GeForce GTX 16 series
Entry-level to mid-range Turing GPUs without RT or Tensor cores.
The GeForce GTX 16 series is a line of graphics processing units from Nvidia that uses the Turing microarchitecture and was first announced in February 2019. It was sold at the same time as the GeForce RTX 20 series, but was designed for the entry-level to mid-range market that the RTX 20 series didn't cover. Reviewers often compared these cards to AMD's Radeon RX 500 series.
The lineup includes the GTX 1650, 1650 Super, 1660, 1660 Super, 1660 Ti, and a later, lower-end GTX 1630. The GTX 1650 came in both GDDR5 and GDDR6 versions. Like the RTX 20 series, the 16 series was succeeded by the GeForce 30 series. It was the last generation of GeForce GPUs without hardware-accelerated real-time ray tracing, which is why it kept the GTX name instead of using RTX.
**Architecture** The GTX 16 series is built on the same Turing architecture as the RTX 20 series, but it leaves out the Tensor (AI) and RT (ray tracing) cores that are exclusive to the RTX 20 series. It does keep the dedicated integer cores that allow integer and floating point operations to run at the same time. In March 2019, Nvidia announced that a driver update in April 2019 would enable DirectX Raytracing on 16 series cards starting with the GTX 1660, along with certain 10 series cards—a feature previously only available on RTX cards.
**Products** The series launched on February 23, 2019, with the announcement of the GeForce GTX 1660 Ti. All cards use a PCIe 3.0 x16 interface to connect to the CPU and are built on TSMC's 12 nm FinFET process. On April 22, 2019, alongside the GTX 1650 announcement, Nvidia also announced laptops with built-in GTX 1650 chipsets. The TU117 chip does not support Nvidia Optical Flow, which is useful for motion interpolation software. All TU117 GPUs use Volta's NVENC encoder (a minor improvement over Pascal) instead of Turing's.
**Desktop** *(No additional facts in source beyond those already covered.)*
**Laptop** *(No additional facts in source beyond those already covered.)*
**Reception** **GTX 1630** After its June 2022 release, the GTX 1630 received negative reviews. Jarred Walton of Tom's Hardware gave it a two-star review, criticizing its price, especially since it launched three years after most of the 16 series while performing only better than the GTX 1050.
Quick Facts
- Manufacturer
- TSMC
- Designfirm
- Nvidia
- Model
- GeForce GTX series
- Codename
- TU11x
- Architecture
- Turing
- Process
- 12 nm (FinFET)
- Entry
- GeForce GTX 1630 · GeForce GTX 1650 · GeForce GTX 1650 Super · GeForce GTX 1650 Ti (laptop only)
- Midrange
- GeForce GTX 1660 · GeForce GTX 1660 Super · GeForce GTX 1660 Ti
- Openglversion
- OpenGL 4.6
- Vulkanapi
- Vulkan 1.4 · SPIR-V 1.6
Facts from the source article.
Lore & Background
The GeForce GTX 16 series launched on February 23, 2019, with the announcement of the GeForce GTX 1660 Ti. Like the GeForce RTX 20 series, it is based on the Turing architecture but omits the Tensor (AI) and RT (ray tracing) cores exclusive to the 20 series. It retains dedicated integer cores for concurrent execution of integer and floating point operations. On March 18, 2019, Nvidia announced that via a driver update in April 2019 they would enable DirectX Raytracing on 16 series cards starting with GTX 1660, together with certain cards in the 10 series, a feature previously reserved to the RTX series.
The series includes the GTX 1650, 1650 Super, 1660, 1660 Super, 1660 Ti, and a lower-end GTX 1630 released later. The GTX 1650 features both a GDDR5 and GDDR6 version. On April 22, 2019, coinciding with the announcement of the GTX 1650, Nvidia announced laptops equipped with built-in GTX 1650 chipsets. The TU117 GPU does not support Nvidia Optical Flow, and all TU117 GPUs use Volta's NVENC encoder instead of Turing's. The 16 series was the last GPU generation of the GeForce series that did not support hardware-accelerated real-time ray tracing and was therefore marketed under the GTX instead of RTX prefix.
Reader's Guide
The GeForce GTX 16 series occupies a distinct position as the last GeForce generation marketed under the GTX prefix, lacking hardware-accelerated real-time ray tracing. Its significance lies in bringing the Turing architecture to the entry-level and mid-range market, a segment not addressed by the RTX 20 series. The series received mixed reception: the GTX 1630, released in June 2022, was criticized for its price and performance, with reviewers noting it performed only better than the GTX 1050 and was outperformed by AMD's Radeon RX 470 from 2016. The GTX 1650 was described as a tempting choice for small, power-efficient PCs, but the AMD RX 570 was considered the better value. The GTX 1650 Super was praised as a 'no-brainer' and a 'super easy recommendation' over the original 1650, being consistently around 30 percent faster for only $10 more, though its 4 GB VRAM was criticized as too small. The GTX 1660 Super was noted to almost make the 1660 Ti redundant and outperformed AMD's offerings in the same price range. Nvidia transitioned driver support for Windows 7 to legacy status in September 2024.
Did You Know?
- The GeForce GTX 16 series is based on the same Turing architecture as the RTX 20 series but omits Tensor and RT cores.
- DirectX Raytracing was enabled on GTX 1660 and certain 10 series cards via a driver update in April 2019.
- All TU117 GPUs in the series use Volta's NVENC encoder instead of Turing's.
More in Nvidia and AMD Graphics Processors 1-24
Spotted an error? Know more?
Reader corrections go straight into our review queue. Suggest an edit · How this site is sourced