NVIDIA GeForce GTX 750 vs NVIDIA GeForce GTX 950M Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 950M records an average benchmark score of 8135, while the NVIDIA GeForce GTX 750 scores 7222. The 950M sits at the 42nd percentile of all GPUs in the database, compared to the 750's 40th percentile.

Q: How do the two cards compare in OpenCL performance?

A: The GTX 950M wins the Geekbench OpenCL test with a score of 9745 against the GTX 750's 9315, a margin of 4.6%. Both cards share the same Maxwell architecture and GM107 chip, but the 950M's higher shading unit count and clock configuration give it the edge in this workload.

Q: Does the GTX 750 win any benchmark?

A: Yes, the GTX 750 takes the Geekbench Vulkan test decisively, scoring 8078 versus the GTX 950M's 6525. That is a 19.2% advantage for the desktop card. The 950M does not have a recorded Vulkan result in the head-to-head comparison beyond its lower score.

Q: What memory configurations do these cards use?

A: The GTX 950M ships with 4 GB of DDR3 memory on a 128-bit bus, delivering 28.80 GB/s of bandwidth. The GTX 750 uses 1024 MB of GDDR5 memory on the same 128-bit bus but reaches 80.19 GB/s, a much higher bandwidth figure.

Q: Are these GPUs still in production?

A: No, both are end-of-life products. The GTX 950M was released in March 2015 as part of the GeForce 900M generation, while the GTX 750 launched in February 2014 in the GeForce 700 series.

Q: What is the power draw difference?

A: The GTX 950M has a TDP of 75 W and is classified as an IGP (integrated graphics package) with no power connectors. The GTX 750 has a 55 W TDP, is a single-slot card, also requires no power connectors, and carries a suggested PSU rating of 250 W.

Architecture Differences

Both the NVIDIA GeForce GTX 950M and the NVIDIA GeForce GTX 750 are built on the same fundamental design: the GM107 chip, fabricated by TSMC on a 28 nm process. Each die contains 1,870 million transistors on a 148 mm² area, giving a transistor density of 12.6M per mm². Both are Maxwell architecture parts with identical API support: DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4.

The core configuration is where the two diverge. The GTX 950M carries 640 shading units, 40 texture mapping units, and 16 ROPs. The GTX 750 has 512 shading units, 32 TMUs, and the same 16 ROPs. That means the 950M has 25% more shading units and 25% more texture units, which directly explains its higher theoretical texture fill rate of 44.96 GTexel/s versus 34.72 GTexel/s. Pixel rates are nearly identical: 17.98 GPixel/s for the 950M and 17.36 GPixel/s for the 750.

Clock speeds tell a more nuanced story. The GTX 950M runs at a base clock of 993 MHz with a boost of 1124 MHz. The GTX 750 has a higher base clock of 1020 MHz but a lower boost of 1085 MHz. So the 950M relies on its boost state to pull ahead, while the 750 starts from a higher idle point. The raw FP32 throughput reflects this: the 950M reaches 1,438.7 GFLOPS, while the 750 manages 1,111.0 GFLOPS.

Memory architecture is a major differentiator. The 950M uses 4 GB of DDR3 with a 128-bit bus, yielding just 28.80 GB/s of bandwidth. The 750 uses 1024 MB of GDDR5, also on a 128-bit bus, but delivers 80.19 GB/s, nearly three times the bandwidth. This is a critical distinction for texture-heavy workloads and high-resolution assets. The 950M compensates with larger capacity, but the 750's memory speed is a clear advantage in bandwidth-sensitive tasks.

The physical designs differ substantially. The 950M is an IGP, meaning it is intended to be soldered onto a motherboard, with display outputs described as "portable device dependent." The 750 is a standalone single-slot card measuring 145 mm (5.7 inches) in length, with 2x DVI and 1x mini-HDMI 1.4a outputs. The bus interfaces also differ: the 950M uses PCIe 3.0 x8, while the 750 uses the full PCIe 3.0 x16 interface. Power delivery is similar in that neither requires external connectors, but the 950M draws 75 W versus the 750's 55 W, and the 750 lists a 250 W suggested PSU.

Generation placement matters. The 950M belongs to the GeForce 900M mobile series, with the GeForce 800M as its predecessor and GeForce 10 Mobile as its successor. The 750 belongs to the GeForce 700 desktop series, preceded by GeForce 600 and succeeded by GeForce 900. Both are end-of-life, but they targeted different market segments from the start.

Head-to-Head Benchmarks

The database records two direct benchmark comparisons between these GPUs, and the results split evenly: one win for each card.

In Geekbench OpenCL, the GTX 950M posts 9745 points against the GTX 750's 9315. That is a 4.6% victory for the mobile part. The margin is modest, but it aligns with the 950M's higher shading unit count and greater FP32 throughput. The 950M's 1,438.7 GFLOPS versus 1,111.0 GFLOPS gives it a theoretical compute advantage that translates into a real, if narrow, OpenCL win. The larger 4 GB frame buffer likely helps as well, allowing larger working sets without spilling to system memory.

The Geekbench Vulkan test flips the script. The GTX 750 scores 8078, while the GTX 950M manages only 6525. That is a 19.2% gap in favor of the desktop card. This is the largest performance delta in either direction. The 750's GDDR5 memory with 80.19 GB/s of bandwidth is the likely culprit for this outcome. Vulkan workloads often stress memory bandwidth heavily, and the 750 has nearly three times the bandwidth of the 950M. The 950M's DDR3 bus simply cannot feed its shader units as effectively in this API.

The overall average benchmark scores reinforce the mixed picture. The 950M averages 8135 across all recorded tests, placing it at the 42nd percentile. The 750 averages 7222, at the 40th percentile. The 950M's nearest rivals include the AMD Radeon R9 M360 (8129, just 0.1% behind) and the NVIDIA GeForce GTX 980 (8167, 0.4% ahead). The 750's nearest rivals include the AMD Radeon Vega 8 Mobile (7203, 0.3% behind) and the NVIDIA GeForce GTX 560 SE (7171, 0.7% behind).

The Vulkan result is worth emphasizing because it is not a close contest. A 19.2% lead for the GTX 750 is substantial in real-world terms, and it suggests that the desktop card's memory subsystem gives it a decisive edge in modern API workloads. The OpenCL result, by contrast, is near parity, with only 4.6% separating the two. Neither card is a dominant winner overall; the data shows a split decision where each GPU leads in a different API environment.

The Verdict

The data paints a clear picture for different use cases. The NVIDIA GeForce GTX 950M is the better choice for compute-oriented OpenCL workloads and for scenarios where 4 GB of memory capacity matters. Its 25% higher shading unit count and 1,438.7 GFLOPS of FP32 throughput give it a real advantage in general compute, and its 4.6% OpenCL win over the 750 confirms that edge. The 950M also posts a higher average benchmark score (8135 versus 7222) and a higher percentile ranking (42nd versus 40th), making it the stronger all-around performer in the database's aggregate metrics.

The NVIDIA GeForce GTX 750 is the pick for Vulkan-based applications and for any workload that depends heavily on memory bandwidth. Its 80.19 GB/s of GDDR5 bandwidth dwarfs the 950M's 28.80 GB/s, and the 19.2% Vulkan victory demonstrates how much that bandwidth matters in practice. The 750 also has a higher base clock (1020 MHz versus 993 MHz), which helps in lightly threaded or latency-sensitive tasks, and its 55 W TDP makes it a more power-efficient desktop option.

For a mobile platform, the 950M makes sense as an integrated part with no power connectors and portable-device-dependent outputs. For a desktop build, the 750 is a compact single-slot card with standard display outputs and a 250 W suggested PSU. The choice comes down to environment and workload priority: compute and capacity favor the 950M, bandwidth and Vulkan favor the 750. The database records one win each in the head-to-head tests, and the average scores favor the 950M, but the 750's Vulkan dominance is too large to ignore for modern API users.

Specification Differences

| Specification | NVIDIA GeForce GTX 950M | NVIDIA GeForce GTX 750 |

|---|---|---|

| Generation | GeForce 900M | GeForce 700 |

| Process Node | 28 nm | 28 nm |

| Transistors | 1,870 million | 1,870 million |

| Die Size | 148 mm² | 148 mm² |

| Base Clock | 993 MHz | 1020 MHz |

| Boost Clock | 1124 MHz | 1085 MHz |

| Memory Clock | 1800 Mbps effective | 5 Gbps effective |

| Memory Size | 4 GB | 1024 MB |

| Memory Type | DDR3 | GDDR5 |

| Memory Bus Width | 128 bit | 128 bit |

| Memory Bandwidth | 28.80 GB/s | 80.19 GB/s |

| Shading Units | 640 | 512 |

| TMUs | 40 | 32 |

| ROPs | 16 | 16 |

| Pixel Rate | 17.98 GPixel/s | 17.36 GPixel/s |

| Texture Rate | 44.96 GTexel/s | 34.72 GTexel/s |

| FP32 | 1,438.7 GFLOPS | 1,111.0 GFLOPS |

| TDP | 75 W | 55 W |

| Slot Width | IGP | Single-slot |

| Power Connectors | None | None |

| Suggested PSU | Not specified | 250 W |

| Bus Interface | PCIe 3.0 x8 | PCIe 3.0 x16 |

| Display Outputs | Portable Device Dependent | 2x DVI, 1x mini-HDMI 1.4a |

| Length | Not specified | 145 mm (5.7 inches) |

| Release Date | 2015-03-12 | 2014-02-17 |

| Predecessor | GeForce 800M | GeForce 600 |

| Successor | GeForce 10 Mobile | GeForce 900 |

| Launch MSRP | Not specified | 119 USD |