NVIDIA GeForce GTX 1660 SUPER vs NVIDIA GeForce GTX 590 Comparison

The data presents a stark generational contrast. The NVIDIA GeForce GTX 1660 SUPER, a 2019 Turing-based card, utterly dominates the 2011 Fermi-based GeForce GTX 590 in the only shared benchmark. The GTX 1660 SUPER’s average benchmark score of 12,986 places it at the 53rd percentile of all GPUs, while the GTX 590’s average of 12,830 lands at the 52nd percentile. However, the head-to-head results tell a far more dramatic story than the average scores suggest, with the newer card delivering a 309.1% performance advantage in the Geekbench OpenCL test.

Where Each One Wins

Based on the benchmark data, the GTX 1660 SUPER wins in every measurable category where both cards have results. The single head-to-head benchmark, Geekbench OpenCL, shows the GTX 1660 SUPER scoring 52,490 against the GTX 590’s 12,830, a delta of 309.1%. This is not a marginal victory; it is a generational obliteration. The GTX 1660 SUPER also has a broader benchmark footprint, with scores in 3DMark Steel Nomad, Passmark DirectX 10/11/12/9, Passmark G2D, Passmark G3D, and Passmark GPU Compute, indicating it was tested across modern workloads that the GTX 590 was never subjected to.

The GTX 590 wins no benchmark in the provided data. Its sole result is the Geekbench OpenCL score, which is less than a quarter of the GTX 1660 SUPER’s performance in the same test. While the GTX 590’s average score of 12,830 is close to the GTX 1660 SUPER’s average of 12,986, this is misleading. The GTX 590’s average is derived from a single test, whereas the GTX 1660 SUPER’s average is pulled from ten diverse workloads. In terms of raw compute throughput, the GTX 1660 SUPER’s FP32 performance of 5.027 TFLOPS dwarfs the GTX 590’s 1,244.2 GFLOPS. For any modern application, the GTX 1660 SUPER is the only viable option.

Architecture Differences

The architectural divide between these two cards is immense. The GTX 1660 SUPER is built on the Turing architecture using the TU116 chip, fabricated on a 12 nm process at TSMC. It packs 6,600 million transistors into a 284 mm² die, achieving a transistor density of 23.2M / mm². In contrast, the GTX 590 uses the older Fermi 2.0 architecture with the GF110 chip, built on a 40 nm process, also at TSMC. It contains only 3,000 million transistors spread across a massive 520 mm² die, resulting in a density of just 5.8M / mm².

Memory configurations reinforce the generational gap. The GTX 1660 SUPER features 6 GB of GDDR6 memory on a 192-bit bus, delivering 336.0 GB/s of bandwidth with an effective speed of 14 Gbps. The GTX 590, by comparison, offers a paltry 1536 MB of GDDR5 memory on a 384-bit bus, yielding only 164.0 GB/s of bandwidth at 3.4 Gbps effective. The newer card has four times the memory capacity and more than double the bandwidth, a critical advantage for modern textures and high-resolution workloads.

Compute resources also favor the newer card. The GTX 1660 SUPER has 1,408 shading units, 88 TMUs, and 48 ROPs, with pixel and texture rates of 85.68 GPixel/s and 157.1 GTexel/s, respectively. The GTX 590 has just 512 shading units, 64 TMUs, and 48 ROPs, with pixel and texture rates of 19.46 GPixel/s and 38.91 GTexel/s. The GTX 1660 SUPER also supports FP16 at 10.05 TFLOPS (2:1 ratio), a feature entirely absent from the GTX 590, which has no FP16 data. In terms of interface and features, the GTX 1660 SUPER uses PCIe 3.0 x16 and supports Vulkan 1.4, while the GTX 590 is limited to PCIe 2.0 x16 and has no Vulkan support listed.

The Verdict

The data is unambiguous: the NVIDIA GeForce GTX 1660 SUPER is the superior product in every tested metric. Its 309.1% lead in Geekbench OpenCL over the GTX 590 is the defining statistic. The GTX 1660 SUPER offers modern API support, including DirectX 12 (12_1) versus the GTX 590’s DirectX 12 (11_0), and Vulkan 1.4 versus none. It also consumes significantly less power, with a 125 W TDP versus the GTX 590’s 365 W, and requires a 300 W PSU versus a 750 W PSU.

For any user, the choice is clear: pick the GTX 1660 SUPER. The GTX 590’s only advantage is its launch MSRP of 699 USD, which was higher than the GTX 1660 SUPER’s launch MSRP of 229 USD. However, the GTX 590’s performance is so far behind that its price is irrelevant. The GTX 1660 SUPER’s average benchmark score of 12,986 is only 0.4% higher than its closest rival, the NVIDIA GeForce RTX 3050 Ti Mobile, but it is 309.1% higher than the GTX 590 in the head-to-head. The GTX 590, with its 52nd percentile ranking, is effectively obsolete. The GTX 1660 SUPER, ranking at the 53rd percentile, is a capable modern entry-level card. Data supports only one verdict: the GTX 1660 SUPER wins outright.

FAQ

Q: How much faster is the GTX 1660 SUPER than the GTX 590 in the head-to-head benchmark?

A: The GTX 1660 SUPER scores 52,490 in Geekbench OpenCL, while the GTX 590 scores 12,830, giving the GTX 1660 SUPER a 309.1% performance advantage.

Q: Which card has more memory and bandwidth?

A: The GTX 1660 SUPER has 6 GB of GDDR6 memory with 336.0 GB/s bandwidth, while the GTX 590 has 1536 MB of GDDR5 memory with 164.0 GB/s bandwidth.

Q: What are the architectural differences between the two cards?

A: The GTX 1660 SUPER uses the Turing architecture on a 12 nm process with 6,600 million transistors, while the GTX 590 uses Fermi 2.0 on a 40 nm process with 3,000 million transistors.

Q: Does the GTX 590 support Vulkan?

A: No, the GTX 590 has no Vulkan support listed, whereas the GTX 1660 SUPER supports Vulkan 1.4.

Q: What is the power consumption difference?

A: The GTX 1660 SUPER has a 125 W TDP and requires a 300 W PSU, while the GTX 590 has a 365 W TDP and requires a 750 W PSU.

Q: Which card has a higher average benchmark score?

A: The GTX 1660 SUPER has an average benchmark score of 12,986, which is slightly higher than the GTX 590’s average of 12,830.

Head-to-Head Benchmarks

The only direct comparison available is Geekbench OpenCL, and it is a landslide. The GTX 1660 SUPER posts a score of 52,490, while the GTX 590 manages just 12,830. This translates to a 309.1% delta in favor of the newer card. To put that in perspective, the GTX 590’s score is roughly equivalent to the GTX 1660 SUPER’s closest rivals; the GTX 590’s average score of 12,830 is within 0.4% of the NVIDIA GeForce GTX 670’s 12,773 and within 0.3% of the AMD Radeon 740M’s 12,870. The GTX 1660 SUPER, by contrast, is 309.1% ahead of the GTX 590 in the same test.

The GTX 1660 SUPER also wins all other benchmarks it was tested in, though the GTX 590 has no results in those tests. For example, the GTX 1660 SUPER scores 12,699 in Passmark G3D and 5,076 in Passmark GPU Compute. Its 3DMark Steel Nomad DX12 score is 1,278. These scores, combined with its 53rd percentile ranking, show a card that is competitive with modern entries like the AMD Radeon RX 580 (which has an average score of 12,928, a 0.4% delta from the GTX 1660 SUPER). The GTX 590, with its 52nd percentile ranking, is barely distinguishable from much older cards like the AMD FirePro W5100 (average score 12,847, -0.1% delta).

The compute performance gap is the most telling. The GTX 1660 SUPER’s FP32 throughput is 5.027 TFLOPS, while the GTX 590’s is 1,244.2 GFLOPS. Even the pixel and texture rates show the older card struggling: 19.46 GPixel/s and 38.91 GTexel/s versus the GTX 1660 SUPER’s 85.68 GPixel/s and 157.1 GTexel/s. In every measurable way, the GTX 1660 SUPER is the definitive winner, with the Geekbench OpenCL result serving as the ultimate proof of its 309.1% dominance.