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

Where Each One Wins

The benchmark data splits the two cards cleanly by workload type. The GTX 960 takes the compute-oriented tests decisively, while the GTX 650 Ti Boost holds a narrower edge in one API-specific workload.

In the Geekbench Metal test, the GTX 960 scores 8773 against 4858 for the GTX 650 Ti Boost, a margin of 80.6%. This is the largest single-test gap in the comparison. The GTX 960 also dominates Geekbench OpenCL, posting 18925 versus 9302, a 103.5% advantage. These two wins indicate the Maxwell architecture handles general-purpose compute and Apple's Metal API far more efficiently than the older Kepler design.

The GTX 650 Ti Boost claims its only victory in Geekbench Vulkan, scoring 10041 against 9231 for the GTX 960, an 8.1% lead. This result is notable because it shows the older card outperforming its successor in a specific low-level graphics API, despite the GTX 960 having superior raw specifications elsewhere. The Vulkan result suggests the GK106 chip's driver optimization or architectural characteristics favor that particular workload.

The overall win count stands at 2 for the GTX 960 and 1 for the GTX 650 Ti Boost. However, the magnitude of the GTX 960's wins dwarfs the GTX 650 Ti Boost's single victory. The GTX 960's average benchmark score of 9273 places it at the 45th percentile of all GPUs in the database, while the GTX 650 Ti Boost averages 8067 at the 42nd percentile. The GTX 960 sits slightly above the GTX 465 in the database's ranking, with a -0.2% delta, while the GTX 650 Ti Boost is nearly identical to the GRID K2, also at -0.2%.

For users prioritizing compute performance, Metal compatibility, or OpenCL workloads, the GTX 960 is the clear choice. For those targeting Vulkan-specific applications, the GTX 650 Ti Boost offers a measurable, if modest, advantage.

Architecture Differences

Both cards use a 28 nm process at TSMC, but they belong to different architectural generations with distinct design priorities.

The GTX 960 uses the GM206 chip based on Maxwell 2.0 architecture, while the GTX 650 Ti Boost uses the GK106 chip based on Kepler. The GTX 960 packs 2,940 million transistors on a 228 mm² die, yielding a transistor density of 12.9M per mm². The GTX 650 Ti Boost contains 2,540 million transistors on a 221 mm² die, with a lower density of 11.5M per mm². The Maxwell design achieves higher density despite the same process node, indicating more efficient circuit layout.

Shading unit counts differ substantially: the GTX 960 has 1024 shading units, while the GTX 650 Ti Boost has 768. Both cards feature 64 texture mapping units, but the GTX 960 has 32 ROPs versus 24 on the GTX 650 Ti Boost. The GTX 960's pixel rate reaches 37.70 GPixel/s compared to 16.51 GPixel/s on the older card, and its texture rate of 75.39 GTexel/s exceeds the GTX 650 Ti Boost's 66.05 GTexel/s.

Memory configurations differ in bus width and bandwidth. The GTX 960 uses a 128 bit bus with 2 GB GDDR5 at 112.2 GB/s, while the GTX 650 Ti Boost uses a wider 192 bit bus with 2 GB GDDR5 at 144.2 GB/s. Despite the narrower bus, the GTX 960 achieves higher overall performance, suggesting its memory compression and caching improvements compensate for the bandwidth deficit.

Clock speeds favor the GTX 960. It runs at 1127 MHz base and 1178 MHz boost, while the GTX 650 Ti Boost runs at 980 MHz base and 1032 MHz boost. Memory clocks also differ: 1753 MHz (7 Gbps effective) on the GTX 960 versus 1502 MHz (6 Gbps effective) on the GTX 650 Ti Boost.

Feature support varies in API levels. The GTX 960 supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 650 Ti Boost supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6. The GTX 960 offers newer display outputs including HDMI 2.0, while the GTX 650 Ti Boost has HDMI 1.4a.

Power characteristics also differ. The GTX 960 has a 120 W TDP, while the GTX 650 Ti Boost draws 134 W. Both require a single 6-pin power connector and a 300 W suggested PSU. The GTX 960 achieves higher performance with lower power consumption, reflecting Maxwell's efficiency gains.

Head-to-Head Benchmarks

The three shared benchmark results provide a complete picture of relative performance, though each test tells a different story.

Geekbench Metal shows the largest disparity. The GTX 960 scores 8773, while the GTX 650 Ti Boost manages only 4858. This 80.6% delta places the GTX 960 far ahead in Metal-based applications. The result aligns with the GTX 960's newer architecture and higher shading unit count, though the magnitude exceeds what raw specification differences would suggest.

Geekbench OpenCL reveals an even bigger gap. The GTX 960 posts 18925, more than double the GTX 650 Ti Boost's 9302. The 103.5% delta means the GTX 960 delivers over twice the OpenCL compute performance of the older card. This is the strongest result in the entire comparison and indicates that Maxwell's compute capabilities represent a generational leap over Kepler.

Geekbench Vulkan reverses the trend. The GTX 650 Ti Boost scores 10041, beating the GTX 960's 9231 by 8.1%. This single result prevents the GTX 960 from achieving a clean sweep. The Vulkan outcome is difficult to predict from specifications alone, since the GTX 960 has more shading units, higher clocks, and newer architecture. The result suggests that Vulkan driver maturity or specific scheduling behavior favors the GK106 chip.

Looking at the broader database context, the GTX 960's average score of 9273 places it near the GTX 465 (9294, -0.2%) and GTX 850M (9302, -0.3%). The GTX 650 Ti Boost's average of 8067 sits close to the GRID K2 (8080, -0.2%) and GTX 650 Ti (8053, 0.2%). The GTX 960 outperforms the GTX 650 Ti Boost by roughly 15% in average score, a meaningful overall margin even though the Vulkan test goes the other way.

The Verdict

The data supports a clear recommendation for most users: the GTX 960 is the superior card. It wins two of three head-to-head benchmarks, and its victories are lopsided. The 103.5% OpenCL advantage and 80.6% Metal advantage demonstrate that the Maxwell architecture delivers dramatically better compute performance. The GTX 960 also achieves this with a lower TDP of 120 W versus 134 W, making it more efficient per watt.

The GTX 650 Ti Boost retains relevance only for Vulkan-specific workloads, where its 8.1% lead could matter for users running Vulkan-based applications exclusively. However, this single test does not compensate for the substantial deficits in Metal and OpenCL performance. The GTX 650 Ti Boost also has a lower average benchmark score (8067 versus 9273) and sits at the 42nd percentile versus the GTX 960's 45th percentile.

Users who prioritize compute performance, modern API support, or power efficiency should choose the GTX 960. Its DirectX 12 (12_1) support, Vulkan 1.4 compatibility, and higher shading unit count make it the more future-proof option. Users with a specific need for Vulkan performance might consider the GTX 650 Ti Boost, but the data shows this is a narrow use case. The GTX 960's combination of higher scores, lower power draw, and newer feature set makes it the stronger overall choice in nearly every scenario.

FAQ

Q: Which card performs better in OpenCL benchmarks?

A: The GTX 960 scores 18925 in Geekbench OpenCL, which is 103.5% higher than the GTX 650 Ti Boost's 9302. This is the largest performance gap between the two cards.

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

A: Yes, the GTX 650 Ti Boost wins the Geekbench Vulkan test with a score of 10041, beating the GTX 960's 9231 by 8.1%. This is the only test where the older card comes out ahead.

Q: How do the memory configurations compare?

A: Both cards have 2 GB of GDDR5 memory. The GTX 960 uses a 128 bit bus with 112.2 GB/s bandwidth, while the GTX 650 Ti Boost uses a 192 bit bus with 144.2 GB/s bandwidth. The GTX 650 Ti Boost has higher raw bandwidth, but the GTX 960 still outperforms it in most tests.

Q: What are the power requirements for each card?

A: The GTX 960 has a 120 W TDP, while the GTX 650 Ti Boost has a 134 W TDP. Both cards require a single 6-pin power connector and a 300 W suggested PSU.

Q: Which card has better API support?

A: The GTX 960 supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 650 Ti Boost supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6. The GTX 960 has newer API versions.

Q: How do the two cards compare in overall average benchmark score?

A: The GTX 960 has an average benchmark score of 9273, while the GTX 650 Ti Boost averages 8067. The GTX 960 sits at the 45th percentile of all GPUs, and the GTX 650 Ti Boost sits at the 42nd percentile.

Specification Differences

| Specification | GTX 960 | GTX 650 Ti Boost |

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

| Chip | GM206 | GK106 |

| 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 | 980 MHz |

| Boost Clock | 1178 MHz | 1032 MHz |

| Memory Clock | 1753 MHz, 7 Gbps effective | 1502 MHz, 6 Gbps effective |

| Memory Bus Width | 128 bit | 192 bit |

| Memory Bandwidth | 112.2 GB/s | 144.2 GB/s |

| Shading Units | 1024 | 768 |

| TMUs | 64 | 64 |

| ROPs | 32 | 24 |

| Pixel Rate | 37.70 GPixel/s | 16.51 GPixel/s |

| Texture Rate | 75.39 GTexel/s | 66.05 GTexel/s |

| FP32 | 2.413 TFLOPS | 1.585 TFLOPS |

| TDP | 120 W | 134 W |

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

| Vulkan | 1.4 | 1.2.175 |

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

| Release Date | 2015-01-21 | 2013-03-25 |

| Launch MSRP | 199 USD | 169 USD |