First time using 24pa

bus33

Member
I shot some footage in 24pa and one tape looked great while the other looked slightly pixelated around the edges of the people, windows, etc. Is there one specific way to upload footage where it all looks good? I've tried to re-upload but got the same results.
I uploaded advanced 2:3:3:2 pulldown removal.

thanks
 
First, a bit of terminology -- the process of transferring from tape to computer is called "capturing." :)

As for the pixelating, would need to see examples to know for sure what's going on. You can use the Member Upload Center (the red up-arrow in the typing box) to upload and display pics, like screen captures.

Which NLE are you using?
 
here's a two pics from the vid
1259947109.jpg

1259947185.jpg

also im using fcp 6
 
Looks like THIN Vertical Detail Frequency. If all of your footage doesn't have this kind of thing, did you change Scene File settings between any shots?
 
i left it on scene file 6 the entire time. would it be better to capture clips or entire tapes? i usually do whole tapes. will these lines go away when i de interlace the video while exporting.
 
Last edited:
You don't deinterlace. You remove pulldown. It's an important distinction; they're completely different things. You never want to deinterlace.

Make sure you're following NoahK's sticky at the top of the FCP section, for "digitizing" 24pA in FCP.

The stairstepping you're seeing doesn't have anything to do with interlacing. It's the level of vertical detail the camera is set to; it can produce that effect on diagonals.

What's your intended purpose for this footage?
 
it's a wedding video. most all of the other footage came out fine. just this one tape had problems. can it be fixed in fcp?
 
You don't deinterlace. You remove pulldown. It's an important distinction; they're completely different things. You never want to deinterlace.
IVTC is a form of deinterlacing, because the original video is interlaced, and must be converted into progressive. It is also known as "film mode deinterlacing" as opposed to "video mode deinterlacing", which you would call simply deinterlacing.

The stairstepping you're seeing doesn't have anything to do with interlacing. It's the level of vertical detail the camera is set to; it can produce that effect on diagonals.
If he deinterlaced in video mode, by throwing one field away or by converting to 60p, this is what he would see, doubled lines in fields-became-frames.

bus33, you need to perform film-mode deinterlacing also known as inverce telecine (IVTC) or reverse telecine or pulldown removal.

David, granted I don't own a DVX, but from what I've read in the op instructions, THIN/PROG simply does not blur fields, while THICK/INTRLCE blur fields to prevent interline twitter when watched on an interlaced TV. I don't see how stairstepping is relevant to non-blurred fine details. If any, details should be fine and sharp, no visible blocking. IMHO, what we see here is classic line doubling.
 
im still new at all this and dont understand about the IVCT or reverse telecine. how do i do that?
 
IVTC is a form of deinterlacing, because the original video is interlaced, and must be converted into progressive. It is also known as "film mode deinterlacing" as opposed to "video mode deinterlacing", which you would call simply deinterlacing.

It's not the same thing. When people refer to "deinterlacing," they're referring to taking inherently interlaced footage and deinterlacing it, i.e., converting it to progressive. This is done through various methods, including "deinterlace filters," which do not put the 24p footage through an inverse telecine process, but instead blend or interpolate frame as if they were always interlaced -- thus sacrificing resolution and, especially in the case of 24p-over-60, such as the DVX records, making a mess of the footage.

Pulldown removal, in contrast, simply reverses the process of telecine and restores the original progressive frames. It does not interpolate. It does not "blend" fields from different moments in time into "frames," it restitches frames which were originally whole back to whole.

While in a general sense, it could be thought of "deinterlacing," it's only in a general, blanket sense, and to call it that is confusing the two concepts, and can be especially confusing to a beginner.

Thus, the simplest way to keep a beginner from making the mistake of applying a general deinterlace process to progressive footage -- which you should never do -- is to phrase it exactly as I did. Learning and understanding the difference will come with experience.

If he deinterlaced in video mode, by throwing one field away or by converting to 60p, this is what he would see, doubled lines in fields-became-frames.
Well, for one thing, no deinterlace filter on a DV timeline would ever convert the footage to 60p. There is no DV 60p.

For another, deinterlacing footage would not cause those kinds of specific artifacts on specific angles without causing it on others. The artifacting you would see would be ghosting, blurring, or staggering in the motion cadence. Any competent deinterlacer would not simply double lines and stack them, in the manner you describe, specifically for that reason.


bus33, you need to perform film-mode deinterlacing also known as inverce telecine (IVTC) or reverse telecine or pulldown removal.
Thus, as mentioned above, this sentence can be horribly confusing to someone who just got into the world of 24p video.

What I said was correct. Remove pulldown, which is easy to do with 24pA footage. Do NOT apply a deinterlace filter, or do anything other than remove pulldown.

David, granted I don't own a DVX, but from what I've read in the op instructions, THIN/PROG simply does not blur fields, while THICK/INTRLCE blur fields to prevent interline twitter when watched on an interlaced TV. I don't see how stairstepping is relevant to non-blurred fine details. If any, details should be fine and sharp, no visible blocking. IMHO, what we see here is classic line doubling.
Experience with DVX footage would help you here; when you see this kind of stairstepping, it's usually associated with THIN mode, and you see it specifically because the lines are not blurred. It most often appears in things like window blinds or staircases. (Which is not to say it usually happens with THIN mode, only that when it does, it's usually with THIN mode.)
 
im still new at all this and dont understand about the IVCT or reverse telecine. how do i do that?

Don't worry about it just yet; it's a couple of steps ahead of where you need to be at the moment. For now, just follow the directions for 24pA capture, as I mentioned above. It's here:

http://www.dvxuser.com/V6/showthread.php?t=23355

Inverse telecine wouldn't even really apply here, because traditional telecine involves inserting a 2:3 pulldown consistent with 24p standard, not 24pA. For now, just make sure to shoot 24pA and follow those directions.

Later, as you become more comfortable with 24p footage and want to learn more about it, you can explore the traditional means of transferring 24 fps footage to video -- telecine -- and learn about reversing it.

But 24pA was designed specifically to removed easily, much more easily than traditional telecine, which is why it's there.
 
it's a wedding video. most all of the other footage came out fine. just this one tape had problems. can it be fixed in fcp?

Well, what I mean is, do you plan to put this on DVD, or deliver to the web, or what?

You may find that there's no problem once you get it on DVD and play it on a TV. But a 1-pixel vertical blur should take care of it if it does.

Because of your cropping of the screen grabs, I can't immediately tell if the footage is Squeeze footage or not. Is it? Is it 4:3 that you plan to crop and convert to widescreen?

If you're going to crop, then you should have no problems at all, and the extra-fine definition afforded by THIN will help you in the process.

But you know, heck -- try recapturing, making sure you're following the steps correctly, and see if it helps, just to make sure the pulldown removal process got started on the right cadence, etc. Maybe it'll straighten itself out.
 
So if someone is trying to get a video on the web, is using THIN mode a bad idea? I heard it's always better to use thin vs thick. Can someone clarify this?
 
No, THIN is what you want; it gives you the most detail. It's rarely a problem, and will most often disappear after rendering, especially if you resize the image.

If it persists, like I said, a tiny vertical blur will take care of it.
 
im exporting to dvd. its 4:3 not squeeze. where do i change from 23.98 to 29.97 for capturing? i cant find it.
 
Last edited:
Back
Top