NVIDIA GeForce GTX 1660 SUPER vs NVIDIA GeForce GTX 570 Comparison

# NVIDIA GeForce GTX 570 vs NVIDIA GeForce GTX 1660 SUPER

The GeForce GTX 570 is a Fermi 2.0-era card from late 2010, while the GeForce GTX 1660 SUPER is a Turing-based card from late 2019. The benchmark data shows a decisive generational shift: the GTX 1660 SUPER wins the only shared head-to-head test by a massive 74.3% margin in Geekbench OpenCL, and its average benchmark score of 12,986 sits just 3.9% below the GTX 570's 13,515 despite the 570 having a higher percentile ranking (54th vs 53rd). The 570 edges the 1660 SUPER in percentile placement, but that is a narrow distinction that does not reflect real-world capability.

Where Each One Wins

The GTX 570 has exactly one benchmark result in the data: Geekbench OpenCL at 13,515. That score places it at the 54th percentile among all GPUs, with its nearest rival being the AMD Radeon HD 7770M at 13,536 (a 0.2% gap) and the NVIDIA P106-090 at 13,470 (0.3% behind). In this single test, the 570 is competitive with low-end mobile and entry-level desktop parts from much later generations, which shows that its raw compute ability has aged but not collapsed.

The GTX 1660 SUPER, by contrast, has ten benchmark results spanning multiple APIs and workloads. Its Geekbench OpenCL score of 52,490 is 74.3% higher than the 570's — a dominant win that indicates the Turing architecture handles OpenCL compute far more efficiently. Beyond that, the 1660 SUPER shows strength across DirectX tests: Passmark DirectX 11 scores 104, DirectX 9 scores 189, and DirectX 12 scores 50. Its Passmark G3D score of 12,699 and GPU Compute score of 5,076 further demonstrate versatility. The 1660 SUPER's percentile rank is 53rd, just one point below the 570, but its nearest rivals include the NVIDIA GeForce RTX 3050 Ti Mobile (12,940, 0.4% ahead) and AMD Radeon RX 580 (12,928, 0.4% ahead) — both much newer and more capable cards.

The use-case split is clear: the 570 wins only on percentile placement among all GPUs, while the 1660 SUPER wins every actual benchmark where both are measured, plus it has data for Vulkan (57,102), DirectX 10 (65), and 2D workloads (805) that the 570 lacks entirely. For any modern workload involving compute, DirectX 11/12, or Vulkan, the 1660 SUPER is the only viable choice.

Architecture Differences

The two cards are separated by nearly a decade of GPU design. The GTX 570 uses the GF110 chip on a 40 nm TSMC process, packing 3,000 million transistors into a 520 mm² die — a transistor density of 5.8 million per mm². The GTX 1660 SUPER uses the TU116 chip on a 12 nm TSMC process, with 6,600 million transistors in a 284 mm² die, achieving 23.2 million transistors per mm². That is a fourfold increase in density and over double the transistor count on a physically smaller chip.

The 570's Fermi 2.0 architecture is built around 480 shading units, 60 texture mapping units, and 40 ROPs. The 1660 SUPER's Turing architecture scales that up to 1,408 shading units, 88 TMUs, and 48 ROPs. Neither card has ray tracing or tensor cores, but Turing brings significant architectural improvements over Fermi in terms of scheduling, cache hierarchy, and instruction efficiency.

Memory architecture differs substantially. The 570 has 1280 MB of GDDR5 on a 320-bit bus, delivering 152.0 GB/s bandwidth at 3.8 Gbps effective. The 1660 SUPER has 6 GB of GDDR6 on a 192-bit bus, delivering 336.0 GB/s bandwidth at 14 Gbps effective. The 1660 SUPER more than doubles bandwidth despite a narrower bus, which is critical for modern textures and compute workloads.

Feature support also diverges. The 570 supports DirectX 12 (11_0) and OpenGL 4.6, but has no Vulkan support listed. The 1660 SUPER supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The 1660 SUPER also uses PCIe 3.0 x16 versus the 570's PCIe 2.0 x16, and its display outputs include HDMI 2.0 and DisplayPort 1.4a, whereas the 570 is limited to DVI and mini-HDMI 1.3a.

FAQ

Q: Which card has a higher average benchmark score?

A: The GTX 1660 SUPER has an average benchmark score of 12,986, while the GTX 570 has 13,515. The 570's average is higher, but that is because the 570 only has one benchmark result (Geekbench OpenCL), while the 1660 SUPER's average spans ten tests including older DirectX APIs where it scores lower.

Q: How much faster is the GTX 1660 SUPER in OpenCL compute?

A: The Geekbench OpenCL score for the GTX 1660 SUPER is 52,490 versus 13,515 for the GTX 570. That is a 74.3% advantage for the 1660 SUPER, the only head-to-head benchmark available in the data.

Q: Does the GTX 570 support Vulkan?

A: No Vulkan support is listed for the GTX 570. The GTX 1660 SUPER supports Vulkan 1.4 and has a Geekbench Vulkan score of 57,102.

Q: What are the memory capacities and bandwidths?

A: The GTX 570 has 1280 MB of GDDR5 with 152.0 GB/s bandwidth on a 320-bit bus. The GTX 1660 SUPER has 6 GB of GDDR6 with 336.0 GB/s bandwidth on a 192-bit bus.

Q: Which card has better DirectX 12 support?

A: The GTX 1660 SUPER supports DirectX 12 (12_1), whereas the GTX 570 only supports DirectX 12 (11_0). The 1660 SUPER also has a Passmark DirectX 12 score of 50, which the 570 lacks.

Q: What is the transistor density difference?

A: The GTX 570 has a transistor density of 5.8 million per mm² on a 40 nm process. The GTX 1660 SUPER has 23.2 million per mm² on a 12 nm process — a fourfold improvement.

Specification Differences

The following specifications differ between the two cards:

| Specification | GTX 570 | GTX 1660 SUPER |

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

| Chip | GF110 | TU116 |

| Architecture | Fermi 2.0 | Turing |

| Generation | GeForce 500 | GeForce 16 |

| Process Node | 40 nm | 12 nm |

| Transistors | 3,000 million | 6,600 million |

| Die Size | 520 mm² | 284 mm² |

| Transistor Density | 5.8M / mm² | 23.2M / mm² |

| Base Clock | None listed | 1530 MHz |

| Boost Clock | None listed | 1785 MHz |

| Memory Clock | 3.8 Gbps effective | 14 Gbps effective |

| Memory Size | 1280 MB | 6 GB |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bus | 320 bit | 192 bit |

| Memory Bandwidth | 152.0 GB/s | 336.0 GB/s |

| Shading Units | 480 | 1408 |

| TMUs | 60 | 88 |

| ROPs | 40 | 48 |

| Pixel Rate | 21.96 GPixel/s | 85.68 GPixel/s |

| Texture Rate | 43.92 GTexel/s | 157.1 GTexel/s |

| FP32 | 1,405.4 GFLOPS | 5.027 TFLOPS |

| FP16 | None listed | 10.05 TFLOPS (2:1) |

| TDP | 219 W | 125 W |

| Power Connectors | 2x 6-pin | 1x 8-pin |

| Suggested PSU | 550 W | 300 W |

| Bus Interface | PCIe 2.0 x16 | PCIe 3.0 x16 |

| Display Outputs | 2x DVI, 1x mini-HDMI 1.3a | 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a |

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

| Vulkan | None listed | 1.4 |

| Length | 267 mm (10.5 inches) | 229 mm (9 inches) |

| Height | None listed | 111 mm (4.4 inches) |

| Width | None listed | 35 mm (1.4 inches) |

| Release Date | 2010-12-06 | 2019-10-28 |

| Launch MSRP | 349 USD | 229 USD |

Head-to-Head Benchmarks

The only direct comparison available is Geekbench OpenCL, where the GTX 1660 SUPER scores 52,490 against the GTX 570's 13,515. That is a 74.3% delta in favor of the 1660 SUPER, which is a decisive win by any measure. The 570's score of 13,515 places it just 0.2% behind the AMD Radeon HD 7770M (13,536) and 0.3% ahead of the NVIDIA P106-090 (13,470) — both are low-end or older parts. In contrast, the 1660 SUPER's OpenCL score is not directly compared to the 570 in the nearestRivals data, but its average score of 12,986 puts it within 0.4% of the RTX 3050 Ti Mobile (12,940) and the RX 580 (12,928), both of which are substantially more modern than the 570.

Looking at the 1660 SUPER's broader benchmark suite, its Passmark G3D score of 12,699 and GPU Compute score of 5,076 indicate strong general 3D rendering and compute performance. Its Geekbench Vulkan score of 57,102 shows excellent low-level API performance, which is entirely absent from the 570's feature set. The 570's only benchmark result is a single OpenCL score that is now closer to mobile parts from 2012-2017 than to any desktop GPU from its own era.

The wins tally is one for the 1660 SUPER and zero for the 570 in the head-to-head data. That single win is categorical: the 1660 SUPER is not merely faster, it is in a different performance class. The 570's higher average score is an artifact of having only one benchmark, and that benchmark is its weakest relative showing.

The Verdict

The data points to a straightforward conclusion: the GTX 1660 SUPER outperforms the GTX 570 in every measurable way. The 74.3% OpenCL advantage is the headline number, but the supporting evidence is just as one-sided. The 1660 SUPER has more than double the shading units (1,408 vs 480), more than double the memory capacity (6 GB vs 1280 MB), more than double the memory bandwidth (336.0 GB/s vs 152.0 GB/s), and a nearly 3.6x higher FP32 throughput (5.027 TFLOPS vs 1,405.4 GFLOPS). It also draws 94 W less power (125 W vs 219 W), requires a smaller PSU (300 W vs 550 W), and fits in a shorter card (229 mm vs 267 mm).

The 570's only advantage is its percentile ranking of 54th versus the 1660 SUPER's 53rd, but that is a statistical artifact of the 570 having a single OpenCL score that happens to land slightly higher in the distribution. The 1660 SUPER's average score of 12,986 is within 4% of the 570's 13,515, yet the 1660 SUPER achieves that average across ten diverse tests, including DirectX 9, 10, 11, 12, Vulkan, and compute workloads. The 570 has no data for any of those.

For a builder choosing between these two, the decision is dictated by workload and platform. If you need modern API support (Vulkan 1.4, DirectX 12_1), higher resolutions with 6 GB of GDDR6, or efficient power draw for a small form factor build, the GTX 1660 SUPER is the only rational pick. If you are restoring a period-correct 2010-era system and want the authentic Fermi experience, the GTX 570 remains a piece of computing history — but it is not a competitive GPU by any modern standard. The benchmark data does not support any scenario where the 570 is the better performer.