NVIDIA GeForce GTX 650 Ti vs NVIDIA GeForce GTX 960 Comparison

Head-to-Head Benchmarks

The recorded data includes two head-to-head comparisons between the NVIDIA GeForce GTX 960 and the NVIDIA GeForce GTX 650 Ti, and the GTX 960 wins both. In the Geekbench OpenCL test, the GTX 960 scores 18,925 against the GTX 650 Ti's 7,877, a delta of 140.3%. That is a decisive margin, more than double the older card's compute output in this workload. The Geekbench Vulkan test is closer but still favors the GTX 960: 9,231 versus 8,229, a 12.2% advantage. The GTX 960's average benchmark score across all recorded tests is 9,273, while the GTX 650 Ti averages 8,053. The percentile rankings reflect this gap: the GTX 960 sits at the 45th percentile of all GPUs in the database, while the GTX 650 Ti sits at the 42nd.

Looking at the nearest rivals in the database provides context for these scores. The GTX 960's average score of 9,273 places it within 0.2% of the NVIDIA GeForce GTX 465 (9,294) and 0.3% of the NVIDIA GeForce GTX 850M (9,302). It is also 0.4% behind the AMD Radeon R7 M380 (9,313) and 0.6% ahead of the AMD Radeon Vega 8 (9,221). These are tight margins, suggesting the GTX 960 sits in a crowded performance band. The GTX 650 Ti's average of 8,053 is similarly clustered: 0.2% behind the NVIDIA GeForce GTX 880M (8,040), 0.2% ahead of the NVIDIA GeForce GTX 650 Ti Boost (8,067), 0.2% ahead of the NVIDIA Quadro P5000 (8,039), and 0.3% behind the NVIDIA GRID K2 (8,080). The two cards do not directly overlap in their rival clusters; the GTX 960 operates in a higher tier.

The OpenCL result is particularly striking. A 140.3% delta means the GTX 960 delivers well over double the GTX 650 Ti's OpenCL score. This is not an incremental improvement; it is a generational leap in raw compute throughput. The Vulkan delta of 12.2% is more modest, indicating that the newer architecture's advantage narrows when both cards are pushed through a modern graphics API. Still, the GTX 960 wins outright in every recorded head-to-head test, with a 2-0 record.

Where Each One Wins

The GTX 960 wins in every benchmark category where both cards have recorded data. In OpenCL, its 18,925 score dwarfs the GTX 650 Ti's 7,877. This suggests the GTX 960 is the clear choice for compute-oriented workloads that leverage OpenCL, such as general-purpose GPU tasks, certain rendering pipelines, and non-gaming acceleration. The 140.3% lead is not a marginal advantage; it indicates the GTX 960 can handle substantially heavier compute loads without bottlenecking.

In Vulkan, the GTX 960 scores 9,231 versus 8,229, a 12.2% lead. Vulkan is a low-overhead API used in modern games and applications, and the GTX 960's advantage here, while smaller, still points to better frame pacing and draw-call handling. For gaming scenarios built around Vulkan, the GTX 960 should deliver smoother performance, though the gap is far less dramatic than in OpenCL.

The GTX 650 Ti has no recorded wins in any head-to-head test. However, the data does not show a scenario where it excels. Its only recorded benchmarks are Geekbench OpenCL (7,877) and Geekbench Vulkan (8,229), both of which trail the GTX 960. There is no evidence in the database of a workload where the older card pulls ahead. Its lower power draw (110 W versus 120 W) and single-slot design might appeal to specific system constraints, but in pure benchmark performance, the GTX 960 dominates.

For users prioritizing compute throughput, the GTX 960 is the only rational choice based on the data. For those locked into older software that does not leverage modern APIs, the GTX 650 Ti's Vulkan score of 8,229 is still respectable relative to its own generation, but it cannot match the GTX 960's 9,231. The GTX 650 Ti's average score of 8,053 places it in the 42nd percentile, while the GTX 960's 9,273 places it in the 45th, a modest overall ranking difference that understates the OpenCL gap.

Architecture Differences

The two cards come from different NVIDIA generations and architectures. The GTX 960 uses the GM206 chip built on Maxwell 2.0, part of the GeForce 900 generation. The GTX 650 Ti uses the GK106S chip built on Kepler, part of the GeForce 600 generation. Both are fabricated on a 28 nm process at TSMC, but the transistor counts differ: the GTX 960 packs 2,940 million transistors on a 228 mm² die, while the GTX 650 Ti has 2,540 million transistors on a 221 mm² die. The transistor density is 12.9 million per mm² for the GTX 960 versus 11.5 million per mm² for the GTX 650 Ti, reflecting a denser, more modern layout.

The GTX 960 has 1,024 shading units, 64 texture mapping units, and 32 ROPs. The GTX 650 Ti has 768 shading units, 64 texture mapping units, and only 16 ROPs. The doubling of ROPs on the GTX 960 is significant for pixel throughput: the GTX 960 delivers 37.70 GPixel/s versus 14.85 GPixel/s on the GTX 650 Ti. Texture fill rates are closer: 75.39 GTexel/s versus 59.39 GTexel/s. Floating-point performance (FP32) is 2.413 TFLOPS on the GTX 960 against 1,425.4 GFLOPS on the GTX 650 Ti, roughly a 69% advantage.

Memory configurations also differ. The GTX 960 has 2 GB of GDDR5 on a 128-bit bus, yielding 112.2 GB/s of bandwidth. The GTX 650 Ti has 1,024 MB (1 GB) of GDDR5 on the same 128-bit bus, yielding 86.40 GB/s. The GTX 960's memory clock is 1,753 MHz (7 Gbps effective), while the GTX 650 Ti runs at 1,350 MHz (5.4 Gbps effective). Both cards use a PCIe 3.0 x16 interface and require a 300 W suggested PSU, but the GTX 960 has a 120 W TDP versus the GTX 650 Ti's 110 W.

API support diverges as well. The GTX 960 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 650 Ti supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The DirectX feature level difference (12_1 versus 11_0) is notable for modern games that use advanced DX12 features. Display outputs differ: the GTX 960 offers 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2, while the GTX 650 Ti offers 2x DVI and 1x mini-HDMI 1.4a. The GTX 960 is dual-slot and 241 mm long, while the GTX 650 Ti is single-slot and 145 mm long.

FAQ

Q: Which card has a higher average benchmark score?

A: The GTX 960 averages 9,273 across all recorded benchmarks, while the GTX 650 Ti averages 8,053.

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

A: The GTX 960 scores 18,925 in Geekbench OpenCL versus 7,877 for the GTX 650 Ti, a 140.3% lead.

Q: Is the GTX 960 ahead in Vulkan as well?

A: Yes, the GTX 960 scores 9,231 in Geekbench Vulkan versus 8,229 for the GTX 650 Ti, a 12.2% advantage.

Q: What are the memory sizes and bandwidths?

A: The GTX 960 has 2 GB of GDDR5 with 112.2 GB/s bandwidth. The GTX 650 Ti has 1,024 MB of GDDR5 with 86.40 GB/s bandwidth.

Q: How do the transistor counts compare?

A: The GTX 960 has 2,940 million transistors on a 228 mm² die. The GTX 650 Ti has 2,540 million transistors on a 221 mm² die.

Q: Which card supports a higher DirectX feature level?

A: The GTX 960 supports DirectX 12 (12_1), while the GTX 650 Ti supports DirectX 12 (11_0).

The Verdict

The data points clearly to the GTX 960 as the superior performer. It wins both head-to-head benchmarks, holds a higher average score (9,273 versus 8,053), and sits at a higher percentile (45th versus 42nd). The OpenCL margin of 140.3% is the single largest differentiator, making the GTX 960 the obvious pick for compute-heavy tasks. The Vulkan margin of 12.2% reinforces the same conclusion for gaming and modern API workloads. The GTX 960 also offers double the ROPs (32 versus 16), higher pixel rate (37.70 GPixel/s versus 14.85 GPixel/s), more shading units (1,024 versus 768), and more memory (2 GB versus 1,024 MB). Its bandwidth advantage (112.2 GB/s versus 86.40 GB/s) supports higher resolutions and texture-heavy scenes.

The GTX 650 Ti is not without merits in the database. Its lower TDP (110 W versus 120 W) and single-slot design make it easier to fit into compact systems, and its 145 mm length is substantially shorter than the GTX 960's 241 mm. Its Vulkan score of 8,229 is respectable and close to the GTX 960's 9,231, suggesting that in Vulkan titles the gap is manageable. But for anyone choosing between these two based on benchmark data, the GTX 960 wins every measurable category. The GTX 650 Ti's nearest rivals include the GTX 650 Ti Boost (8,067) and the GTX 880M (8,040), placing it in a lower performance band. The GTX 960's nearest rivals, such as the GTX 465 (9,294) and GTX 850M (9,302), are all slightly above or near its score, confirming it belongs to a higher tier.

Users who need raw compute throughput, higher memory capacity, and modern API support should select the GTX 960. Users constrained by slot width or power limits might consider the GTX 650 Ti, but the recorded data shows no performance scenario where it wins. The GTX 960 is the benchmark champion in this comparison.

Specification Differences

| Field | NVIDIA GeForce GTX 960 | NVIDIA GeForce GTX 650 Ti |

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

| Chip | GM206 | GK106S |

| Architecture | Maxwell 2.0 | Kepler |

| Generation | GeForce 900 | GeForce 600 |

| Transistors | 2,940 million | 2,540 million |

| Die Size | 228 mm² | 221 mm² |

| Transistor Density | 12.9M / mm² | 11.5M / mm² |

| Base Clock | 1127 MHz | Not specified |

| Boost Clock | 1178 MHz | Not specified |

| Memory Clock | 1753 MHz (7 Gbps effective) | 1350 MHz (5.4 Gbps effective) |

| Memory Size | 2 GB | 1024 MB |

| Memory Bandwidth | 112.2 GB/s | 86.40 GB/s |

| Shading Units | 1024 | 768 |

| ROPs | 32 | 16 |

| Pixel Rate | 37.70 GPixel/s | 14.85 GPixel/s |

| Texture Rate | 75.39 GTexel/s | 59.39 GTexel/s |

| FP32 | 2.413 TFLOPS | 1,425.4 GFLOPS |

| TDP | 120 W | 110 W |

| Slot Width | Dual-slot | Single-slot |

| Display Outputs | 1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.2 | 2x DVI, 1x mini-HDMI 1.4a |

| DirectX | 12 (12_1) | 12 (11_0) |

| Vulkan | 1.4 | 1.2.175 |

| Length | 241 mm (9.5 inches) | 145 mm (5.7 inches) |

| Release Date | 2015-01-21 | 2012-10-08 |

| Predecessor | GeForce 700 | GeForce 500 |

| Successor | GeForce 10 | GeForce 700 |

| Launch MSRP | 199 USD | 149 USD |