NVIDIA GeForce GTX 1660 vs NVIDIA GeForce GTX 1660 SUPER Comparison

Head-to-Head Benchmarks

The recorded data leaves no ambiguity: the NVIDIA GeForce GTX 1660 SUPER wins all ten head-to-head benchmark comparisons. The most decisive victory comes in the Passmark DirectX 11 test, where the SUPER scores 104 against the standard GTX 1660's 79, a delta of 31.6%. This is the single largest performance gap between the two cards in any measured workload, and it suggests that the SUPER's advantage is especially pronounced in legacy DirectX 11 titles, which still constitute a large share of the active gaming library.

The 3DMark Steel Nomad DX12 test tells a similar story, though with a smaller margin. Here the SUPER records 1278 points versus 1065 for the GTX 1660, a 20% advantage. This is a modern, low-level API workload, and the SUPER's lead indicates that its architectural advantages translate to contemporary rendering paths as well. The Geekbench Vulkan result reinforces this: 57102 for the SUPER versus 50137, a 13.9% delta. Vulkan is another low-overhead API, and the consistency across DX12 and Vulkan suggests the SUPER's memory subsystem improvements are not tied to any single API generation.

Compute workloads show a narrower but still positive gap. In Geekbench OpenCL, the SUPER scores 52490 versus 47850, a 9.7% advantage. Passmark GPU Compute shows only 5076 against 4963, a 2.3% delta. This pattern is notable: the gap narrows as the workload becomes more purely arithmetic and less memory-bandwidth-bound. The Passmark DirectX 12 test shows the smallest margin of all, 50 versus 49, just 2%. The DirectX 9 test shows a 6.8% lead (189 vs 177), and DirectX 10 shows 6.6% (65 vs 61). The 2D test, Passmark G2D, shows a modest 3.7% advantage (805 vs 776).

The overall average benchmark score reflects this comprehensive sweep. The SUPER sits at 12986, while the GTX 1660 trails at 11680. That is an 11.2% gap in the aggregate. The SUPER's percentile ranking among all GPUs is 53, compared to 51 for the GTX 1660. Both cards sit near the middle of the overall distribution, but the SUPER is meaningfully closer to the median of all recorded GPUs. The nearest rivals in the database further contextualize the SUPER's position: it sits within 0.4% of the RTX 3050 Ti Mobile and the RX 580, and within 0.7% of the Tesla M2090. The GTX 1660, by contrast, is within 0.5% of the RX 7800 XT and 1.3% of the Radeon Pro 5500M, a slightly more dispersed rivalry cluster.

Architecture Differences

Both cards share the same fundamental silicon. The chip is the TU116, built on TSMC's 12 nm process, with 6,600 million transistors on a 284 mm² die, yielding a transistor density of 23.2M per mm². The architecture is Turing, the generation is GeForce 16, and the predecessor and successor are GeForce 10 and GeForce 20 respectively. The shading unit count is identical at 1408, as are the texture mapping units at 88 and the raster operation units at 48. The pixel rate is 85.68 GPixel/s and the texture rate is 157.1 GTexel/s for both. FP32 throughput is 5.027 TFLOPS, and FP16 is 10.05 TFLOPS (2:1) for both cards. Neither card has ray tracing cores or tensor cores. The base clock is 1530 MHz and the boost clock is 1785 MHz for both.

The divergence begins with memory. The SUPER uses GDDR6, while the GTX 1660 uses GDDR5. The memory clock is listed as 1750 MHz with 14 Gbps effective for the SUPER, versus 2001 MHz with 8 Gbps effective for the GTX 1660. The bus width is the same, 192 bit, and the capacity is the same, 6 GB. But the resulting bandwidth is dramatically different: 336.0 GB/s for the SUPER versus 192.1 GB/s for the GTX 1660. That is a 75% increase in memory bandwidth, and it is the single most important architectural distinction between the two cards. The core compute resources are identical, so the performance differences observed in the benchmark data must be attributed almost entirely to this memory bandwidth advantage.

The power envelope differs slightly. The SUPER has a TDP of 125 W, the GTX 1660 is rated at 120 W. Both use a single 8-pin power connector and recommend a 300 W power supply. Physical dimensions are identical: 229 mm length (9 inches), 111 mm height (4.4 inches), and 35 mm width (1.4 inches). Both are dual-slot cards. Display outputs are the same: one DVI, one HDMI 2.0, and one DisplayPort 1.4a. API support is identical: DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The release dates differ: the GTX 1660 launched on 2019-03-13, while the SUPER arrived later on 2019-10-28. Both are marked as end-of-life. The launch MSRP for the SUPER is 229 USD, and for the GTX 1660 it is 219 USD.

The Verdict

The data points to a clear hierarchy: the GTX 1660 SUPER is the faster card in every recorded benchmark. The margin ranges from a barely perceptible 2% in DirectX 12 to a commanding 31.6% in DirectX 11. The average benchmark score gap of 11.2% (12986 vs 11680) is substantial and consistent across APIs, resolutions, and workload types. The SUPER's percentile rank of 53 versus 51 for the GTX 1660 confirms that the gap is meaningful in the broader GPU landscape.

Who should choose the GTX 1660? The only argument from the data is that it is the lower-power option, at 120 W versus 125 W, and it carries a lower launch MSRP of 219 USD versus 229 USD. For a system where power draw is the absolute priority and the performance difference is acceptable, the GTX 1660 is the more conservative choice. The 5 W power savings is real, but the performance cost is not: in every single benchmark, the GTX 1660 trails, and in some cases by a wide margin.

Who should choose the SUPER? Anyone who wants the best performance within this specific product pair. The 20% lead in 3DMark Steel Nomad DX12 and the 31.6% lead in DirectX 11 are not marginal differences. The SUPER's memory bandwidth advantage, 336.0 GB/s versus 192.1 GB/s, is the likely driver, and it shows up across the board. The 10 USD launch MSRP difference is small relative to the performance delta. The data makes the SUPER the straightforward recommendation for gaming and general GPU compute.

FAQ

Q: How much faster is the GTX 1660 SUPER than the GTX 1660 in the average benchmark?

A: The SUPER has an average benchmark score of 12986, while the GTX 1660 scores 11680. That is an 11.2% advantage for the SUPER.

Q: Which card has better memory bandwidth?

A: The GTX 1660 SUPER has a bandwidth of 336.0 GB/s using GDDR6 memory. The GTX 1660 has 192.1 GB/s using GDDR5. The SUPER leads by 75%.

Q: Are the core specifications identical between the two cards?

A: Yes. Both have 1408 shading units, 88 TMUs, 48 ROPs, a base clock of 1530 MHz, a boost clock of 1785 MHz, and 5.027 TFLOPS of FP32 throughput.

Q: In which benchmark does the SUPER have the largest lead?

A: The Passmark DirectX 11 test shows the largest delta at 31.6%, with the SUPER scoring 104 versus 79 for the GTX 1660.

Q: In which benchmark is the performance gap the smallest?

A: The Passmark DirectX 12 test shows the smallest gap at 2%, with scores of 50 and 49 for the SUPER and GTX 1660 respectively.

Q: Do both cards have the same power connector and power supply recommendation?

A: Yes. Both use a single 8-pin power connector and recommend a 300 W power supply. The TDP differs slightly: 125 W for the SUPER, 120 W for the GTX 1660.

Where Each One Wins

The GTX 1660 SUPER wins in every measured category, but the magnitude of the victory varies by workload. For users prioritizing modern low-level APIs, the SUPER's 20% lead in 3DMark Steel Nomad DX12 and 13.9% lead in Geekbench Vulkan are the strongest arguments. These are the workloads that stress memory bandwidth most heavily, and the SUPER's GDDR6 memory at 336.0 GB/s versus 192.1 GB/s is the decisive factor.

For users running older DirectX titles, the SUPER's 31.6% lead in DirectX 11 is the standout result. The 6.8% lead in DirectX 9 and 6.6% lead in DirectX 10 are smaller but still consistent. The SUPER is the better choice for anyone with a legacy game library.

For compute-oriented tasks, the SUPER's advantage is more modest. The 9.7% lead in OpenCL and 2.3% lead in GPU Compute suggest that raw arithmetic throughput is largely similar between the two cards, which makes sense given the identical core configuration. The memory bandwidth advantage matters less in these workloads.

The GTX 1660 has no benchmark wins. Its only advantages in the recorded data are its lower TDP of 120 W versus 125 W and its lower launch MSRP of 219 USD versus 229 USD. For users building a system with strict power limits or a tighter initial budget, the GTX 1660 remains a viable option, but the performance data consistently favors the SUPER.

Specification Differences

| Specification | GTX 1660 SUPER | GTX 1660 |

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

| Memory Type | GDDR6 | GDDR5 |

| Memory Clock | 1750 MHz, 14 Gbps effective | 2001 MHz, 8 Gbps effective |

| Memory Bandwidth | 336.0 GB/s | 192.1 GB/s |

| TDP | 125 W | 120 W |

| Release Date | 2019-10-28 | 2019-03-13 |

| Launch MSRP | 229 USD | 219 USD |

| Average Benchmark Score | 12986 | 11680 |

| Percentile vs All GPUs | 53 | 51 |