NVIDIA GeForce GTX 1660 vs NVIDIA GeForce GTX 590 Comparison

NVIDIA’s GeForce GTX 590 and GeForce GTX 1660 represent two very different eras of the company’s GPU lineup, separated by eight years of architectural evolution. The GTX 590, a dual-GPU flagship from the Fermi generation, was built for absolute performance at the top of the product stack, while the GTX 1660 is a Turing-based mainstream part designed for efficiency and modern feature support. Benchmark data shows a clear generational shift, but the comparison is more nuanced than a simple score difference.

Head-to-Head Benchmarks

The only directly comparable benchmark between the two cards is Geekbench OpenCL, and the results are decisive. The GTX 1660 scores 47,850 points, while the GTX 590 manages just 12,830 points. That translates to a delta of -73.2% for the older card, meaning the GTX 1660 delivers nearly 3.7 times the compute performance in this workload. This is not a marginal improvement; it is a complete overhaul of compute capability.

In the context of their respective peers, the GTX 590’s score of 12,830 places it in a tight cluster. Its nearest rival, the AMD Radeon Pro 455, scores 12,831 with a delta of 0%. The AMD FirePro W5100 is just ahead at 12,847 (+0.1%), while the AMD Radeon 740M posts 12,870 (+0.3%). The only card in its vicinity that it beats is the NVIDIA GeForce GTX 670, which scores 12,773 with a delta of -0.4%. The GTX 590 is, therefore, essentially a peer of entry-level workstation and integrated-class GPUs from a compute perspective today.

The GTX 1660, by contrast, sits in a different league. Its 47,850 OpenCL score is not close to any of its nearest rivals. The AMD Radeon RX 6500 XT scores 11,842, a delta of -1.4% relative to the GTX 1660’s average, but that is comparing the average benchmark score, not the OpenCL result. For the specific Geekbench OpenCL test, the GTX 1660’s advantage is overwhelming. Its average benchmark score across all tests is 11,680, with a percentile rank of 51, while the GTX 590’s average is 12,830 with a percentile of 52. Interestingly, the GTX 590 actually has a higher average score and percentile than the GTX 1660, but this is misleading because the GTX 590 has only one benchmark entry, while the GTX 1660 has ten, including several DirectX and compute tests where it performs differently.

The GTX 1660 also has additional benchmark data that the GTX 590 lacks entirely. In 3DMark Steel Nomad DX12, it scores 1,065. In Geekbench Vulkan, it hits 50,137. Passmark results show 11,646 in G3D, 4,963 in GPU Compute, and 776 in G2D. These scores are not comparable to anything on the GTX 590 because no such tests were run for the older card, but they demonstrate that the GTX 1660’s modern feature set is measurable across a wider range of workloads.

FAQ

Q: Which card has a higher Geekbench OpenCL score?

A: The NVIDIA GeForce GTX 1660 is significantly ahead, scoring 47,850 versus the GTX 590’s 12,830, a delta of -73.2% in favor of the newer card.

Q: Are the two cards comparable in overall performance percentile?

A: No. The GTX 590 holds a 52nd percentile rank among all GPUs, while the GTX 1660 sits at the 51st percentile. Despite the massive OpenCL gap, their percentile ranks are nearly identical because the GTX 590 has only one benchmark score (12,830) and the GTX 1660 has an average of 11,680 across many tests.

Q: What is the closest rival to the GTX 590 based on average score?

A: The AMD Radeon Pro 455 is the closest, with an average score of 12,831 and a delta of 0% from the GTX 590’s 12,830. The AMD FirePro W5100 and AMD Radeon 740M are also within 0.3% of the GTX 590’s score.

Q: Does the GTX 1660 have a higher transistor density than the GTX 590?

A: Yes. The GTX 1660 has a transistor density of 23.2M per mm², while the GTX 590 has only 5.8M per mm². This reflects the move from a 40 nm process to a 12 nm process.

Q: Which card supports Vulkan natively?

A: The GTX 1660 supports Vulkan 1.4, while the GTX 590’s API list shows no Vulkan support. The GTX 590 is limited to DirectX 12 (11_0) and OpenGL 4.6.

Q: How do the memory configurations differ?

A: The GTX 590 has 1536 MB of GDDR5 on a 384-bit bus, with 164.0 GB/s bandwidth. The GTX 1660 has 6 GB of GDDR5 on a 192-bit bus, with 192.1 GB/s bandwidth, offering four times the capacity and higher total bandwidth.

Architecture Differences

The two cards are built on fundamentally different architectures. The GTX 590 uses the GF110 chip based on Fermi 2.0, fabricated on a 40 nm process at TSMC. It packs 3,000 million transistors into a 520 mm² die, resulting in a transistor density of 5.8M per mm². The GTX 1660, in contrast, uses the TU116 chip based on Turing, built on a 12 nm process, also at TSMC. It contains 6,600 million transistors on a 284 mm² die, achieving a density of 23.2M per mm². This is a fourfold increase in density, enabling far more complex logic in a smaller physical footprint.

The GTX 590 is a dual-GPU card, as evidenced by its 512 shading units, 64 texture mapping units, and 48 ROPs. The GTX 1660 has 1,408 shading units, 88 TMUs, and 48 ROPs, meaning the newer card has nearly three times the shader count in a single GPU. The GTX 590’s pixel rate is 19.46 GPixel/s and its texture rate is 38.91 GTexel/s, while the GTX 1660 achieves 85.68 GPixel/s and 157.1 GTexel/s, respectively. FP32 compute is also drastically different: the GTX 590 delivers 1,244.2 GFLOPS, while the GTX 1660 delivers 5.027 TFLOPS, or roughly 4x more. The GTX 1660 also supports FP16 at 10.05 TFLOPS with a 2:1 ratio, a feature entirely absent from the GTX 590’s spec sheet.

Memory architecture also diverges. The GTX 590 uses 1536 MB of GDDR5 across a 384-bit bus, yielding 164.0 GB/s of bandwidth. The GTX 1660 uses 6 GB of GDDR5 across a 192-bit bus, yielding 192.1 GB/s. Despite the narrower bus, the newer card has higher bandwidth and four times the memory capacity. The GTX 590 requires a 365 W TDP and dual 8-pin power connectors, with a suggested PSU of 750 W. The GTX 1660 is far more efficient, drawing 120 W from a single 8-pin connector and requiring a 300 W PSU.

The process node difference is stark: 40 nm versus 12 nm. This explains the efficiency gap. The GTX 590 is 279 mm long, 111 mm tall, and 40 mm wide, while the GTX 1660 is 229 mm long, 111 mm tall, and 35 mm wide, making the older card physically larger as well. The GTX 590 uses PCIe 2.0 x16, while the GTX 1660 uses PCIe 3.0 x16. Display outputs also differ: the GTX 590 offers 3x DVI and 1x mini-DisplayPort, while the GTX 1660 provides 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a. The GTX 590 supports DirectX 12 (11_0) and OpenGL 4.6, with no Vulkan listing. The GTX 1660 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4.

The Verdict

The data makes the choice clear for almost any workload. The GTX 1660 is the superior card in compute performance, memory capacity, feature support, and power efficiency. Its Geekbench OpenCL score of 47,850 is 373% higher than the GTX 590’s 12,830, and it does so while consuming less than a third of the power (120 W versus 365 W). The GTX 1660 also supports Vulkan 1.4 and DirectX 12 (12_1), whereas the GTX 590 is limited to DirectX 12 (11_0) and no Vulkan. For gaming, the GTX 1660’s 6 GB VRAM is four times the GTX 590’s 1.5 GB, eliminating the capacity bottlenecks that would cripple modern titles on the older card.

The GTX 590’s only statistical edge is its average benchmark score of 12,830 versus the GTX 1660’s 11,680, and its 52nd percentile versus 51st. However, this is an artifact of test selection. The GTX 590 has a single benchmark entry, while the GTX 1660 has ten, including lower-scoring DirectX 9 and G2D tests. In any DirectX 10 or 11 test, the GTX 1660 scores 61 and 79 respectively on Passmark, which are low absolute numbers but represent modern driver support. The GTX 590 has no such data.

Who should pick which? The GTX 1660 is the only rational choice for anyone building a system today. It offers modern API support, higher compute throughput, and vastly better efficiency. The GTX 590 is an end-of-life product from 2011, with a successor in the GeForce 600 series and a predecessor in the GeForce 400 series. Its 365 W TDP and dual 8-pin requirement make it impractical for modern PSUs and cases. The GTX 1660, also end-of-life but from 2019, is a far more viable secondary or budget card. Its 120 W TDP and single 8-pin connector are manageable, and its 6 GB memory is still relevant for 1080p gaming. The GTX 590 should be avoided except for legacy collection purposes, as its performance is now comparable to an AMD Radeon 740M integrated GPU.

Specification Differences

The core specifications differ substantially between the two cards. The GTX 590 uses the GF110 chip on a 40 nm process, with 3,000 million transistors on a 520 mm² die. The GTX 1660 uses the TU116 chip on a 12 nm process, with 6,600 million transistors on a 284 mm² die. Transistor density jumps from 5.8M per mm² to 23.2M per mm².

Memory is another major split. The GTX 590 has 1536 MB of GDDR5 on a 384-bit bus with 164.0 GB/s bandwidth, operating at 854 MHz (3.4 Gbps effective). The GTX 1660 has 6 GB of GDDR5 on a 192-bit bus with 192.1 GB/s bandwidth, operating at 2001 MHz (8 Gbps effective). The GTX 590 has 512 shading units, 64 TMUs, and 48 ROPs, while the GTX 1660 has 1,408 shading units, 88 TMUs, and 48 ROPs. Pixel rate rises from 19.46 GPixel/s to 85.68 GPixel/s, and texture rate rises from 38.91 GTexel/s to 157.1 GTexel/s. FP32 compute is 1,244.2 GFLOPS versus 5.027 TFLOPS, with the GTX 1660 adding FP16 at 10.05 TFLOPS.

Power and physical specs also diverge. The GTX 590 has a 365 W TDP, dual 8-pin connectors, and a 750 W suggested PSU, measuring 279 mm by 111 mm by 40 mm. The GTX 1660 has a 120 W TDP, a single 8-pin connector, and a 300 W suggested PSU, measuring 229 mm by 111 mm by 35 mm. The GTX 590 uses PCIe 2.0 x16, while the GTX 1660 uses PCIe 3.0 x16. Display outputs are 3x DVI and 1x mini-DisplayPort on the GTX 590, versus 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a on the GTX 1660. API support differs, with the GTX 590 limited to DirectX 12 (11_0) and OpenGL 4.6, while the GTX 1660 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 590 launched on 2011-03-23 with an MSRP of 699 USD, while the GTX 1660 launched on 2019-03-13 with an MSRP of 219 USD.