Differnce between 720p/24pn 60fps and 720p/60p [Archive] - DVXuser.com -- The online community for filmmaking

andy_starbuck

05-31-2006, 04:13 AM

720 refers to the height of the image in pixels. In this case the image is 720 lines of 1280 pixels each.

24p or 60p refers to how many frames per second are being recorded. The 60 is using a lot more memory a lot faster because it is actually recording that many more frames every second.

The 'p' in 24p or in 60p refers to 'progressive'. The 720 lines can be captured in two ways; interlaced or progressive. In progressive, the camera starts at the top and takes each line in sequence to the bottom of the frame. The look of a progressively captured frame is similar to film, because the entire frame is caught in an instant. In interlacing, the lines are numbered, and all the odd numbered lines form one field and the even numbered lines form another field. The camera captures all of the lines in one field before it captures all of the lines in the other field. This changes the physical requriements of capturing the image (less information more often), and also changes the look of the image. The interlaced image appears sharper, more like television.

The 'N' in 24pN stands for 'Native' mode. 24 frames per second is a film rate, not a video rate. In order to pass 24 frame per second recordings over equipment such as firewire cables that were intended for 30 frame per second video, the camera must actually add in 6 extra frames to bring it up to the standard 30 frames per second. There are several different ways it can create these frames, called "pulldown", but essentially the camera either repeats frames or repeats fields to top-up the number. In "Native" mode, the camera records only the real frames and doesn't add the extra ones. This saves space (6 frames out of every 30). And since the camera doesn't have to invent frames, it simplifies the process of storing the captured image. The HVX200 uses this mode for variable frame rate recording.

Now imagine if you have a file that is labeled as containing frames to be played back at 24 frames per second. But just because it will be played back at that speed doesn't mean you have to record them at that speed. You could record 20 frames every second into that 24fps file. When it is played back its played back at 24fps even though it was recorded at 20fps. Result: fast motion effect. (undercranking)

Likewise, you could go the other way and record at 26 frames per second into the file. When it is played back at 24fps things will be slightly slowed down: slow motion effect. (overcranking) The fewer frames per second you actually record, the faster the motion, and the more frames you record the slower the motion.

Recording at 60 frames per second and playing back at 24fps is about as slow as you can get.

A note about audio: Variable frame rate recording breaks the time relationship between what is captured and what is being played back. There is no equivalent for audio; the audio is a continuous stream that is normally mapped to the timing of the frames. Since the frames now don't represent 'real' time, there is no way to map the audio to them to be recorded. So when you record in variable frame rate, you lose the ability to record sound. Either go back and record the event a second time at normal speed to capture the sound, or record the sound using a second sound unit (or a second camera). Then you can use a sound editor in post to alter the sound and mix it for the slow or fast motion footage.

You can get similar fast motion or slow motion effects using Time Remapping to change the speed of footage in post. However, Time Remapping either drops or creates frames to accomplish the effect, which creates motion artifacts: it is not a perfectly smooth look. In-camera undercranking and overcranking records the actual event at the chosen speed, giving 'real' frames without artificially dropping or inventing them. The result is real-looking motion without artifacts. And especially, glass-smooth slow motion that looks like it was shot on film.

So that's the first part: 720p/24pN 60fps.

The second part you mentioned is 720p/60p.

Same size frame, 720 x 1280, recorded in the same way, progressive lines rather than in alternating fields. But this time the frames are recorded into a file marked for 60 frame per second playback. And the frames are being captured 60 times a second.

The number of times a second that a frame is recorded controls motion blur. At 24 frames per second, someone swinging their arms quickly while walking is a fast enough motion that the ends of their hands will move while the frame is being captured, resulting in blurring of their hands. At 60 frames per second, the same hand movement will be sharp and unblurred. So 60p is less film-like because it doesn't represent motion in the same way, but it provides a very high quality recording of the motion. The tradeoff: film-like motion blur and less memory consumed per second, or very little motion blur and a lot more memory consumed.

Hope that helps.