Light from a streetlamp passing through blinds flickers: the stripes shift, and brightness rises and falls. From that rhythm, you can guess how wide the lamp is. The new method for measuring stars works similarly: instead of capturing a sharp image, astronomers monitor the "twinkling" of light collected by two or more telescopes simultaneously.

Usually, the size of a distant star is determined by combining light from different telescopes into one sharp picture, as if building a giant mirror as wide as the distance between them. But there's another way: don't combine the beams, just record how quickly brightness changes in each telescope. From the speed of these changes, you can compute the star's apparent diameter. This technique is called "intensity interferometry," and it doesn't require perfect synchronization — ordinary clocks suffice.

The study shows: for the star HD 17652, a red giant, two hours of observations at an infrared wavelength can yield a size with an error of less than one percent. This uses a baseline, the distance between telescopes, of about one hundred meters, like the existing PIONIER instrument. If you observe at shorter wavelengths, a second "peak" appears in the signal, providing an independent check that is nearly unaffected by how we model the limb darkening of the star's disk.

The main advantage of the new approach is the ability to use many telescopes and multiple color filters simultaneously. For instance, the future Cherenkov Telescope Array will consist of dozens of mirrors spread over kilometers. This would cut observation time to a few minutes per star. Then the method becomes practical for checking the sizes of a whole sample of red giants, which serve as "standard candles"—reference points for measuring distances in the universe.