1

Many action cameras claim to record 4K UHD but only meet the spatial characteristics, i.e., 3840 x 2160, but not the temporal characteristics, i.e., minimum 23.976fps. Instead they record at 12fps (or 15fps) and playback at 1x speed but 24fps (or 30fps). My understanding is that this is accomplished by playing each frame twice and these added frames are not interpolated but rather are exact copies. In action cam forums this is called frame doubling. I have two questions.
1. Is it in fact correct that the added frames are duplicates, not interpolated?
2. If yes, is frame doubling the correct technical term?

EDIT: I should clarify that when I view these videos frame-by-frame it appears the frames are duplicates. However, I'm viewing 4K video on a 1080p monitor so I'm not sure if I'm missing something.

  • Different brands of cameras will probably handle this differently. Some may duplicate; others may interpolate. Did you have a question about a specific camera? – BrettFromLA Mar 8 '17 at 22:00
  • Thanks for responding. I didn't have a question about a specific camera. It's a general question about all the "fake 4K" cameras out there. So many are not 4K native resolution or frame rate. Can you tell me please if this is correct usage of the term "frame doubling" regardless of whether the extra frames are duplicated or interpolated? I even had a camera that claimed to shoot 720p at 240fps but it was actually 30fps at octuple frame rate! – Joe Mar 8 '17 at 23:31
  • I'm honestly not sure of the right definition of "frame doubling", but I understand what you mean when you're using it the way you do. I'm much more of a producer / director / practical effect guy. – BrettFromLA Mar 9 '17 at 18:40
  • Thanks BrettFromLA. I just like to be sure I'm using the terminology correctly. I hate to sound stupid (but often fail anyway!)! Cheers :-) – Joe Mar 10 '17 at 17:09
  • A similar question, and an answer: video.stackexchange.com/questions/23671/… – Rusty Core Sep 7 '18 at 3:02
1

The term "frame doubling" is occasionally used to describe the result of animators who "shoot on twos". Namely, when an animator draws only every other image for a given frame rate, the image is displayed twice before the next image is displayed. That "frame doubling" is an observable property of the animation, but it is not the key idea.

The key idea is that frame rates can be subdivided (shot on ones, twos, threes, or even fours) and the look and effect of each can be appropriate for a given instant. Generally, lead animators generate key frames and junior animators (or computers) generate the in-between frames. Shooting on ones requires generating twice as many in-between images as shooting on twos. But it also creates a smoothness that may or may not better serve the aesthetic of the animation. Nick Park's classic Wallace and Grommit animations look "right" because of frame doubling. Pixar's Toy Story films look "right" because they shoot on the ones (and integrate motion blur on every frame to look as if shot on a cinema camera).

I would argue that a camera that shoots at 1/15th of a second and then doubles those frames for a 30fps playback will create a highly stylized look that combines (in a "wrong" way) the motion blur one would expect from shooting on the ones with the pacing of shooting on the twos. As Blues music teaches, the wrong note at the right time could be the right note. But generally, animation shot on twos rarely integrates motion blur as such would be "too obvious". So if going for a frame doubling effect, one should use a small shutter angle, like 30 or 45 degrees, not 180 or 360, when shooting 15 fps.

0

Do you mean recording at a Shutter Speed slower than the Frame Rate:

?

Panasonic Cameras call it "Creative Movie Mode". This Blog has a few entries covering this technique: http://m43photo.blogspot.ca/2011/01/new-years-eve-fireworks-with-gh2-8mm.html and http://m43photo.blogspot.ca/2010/02/gh1-video-recording-at-slow-shutter.html .

If most of the Scene is static and any movement is slow (or OK if it is blurred) then holding open the Shutter longer than would be normal for the Frame Rate (IE: Less than 1/30 sec. for 30FPS) allows in more light.

A 30FPS Video with a 1/15th of a second Shutter lets in twice the light with two frames of blur on the moving objects. That would be "doubling" but it's not restricted to 50%.

  • Thanks for responding Bob. I did have a response to your comment but it's too long to post. Basically it would have said, "I think I'm getting confused!" I know that in action cameras each frame is made up of several exposures that are put together by the processor and made into a single frame. I wanted to post a much better explanation of this from a forum on the website DashCamTalk dashcamtalk.com/forum but it's too long to post and I can't link to specific comments there. Basically the issue is not what goes into producing a single frame but how they are played back. (cont'd) – Joe Mar 11 '17 at 2:23
  • When a manufacturer claims their action camera can record 4K at 30fps but actually records at 15fps, what are the other 15 frames that are played back every second? Is every other one an exact duplicate (what I've normally been calling frame doubling) or is it a combination of the preceding and following frame, i.e., interpolated, and would that still be considered frame doubling? Perhaps I'm splitting hairs or perhaps "frame doubling" is one of those imprecise terms that's used in more than one way. Perhaps you are in fact describing what should be properly considered frame doubling. – Joe Mar 11 '17 at 2:30
  • Rob, here's the post from Nigel on DashCamTalk set out in three comments: ". . . assuming it is night time the frame rate will be 30 fps (if you have it set to 60 fps then it actually records 30fps when it is dark to get a brighter exposure), the 30th of a second will contain something like one exposure of a 35th second, one of about a 350th and one of about a 3500th. That almost fills the 30th second and allows correct exposure for everything from the shadows to full headlights (although not quite HID headlights) (continued) – Joe Mar 11 '17 at 2:36
  • the 3 exposures are then combined and recorded while the exposures for the next frame are being collected. During daylight the same thing happens, even if using 60fps which is genuine 60fps during daylight." - AND - "I did say "something like" before my sequence, I don't know the exact exposure times or even number that are used and obviously it changes depending on lighting conditions. As far as I can tell by analysing the results, it does use 3 exposures at least most of the time, you can sometimes see the separate exposures when there is very fast motion somewhere in the image.(continued) – Joe Mar 11 '17 at 2:38
  • The longest exposure can't be slower than about 1/35th due to the frame rate and I doubt the fastest would be much faster than 1/5000th as I don't expect it is capable of reading the data out of the sensor faster than that and the data isn't buffered, it is read out a line at a time with the "rolling shutter". As I said, it doesn't quite cope with the brightness of HID headlamps so it obviously isn't very fast. (END POST) – Joe Mar 11 '17 at 2:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.