user_mobilelogo

The Book

Book front cover

The ultimate resource:
A Field Guide To Time-Varying Light Sources

Learn more...

Purchase the book on CreateSpace!

Purchase the book on Amazon!

You might think that video is the perfect medium for showing time-varying lights. However, it turns out that this is almost certainly not true. The reason is that video is itself a time-varying medium, and capturing a time-varying light source with a time-varying device, and then displaying it on yet another time-varying device is fraught with the risk of misrepresenting the original light source.

In fact, you are much more likely to suffer from the side-effects of time-varying light sources on videography. Let me illustrate this with the two videos below. Both videos show a Korg M3 synthesizer (I would love to own one of these!). This synthesizer has red LED indicators which appear to the human eye to be steadily lit, but are in fact flashing rapidly. This is due to the design of the electronic circuit which powers the LEDs, and is a very common implementation: many products show the same behavior.

This first video was taken with a video camera that represents the LEDs as being always on. Look at the red indicators on the control panel.

(Embedded by permission: Pirkka Maksimainen)

 

This second video is very different. Go to 1:10 or later in the video, and you will see the LEDs appear to flash. You might think that these indicators would be unbearably distracting when playing the M3 synthesizer - but in fact the video camera is lying. To the human eye, these LEDs appear stable.

(Embedded by permission: K-Sounds)

 

Why is this happening, and why are the two videos so different?

A video camera operates by capturing frames, which are essentially photographs. It captures many frames per second, typically 25 or 30. Imagine that an LED is flashing at, say, 30 times per second. To the human eye, it will look like it's permanently on (unless you're a seasoned observer of time-varying light sources). But the camera is taking "photographs" at, say, 25 per second. It follows that it might capture a series of frames during "off" times of the LED, followed by a series of frames during "on" times. If this happens, the recorded video will show an apparently slow flashing rate, with several "on" frames followed by several "off" frames.

This is made even more uncertain by the exposure time of the video camera, which is the duration that shutter is open. A longer exposure time increases the likelihood that the LED came on at some point during the exposure, which would result in a frame where the LED is "on". Perhaps the first video was taking with a camera that has a very long exposure time relative to the frame rate.

The point of this illustration is to show that when videoing a time-varying light source, different video cameras can deliver very different results. This means that it's probably very hard to use a video camera to document a time-varying light source accurately. Not only that, but when the video is watched on your computer, TV or cellphone, that device's display is itself a time-varying light source and may confuse what you see even further!

Site design

This site is powered by Joomla! CMS. The template is "Cast" from Joomlage, modified here and there. The gallery is powered by RokGallery. The forum is powered by Kunena. All design by Dan Bennett.

Google+

Copyright

All content on this website, including images, videos, text and material from "A Field Guide To Time-Varying Light Sources" is © 2015 Daniel H. Bennett / timevaryinglights.com / A Bear Peering Round A Rock, and may not be reproduced without permission. All rights reserved to the extent of applicable law. Exceptions: third-party images, which are credited as applicable, third-party embedded videos and public domain images.