NVIDIA GeForce GTX 660 Ti vs NVIDIA GeForce RTX 4060 Comparison

The Verdict

The data presents a decisive generational shift. The NVIDIA GeForce RTX 4060 outperforms the NVIDIA GeForce GTX 660 Ti by a staggering margin in every recorded benchmark. Across the two shared tests, the RTX 4060 wins both, with an average benchmark score of 17,639 compared to the GTX 660 Ti's 15,063. This represents a 17% advantage in aggregate performance, but the head-to-head results reveal a much wider gap in raw compute tasks.

The RTX 4060 is the clear choice for any modern workload. Its Geekbench OpenCL score of 95,057 is 529% higher than the GTX 660 Ti's 15,113, indicating a massive leap in general-purpose compute capability. Similarly, its Vulkan score of 48,643 is 224% ahead of the older card's 15,012. The GTX 660 Ti, while a capable card in its era, is now firmly in legacy territory. The percentile rankings reinforce this: the RTX 4060 sits in the 61st percentile of all GPUs, while the GTX 660 Ti is in the 57th percentile. For users seeking any degree of future-proofing or the ability to handle modern APIs, the RTX 4060 is the only viable option. The GTX 660 Ti should only be considered for legacy systems where its 2 GB memory and older feature set are sufficient, and even then, the performance delta is so large that it is difficult to recommend outside of a purely retro build.

Architecture Differences

The architectural gap between these two cards is vast, representing over a decade of GPU evolution. The RTX 4060 is built on the AD107 chip using the Ada Lovelace architecture, fabricated on a 5 nm process at TSMC. It packs 18,900 million transistors into a die size of 159 mm², yielding a transistor density of 118.9M per mm². In contrast, the GTX 660 Ti uses the GK104 chip with the Kepler architecture, on a much older 28 nm process, also at TSMC. It contains 3,540 million transistors spread across a larger 294 mm² die, giving a density of just 12.0M per mm². This fundamental difference in manufacturing technology explains the massive efficiency and performance gains.

The RTX 4060 is equipped with 24 RT cores and 96 tensor cores, features entirely absent from the GTX 660 Ti. This enables hardware-accelerated ray tracing and AI-driven technologies like DLSS, which the older card cannot access. The RTX 4060 also supports DirectX 12 Ultimate (12_2), while the GTX 660 Ti is limited to DirectX 12 (11_0). Vulkan support also differs, with the newer card at version 1.4 versus the older card's 1.2.175. The RTX 4060 features a dual-slot design with a single 12-pin power connector and a suggested 300 W PSU, compared to the GTX 660 Ti's dual-slot design with two 6-pin connectors and a 450 W suggested PSU. The newer card is also more power-efficient, with a 115 W TDP versus the older card's 150 W, despite delivering vastly more performance.

Head-to-Head Benchmarks

The head-to-head results are overwhelmingly in favor of the RTX 4060. In Geekbench OpenCL, the RTX 4060 scores 95,057 against the GTX 660 Ti's 15,113, a delta of 529%. This is not a marginal improvement; it is a 6.3-fold increase in raw compute throughput. The RTX 4060's shading units, 3,072 versus 1,344, and its FP32 performance of 15.11 TFLOPS versus 2.634 TFLOPS, directly translate into this dominant OpenCL result. The GTX 660 Ti, with its older architecture and lower core count, simply cannot compete in parallel processing workloads.

In Geekbench Vulkan, the RTX 4060 again prevails with a score of 48,643 versus 15,012, a delta of 224%. While this is a smaller relative gap than OpenCL, it still represents a 3.2-fold performance advantage. The RTX 4060's modern architecture, with its dedicated RT and tensor cores, allows it to handle the Vulkan API's advanced features far more efficiently. The GTX 660 Ti's Vulkan support, capped at version 1.2.175, is functional but lacks the optimizations and hardware features of the newer card. Overall, the data shows that the RTX 4060 is not just faster; it is in a completely different performance class.

Specification Differences

The specification sheets for these two cards show a clear generational divide. The RTX 4060 has 3,072 shading units, 96 TMUs, and 48 ROPs, while the GTX 660 Ti has 1,344 shading units, 112 TMUs, and 24 ROPs. Interestingly, the older card has more TMUs, but this does not offset the RTX 4060's overwhelming advantages in raw compute and clock speed. The RTX 4060 operates at a base clock of 1830 MHz and a boost clock of 2460 MHz, while the GTX 660 Ti runs at 915 MHz base and 980 MHz boost. This clock speed difference is substantial, directly contributing to the newer card's higher pixel rate of 118.1 GPixel/s versus 27.44 GPixel/s, and texture rate of 236.2 GTexel/s versus 109.8 GTexel/s.

Memory configurations also differ significantly. The RTX 4060 comes with 8 GB of GDDR6 memory on a 128-bit bus, providing 272.0 GB/s of bandwidth. The GTX 660 Ti has 2 GB of GDDR5 memory on a 192-bit bus, delivering 144.2 GB/s. Despite the narrower bus, the RTX 4060's faster memory clock (2125 MHz versus 1502 MHz) and higher effective speed (17 Gbps versus 6 Gbps) give it almost double the bandwidth. The RTX 4060 also features a PCIe 4.0 x8 interface, while the GTX 660 Ti uses PCIe 3.0 x16. Display outputs differ as well: the newer card offers 1x HDMI 2.1 and 3x DisplayPort 1.4a, while the older card has 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. Both cards are end-of-life and launched at an MSRP of 299 USD, though the RTX 4060 launched in May 2023, while the GTX 660 Ti launched in August 2012.

FAQ

Q: Which card is faster in OpenCL compute workloads?

A: The NVIDIA GeForce RTX 4060 is dramatically faster, scoring 95,057 in Geekbench OpenCL compared to the GTX 660 Ti's 15,113, a 529% difference.

Q: Does the GTX 660 Ti support ray tracing?

A: No. The GTX 660 Ti has no RT cores listed in its specifications, while the RTX 4060 includes 24 RT cores for hardware-accelerated ray tracing.

Q: How much memory does each card have?

A: The RTX 4060 has 8 GB of GDDR6 memory, while the GTX 660 Ti has 2 GB of GDDR5 memory. The newer card also offers 272.0 GB/s of bandwidth versus 144.2 GB/s.

Q: What is the power consumption difference?

A: The RTX 4060 has a TDP of 115 W, while the GTX 660 Ti has a TDP of 150 W. The newer card is more power-efficient despite being significantly faster.

Q: Which card has better Vulkan performance?

A: The RTX 4060 scores 48,643 in Geekbench Vulkan, which is 224% higher than the GTX 660 Ti's 15,012. The RTX 4060 also supports Vulkan 1.4, a newer version than the GTX 660 Ti's 1.2.175.

Q: Are both cards the same price at launch?

A: Yes, both the RTX 4060 and the GTX 660 Ti had a launch MSRP of 299 USD, though they were released over a decade apart.