What is the Workflow to convert 1080i 24p to true progressive??

C

chrismartin

Member
I've shot some footage at 1080i 24p and I've read I can convert that to true 24fps 1080p in Post.

I've tried a number of different ways to convert what I shot but I am still seeing interlacing in moving objects when I'm done. Reverse telecine didn't work for me, etc.

I know how to log and transfer, use compressor, use FCP etc. I just don't understand how to do the the conversion from the HPX 170 to true progressive frames on the computer.

I don't need help using the programs - I'm ok there - But if someone could give me a step by step of the conversion process from log and transfer to at clean 1080 24p that would really help me out.

When I log and transfer the files into FCP 7 the vid rate is 29.97. I'm not sure if that is how they are supposed to be or if that is part of my problem. I shot at 1080i 24p... why would they be 29.97?


Thanks for any help!
 
David thanks for the link!

That is the best description I've read so far on what 1080i 24p (and pA) are doing. I think you cleared up a lot of confusion I've had about how the i and p work together. But to make sure I really understand what you were saying, do I have it right when put it this way:

When shooting 1080i 24pA: The camera is initially grabbing 24 moments of time within every second the camera is recording. Each moment grabbed is 1/48ths of a second (assuming you have set the shutter speed to that). But then each of the 24 moments of time from one of those seconds is then split up and saved 60 times. Moment 1 split and saved in 2 fields. Moment 2 is split and saved into 2 fields(and the first half saved again for 3 fields). Repeat.

Then in FCP you basically put it all back together. So you do get a very true 24p.

If I have that all correct. Is this then also true:

The end result of shooting when shooting, on a HPX170, 720 24p and 1080i 24pA is exactly the same. Except in the later you have captured a higher resolution. The only real difference is the way you got there. With 720 24p if was a nice straight path where with 1080i 24pA is was a bit of a round-about. But with both you have nice clean true progressive images (unless my question #2 below has some merit).

If I have that correct also. Then I have a couple follow up questions:

1) FCP can do the reverse Advance pull down. Can anything else do it? Compressor? Sorrenson Squeeze, etc? Or do you have to generate your 24p from FCP? I have DPX files (from 16mm) that I will be using in Da Vinci Resolve and might want to convert the 1080i 24pA to that format. Any thoughts on the best route to get there?

2) I'm getting into the weeds of 24pA now......But this question keep nagging at me: When you shoot 24fps with a movie camera, you are capturing light for 1/48th of a second and then the film is advancing in darkness for 1/48th of a second. The time period when the film is advancing and no light is hitting the film is very even. But with Advanced pulldown. You are capturing light at 1/48th of a second. But what about the other 1/48th of a second? Since you are writing/saving frame 1 in 2 fields, then frame 2 in 3 fields, and so on..... Is the time period between the moments light being captured still an even 1/48th of a second? It seems to me that to save 2 fields of date then 3, then 2, then 3, etc., might create a slight stuttering effect.... Is that a crazy question? I'm i thinking too mechanically?

Thanks for the help!
 
chrismartin said:
When shooting 1080i 24pA: The camera is initially grabbing 24 moments of time within every second the camera is recording.
Click to expand...

Yes.

Each moment grabbed is 1/48ths of a second (assuming you have set the shutter speed to that).
Click to expand...

Yes, though it's not important for this. All of this is true of any shutter speed.

But then each of the 24 moments of time from one of those seconds is then split up and saved 60 times. Moment 1 split and saved in 2 fields. Moment 2 is split and saved into 2 fields(and the first half saved again for 3 fields). Repeat.
Click to expand...

Yes. I wouldn't put it exactly that way myself, but I believe you understand it, yes.

Then in FCP you basically put it all back together. So you do get a very true 24p.
Click to expand...

Yes, though in the pattern you describe (2:3), FCP can't do it automatically. You need the 24pA pattern (2:3:3:2) for FCP to be able to do it. Otherwise, you need to do it manually in Cinema Tools.

(Note: this is not true of other NLEs; others can handle removing either pulldown pattern.)


If I have that all correct. Is this then also true:

The end result of shooting when shooting, on a HPX170, 720 24p and 1080i 24pA is exactly the same. Except in the later you have captured a higher resolution. The only real difference is the way you got there. With 720 24p if was a nice straight path where with 1080i 24pA is was a bit of a round-about. But with both you have nice clean true progressive images (unless my question #2 below has some merit).
Click to expand...

Yes, however in 720p, you have the option of shooting 24pN, which will not introduce any pulldown and will record straight 24p. Nothing needs to be removed; the files are ready to go.



1) FCP can do the reverse Advance pull down. Can anything else do it? Compressor? Sorrenson Squeeze, etc?
Click to expand...

I know at least SOME other software will, but I'd have to review it all again to see what's what. After Effects will.


Or do you have to generate your 24p from FCP? I have DPX files (from 16mm) that I will be using in Da Vinci Resolve and might want to convert the 1080i 24pA to that format. Any thoughts on the best route to get there?
Click to expand...

The pulldown removal process happens during Log & Transfer, so as long as you did it properly, the MOV files you create will be 23.976. You should be able to use them in anything, not just FCP.


2) I'm getting into the weeds of 24pA now......But this question keep nagging at me: When you shoot 24fps with a movie camera, you are capturing light for 1/48th of a second and then the film is advancing in darkness for 1/48th of a second. The time period when the film is advancing and no light is hitting the film is very even. But with Advanced pulldown. You are capturing light at 1/48th of a second. But what about the other 1/48th of a second? Since you are writing/saving frame 1 in 2 fields, then frame 2 in 3 fields, and so on..... Is the time period between the moments light being captured still an even 1/48th of a second? It seems to me that to save 2 fields of date then 3, then 2, then 3, etc., might create a slight stuttering effect.... Is that a crazy question? I'm i thinking too mechanically?
Click to expand...

Again, shutter speed has no effect on pulldown. When the whole frames are split into fields, it's not a temporal split -- it doesn't create the first field from a 1/48th-of-a-second period and then the next field from the next. If it did that, it wouldn't even be 24 fps, it would be 48. Everything about the footage would be different.
 
Thank you very much David for taking the time to explain this all in such detail!

I have understood how movie cameras work for some time and I have understood the basics of interlacing for some time.

I have read a lot on the internet on the subject but until I read your explanation of 1080i 24p it never really understood that some cameras capture progressive and save interlaced. I must have read it many times, but I really needed it spelled out because whenever I saw "1080i" I just didn't see any kind of capture possible other than capturinng in feilds first.

I'm sure with this breakthrough other areas of digital filmmaking I still find confusing might start to make a bit more sense.

Thanks again!!!! :thumbsup:
 
Last edited:
You must log in or register to reply here.
Back
Top