Why timelapse photography?

Messages
290
Edit My Images
No
Ok so i'm going to start by saying that this is a genuine question that i can't figure the answer to and i'd appreciate it if someone could educate me.

What is the point of timelapse photography? I mean wouldn't it be a lot easier to use a video camera or am i missing something here? Are video cameras that capture in that sort of detail really expensive? Is a video camera incapable of being manipulated like a DSLR? I've seen many awesome timelapse films but fail to understand why you would go through the bother of capturing XXXX amount of single shots and then go through the mundane ritual of putting them all together. I know that there is software that does this but still. On top of all that surely it hammers the shutter count on the camera?

I understand that it could be a cost issue but people that produce the best timlelapse videos must be selling them so why not make your life easier by using a HD digital video camera.

I also understand that people may in fact just like to produce these sort of videos.

Thanks people.
 
Not sure video cameras can be set to take single frames every x seconds, minutes or hours, whereas some cameras can be set to do so.
 
It's a way of compressing time. If you take one frame every minute for eight hours you will have 480 frames. When played at 30 frames a second that gives 8 hours of time squeezed into 16 seconds of viewing. You can of course take as many as you want at any time interval you choose to achieve different effects.
 
Last edited:
Thanks for your comments guys. The way i see it it must be continuous shooting so that the video is smooth. Why not just use a digital video camera and speed it up?
 
To get a whole days action or at least many hours would be an awful lot of video and it would need to be played at a terrific speed to compress to the much shorter time scale used in most of these.
 
Thanks for your comments guys. The way i see it it must be continuous shooting so that the video is smooth. Why not just use a digital video camera and speed it up?

That's the point its not.

Go back to the first 'moving pictures' and they were a series of stills, viewed in sequence very quickly, like a 'flick book'.

Movie Cameras are no different; they shoot 'stills' at very rapid intervals, about 30 or so a second; shown, rapidly, in sequence at the same frame rate, you see a moving picture.

VIDEO... is just the same... except it uses 'sampling'.

Using an electronic sensor, rather than a film, that has to be changed every frame, it can continually 'sense' the scene... but some-how that signal has to be put onto a capture medium...

TV... you have a SINGLE electron beam created by a big electro-magnet and a cathode emmitter... electro magnet drags electrons off the cathode and 'shoots' them in a stream. More magnets,m field coils then bend that stream to hit any spot on the screen in front, where, striking a bit of phosphorescent paint... the electrons will make it glow.

So; by controling the electric current to the catrhode, making more or less electrons, and then the voltage on the field coils, you can 'steer' the beam around the screen and light or not light points on it pretty much at will.

TV... the signal, recieved through the areil then controls the cathode voltage & beam direction coils; the beam moves accross the screen from left to right, in rows, at the end of each row moving down and doing one beneath.

To make the screen glow evenly, the beans sweeps every third row, so makes three passes up and down to make ONE full picture on the screen and used to do this at a frequency of fifty times a second.

However, continiousely sampling, yes, you would get in the bottom of the picture the bit of picture from the last sweep while the top is being painted with the next.

It is NOT continuiosely 'seeing' though, its sampling, at about 20 frames a second.....

Why dont they use a Video camera for time lapse?

Why do they use Stills SLR's for Video?

The technologies are very similar, especially these days with Digital SLRS and Digital Video cameras.

BUT Digital Stills cameras, tend to be easier to set up to do time-lapse type filming, and they have a much higher Image Quality for Stills.

Take an M-Peg and it ISN'T like a series of movie frames; there is a lot of process compression in it, camera only records perhaps one frame in thirty, and short-hands big changes to that still image to 'best guess' frames between, to make a motion picture. And can get away with it, as they are never on screen long enough for any-one to notice.

If you let a video camera run for say 24 hours, and then tried to speed it up, it would speed up the film SIMPLY by 'dropping' frames, and not showing them to you, NOT showing them all, but faster.....

So.. your stills camera, taking say 1s interval stills, does the same thing.... a lot more efficiently... not capturing 'best guess' intermediate images, that are probably going to get thrown away.... and in fact, chances are you are as likely to throw away all or most of the reference images and ONLY show best guess compression interpolated ones....

So

the quality is 'better'.
Its more efficient
Its easier
 
The resulting files will be Smaller and more manageable too, rather than dealing with a whole video file. Also remember that time lapse can be over weeks, months or years. There wouldn't be enough storage to adequately store a continuous video file that covered more than a day or two's worth of work.
 
Thanks for your comments guys. The way i see it it must be continuous shooting so that the video is smooth. Why not just use a digital video camera and speed it up?

Firstly, file size.
8 hours of digital video will be massive. Huge.

Second, quality.
My 5D3 has a 3840 pixel vertical resolution. A Canon 550D has a 3400 pixel vertical resolution. Those are well above HD res...

Third. Quality (again)
Frankly, a DSLR allows such good depth of field control, and has excellent low light peformance. Much better than a video camera. There is a good reason why DSLR's are being used for a lot of movies these days...
 
8 hours of digital video will be massive. Huge.
Standard Video, single density, 4.5Gb DVD is 2Hrs at HQ, 4Hrs at LQ... so about 9Gb

At Reduced frame rate & resolution, with higher compression, typical compact camera's 'video clip' m-peg, 600x480 or so? Might be as little as 2 or 3Gb.

Uncompressed, edit quality AVI? About 100Mb per minute.... 6Gb per hour, depending on frame size.

If you shot j-peg with a DSLR in time-lapse, you'd be looking at something in the order of 1-10Mb per frame, 24 frames per second? 24-240Mb per second, or about between 85 & 850Gb per hour of screen-time.

If you were going to make a time lapse film, of I don't know, rose buds opening up... takes maybe four hours; and you wanted to speed that up to screen over 4-minutes....

Your viewing frame rate would be 24 frames per second.... 1440 per minute, 5760 for clip. (One frame every two and a half seconds of 'real time')

On my camera, I get 100 J-Peg stills or 5min 1080HD M-Peg video per Gb.

So to make my 4-minute time-lapse movie, I would need, 60Gb of stills, or 48Gb for the full four-hours of 'capture' video.

Ie, it would actually take LESS memory to capture in Video than in Stills....

But would result in higher quality video, and it would almost certainly be less painstaking to piece together to make a video from stills.

Your four hours of raw, real time footage would contain aprox 345,000 frames... and you would have to go through them and snip out 60 frames for every one you kept, nearly 6000 times!

The shorter the screen-time you compress down to, the fewer stills you would need. The longer the real-time duration; the more real time video you would need.

So, the more 'real-time' to 'screen-time' time lapse compression you do; the more bytes of real-time video you would need to capture compared to the bytes of j-peg stills to make the same movie; so the longer you time lapse over, so the more manageable using DSLR stills becomes.
 
Uncompressed, edit quality AVI? About 100Mb per minute.... 6Gb per hour, depending on frame size.

Depends on the codec used. Most companies DV codecs use 13Gb/hour

if we are talking true uncompressed video the figures are much higher.
625 pal uncompressed is
8 bit @ 720 x 576 @ 25fps = 20 MB per/sec, or 70 GB per/hr.
10 bit @ 720 x 576 @ 25fps = 26 MB per/sec, or 93 GB per/hr.

move that upto HD and 1080 is around
8 bit @ 1920 x 1080 @ 25fps = 99 MB per/sec, or 348 GB per/hr.
10 bit @ 1920 x 1080 @ 25fps = 132 MB per/sec, or 463 GB per/hr.



Editing this takes some serious storage and generally some fast computing time to handle, especially when its rendered down.
 
Last edited:
Your four hours of raw, real time footage would contain aprox 345,000 frames... and you would have to go through them and snip out 60 frames for every one you kept, nearly 6000 times!

But the point is you take one picture every minute, hour or whatever, you do not need to take 24 fps.

For example for a 24 hour time lapse movie of a day, if you took 1 picture per minute 60*24 = 1488 pictures. You then join them into a movie and you would get a time-lapse movie that plays for 1 minute but shows 24 hours.
 
But the point is you take one picture every minute, hour or whatever, you do not need to take 24 fps..
24 frames per second VIELWING rate...

You want 4 minutes of video, at 24 frames per second you would have to take aprox 6000 pictures to make that length of movie, over whatever time you want to lapse.
 
But would result in higher quality video, and it would almost certainly be less painstaking to piece together to make a video from stills.

Your four hours of raw, real time footage would contain aprox 345,000 frames... and you would have to go through them and snip out 60 frames for every one you kept, nearly 6000 times!

.

A lot of what you say makes sense. I can't speak for the video file sizes but one thing I can say is on your camera, you cannot record more than 20 mins of video in one go... makes 8 hours trickier without little gaps :)

Not sure how you got to stating recording video direct would result in higer quality video. This makes no sense when the vertical resolution is ~ 1/3 of the vertical resolution delivered by shooting stills. If you reduce your still quality down to the same vertical as HD then you significantly reduce the stills file sizes too.

The second point, I assume you are referring to video?
 
Thanks for your comments guys. The way i see it it must be continuous shooting so that the video is smooth. Why not just use a digital video camera and speed it up?

Time lapse photography is normally used on things that happen very slowly. Like a plant growing for instance. Or a spider making a web. When you play it back at "movie speeds" of say 24 images per second you can actually see the plant grow etc and it looks like it was shot on video and looks smooth like normal video.
 
Ok so all of the above makes sense and answers a lot of the questions that i originally asked. Here's the video that aroused my curiosity in the first place. For the people in the know, how long was the gap between each shot do you reckon?

http://vimeo.com/62980495
 
Probably over a few hours, image every 10 seconds or so, perhaps longer.

On a seperate note that videos opening scenes and text lay outs are a rip of off dustin farrels work.
 
For most, I would say it is more a style of shooting and processing over anything else.
 
Last edited:
Back
Top