Oh boy... here we go... if you shoot in 2:3 24P you should edit in 29.97... if you shoot in 2:3:3:2 you should CAPTURE removing pulldown and edit in 23.98...

These settings will make your video look the best, many people break these rules and IMHO that is what causes many of the problems people talk about (interlacing, jittery moves, etc.)


ash =o)

damn Ash, i need you on speed-dial...And by the way. Do you ever go on AIM cause you've been on my BL for awile and you've never been on...

Perhaps we should have a special permanent thread on acquisition and capture of 24p concerning each camera with each NLE? This issue seems to warrant such a thing, as I myself have asked similar questions along with countless others. Just an idea.

I need to add you to my AIM list... There is lots of confusion over the 24P workflow, I personally find it best to edit and shoot in the format you plan on delivering. This will cause fewer problems with conversions and fewer problems in general.



ash =o)

DV only records one format: 60i. When you shoot 24p with 2:3:3:2 pulldown the fields written are Aa Bb bC Cc Dd Ee, Ff, Fg, Gg, Hh ... with the large letter signifying one field of an image and the small letter the other. If you tell the NLE that you have shot 24p it will take the first pair of fields and put them together to make the full image frame Aa. It will do the same with the next 2 fields Bb. The third pair of fields, bC, it will throw away. It then keeps Cc, Dd, Ee and so on. Thus it reproduces the 24 frames from the sensor but throws away the mixed frames. If you look at the dv sequence on a progressive (computer) monitor you will see a split frame on the 3rd, 8th, 13th... frames.

In 2:3 pulldown the cadence is Aa Bb BC cD dD Ee Ff FG... Again Aa, Bb,Cc can be reconstructed from the individual fields but if viewed on a 60i TV you will now see split frames in nos 3 & 4, 7 & 8 etc. i.e. more frequently. So if you are going to go to film then use 24p without fear of these funny looking split frames. If you will be going back to dv you'll have the funny frames to deal with if viewing on a progressive monitor but fewer in 2:3:3:2 than 2:3.

Ok, ive run into a problem...I've been Editing on the 23.976 for the past day and already completed alot of the video. I then read all the posts and converted it back to 29.97. Is that going to be a problem. And another thing, I noticed on some of the parts of the video, alot of the Reds are pixels. Big ones at that. Im working on 3 monitors and it looks the same on all 3 monitors. I also rendered the video as 24p and rendered it as 29.97. Could it be the graphics card? Its not my video i dont think because i tweaked the video via in camera to get it a little undersaturated and then in post i made it a little more undersaturated. Im not really sure what i could be doing wrong here though. I captured my video the normal way if that matters either. I shot using 24p and the normal pull down setting, also its 16:9. Im going to the studio now and ill grab some frames and should be posting them sometime tonight.

Now I have been using adobe premiere pro and an Xl2 for nearly a year now. I know all about the 24p stuff.

However, I ran into a snag that maybe Ash can help with. I have been shooting in 24p 2:3 (not advanced). However, I recently switched over to adobe premiere pro 1.5 and even when I don't start a 24p project, and the timeline is 60i, it still reads the 2:3 footage as 23.976 fps.

Odd?

Matt

Do you have all the patches/upgrades? It sees my stuff as 29.97... maybe ask in the Premiere forum...



ash =o)

Those reds are normal... that is a DV issue, not an XL2 issue. The XL2 is actually better than the DVX with reds. You can add some chroma blur to help cover it. It has something to do with interlace/deinterlace and the red channel. Check the NLE forums, as good as I am with the camera... I am just a hack editor =o)



ash =o)

deLange has nailed it, but here's some more...

The DVX100 and the XL2 both have progressive CCDs and can both record at 24 fps. Both cameras save this TRUE 24p to the MiniDV format at 60i by inserting pulldown. If you record to 2:3, then you will have to edit in a 60i timeline. IMHO, you'd be WAY BETTER off recording in advance mode (2332) so that your NLE can strip the 60i back down to true 24p, then you can edit in 24p in the timeline. From there, you have a bunch of options for rednering. Render out to 2-3 pulldown 60i if you need NTSC, or render out to 24p mpeg2 for DVD burn, or encode out to WMV or MOV or MP4 at 24p if your delivery is via the web.

Basically, by using 2332 pulldown during acquisition, you can strip it back to 24p from 60i, and after that you can always render out back to 23 pulldown if needed later. Where with 23 pulldown, it's harder to get the true 24p back (you can still, but it's more of a pain in the arse), and usually you're stuck editing in 60i with weird bc frames and all. This keeps everything very clean through the entire workflow, and will also help keep any fx and titling you may be doing (you can do the credits/titles in 24p this way) very tight with no flicker and whatnot.

If you have any designs on editing your footage, I see no reason not to shoot with advance 2332 mode if you are going to shoot any 24p at all. And I think I would still say that even if some of the footage will end up mixed into a 60i timeline with other 60i footage, however, at times that might be another case where just shooting 23 might be warranted.

2332 is where it is at for sure if you are planning on editing...

Here goes the debate... like I said, I dont claim to be a great editor but 24P shot and edited natively in DV world is not as pleasing to the viewer. It tends to get more choppy and remember in almost ALL cases you are going to be showing it on an interlaced TV at 60i. Those extra frames are going to be there no matter what and I have found in DV world that it is best to do it in camera. The Varicam stuff I do, is all shot and edited in 24P and I think it looks great when output to 29.96 60i but that is a much different and more expensive format. I cant post a bunch of numbers to support my opinion but I have shot more DV than anyone I know and these are my observations...



ash =o)

This thread is exactly what I needed for a studdering issue I am seeing on DVD. I shot the Iguazu Falls in South America in 24p (2:3) 16:9 and decided against 60i in favor of the film look.
All looks well in FCP5 on a 29.97 timeline and I do notice the interlacing every 5th frame or so and there is no studdering when played on the timeline. when I output to DVD using DVD Studio Pro I notice on some shots a studdering like every 5th of a second or so. When I moved the footage to a 23.98 timeline, all gets way jerky, so I deleated that sequence.

I captured the footage as I would 60i footage with settings for anamorphic (no pulldown etc.) I need some direct answers/techniques for this issue....do I record with 2:3 or 2:3:3:2 ?? I am not going to film (YET) and I want the progressive look. I am outputting to DVD only at this point. Maybe I should stick with 60i for nature footage from now on (but it looks awesome in my camera and timeline.) This whole scenario is very confusing for me. Where am I going wrong? Maybe it's a setting in DVD Studio Pro. Thanks, Larry

If you shot it in 2:3 then I would leave everything in a 60i 29.97fps timeline. Do not author it as a progressive DVD...

Incidentally, 30p looks GREAT for DVD or streaming internet stuff... it CAN be an issue going to film and IS an issue going to PAL...



ash =o)

Shooting 24p or 24pa and motion graphics
 

Hi guys I'm new here, I just recently purchased a canon xl2.

If this has been covered before, I apologize. The search function isn't finding any of my key words.

I am about to shoot my first 24p project and I am going to have to decide weather to shoot 24p or 24pa. The output media is DVD.

If I play it safe and shoot 24p, how will my 30p motion graphics look overlaid on the 24p with 3:2 pull down? Will it look strange?

Is there a way to add 3:2 pull down to 24p renders coming out of after effects?

Should I just go ahead and shoot 24pa, edit in 24p timeline and do all my graphics in 24p then out-put to a 24p DVD? This sounds the best, but I have never done it and this is a one time event (kind of. It would be not as ideal to do it some other time), and I don't want to end up with footage that’s jerky, strobey etc...


The shoot is this Saturday, please some one help me make sense of all this for me.

I create my motion graphics files in aftereffects at 24 fps to import into a finacut 24pa timeline. You get frame-for-frame accuracy that way and it looks spectacular. Definitely try to keep all of the sources the same fps if you can.