NVIDIA GeForce GTX 1660 vs NVIDIA GeForce GTX 570 Comparison

Head-to-Head Benchmarks

The database contains one directly comparable benchmark between these two cards: Geekbench OpenCL. The results are decisive. The GTX 1660 scored 47,850, while the GTX 570 scored 13,515. This gives the GTX 1660 a 71.8% advantage in the OpenCL workload, a massive gap that reflects the generational shift in compute capability.

In the context of their respective peer groups, the GTX 570 sits in the 54th percentile of all GPUs, while the GTX 1660 sits in the 51st percentile. Despite the lower overall percentile ranking, the GTX 1660's raw OpenCL score is more than triple that of the GTX 570. The GTX 570's nearest rivals include the AMD Radeon HD 7770M (13,536, a -0.2% delta) and the AMD Radeon RX 9070 XT (13,543, also -0.2%), showing that the older card is clustered with much weaker modern parts in compute tests. The GTX 1660's nearest rivals include the AMD Radeon RX 7800 XT (11,627, a 0.5% delta) and the AMD Radeon RX 6500 XT (11,842, a -1.4% delta), placing it in a performance tier far above the GTX 570's competition.

The delta between the two cards in the head-to-head test is -71.8% from the GTX 1660's perspective, meaning the GTX 570 trails by nearly three-quarters of the winner's score. No other benchmark results exist for the GTX 570 in the database, so the OpenCL result stands as the primary metric for comparison. The GTX 1660 also has a broader benchmark suite recorded, including a 3DMark Steel Nomad DX12 score of 1,065, a Geekbench Vulkan score of 50,137, and a Passmark G3D score of 11,646, but none of these have corresponding GTX 570 results for a direct comparison.

FAQ

Q: Which card has the higher average benchmark score?

A: The GTX 570 has an average benchmark score of 13,515, while the GTX 1660 has an average of 11,680. The GTX 570's average is higher only because it has a single recorded benchmark, the Geekbench OpenCL test, whereas the GTX 1660's average includes a wider range of tests, some of which score much lower.

Q: In the Geekbench OpenCL test, what is the exact score difference?

A: The GTX 1660 scored 47,850 against the GTX 570's 13,515. The GTX 1660 wins by a delta of -71.8%, meaning the GTX 570's score is 71.8% lower than the GTX 1660's score.

Q: How do these cards rank against all other GPUs in the database?

A: The GTX 570 is in the 54th percentile of all GPUs, and the GTX 1660 is in the 51st percentile. Despite the GTX 570 ranking slightly higher in percentile, that ranking is based on a much narrower set of recorded data.

Q: What are the closest rivals for each card?

A: For the GTX 570, the closest rivals are the AMD Radeon HD 7770M with an average score of 13,536 (-0.2%), the AMD Radeon RX 9070 XT with 13,543 (-0.2%), the NVIDIA P106-090 with 13,470 (0.3%), and the AMD Radeon Pro 555 with 13,407 (0.8%). For the GTX 1660, the closest rivals are the AMD Radeon RX 7800 XT with 11,627 (0.5%), the AMD Radeon Pro 5500M with 11,528 (1.3%), the AMD Radeon RX 6500 XT with 11,842 (-1.4%), and the NVIDIA Tesla K20c with 11,479 (1.8%).

Q: Which card supports Vulkan?

A: The GTX 1660 lists Vulkan 1.4 support in its API specifications. The GTX 570's API list does not include a Vulkan version, though it does support DirectX 12 (11_0) and OpenGL 4.6.

Q: What is the difference in DirectX feature level support?

A: The GTX 570 supports DirectX 12 (11_0), while the GTX 1660 supports DirectX 12 (12_1). The 12_1 feature level includes additional features that the older Fermi architecture cannot expose.

Architecture Differences

The GTX 570 is built on the GF110 chip using the Fermi 2.0 architecture, fabricated on a 40 nm process at TSMC. The GTX 1660 uses the TU116 chip with the Turing architecture, also from TSMC but on a 12 nm process. The node shrink is significant: the GTX 570 packs 3,000 million transistors into a 520 mm² die, giving a transistor density of 5.8M per mm². The GTX 1660 packs 6,600 million transistors into a smaller 284 mm² die, achieving a density of 23.2M per mm². That is roughly four times the density, which explains how the newer card delivers far higher compute throughput while drawing less power.

The GTX 570 has 480 shading units, 60 texture mapping units, and 40 ROPs. The GTX 1660 has 1,408 shading units, 88 TMUs, and 48 ROPs. Neither card has ray tracing cores or tensor cores, so both rely purely on traditional rasterization and compute hardware. The GTX 1660 supports FP16 compute at a 2:1 ratio with 10.05 TFLOPS, while the GTX 570 lists no FP16 capability. FP32 performance tells the story: the GTX 570 delivers 1,405.4 GFLOPS, while the GTX 1660 delivers 5.027 TFLOPS, a generational leap in raw shader throughput.

The GTX 570 uses a PCIe 2.0 x16 interface, while the GTX 1660 uses PCIe 3.0 x16, doubling the available bus bandwidth for data transfers. Display outputs also differ: the GTX 570 offers 2x DVI and 1x mini-HDMI 1.3a, while the GTX 1660 offers 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a. The newer card supports modern display standards, which matters for contemporary monitors. The GTX 570's DirectX 12 support is limited to the 11_0 feature level, while the GTX 1660 reaches 12_1, and the GTX 1660 also adds Vulkan 1.4 support, which the GTX 570 completely lacks.

Specification Differences

The two cards differ in nearly every measurable specification. The GTX 570 has 1280 MB of GDDR5 memory on a 320-bit bus, while the GTX 1660 has 6 GB of GDDR5 on a 192-bit bus. Memory bandwidth is 152.0 GB/s for the GTX 570 and 192.1 GB/s for the GTX 1660, so the newer card wins on bandwidth despite a narrower bus. Memory clock is 950 MHz (3.8 Gbps effective) on the GTX 570, versus 2001 MHz (8 Gbps effective) on the GTX 1660.

The GTX 570 has no base or boost clock listed in the database, while the GTX 1660 runs at 1530 MHz base and 1785 MHz boost. Pixel rate is 21.96 GPixel/s for the GTX 570 and 85.68 GPixel/s for the GTX 1660. Texture rate is 43.92 GTexel/s for the GTX 570 and 157.1 GTexel/s for the GTX 1660. Power consumption is dramatically different: the GTX 570 has a 219 W TDP and requires a 550 W suggested PSU with 2x 6-pin power connectors. The GTX 1660 has a 120 W TDP, a 300 W suggested PSU, and a single 8-pin connector.

Physical dimensions differ as well. The GTX 570 is 267 mm (10.5 inches) long, while the GTX 1660 is 229 mm (9 inches) long, 111 mm (4.4 inches) tall, and 35 mm (1.4 inches) wide. Both are dual-slot cards. The GTX 570 launched with a 349 USD MSRP, while the GTX 1660 launched with a 219 USD MSRP. Release dates are far apart: the GTX 570 came out on December 6, 2010, and the GTX 1660 on March 13, 2019. The GTX 570's predecessor is the GeForce 400 series and its successor is the GeForce 600 series. The GTX 1660's predecessor is the GeForce 10 series and its successor is the GeForce 20 series.

The Verdict

The data points to one clear conclusion: the GTX 1660 is the superior card for compute-heavy workloads. The OpenCL benchmark shows a 71.8% lead over the GTX 570, and the GTX 1660 also brings a modern feature set including Vulkan 1.4, DirectX 12_1, and HDMI 2.0 with DisplayPort 1.4a. The GTX 570's only advantage in the recorded data is its percentile ranking, 54th versus 51st, but that ranking is based on a single benchmark score of 13,515. The GTX 1660's average score of 11,680 is dragged down by lower-scoring tests like Passmark DirectX 9 (177), Passmark DirectX 10 (61), and Passmark DirectX 12 (49), which do not reflect its actual compute strength.

For a user choosing between these two cards today, the GTX 1660 is the obvious pick. It has more than four times the FP32 throughput, triple the memory capacity, higher bandwidth, lower power draw, and a smaller physical footprint. The GTX 570 is an end-of-life product from 2010 with no Vulkan support and limited DirectX 12 feature levels. The GTX 1660 is also end-of-life as of the database records show, but it is a much newer design that remains relevant for modern workloads.

Where Each One Wins

The GTX 570 wins in exactly one recorded category: percentile ranking against all GPUs in the database, where it sits at 54th place compared to the GTX 1660's 51st. That is a statistical artifact of the limited benchmark data rather than a practical performance win. In raw compute, the GTX 1660 wins the only direct head-to-head test by a 71.8% margin.

The GTX 1660 wins in every other practical category recorded in the database. It has higher memory bandwidth (192.1 GB/s versus 152.0 GB/s), higher pixel rate (85.68 GPixel/s versus 21.96 GPixel/s), higher texture rate (157.1 GTexel/s versus 43.92 GTexel/s), and vastly higher FP32 throughput (5.027 TFLOPS versus 1,405.4 GFLOPS). It also supports Vulkan 1.4, has a PCIe 3.0 interface, and offers modern display outputs including HDMI 2.0 and DisplayPort 1.4a. The GTX 570 lacks Vulkan entirely and is limited to PCIe 2.0 and older display connectors.

For gaming, the GTX 1660's higher pixel and texture rates translate directly to better rasterization performance, and its 6 GB frame buffer is far more practical than the GTX 570's 1280 MB. For compute tasks like OpenCL workloads, the GTX 1660's 47,850 score versus the GTX 570's 13,515 score speaks for itself. The GTX 570's lower TDP is not an advantage, since the GTX 1660 draws 120 W versus 219 W while delivering more performance. The GTX 1660 also requires a smaller PSU, 300 W versus 550 W, and a single 8-pin connector instead of two 6-pin connectors. There is no workload category in the recorded data where the GTX 570 outperforms the GTX 1660 except the ambiguous percentile metric, and even that is based on incomplete testing.