difference is substantial: 14 nm versus 28 nm means the MX350 packs transistors far more densely.

The transistor counts reflect this. The MX350 contains 3,300 million transistors on a 132 mm² die, yielding a density of 25.0 million transistors per mm². The GTX 960M has 1,870 million transistors on a larger 148 mm² die, resulting in a much lower density of 12.6 million per mm². The MX350 fits nearly twice the transistor density into a smaller physical package.

Both GPUs have the same number of shading units at 640, but the MX350 pairs them with 32 texture mapping units (TMUs) and 16 raster operations pipelines (ROPs). The GTX 960M has more TMUs at 40, but the same 16 ROPs. The MX350's pixel rate comes in at 23.49 GPixel/s, ahead of the GTX 960M's 18.82 GPixel/s. Texture rates are nearly identical: 46.98 GTexel/s for the MX350, 47.04 GTexel/s for the GTX 960M.

Compute throughput also diverges. The MX350 delivers 1.879 TFLOPS of FP32 performance, while the GTX 960M delivers 1.505 TFLOPS. The MX350 also supports FP16 at 29.36 GFLOPS, a 1:64 ratio, whereas the GTX 960M lists no FP16 capability. The MX350's Pascal architecture supports DirectX 12 (12_1), while the GTX 960M supports DirectX 12 (11_0), a meaningful API feature gap.

The MX350 is a lower-power design with a 20 W TDP and no power connectors, relying on the PCIe slot. The GTX 960M is a 75 W part packaged as an MXM Module, also without power connectors. The MX350 uses a PCIe 3.0 x4 interface, while the GTX 960M uses an MXM-B (3.0) bus interface.

Specification Differences

The memory subsystems diverge sharply. The MX350 comes with 2 GB of GDDR5 on a 64-bit bus, delivering 56.06 GB/s of bandwidth. The GTX 960M doubles the capacity to 4 GB, uses a 128-bit bus, and achieves 80.19 GB/s. The GTX 960M's memory bandwidth is 43% higher, which helps explain its OpenCL advantage.

Clock speeds also differ. The MX350 has a base clock of 1354 MHz and a boost clock of 1468 MHz. The GTX 960M runs at 1097 MHz base and 1176 MHz boost. The MX350's clocks are roughly 23% higher at base and 25% higher at boost. Memory clocks follow suit: the MX350 runs at 1752 MHz (7 Gbps effective), while the GTX 960M runs at 1253 MHz (5 Gbps effective).

The MX350 is built on 14 nm with 3,300 million transistors on a 132 mm² die, while the GTX 960M uses 28 nm with 1,870 million transistors on a 148 mm² die. The MX350's TDP is 20 W versus 75 W for the GTX 960M, a 3.75x difference in power envelope. The MX350 uses a PCIe 3.0 x4 interface; the GTX 960M uses MXM-B (3.0). The MX350's form factor is portable-device dependent, while the GTX 960M is an MXM Module.

DirectX support differs: the MX350 supports DirectX 12 (12_1), the GTX 960M supports DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan 1.4. The MX350 was released in 2020, five years after the GTX 960M's 2015 debut. Both are end-of-life, and neither has a launch MSRP in the database.

The Verdict

The data points to a straightforward conclusion: pick based on the workload, not brand loyalty. The GTX 960M is the OpenCL specialist, with a 21.3% lead over the MX350 in that specific test. It also offers double the VRAM (4 GB versus 2 GB) and significantly more memory bandwidth (80.19 GB/s versus 56.06 GB/s). For compute tasks that rely on OpenCL and need larger frame buffers, the GTX 960M is the stronger choice.

The MX350 is the Vulkan champion, beating the GTX 960M by 58.6% in that API. Its higher clocks (1468 MHz boost versus 1176 MHz), higher FP32 throughput (1.879 TFLOPS versus 1.505 TFLOPS), and superior pixel rate (23.49 GPixel/s versus 18.82 GPixel/s) make it the more modern, efficient design. It also draws only 20 W versus 75 W, making it far better suited to thin-and-light portables.

The average benchmark score favors the MX350: 10,883 versus 9,645. The percentile ranking also favors it: 49th versus 46th. For general-purpose use where Vulkan is the primary API, the MX350 is the better all-rounder. For legacy OpenCL applications or workloads that need more VRAM, the GTX 960M remains competitive.

Users should note the MX350's 64-bit memory bus and 2 GB capacity as potential bottlenecks, despite its architectural advantages. The GTX 960M's 128-bit bus and 4 GB capacity are meaningful for texture-heavy workloads.

Where Each One Wins

NVIDIA GeForce MX350 wins in:

• Vulkan performance: 13,077 versus 8,245, a 58.6% lead
• Average benchmark score: 10,883 versus 9,645
• Database percentile: 49th versus 46th
• Compute throughput: 1.879 TFLOPS FP32 versus 1.505 TFLOPS
• Pixel fill rate: 23.49 GPixel/s versus 18.82 GPixel/s
• Power efficiency: 20 W TDP versus 75 W
• Process technology: 14 nm versus 28 nm
• Transistor density: 25.0M per mm² versus 12.6M per mm²
• DirectX 12 feature level: 12_1 versus 11_0
• FP16 support: 29.36 GFLOPS versus none listed

NVIDIA GeForce GTX 960M wins in:

• OpenCL performance: 11,045 versus 8,689, a 21.3% lead
• Memory capacity: 4 GB versus 2 GB
• Memory bus width: 128-bit versus 64-bit
• Memory bandwidth: 80.19 GB/s versus 56.06 GB/s
• Texture mapping units: 40 versus 32
• Texture fill rate: 47.04 GTexel/s versus 46.98 GTexel/s

The GTX 960M's memory advantages are the key differentiator in its win column. The MX350's architectural modernity and Vulkan performance define its wins. Neither GPU sweeps, and the choice hinges entirely on which API and memory profile matters more for the intended use case.