Field dominance question/best workflow?

[Tom Brooks]

Hey Paul,
Your material should all be compatible with a normal, lower-first, interlaced sequence with a frame rate of 29.97. You should not be using a sequence with a 23.98 frame rate.

Perhaps your problem has more to do with how you are monitoring your sequence. Do you have an analog I/O device so that you can ultimately view your rendered sequence on an interlaced TV? In a case like this, with mixed footage and TV being your desired end point, it’s very important. Viewing only in the Canvas will not show you everything you need to see in the final product.
-Tom

[Paul Campbell]

Ah, good morning! Nice to be back. Hi, Tom. I do not have an I/O device w/ a monitor. I know how vitally important it is to use this, and I’m working on getting my rig setup to have this as soon as I can afford it. I do some side work for a friend who does have a nice broadcast monitor, and the problems I’m describing here do indeed show up on the monitor.

Regarding why I worked with the 23.98 frame rate, my plan was to nest my downconverted HD footage into my SD timeline. However, when I tried this the first time FCP gave me an error about not being able to nest different frame rates. So (and I know I’m screwing this up totally, but I feel like I’m getting close) I started up a new sequence, dropped my 23.98 HD in it (and answered yes to the question about setting the sequence to the clip), did a quick edit with about 10 seconds worth of stuff, and then nested this as an NTSC clip. I then brought that nested clip into my other sequence with all my SD graphics.

As I mentioned, I thought I nailed it. Both the SD and HD stuff looked killer, that is until I tried putting a cross dissolve on the nested HD clip. The point where the cross dissolve started is where the HD clip got blurry again. I just can’t seem to get there from here. (Maybe I was just getting tired last night 🙂

[Paul Campbell]

Ok, I’m going nuts here. I’m not able to bring my HD footage into my SD timeline and have the HD stuff still look good. I’ve looked for anything out there that might explain what I’m doing wrong. Man, this is driving me to drink. If anyone’s reading this, can I recap my scenario here? I’ve got:

-1080p footage from a Sony XDCAM camera
(the FCP format info on this footage tells me that the frame rate is 23.98, the Compressor is
XDCAM EX1080p24 @ 35 Mb/s VBR, field dominance is “none” and the pixel aspect ratio is
square. I’m not sure how necessary it was for me to list all of that, but I’m losing my mind)

-720×486 animated graphic clips
(the FCP format info on these clips is frame rate of 29.97, the Compressor is Animation,
field dominance is Lower, and pixel aspect ratio is NTSC-CCIR 601)

-720×486 pre-produced commercial
(info on the commercial is frame rate of 29.97, Compressor is ProRes 422, field dominance
is lower, and pixel aspect ratio is NTSC-CCIR 601)

-720×480 music videos
(info on the videos is frame rate of 29.97, Compressor is ProRes 422HQ, field dominance is
lower, and pixel aspect ratio is NTSC-CCIR 601)

I don’t have an external broadcast monitor, just the Canvas. I know how you guys feel about this monitor issue, but the problem I’m having here goes beyond not having an external monitor. My troubles here aren’t with color shifts or contrast, etc. My HD footage in my SD timeline just looks horrendous and blurry. If I had an external monitor, it would look horrendous and blurry on that. I’m being told that the HD stuff should look good in my SD timeline, but it just doesn’t. I’ve tried every combination of settings I can think of. If you’ve had the stomach to read this entire thread, you know the problems I’m having with nesting. It’s almost like I have two choices here: use a no-field sequence and have my HD stuff look good with everything else looking blurry, or use a lower-field sequence and go with bad HD/everthing else nice. I can’t get these guys to place nice in the same sequence. I feel like I’ve adopted two dogs from different SPCA’s and they won’t get along in my house.

Please accept my apologies for beating this to death. I’m trying really hard to figure this out, but even Shazaam needed to consult with the elders from time to time, right? (Too dated?)

[Rafael Amador]

Hi Paul,
It seems that yesterday I wasn’t much awake either.
I didn’t realize you were working with two different time bases.
I agree with Tom, I think this is the problem.
Not the size, pixels or field order of the HD, but the time base.
You need to convert the EX-1 stuff to 29.97 and I think everything will work. FC can cope with different formats (codec, size, pixels,..) in the same time line, but no with different time bases.
Try converting your HD sequence with Compressor or Nattress Standard converter.
Just to add, If you have not an external monitor, at least try to set your Canvas window to 100%.
Cheers,
Rafa el

http://www.nagavideo.com

[Tom Brooks]

Hey Paul,
I’m not familiar with the XDCAM workflow, so I can’t give you much advice on the specifics. In general, what you need is to either not remove pulldown from the original footage when it’s captured or re-add pulldown so that you end up with 29.97 footage that can be more easily combined with your SD material. Nattress has a plug-in called “G Add Pulldown” that might work for you in part. Perhaps the XDCAM stuff can be conformed to 29.97 in Compressor and then, using the Nattress plug-in, have the pulldown re-added. Re:Vision FieldsKit also does this. It’s going to take some learning on your part to get it right.

Or, another strategy would be to work with a post house that can convert the HD stuff properly for you.

I meant to also mention to be sure your sequence is fully rendered–not just previewing in Unlimited RT.
Unlimited RT will give you blurry playback with some settings (check manual).
-Tom

[Paul Campbell]

Gents, thanks so much for the info. I’m sorry for not making that framerate difference more obvious from the start. I thought I was providing enough, but I clearly left the most important thing out.

I’m in the process now of trying to get Compressor to convert my HD stuff up to 29.97. No luck yet, but I only started 5 minutes ago. I’m quite encouraged by this latest advice, so I’ll be plugging away at this for a while no doubt. I wish I’d checked for your replies this morning instead of now! (Of course, my garage would still be a mess if I did)

Perhaps I’ll ask my friend who’s shooting the footage if he’d consider shooting at 29.97 instead of 23.98. That’s certainly an option!

[Paul Campbell]

Hey, something just came to me here. If this whole thing here is simply a timebase issue, why am I able to drop 24 frame footage into a 30 frame timeline and have it look good once I set the sequence field dominance to “none”? Shouldn’t the 24 frame stuff look weird?