What are the present real-time capabilities of various modern telescopes as video cameras? How do these compare with the capabilities various of space telescopes?

For example, if it was known that there would soon be a Saturn-comet impact (akin to the Shoe-Maker levy impact into Jupiter), what would the "frames per second" of the captured imagery be, regardless of imaging resolution?

Video of the Shoemaker-Levy impact: https://youtu.be/p7RP2SW_gSw?t=92. Is this video presented in real time?

Animation of the Shoemaker-Levy impact:


Live Streaming Jupiter (from Earth):


Shoemaker-Levy image:


ISS Live Stream [for comparison]:


  • $\begingroup$ This is an interesting question! Answers to "what would the 'frames per second' of the captured imagery be" may be different than answers to "Frames per second capabilities of Modern Telescopes". I see that the Shoe-Maker levy impact is what got you interested (background of the question) but are you more interested in capabilities or of what the frame rate would likely be for this example? $\endgroup$
    – uhoh
    Commented May 10, 2020 at 5:57
  • 1
    $\begingroup$ The biggest factor is how much light you can get from the target on the sensor/film. You can make a very high frame rate with any telescope if you're happy with basically completely dark frames because the exposure was not long enough to capture useful data. $\endgroup$ Commented May 10, 2020 at 6:38
  • $\begingroup$ Especially modern earth-based telescopes don't have "that one sensor", but are equipped such that the sensor pack can be exchanged - depending on the needs in terms of observed wavelength and possibly also sensitivity / speed. $\endgroup$ Commented May 10, 2020 at 8:24
  • 1
    $\begingroup$ Rather than frames per second, a better metric in many cases is exposure time, which is measured in minutes or even hours per exposure. $\endgroup$ Commented May 10, 2020 at 17:24
  • $\begingroup$ Hope I clarified your title. $\endgroup$ Commented May 11, 2020 at 17:33

3 Answers 3


The types of sensors that are used for "Lucky Imaging" can work at frame rates easily capable of sampling changes in the speckle pattern due to turbulence in the atmosphere; 50 Hz is certainly possible.

The detectors used are electron-multiplying CCDs and are available on a number of telescopes around the world.

A video camera can of course be put at the focus of any ground-based telescope.

Such devices are not usually put on space telescopes because they have no need to deal with a turbulent atmosphere. Space telescopes are usually concerned with taking pictures of very faint objects, where minimising the readout noise of the device (to a few electrons) is a priority, and necessitates longer readout times - often tens of seconds.


As several people have said, the capabilities of the sensor are probably less important than the problem of getting enough light to have anything to record in a short exposure. We find that a magnitude zero star (one of the brightest) corresponds to a visible light flux of about $3.6\times 10^{−20} erg/(s·cm^2·Hz)$.

Visible light corresponds to a bandwidth of about $4\times 10^{14} Hz$ and a large ground based telescope might have a collecting area of about $50 m^2$ (or $5\times 10^5 cm^2$). So it intercepts about $6 erg/s$ of visible light from the star. Assume this is spread across a 2000x2000 pixel sensor, and each pixel and switch to SI units and this is $1.5\times 10^{-13} J/s/pixel$. This is actually a few $10^5$ of photons, so for a very bright source like this, quite a rapid frame rate would be possible, probably thousands of frames per second, and, rather to my surprise, the readout electronics might be the main problem, rather than light intensity.

I've made lots of assumptions here, about sensor sensitivity, number of pixels, and many other things, but it seems like, at least from Earth, with an 8m class telescope you could image a magnitude zero source like Saturn at a pretty high frame rate.

  • 2
    $\begingroup$ Modern CCD or CMOS are surprisingly sensitive. Once I even managed to take an image sequence with a dedicated high-speed camera (Photron SA-3, not made for astrophysics) and get a read-out of Aldebaran with like ~300 fps - sufficient to start getting the temporal diffraction pattern when it was being obstructed by the moon - and that was only a 50cm dish, not a bigger, professional one. $\endgroup$ Commented May 11, 2020 at 10:32

Asrto video and near live astro viewing cameras.


Mallincam forum See photo section for examples


NightSkiesNetwork. Members broadcasting their observing sessions


Frame per second are determined by the camera not the telescope or lens. The Mallincam video cameras are 60HZ interlaced.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .