NVIDIA GeForce GTX 880M vs NVIDIA GeForce GTX 960M Comparison

Head-to-Head Benchmarks

The database contains one directly comparable compute workload for these two mobile graphics processors: the Geekbench OpenCL test. In that test, the NVIDIA GeForce GTX 960M scores 11,045 points, while the NVIDIA GeForce GTX 880M scores 5,622 points. That is a 96.5% advantage for the GTX 960M, a massive margin that nearly doubles the older card’s output. The result is striking because the GTX 880M carries a much larger GPU die and more shading units, yet the Maxwell-based GTX 960M decisively outperforms it in this synthetic compute benchmark.

The GTX 960M also holds a higher average benchmark score across all recorded tests. Its average is 9,645 points, compared to 8,040 points for the GTX 880M. That gap is roughly 20%, which is substantial, though the head-to-head OpenCL result shows an even larger spread. The average includes other workloads where the GTX 880M appears, notably a Metal benchmark score of 10,458 points, but the GTX 960M has no recorded Metal result in the database. The OpenCL test is the only direct comparison available, and it strongly favors the newer architecture.

Looking at the percentile rankings, the GTX 960M sits at the 46th percentile of all GPUs in the database, while the GTX 880M sits at the 42nd percentile. That four-point gap reflects the average score difference and places the GTX 960M closer to the middle of the field. For context, the GTX 960M’s nearest rivals in average score are the NVIDIA Quadro K5000 at 9,637 points (0.1% ahead), the AMD Radeon Pro WX 2100 at 9,653 points (0.1% behind), and the NVIDIA Quadro P4000 at 9,665 points (0.2% behind). The GTX 880M’s nearest rivals are the NVIDIA Quadro P5000 at 8,039 points (0% difference), the NVIDIA GeForce GTX 650 Ti at 8,053 points (0.2% behind), and the NVIDIA GeForce GTX 650 Ti Boost at 8,067 points (0.3% behind). These rival comparisons show that the GTX 960M trades blows with professional workstation cards, while the GTX 880M sits in the same performance class as older desktop GeForce parts.

The single head-to-head benchmark result is unambiguous. The GTX 960M wins the only direct test, and the magnitude of that win, 96.5%, dwarfs the differences seen in the average score comparison. Even accounting for the GTX 880M’s strong Metal showing, which is not a shared test, the OpenCL data suggests the GTX 960M is the faster part for general-purpose compute workloads.

Where Each One Wins

The GTX 960M wins the only directly comparable benchmark, the Geekbench OpenCL test. That gives it a clean 1-0 record in head-to-head competition. Its advantage comes from a combination of higher clock speeds and a more efficient architecture. The GTX 960M runs at a base clock of 1,097 MHz with a boost clock of 1,176 MHz, while the GTX 880M runs at 954 MHz base and 993 MHz boost. That is a 143 MHz difference at base and 183 MHz at boost, which helps the smaller chip overcome its lower core count.

The GTX 880M, however, is not without strengths. It offers 8 GB of GDDR5 memory on a 256-bit bus, delivering 160.0 GB/s of bandwidth. The GTX 960M has 4 GB on a 128-bit bus, giving 80.19 GB/s. For workloads that depend on memory capacity or raw bandwidth, such as large texture sets or data-intensive compute, the GTX 880M holds a clear spec advantage. Its pixel rate is 31.78 GPixel/s versus 18.82 GPixel/s for the GTX 960M, and its texture rate is 127.1 GTexel/s versus 47.04 GTexel/s. Those numbers indicate that the GTX 880M should excel in fill-rate-bound scenarios, even though the OpenCL result does not reflect that advantage.

The GTX 880M also has a higher raw FP32 throughput at 3.050 TFLOPS, compared to 1.505 TFLOPS for the GTX 960M. That is exactly double the floating-point capability, yet the GTX 960M still wins the OpenCL test. This suggests the benchmark is sensitive to driver optimization, clock efficiency, or architectural improvements in Maxwell rather than pure theoretical throughput.

In practical terms, the GTX 960M is the better choice for compute-oriented tasks like OpenCL acceleration. The GTX 880M may be preferable for gaming scenarios that require large frame buffers or heavy texture filtering, given its 8 GB capacity and higher fill rates. The database does not include gaming benchmarks, so these conclusions rest on the recorded specs and the one compute test. The GTX 960M’s win count stands at one, while the GTX 880M has zero direct wins.

Architecture Differences

The two GPUs come from different NVIDIA generations and use different chip designs. The GTX 960M is built on the GM107 chip, part of the Maxwell architecture, and belongs to the GeForce 900M series. The GTX 880M uses the GK104 chip, part of the older Kepler architecture, and belongs to the GeForce 800M series. Both are manufactured by TSMC on a 28 nm process, so the lithography is identical, but the transistor counts and die sizes differ significantly.

The GTX 960M has 1,870 million transistors on a 148 mm² die, yielding a transistor density of 12.6 million per square millimeter. The GTX 880M has 3,540 million transistors on a 294 mm² die, with a slightly lower density of 12.0 million per square millimeter. The GTX 880M’s die is nearly twice as large, and it packs almost twice the transistors, but the Maxwell architecture in the GTX 960M achieves better compute performance per transistor in the recorded benchmark.

Core configuration also diverges sharply. The GTX 960M has 640 shading units, 40 texture mapping units, and 16 raster output units. The GTX 880M has 1,536 shading units, 128 TMUs, and 32 ROPs. That is 2.4 times the shading units, 3.2 times the TMUs, and twice the ROPs for the GTX 880M. Despite this, the GTX 960M wins the OpenCL test, which points to architectural efficiency gains in Maxwell, including better scheduling and lower overhead per instruction.

Memory subsystems differ as well. The GTX 960M uses 4 GB of GDDR5 with a 128-bit bus and 80.19 GB/s bandwidth. The GTX 880M uses 8 GB of GDDR5 with a 256-bit bus and 160.0 GB/s bandwidth. The GTX 880M offers double the capacity and double the bandwidth, which is a major advantage for memory-heavy workloads. The memory clocks are nearly identical: 1,253 MHz for the GTX 960M and 1,250 MHz for the GTX 880M, both yielding 5 Gbps effective. The bandwidth difference comes entirely from the wider bus on the GTX 880M.

Clock speeds favor the GTX 960M. Its base clock of 1,097 MHz and boost clock of 1,176 MHz are both higher than the GTX 880M’s 954 MHz base and 993 MHz boost. The GTX 960M also has a lower TDP at 75 W, compared to 122 W for the GTX 880M, meaning it delivers better performance per watt in the OpenCL test.

Both cards use the MXM-B (3.0) bus interface and come as MXM modules, with no power connectors listed and portable-device-dependent display outputs. API support shows differences: the GTX 960M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4, while the GTX 880M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The GTX 960M has a newer Vulkan version, though both are end-of-life products. The GTX 960M was released on March 12, 2015, succeeding the GeForce 800M series and preceding the GeForce 10 Mobile series. The GTX 880M was released on March 11, 2014, succeeding the GeForce 700M series and preceding the GeForce 900M series, which includes the GTX 960M.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 960M has an average benchmark score of 9,645 points, while the NVIDIA GeForce GTX 880M has an average of 8,040 points. The GTX 960M is about 20% higher.

Q: How much faster is the GTX 960M in the OpenCL test?

A: The GTX 960M scores 11,045 points versus 5,622 points for the GTX 880M, a 96.5% difference. The GTX 960M nearly doubles the GTX 880M’s OpenCL score.

Q: Does the GTX 880M have any advantages in memory?

A: Yes. The GTX 880M has 8 GB of GDDR5 memory on a 256-bit bus with 160.0 GB/s bandwidth, while the GTX 960M has 4 GB on a 128-bit bus with 80.19 GB/s. The GTX 880M offers double the capacity and double the bandwidth.

Q: What are the transistor counts and die sizes?

A: The GTX 960M has 1,870 million transistors on a 148 mm² die. The GTX 880M has 3,540 million transistors on a 294 mm² die. Both use a 28 nm TSMC process.

Q: Which GPU has more shading units?

A: The GTX 880M has 1,536 shading units, while the GTX 960M has 640. The GTX 880M also has 128 TMUs and 32 ROPs, compared to 40 TMUs and 16 ROPs for the GTX 960M.

Q: What is the TDP difference?

A: The GTX 960M has a TDP of 75 W, while the GTX 880M has a TDP of 122 W. The GTX 960M consumes significantly less power.

The Verdict

The data points to a clear winner for compute workloads: the NVIDIA GeForce GTX 960M. Its OpenCL score of 11,045 against 5,622 for the GTX 880M is a 96.5% margin, and its average benchmark score of 9,645 versus 8,040 confirms a consistent advantage. The GTX 960M achieves this with fewer shading units, fewer TMUs, fewer ROPs, half the FP32 throughput, and half the memory bandwidth. That outcome confirms the Maxwell architecture’s efficiency over Kepler.

The GTX 880M is not without merit. Its 8 GB frame buffer and 160.0 GB/s bandwidth make it better suited for workloads that require large memory pools or high fill rates. Its pixel rate of 31.78 GPixel/s and texture rate of 127.1 GTexel/s are both far above the GTX 960M’s figures. For gaming at high resolutions with heavy texture detail, the GTX 880M’s specs suggest it could hold its own, though the database does not include gaming benchmarks to confirm this.

The GTX 960M also wins on power efficiency. Its 75 W TDP is nearly half the GTX 880M’s 122 W, and it still delivers superior compute performance. For laptop designers, that lower power draw translates to less heat and longer battery life. The GTX 960M’s higher clock speeds, 1,097 MHz base and 1,176 MHz boost, contribute to its performance advantage despite the smaller core.

The percentile ranks reinforce the hierarchy: the GTX 960M sits at the 46th percentile of all GPUs, while the GTX 880M sits at the 42nd. Both are older, end-of-life parts, but the GTX 960M is the more capable one in the recorded metrics. Its nearest rivals include professional cards like the Quadro K5000 and Radeon Pro WX 2100, while the GTX 880M aligns with desktop cards like the GTX 650 Ti and GTX 650 Ti Boost.

For users choosing between these two mobile GPUs today, the GTX 960M is the safer pick for general compute and modern API support, including Vulkan 1.4 versus 1.2.175. The GTX 880M only makes sense if the use case explicitly demands 8 GB of VRAM or maximum memory bandwidth, where its 256-bit bus provides a clear spec advantage. Otherwise, the benchmark data favors the GTX 960M by a wide margin. The GTX 960M wins the head-to-head test, holds a higher average score, ranks higher in the percentile distribution, and does so at lower power. That is the verdict from the recorded data.