NVIDIA GeForce GTX 1660 SUPER vs NVIDIA GeForce GTX 660 Ti Comparison

Where Each One Wins

The data splits this comparison into two very different eras of NVIDIA graphics. The GTX 660 Ti, a Kepler-generation card from the GeForce 600 series, does not win a single recorded head-to-head benchmark in the database. The GTX 1660 SUPER, built on the Turing architecture from the GeForce 16 series, wins both recorded comparisons outright. The benchmark results indicate a complete sweep for the newer card.

The GTX 660 Ti occupies the 57th percentile among all GPUs in the database, while the GTX 1660 SUPER sits at the 53rd percentile. This is a notable inversion: the older card holds a higher percentile ranking relative to all GPUs, yet loses every direct comparison. The explanation lies in the different benchmark suites each card was tested under. The GTX 660 Ti has an average benchmark score of 15,063 from its two recorded tests, while the GTX 1660 SUPER averages 12,986 across a much wider range of tests, including legacy DirectX 9 and DirectX 10 workloads that drag its average down despite its modern strengths.

For use-case selection, the GTX 1660 SUPER is the clear winner in compute-heavy and modern API workloads. Its Geekbench OpenCL and Vulkan scores are dramatically higher than the GTX 660 Ti's. The older card, however, shows more competitive standing relative to its own era, sitting within 1.4% of the AMD Radeon 680M and only 0.9% behind the NVIDIA GeForce RTX 3050 OEM in average score. The GTX 660 Ti's nearest rivals in the database are all much newer parts, suggesting its legacy performance still holds a recognizable tier. But when directly compared, the GTX 1660 SUPER dominates in every measured category.

Architecture Differences

The architectural gap between these two cards is generational and profound. The GTX 660 Ti uses the GK104 chip built on Kepler architecture, fabricated on a 28 nm process at TSMC. The GTX 1660 SUPER uses the TU116 chip on Turing architecture, fabricated on a 12 nm process, also at TSMC. The transistor counts tell the story: the GTX 660 Ti packs 3,540 million transistors on a 294 mm² die, yielding a density of 12.0M transistors per mm². The GTX 1660 SUPER crams 6,600 million transistors onto a slightly smaller 284 mm² die, achieving 23.2M transistors per mm², nearly double the density.

Shading unit counts are similar on paper, with the GTX 660 Ti at 1,344 and the GTX 1660 SUPER at 1,408, but the execution efficiency differs enormously. The GTX 1660 SUPER has 88 texture mapping units versus 112 on the GTX 660 Ti, yet delivers higher texture rate. The newer card also doubles the render output units, 48 versus 24, which explains its vastly higher pixel throughput.

The GTX 1660 SUPER supports DirectX 12 (12_1), while the GTX 660 Ti is limited to DirectX 12 (11_0). Both support OpenGL 4.6, but the Vulkan support differs: the GTX 660 Ti runs Vulkan 1.2.175, while the GTX 1660 SUPER runs Vulkan 1.4. Neither card has ray tracing cores or tensor cores. The GTX 1660 SUPER adds FP16 support at 10.05 TFLOPS with a 2:1 ratio, while the GTX 660 Ti has no recorded FP16 capability.

Head-to-Head Benchmarks

The database records two direct comparisons, both in Geekbench tests, and both are decisive victories for the GTX 1660 SUPER.

In Geekbench OpenCL, the GTX 660 Ti scores 15,113, while the GTX 1660 SUPER scores 52,490. The delta is 71.2% in favor of the newer card. This is not a marginal improvement; the GTX 1660 SUPER delivers more than three times the compute performance in this OpenCL workload. For context, the GTX 660 Ti's nearest rivals in this score range include the AMD Radeon Pro 560X at 15,082 (0.1% behind) and the AMD Radeon RX 7600 at 15,171 (0.7% ahead), meaning the older card is bracketed by modern midrange and entry-level parts.

In Geekbench Vulkan, the gap widens further. The GTX 660 Ti scores 15,012, while the GTX 1660 SUPER scores 57,102. The delta is 73.7% in favor of the GTX 1660 SUPER. The Vulkan result shows the newer architecture's superior API efficiency and raw throughput. The GTX 660 Ti's Vulkan score is nearly identical to its OpenCL score, suggesting the Kepler architecture handles both APIs with similar capability, but the Turing card gains an additional 4,612 points moving from OpenCL to Vulkan, indicating better optimization for modern graphics APIs.

The wins column confirms the sweep: the GTX 1660 SUPER wins 2 out of 2 recorded head-to-head benchmarks, while the GTX 660 Ti wins none.

Specification Differences

The recorded specifications show several key differences beyond architecture. Memory configuration is a major separator. The GTX 660 Ti has 2 GB of GDDR5 memory running at 1502 MHz with 6 Gbps effective speed, delivering 144.2 GB/s bandwidth over a 192-bit bus. The GTX 1660 SUPER has 6 GB of GDDR6 memory running at 1750 MHz with 14 Gbps effective speed, delivering 336.0 GB/s bandwidth over the same 192-bit bus. The newer card has triple the capacity and more than double the bandwidth.

Clock speeds also favor the GTX 1660 SUPER substantially. The GTX 660 Ti runs at a 915 MHz base and 980 MHz boost. The GTX 1660 SUPER runs at a 1530 MHz base and 1785 MHz boost, representing a 67% higher base clock and an 82% higher boost clock. The compute throughput reflects this: the GTX 660 Ti delivers 2.634 TFLOPS FP32, while the GTX 1660 SUPER delivers 5.027 TFLOPS, nearly double. Pixel rate jumps from 27.44 GPixel/s to 85.68 GPixel/s, and texture rate from 109.8 GTexel/s to 157.1 GTexel/s.

Power characteristics are reversed from what one might expect. The older GTX 660 Ti has a 150 W TDP and requires 2x 6-pin power connectors with a suggested 450 W PSU. The newer GTX 1660 SUPER has a lower 125 W TDP, requires only 1x 8-pin power connector, and suggests a 300 W PSU. The newer card is more power-efficient while delivering significantly more performance.

Physical dimensions differ slightly. The GTX 660 Ti is 241 mm long, while the GTX 1660 SUPER is 229 mm long. Both are dual-slot cards. The GTX 1660 SUPER has recorded height (111 mm) and width (35 mm) dimensions, while the GTX 660 Ti has no recorded height or width. Display outputs also differ: the GTX 660 Ti has 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2, while the GTX 1660 SUPER has 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a.

The launch MSRP for the GTX 660 Ti was 299 USD. The launch MSRP for the GTX 1660 SUPER was 229 USD.

FAQ

Q: Which card has higher raw compute performance?

A: The GTX 1660 SUPER delivers 5.027 TFLOPS FP32, nearly double the GTX 660 Ti's 2.634 TFLOPS. This matches the head-to-head results where the GTX 1660 SUPER leads by 71.2% in OpenCL and 73.7% in Vulkan.

Q: How do the memory configurations compare?

A: The GTX 1660 SUPER has 6 GB of GDDR6 memory with 336.0 GB/s bandwidth, while the GTX 660 Ti has 2 GB of GDDR5 memory with 144.2 GB/s bandwidth. Both use a 192-bit bus, but the newer card's GDDR6 runs at 14 Gbps effective versus 6 Gbps effective.

Q: Which card consumes less power?

A: The GTX 1660 SUPER has a 125 W TDP and suggests a 300 W PSU, while the GTX 660 Ti has a 150 W TDP and suggests a 450 W PSU. The newer card also requires only one 8-pin power connector versus two 6-pin connectors.

Q: What are the API support differences?

A: The GTX 660 Ti supports DirectX 12 (11_0) and Vulkan 1.2.175, while the GTX 1660 SUPER supports DirectX 12 (12_1) and Vulkan 1.4. Both support OpenGL 4.6. Neither card has ray tracing or tensor cores.

Q: How does the GTX 660 Ti compare to its nearest rivals in the database?

A: The GTX 660 Ti's average benchmark score of 15,063 puts it within 0.1% of the AMD Radeon Pro 560X (15,082), 0.7% behind the AMD Radeon RX 7600 (15,171), 0.9% behind the NVIDIA GeForce RTX 3050 OEM (15,199), and 1.4% ahead of the AMD Radeon 680M (15,270, with the delta expressed as a negative 1.4% for the GTX 660 Ti).

Q: Which card has a higher pixel rate?

A: The GTX 1660 SUPER achieves 85.68 GPixel/s, more than three times the GTX 660 Ti's 27.44 GPixel/s, due to its 48 ROPs versus 24 ROPs.

The Verdict

The data is unambiguous: the GTX 1660 SUPER is the superior card in every measured category. It wins both head-to-head benchmarks by margins exceeding 71%, delivers nearly double the FP32 compute, offers triple the memory capacity, more than double the memory bandwidth, and does so while consuming 25 W less power and requiring a 300 W PSU instead of 450 W. The GTX 1660 SUPER also supports newer API versions, including DirectX 12 (12_1) and Vulkan 1.4, making it the better choice for modern games and applications.

The GTX 660 Ti remains a historically relevant part, sitting at the 57th percentile among all GPUs, which is higher than the GTX 1660 SUPER's 53rd percentile. This reflects the different benchmark sets each card was subjected to, not relative performance. The GTX 660 Ti's recorded scores place it alongside modern parts like the AMD Radeon RX 7600 and NVIDIA GeForce RTX 3050 OEM, but only in the aggregate average, not in direct comparison.

For anyone choosing between these two cards based on the recorded data, the GTX 1660 SUPER is the only rational pick. It wins every direct test, offers superior specifications across memory, compute, pixel rate, texture rate, and API support, and does so with lower power requirements. The GTX 660 Ti is an end-of-life product from the Kepler era with no recorded wins against its newer counterpart. The verdict from the database is clear: the GTX 1660 SUPER is the definitive choice.