NVIDIA GeForce GTX 1660 Ti vs NVIDIA GeForce GTX 460 Comparison

The GeForce GTX 460 and the GeForce GTX 1660 Ti represent two distinct eras of NVIDIA’s GPU design, separated by nearly a decade of architectural evolution. The data shows a fundamental gap in raw compute capability, with the newer card delivering a Geekbench OpenCL score of 51,029 compared to the GTX 460’s 7,925. This translates to a commanding 84.5% lead for the GTX 1660 Ti in the sole head-to-head benchmark available, illustrating the massive performance leap achieved from the Fermi to Turing generations.

Head-to-Head Benchmarks

The only direct benchmark comparison available is Geekbench OpenCL, and it paints a stark picture of generational progress. The GTX 1660 Ti scores 51,029 points, a figure that dwarfs the GTX 460’s 7,925 points. The delta of -84.5% for the older card confirms that the GTX 460 operates in a completely different performance tier. In practical terms, the GTX 1660 Ti delivers over six times the compute throughput of its predecessor, making it suitable for workloads that would be entirely impractical on the GTX 460.

This single benchmark result is informative because OpenCL is a general-purpose compute API that stresses raw shader throughput and memory bandwidth. The GTX 1660 Ti’s 5.437 TFLOPS of FP32 performance and 288.0 GB/s of memory bandwidth are the underlying reasons for its dominance. In contrast, the GTX 460’s 907.2 GFLOPS and 86.40 GB/s bandwidth are figures that were competitive in 2010 but are now considered entry-level. The benchmark result is not a close race; it is a clear demonstration of how far GPU architecture has advanced in the intervening years.

Looking at the broader context, the GTX 460’s average benchmark score of 7,925 places it in the 41st percentile of all GPUs, a position it shares with the GTX 1660 Ti, which also holds a 41st percentile ranking despite its far higher raw score. This paradox is explained by the fact that the percentile is relative to the entire historical database of GPUs, where the GTX 460 was a high-end part at launch, while the GTX 1660 Ti is a mid-range part from a later generation. The GTX 460’s nearest rivals, such as the NVIDIA Quadro K4100M (7,906, +0.2%) and the GeForce GTX 880M (8,040, -1.4%), show that it performs in line with early-2010s mobile and workstation parts. The GTX 1660 Ti’s rivals, including the AMD Radeon 540 (7,673, +0.7%) and Intel Arc A310 (7,550, +2.3%), are much lower-scoring, indicating that its 51,029 Geekbench score is an outlier relative to its average score across multiple tests, which includes older DirectX and compute benchmarks where it scores lower.

Where Each One Wins

The GTX 1660 Ti wins decisively in every measurable category, with the single head-to-head benchmark being a landslide. Its strengths lie in modern compute workloads, as evidenced by its Geekbench OpenCL score. The card’s 1,536 shading units, 96 TMUs, and 48 ROPs provide a robust foundation for both gaming and content creation. Its 6 GB of GDDR6 memory on a 192-bit bus delivers 288.0 GB/s of bandwidth, which is essential for high-resolution textures and compute tasks. The 12 nm Turing architecture also brings features like DirectX 12 (12_1) and Vulkan 1.4 support, making it compatible with modern APIs that the GTX 460 cannot fully utilize.

The GTX 460, by contrast, has no benchmark wins against the GTX 1660 Ti. Its role is purely historical, representing a capable GPU for its era. With 336 shading units, 56 TMUs, and 24 ROPs, it was a strong performer for 2010-era games and applications. Its 768 MB of GDDR5 memory was sufficient for the resolutions and texture sizes of that time, but is now a severe limitation. The card’s DirectX 12 (11_0) support is a compatibility footnote, as it cannot handle feature levels required by modern titles. The GTX 460’s only advantage is its physical size and power draw: it is a 210 mm dual-slot card with a 160 W TDP and two 6-pin connectors, which is smaller and less power-hungry than some modern cards, but this does not translate into any performance benefit.

Architecture Differences

The architectural gap between the GTX 460 and GTX 1660 Ti is immense, stemming from a complete redesign of NVIDIA’s GPU core. The GTX 460 uses the GF104 chip based on the Fermi architecture, manufactured on TSMC’s 40 nm process. This chip contains 1,950 million transistors on a 332 mm² die, yielding a transistor density of 5.9M per mm². The GTX 1660 Ti, in contrast, uses the TU116 chip based on the Turing architecture, built on TSMC’s 12 nm process. This newer chip packs 6,600 million transistors into a smaller 284 mm² die, achieving a far higher density of 23.2M per mm². This jump from 40 nm to 12 nm is a primary driver of the performance difference, allowing for nearly 3.4 times more transistors in a smaller physical area.

The memory subsystem is another major differentiator. The GTX 460 features 768 MB of GDDR5 memory clocked at 900 MHz (3.6 Gbps effective), which provides 86.40 GB/s of bandwidth across a 192-bit bus. The GTX 1660 Ti features 6 GB of GDDR6 memory clocked at 1500 MHz (12 Gbps effective), which provides 288.0 GB/s of bandwidth on the same 192-bit bus. The combination of faster memory technology and higher clock speeds results in a 3.3-fold increase in memory bandwidth, which is critical for modern game assets and compute workloads. The GTX 460’s 24 ROPs and 9.450 GPixel/s pixel rate are dwarfed by the GTX 1660 Ti’s 48 ROPs and 84.96 GPixel/s, highlighting the latter’s superior fill rate capabilities.

The compute capabilities differ by an order of magnitude. The GTX 460’s 907.2 GFLOPS of FP32 performance is a fraction of the GTX 1660 Ti’s 5.437 TFLOPS, a six-fold increase. The GTX 1660 Ti also supports FP16 compute at 10.87 TFLOPS with a 2:1 ratio, a feature absent from the GTX 460. Furthermore, the GTX 1660 Ti supports modern APIs like Vulkan 1.4, while the GTX 460 has no Vulkan support. The bus interface also improves from PCIe 2.0 x16 on the GTX 460 to PCIe 3.0 x16 on the GTX 1660 Ti, doubling the potential data transfer rate between the GPU and the host system. These architectural differences collectively explain why the GTX 1660 Ti is an entirely different class of hardware.

FAQ

Q: How much faster is the GTX 1660 Ti than the GTX 460 in compute benchmarks?

A: In the Geekbench OpenCL test, the GTX 1660 Ti scores 51,029 compared to the GTX 460’s 7,925, a difference of -84.5% for the GTX 460, meaning the GTX 1660 Ti is approximately 6.4 times faster.

Q: Do both cards support the same level of DirectX?

A: No. The GTX 460 supports DirectX 12 (11_0), while the GTX 1660 Ti supports DirectX 12 (12_1), which includes more advanced features. The GTX 1660 Ti also adds Vulkan 1.4 support, which the GTX 460 lacks.

Q: What are the memory configurations of each card?

A: The GTX 460 has 768 MB of GDDR5 memory with a 192-bit bus, providing 86.40 GB/s of bandwidth. The GTX 1660 Ti has 6 GB of GDDR6 memory with a 192-bit bus, providing 288.0 GB/s of bandwidth.

Q: Which card has a higher transistor density?

A: The GTX 1660 Ti has a far higher transistor density of 23.2M per mm², compared to the GTX 460’s 5.9M per mm², due to its more advanced 12 nm process node versus the GTX 460’s 40 nm node.

Q: What is the TDP difference between the two cards?

A: The GTX 460 has a TDP of 160 W and requires two 6-pin power connectors, while the GTX 1660 Ti has a lower TDP of 120 W and requires only one 8-pin connector.

Q: Which card has a higher pixel fill rate?

A: The GTX 1660 Ti has a pixel rate of 84.96 GPixel/s, which is significantly higher than the GTX 460’s 9.450 GPixel/s, thanks to its 48 ROPs versus the GTX 460’s 24 ROPs.

Specification Differences

The following table outlines the key specifications where the two NVIDIA cards differ, based on the available data.

| Specification | NVIDIA GeForce GTX 460 | NVIDIA GeForce GTX 1660 Ti |

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

| Architecture | Fermi | Turing |

| Generation | GeForce 400 | GeForce 16 |

| Process Node | 40 nm | 12 nm |

| Transistors | 1,950 million | 6,600 million |

| Die Size | 332 mm² | 284 mm² |

| Transistor Density | 5.9M / mm² | 23.2M / mm² |

| Base Clock | N/A | 1500 MHz |

| Boost Clock | N/A | 1770 MHz |

| Memory Clock | 900 MHz (3.6 Gbps effective) | 1500 MHz (12 Gbps effective) |

| Memory Size | 768 MB | 6 GB |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bandwidth | 86.40 GB/s | 288.0 GB/s |

| Shading Units | 336 | 1536 |

| TMUs | 56 | 96 |

| ROPs | 24 | 48 |

| Pixel Rate | 9.450 GPixel/s | 84.96 GPixel/s |

| Texture Rate | 37.80 GTexel/s | 169.9 GTexel/s |

| FP32 Performance | 907.2 GFLOPS | 5.437 TFLOPS |

| FP16 Performance | N/A | 10.87 TFLOPS (2:1) |

| TDP | 160 W | 120 W |

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

| Suggested PSU | 450 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 Support | 12 (11_0) | 12 (12_1) |

| Vulkan Support | N/A | 1.4 |

| Dimensions | 210 mm (8.3 inches) | 229 mm (9 inches) x 111 mm (4.4 inches) x 35 mm (1.4 inches) |

| Release Date | 2010-07-11 | 2019-02-21 |

| Launch MSRP | 199 USD | 279 USD |