NVIDIA GeForce GTX 880M vs NVIDIA GeForce GTX 950M Comparison

The NVIDIA GeForce GTX 950M and NVIDIA GeForce GTX 880M represent two distinct approaches to mobile graphics within NVIDIA’s lineup, separated by a generation of architecture and a fundamental shift in design priorities. The GTX 950M, built on the Maxwell architecture, is the newer part, while the GTX 880M is a Kepler-based flagship from the previous generation. The benchmark data shows a clear split: the GTX 950M wins the single available head-to-head test in OpenCL, but the GTX 880M counters with a massive advantage in raw hardware resources and memory bandwidth. This analysis walks through the numbers to determine which card suits which workload, based strictly on the provided facts.

Head-to-Head Benchmarks

The only direct benchmark comparison available is the Geekbench OpenCL test, and the results are starkly one-sided. The GTX 950M scores 9,745, while the GTX 880M scores 5,622. That is a delta of 73.3% in favor of the GTX 950M. This is not a marginal victory; it is a dominant performance gap where the newer Maxwell part nearly doubles the compute output of the older Kepler flagship in this specific workload.

Breaking that result down, the GTX 950M’s OpenCL score of 9,745 places it in the 42nd percentile of all GPUs, with an average benchmark score of 8,135. Its nearest rival, the AMD Radeon R9 M360, scores 8,129, a delta of just 0.1% — essentially a statistical tie. The GTX 950M also edges out the NVIDIA GeForce 945M (8,099, +0.4%) and the GRID K2 (8,080, +0.7%). However, it sits slightly behind the NVIDIA GeForce GTX 980 (8,167, -0.4%) in the same average score ranking. This shows that while the GTX 950M wins the OpenCL test against the GTX 880M, its overall average benchmark score of 8,135 is only marginally above the GTX 880M’s 8,040.

The GTX 880M’s own benchmark scores tell a different story. Its only listed benchmark result is in Geekbench Metal, where it scores 10,458 — a number that is not directly comparable to the OpenCL scores but indicates strong performance in Apple’s graphics API. Its OpenCL score of 5,622 is the weak point, dragging its average down. The GTX 880M’s average benchmark score of 8,040 places it in the same 42nd percentile as the GTX 950M, and its nearest rivals are the NVIDIA Quadro P5000 (8,039, 0% delta), the GTX 650 Ti (8,053, -0.2%), and the GTX 650 Ti Boost (8,067, -0.3%). The data indicates that in average terms, these two GPUs are nearly indistinguishable, despite the massive OpenCL delta.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The GTX 950M has an average benchmark score of 8,135, which is 1.2% higher than the GTX 880M’s 8,040. Both GPUs sit at the 42nd percentile of all GPUs.

Q: How much faster is the GTX 950M in Geekbench OpenCL?

A: The GTX 950M scores 9,745 versus the GTX 880M’s 5,622 in the OpenCL test, a delta of 73.3% in favor of the GTX 950M.

Q: Does the GTX 880M have any benchmark where it wins?

A: The head-to-head data lists only one test (Geekbench OpenCL) and the GTX 880M has zero wins. However, its Geekbench Metal score of 10,458 is the single highest individual benchmark number recorded for either GPU.

Q: What is the memory bandwidth difference between the two?

A: The GTX 880M provides 160.0 GB/s of bandwidth over a 256-bit bus with GDDR5 memory, whereas the GTX 950M provides 28.80 GB/s over a 128-bit bus with DDR3 memory. The GTX 880M’s bandwidth is over 5.5 times higher.

Q: Which GPU has more shading units?

A: The GTX 880M has 1,536 shading units, which is 2.4 times the 640 shading units found on the GTX 950M.

Q: What are the transistor counts for each chip?

A: The GTX 880M’s GK104 chip contains 3,540 million transistors on a 294 mm² die, while the GTX 950M’s GM107 chip contains 1,870 million transistors on a 148 mm² die.

Architecture Differences

The two GPUs are built on fundamentally different architectures. The GTX 950M uses the Maxwell architecture with the GM107 chip, while the GTX 880M uses the Kepler architecture with the GK104 chip. Both are fabricated on a 28 nm process at TSMC, but the design philosophies diverge sharply.

The GTX 880M’s GK104 is a much larger and more complex chip. It packs 3,540 million transistors onto a 294 mm² die, yielding a transistor density of 12.0M per mm². In contrast, the GTX 950M’s GM107 has 1,870 million transistors on a 148 mm² die, with a slightly higher density of 12.6M per mm². The GTX 880M’s chip is 2.4 times larger by die area and holds 1.9 times more transistors, reflecting its role as a high-end part with more execution resources.

The shading unit count confirms this: the GTX 880M has 1,536 shading units, 128 texture mapping units (TMUs), and 32 render output units (ROPs). The GTX 950M has only 640 shading units, 40 TMUs, and 16 ROPs. This means the GTX 880M has 2.4 times the shading units, 3.2 times the TMUs, and 2 times the ROPs. The compute throughput reflects this: the GTX 880M delivers 3.050 TFLOPS of FP32 performance, while the GTX 950M delivers 1,438.7 GFLOPS — the 880M is more than twice as fast in raw shader math.

Clock speeds tell a different story. The GTX 950M runs at a base clock of 993 MHz with a boost of 1124 MHz, while the GTX 880M runs at 954 MHz base and 993 MHz boost. The GTX 950M’s clocks are 4% higher at base and 13% higher at boost, which helps explain its superior OpenCL efficiency despite fewer cores.

The memory subsystems are entirely different in type and scale. The GTX 880M uses 8 GB of GDDR5 memory on a 256-bit bus, achieving 160.0 GB/s of bandwidth. The GTX 950M uses 4 GB of DDR3 on a 128-bit bus, achieving only 28.80 GB/s. The GTX 880M’s bandwidth advantage is massive — over 5.5 times higher — which is critical for texture-heavy and high-resolution workloads.

Both support DirectX 12 (11_0), OpenGL 4.6, and Vulkan, though the Vulkan versions differ: the GTX 950M supports Vulkan 1.4, while the GTX 880M supports Vulkan 1.2.175. Neither GPU has ray tracing or tensor cores.

Specification Differences

The specification sheets reveal clear divergences in every major category except manufacturing and API support. The most obvious difference is memory capacity and type: the GTX 950M has 4 GB of DDR3, while the GTX 880M has 8 GB of GDDR5. The bus widths are 128-bit versus 256-bit, respectively, and bandwidth is 28.80 GB/s versus 160.0 GB/s.

The compute resources differ by roughly a factor of two to three. The GTX 880M has 1,536 shading units, 128 TMUs, and 32 ROPs, compared to the GTX 950M’s 640 shading units, 40 TMUs, and 16 ROPs. Pixel rate is 31.78 GPixel/s for the GTX 880M versus 17.98 GPixel/s for the GTX 950M. Texture rate is 127.1 GTexel/s versus 44.96 GTexel/s. FP32 compute is 3.050 TFLOPS versus 1,438.7 GFLOPS.

Clock speeds are slightly higher on the GTX 950M: 993 MHz base and 1124 MHz boost, compared to 954 MHz base and 993 MHz boost on the GTX 880M. Memory clocks are also different, with the GTX 950M running at 900 MHz (1800 Mbps effective) and the GTX 880M at 1250 MHz (5 Gbps effective).

Power and physical design differ substantially. The GTX 950M has a TDP of 75 W and uses an IGP slot width with no power connectors, indicating it is designed for thinner, lower-power laptops. The GTX 880M has a TDP of 122 W and uses an MXM Module form factor with a bus interface of MXM-B (3.0), indicating a larger, more power-hungry mobile workstation or gaming laptop design. The GTX 950M uses PCIe 3.0 x8, while the GTX 880M uses MXM-B (3.0).

The chips themselves are different: GM107 for the GTX 950M and GK104 for the GTX 880M. Transistor counts are 1,870 million versus 3,540 million, and die sizes are 148 mm² versus 294 mm². The GTX 950M has a slightly higher transistor density at 12.6M / mm² versus 12.0M / mm².

The GTX 950M is from the GeForce 900M generation, released on 2015-03-12, while the GTX 880M is from the GeForce 800M generation, released on 2014-03-11. Both are end-of-life products. The GTX 950M’s predecessor is the GeForce 800M and its successor is the GeForce 10 Mobile, while the GTX 880M’s predecessor is the GeForce 700M and its successor is the GeForce 900M.

Where Each One Wins

The GTX 950M wins decisively in compute-bound workloads that stress the OpenCL pipeline. Its Geekbench OpenCL score of 9,745 is 73.3% higher than the GTX 880M’s 5,622. This indicates that for applications leveraging OpenCL — such as certain content creation tools, physics simulations, or compute offload — the GTX 950M is the superior choice despite its smaller hardware footprint. Its higher boost clock of 1124 MHz and newer Maxwell architecture likely contribute to this efficiency. The GTX 950M also has the advantage of a lower TDP of 75 W, making it suitable for thinner, more portable laptops where power draw is a constraint.

The GTX 880M wins in raw graphics throughput and memory-intensive scenarios. Its 160.0 GB/s of memory bandwidth and 8 GB of GDDR5 memory are more than 5.5 times higher bandwidth than the GTX 950M’s 28.80 GB/s. For high-resolution textures, large frame buffers, or multi-monitor setups, the GTX 880M is clearly better equipped. Its higher pixel rate (31.78 GPixel/s vs 17.98 GPixel/s) and texture rate (127.1 GTexel/s vs 44.96 GTexel/s) mean it can fill frames faster when the shading workload is not the bottleneck. The GTX 880M also has a higher Geekbench Metal score of 10,458, suggesting it performs well in Metal-based applications on compatible platforms.

The GTX 880M’s larger transistor budget (3,540 million vs 1,870 million) and more than double the shading units (1,536 vs 640) make it the part for situations where raw shader complexity is paramount, such as demanding game titles at high settings, provided the software can utilize its Kepler architecture efficiently. Its 122 W TDP and MXM Module form factor indicate it is intended for larger, more power-tolerant laptops where that performance can be sustained.

The Verdict

The data presents a clear but nuanced picture. For users whose primary workload involves OpenCL compute, the GTX 950M is the unambiguous winner. Its 73.3% lead in the Geekbench OpenCL test is the single largest performance gap in the dataset, and its lower TDP of 75 W makes it the more practical choice for modern, slim laptops. The GTX 950M’s average benchmark score of 8,135 is also higher than the GTX 880M’s 8,040, even though both sit at the 42nd percentile.

However, for users who need sustained graphics performance in memory-hungry scenarios, the GTX 880M is the stronger candidate. Its 160.0 GB/s bandwidth and 8 GB of GDDR5 memory are overwhelming advantages that cannot be overlooked. The GTX 880M’s higher pixel and texture rates (31.78 GPixel/s and 127.1 GTexel/s) indicate it can drive more complex scenes at higher resolutions. Its Geekbench Metal score of 10,458 is the highest individual benchmark number recorded for either GPU.

The decision ultimately hinges on the application mix. If the workload is compute-centric and benefits from Maxwell’s efficiency, choose the GTX 950M. If the workload is graphics-centric with large textures and high fill rates, the GTX 880M’s superior memory subsystem and raw throughput make it the better pick. The GTX 950M offers better compute efficiency and lower power draw; the GTX 880M offers brute-force graphics capability. Neither is a universal winner, but the GTX 950M’s decisive OpenCL victory and higher average score give it a slight edge in the overall benchmark data.