NVIDIA GeForce GTX 650 Ti Boost vs NVIDIA GeForce GTX 960M Comparison

Head-to-Head Benchmarks

The database records two compute benchmarks for this pair, and the results are split exactly one win apiece. The NVIDIA GeForce GTX 960M takes the OpenCL test, while the NVIDIA GeForce GTX 650 Ti Boost claims the Vulkan test.

In Geekbench OpenCL, the GTX 960M scores 11045 against the GTX 650 Ti Boost's 9302. That is an 18.7% advantage for the mobile Maxwell part. This is the larger of the two deltas recorded between these cards, and it is a decisive margin in a compute workload. The GTX 960M's score of 11045 also sits comfortably above its own average benchmark score of 9645, indicating that OpenCL is a strong workload for this chip.

In Geekbench Vulkan, the result flips. The GTX 650 Ti Boost scores 10041, while the GTX 960M manages 8245. The delta is 17.9% in favor of the older desktop card. Notably, the GTX 650 Ti Boost's Vulkan score of 10041 is its best recorded benchmark result across all tests in the database, exceeding its OpenCL score of 9302 by a wide margin. The GTX 960M's Vulkan score of 8245 is its weakest recorded result, landing well below its OpenCL score.

Looking at the broader averages, the GTX 960M holds a clear overall lead. Its average benchmark score is 9645, compared to 8067 for the GTX 650 Ti Boost. That is a difference of roughly 19.6% in favor of the GTX 960M when averaging across all recorded tests. The GTX 960M also sits in the 46th percentile of all GPUs in the database, while the GTX 650 Ti Boost sits in the 42nd percentile.

The GTX 960M's nearest rivals in the database include the NVIDIA Quadro K5000 at 9637 (0.1% behind), the AMD Radeon Pro WX 2100 at 9653 (0.1% ahead), and the NVIDIA Quadro P4000 at 9665 (0.2% ahead). The GTX 650 Ti Boost's nearest rivals include the NVIDIA GRID K2 at 8080 (0.2% ahead), the NVIDIA GeForce GTX 650 Ti at 8053 (0.2% behind), and the NVIDIA GeForce GTX 880M at 8040 (0.3% behind). These clustering patterns show that each card is firmly placed within its own performance tier, and the gap between the two tiers is substantial.

Architecture Differences

The two GPUs come from different NVIDIA architectures and are built on different chips. The GTX 960M uses the GM107 chip, which is based on the Maxwell architecture. The GTX 650 Ti Boost uses the GK106 chip, which is based on the older Kepler architecture. Both are manufactured on the same 28 nm process node at TSMC, but the similarities end there.

The transistor counts and die sizes differ significantly. The GTX 960M's GM107 packs 1,870 million transistors on a 148 mm² die, giving a transistor density of 12.6 million per square millimeter. The GTX 650 Ti Boost's GK106 is a larger chip, with 2,540 million transistors on a 221 mm² die, resulting in a lower density of 11.5 million per square millimeter. The Kepler chip is physically larger and has more transistors overall, but the Maxwell chip achieves a higher transistor density.

Clock speeds tell a different story. The GTX 960M runs at a base clock of 1097 MHz with a boost clock of 1176 MHz. The GTX 650 Ti Boost runs at a base clock of 980 MHz with a boost clock of 1032 MHz. The GTX 960M therefore has a higher base clock by 117 MHz and a higher boost clock by 144 MHz. The memory clocks also differ: the GTX 960M's GDDR5 memory runs at 1253 MHz (5 Gbps effective), while the GTX 650 Ti Boost's GDDR5 memory runs at 1502 MHz (6 Gbps effective).

The memory subsystems are configured differently. The GTX 960M has 4 GB of GDDR5 on a 128-bit bus, yielding 80.19 GB/s of bandwidth. The GTX 650 Ti Boost has 2 GB of GDDR5 on a 192-bit bus, yielding 144.2 GB/s of bandwidth. The desktop card has nearly 80% more memory bandwidth, which likely explains its strong Vulkan showing. The GTX 960M compensates with double the capacity.

Compute unit counts are mixed. The GTX 960M has 640 shading units, 40 texture mapping units, and 16 ROPs. The GTX 650 Ti Boost has 768 shading units, 64 texture mapping units, and 24 ROPs. The Kepler card leads in every one of these counts. However, the Maxwell architecture's efficiency improvements show up in the rates: the GTX 960M achieves a pixel rate of 18.82 GPixel/s and a texture rate of 47.04 GTexel/s, while the GTX 650 Ti Boost achieves 16.51 GPixel/s and 66.05 GTexel/s. The GTX 960M wins on pixel throughput, while the GTX 650 Ti Boost wins decisively on texture throughput. In raw FP32 compute, the GTX 960M delivers 1.505 TFLOPS, slightly behind the GTX 650 Ti Boost's 1.585 TFLOPS.

Power and physical design are major differentiators. The GTX 960M is rated at 75 W TDP, has no power connectors, and comes as an MXM module with an MXM-B (3.0) interface. The GTX 650 Ti Boost is rated at 134 W TDP, requires a single 6-pin power connector, has a suggested PSU of 300 W, and is a dual-slot PCIe 3.0 x16 card measuring 241 mm (9.5 inches) in length. The GTX 960M is designed for laptops and portable devices, with display outputs that are "Portable Device Dependent". The GTX 650 Ti Boost offers 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2.

Software support differs slightly. Both support DirectX 12 (11_0) and OpenGL 4.6. The GTX 960M supports Vulkan 1.4, while the GTX 650 Ti Boost supports Vulkan 1.2.175. This newer Vulkan version on the GTX 960M does not translate into better Vulkan benchmark performance, as the recorded data shows the opposite outcome.

Where Each One Wins

The GTX 960M is the clear winner in OpenCL compute workloads. Its score of 11045 versus 9302 gives it an 18.7% edge, and this is the kind of result that matters for general-purpose GPU compute tasks such as OpenCL-accelerated applications. The GTX 960M also holds the advantage in overall average benchmark score (9645 versus 8067) and percentile ranking (46th versus 42nd). If you are looking at the broader picture across all recorded workloads, the GTX 960M is the stronger card.

The GTX 960M also wins on memory capacity with 4 GB versus 2 GB, which matters for modern workloads that exceed the 2 GB frame buffer limit. It has a higher pixel rate (18.82 GPixel/s versus 16.51 GPixel/s), which can benefit fill-rate-bound scenarios. It consumes far less power at 75 W versus 134 W, and it does not require external power connectors.

The GTX 650 Ti Boost wins in Vulkan compute, scoring 10041 versus 8245, a 17.9% margin. This is a substantial advantage in a workload that is increasingly relevant for modern games and compute APIs. The GTX 650 Ti Boost also leads in memory bandwidth (144.2 GB/s versus 80.19 GB/s), texture rate (66.05 GTexel/s versus 47.04 GTexel/s), and raw FP32 throughput (1.585 TFLOPS versus 1.505 TFLOPS). It has more shading units (768 versus 640), more TMUs (64 versus 40), and more ROPs (24 versus 16).

The GTX 650 Ti Boost also has the advantage of being a desktop card with a standard PCIe 3.0 x16 interface and dedicated display outputs. It was released with a launch MSRP of 169 USD. Its higher TDP of 134 W and single 6-pin power connector reflect a design that can draw more power and sustain higher sustained clocks under load. The GTX 960M's MXM form factor means it is tied to a specific laptop or portable device implementation, which can limit its practical usability.

The Verdict

The data points to a clear split decision, but the overall average favors the GTX 960M. The GTX 960M's average benchmark score of 9645 is roughly 19.6% higher than the GTX 650 Ti Boost's 8067. Its OpenCL win is decisive at 18.7%, and its percentile ranking is four points higher. For users who prioritize compute performance in OpenCL, or who need 4 GB of memory capacity, the GTX 960M is the better choice from the recorded data.

However, the GTX 650 Ti Boost is the better choice for Vulkan-based workloads. Its Vulkan score of 10041 beats the GTX 960M's 8245 by 17.9%, and that margin is nearly as large as the GTX 960M's OpenCL advantage. The GTX 650 Ti Boost also offers substantially higher memory bandwidth (144.2 GB/s versus 80.19 GB/s), which is a critical factor for many graphics workloads. Its higher texture rate and FP32 throughput give it an edge in raw shader-heavy tasks.

The form factor difference is fundamental. The GTX 960M is an MXM module for portable devices, with a 75 W TDP and no external power requirement. The GTX 650 Ti Boost is a dual-slot desktop card, 241 mm long, requiring a 6-pin power connector and a 300 W PSU. These are not interchangeable products. The GTX 960M belongs in a laptop or compact MXM-based system. The GTX 650 Ti Boost belongs in a desktop tower with available PCIe slots and power headroom.

For a laptop user who wants the better average compute performance and more memory, the GTX 960M is the pick. For a desktop user who values Vulkan performance, memory bandwidth, and raw texture throughput, the GTX 650 Ti Boost has the edge. The GTX 960M's higher percentile (46 versus 42) and higher average score indicate it is the stronger all-around card in the database.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 960M has an average benchmark score of 9645, compared to 8067 for the NVIDIA GeForce GTX 650 Ti Boost.

Q: How large is the GTX 960M's OpenCL advantage?

A: The GTX 960M scores 11045 in Geekbench OpenCL, while the GTX 650 Ti Boost scores 9302. That is an 18.7% lead for the GTX 960M.

Q: Does the GTX 650 Ti Boost win any benchmark?

A: Yes, it wins the Geekbench Vulkan test with a score of 10041, beating the GTX 960M's 8245 by 17.9%.

Q: Which card has more memory bandwidth?

A: The GTX 650 Ti Boost has 144.2 GB/s of bandwidth on a 192-bit bus, while the GTX 960M has 80.19 GB/s on a 128-bit bus.

Q: What are the power requirements for each card?

A: The GTX 960M has a 75 W TDP and requires no power connectors. The GTX 650 Ti Boost has a 134 W TDP, requires a single 6-pin power connector, and has a suggested PSU of 300 W.

Q: Which card supports a newer Vulkan version?

A: The GTX 960M supports Vulkan 1.4, while the GTX 650 Ti Boost supports Vulkan 1.2.175. Despite this, the GTX 650 Ti Boost scores higher in the Vulkan benchmark.

Specification Differences

| Specification | NVIDIA GeForce GTX 960M | NVIDIA GeForce GTX 650 Ti Boost |

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

| Chip | GM107 | GK106 |

| Architecture | Maxwell | Kepler |

| Generation | GeForce 900M | GeForce 600 |

| Transistors | 1,870 million | 2,540 million |

| Die Size | 148 mm² | 221 mm² |

| Transistor Density | 12.6M / mm² | 11.5M / mm² |

| Base Clock | 1097 MHz | 980 MHz |

| Boost Clock | 1176 MHz | 1032 MHz |

| Memory Clock | 1253 MHz, 5 Gbps effective | 1502 MHz, 6 Gbps effective |

| Memory Size | 4 GB | 2 GB |

| Memory Bus Width | 128 bit | 192 bit |

| Memory Bandwidth | 80.19 GB/s | 144.2 GB/s |

| Shading Units | 640 | 768 |

| TMUs | 40 | 64 |

| ROPs | 16 | 24 |

| Pixel Rate | 18.82 GPixel/s | 16.51 GPixel/s |

| Texture Rate | 47.04 GTexel/s | 66.05 GTexel/s |

| FP32 | 1.505 TFLOPS | 1.585 TFLOPS |

| TDP | 75 W | 134 W |

| Slot Width | MXM Module | Dual-slot |

| Power Connectors | None | 1x 6-pin |

| Suggested PSU | Not specified | 300 W |

| Bus Interface | MXM-B (3.0) | PCIe 3.0 x16 |

| Display Outputs | Portable Device Dependent | 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2 |

| Vulkan Version | 1.4 | 1.2.175 |

| Length | Not specified | 241 mm (9.5 inches) |

| Release Date | 2015-03-12 | 2013-03-25 |

| Launch MSRP | None | 169 USD |