Converting a 24x session to 29.97

[c-post]

Are there any issues with converting a session built for 24p (24 frame
picture on my Protools screen to a session at 29.97 chasing NTSC video?
I'm going to find out in the next few days...

-John

[Paul Greyson]

As far as I know, it should be fine. Just change the TCF to 29.97ND and run at pull-down.

[c-post]

Thanks, Paul, will try it.

-John

[c-post]

29.97 non drop with pulldown was close but no cigar. Audio seems to drift about 4 frames per eight minutes. I have tried every kind of alternative, but the 29.97 with pull down seemed to be the best, but still not there.

-John

[cmaynes]

How are you running 24 frame video in PT?

Paul's method has worked for me in the past, but I was using an unsupported method. (Digitizing at non-pulldown rate into a Miro card from the film composer). It netted some funny fields, but the sync was good.

Once you are in NTSC land, the digitized video will like be unusable.

charles maynes

[c-post]

Oh boy, now I'm having some more serious problems with the 24p to 29.97 conversion and the client and I have started mixing together and I'm in trouble. OK here goes.

The client gave me a 29.97 videotape of his 24p project. After digitizing and importing into Protools, it appeared to drift about 5 frames per 8 minutes. But I've discovered there's a lot more going on here than just converting the project from 24 to 29.97 and non drop with pulldown.

Several problems are cropping up. I digitized the first reel of video using Final Cut Pro and an Aurora card, fine, 600K per second. I "import movie" into my 29.97 non drop project with pull down. I sync up the 2 pop and begin resyncing the movie. Except I have discovered my resync is no good because there is some kind of drift going on. Evidenced by:

1. After a number of minutes into the project, the video appears to sync up with the audio at any given spot you start at, but you won't see a problem unless you PLAY THE SESSION CONTINUOUSLY WITHOUT STOPPING and you will see the audio drifts out of sync with the video by about a second or two over 15 minutes.

2. At any given spot in the movie, if you press play, the video appears to sync with the audio; however, use the scrub bar to try to zero in precisely on a specific cut, and you will find that what you thought was the cut point isn't the cut point. There is a discrepancy between the cut point that appears on playback vs the cutpoint that appears when scrubbing, which makes for very imprecise point location.

3. WITH PULLDOWN, if I run the digitized videotape in protools against the actual videotape I digitized from, a significant drift appears between the two video images, which increases as the minutes pass by, however WITHOUT PULLDOWN the two video pictures remain locked.

Conclusion: Pulldown not only pulls down the sample rate but it also pulls down the video rate, throwing everything askew. While the videotape I digitized from did not have LTC on it, it seems clear that a PT project with pulldown will not sync with the video in this situation.

As I mentioned before, I have tried a variety of timecode and pulldown combinations, and I would have preferred to have the video master on a 3/4 with LTC on it, since it probably would have helped me to locate this problem more quickly; nevertheless pulldown seems to be causing a variety of problems here. Comments?

-John

[SteinT]

Is your PT rig resolved to a video reference?

What was the picture editted on nonlinear system and if so was the sytem and VTR's resolved to video reference?

I have run into somewhat similar problems on a film once only to find that the brand X non-linear video ediitng system and the VTR's were free running. I was able to get it pretty close but it wasnt ever perfect. For that matter neither
was the film.

Good Luck

[c-post]

Yes, I'm resolved to a house sync generator, thanks.

-J

[The Chinese]

Ok, I'm not suire if this is the prob, but there is something here-you mentioned your conversion is from 24P-this sounds like a HD video source. The first thing I would do is to check the HD master- It COULD be a 23.98 master- It which case the audio/video will not be a Pull down!!! I've worked alot with D5 machines alot and this could be what the issue is. If it is, then you should do a down conversion to a NTSC master, then digitize with NO pull down. Hope this helps-
-Todd A.

[c-post]

We finished the mix, for now. Call it a temp mix. It's actually dropframe with no pulldown, 29.97, a sample rate of 48, and it syncs to an NTSC (mini-DV) ref tape which came from the client's setup. However the client is going to go back to his SONY HD 24x setup (which is overseas -- on the other side of the world, in fact) with my NTSC mix, and I'm fairly certain that my mix, which I massaged like hell to get it to sync, isn't going to sync up with his 24 frame setup once he gets it back there. He'll probably have to massage it like hell to get it back in sync at 24, and I told him so, and that seems to be OK with him. What a crazy way to work!

I am investigating the Aurora Igniter 24x hardware option to see if I can get my Aurora card to play back 24x video files on an NTSC monitor. I know Digi have a wonderful $$$$ option that would also do this, but I know nothing about it, so if anyone can point me to any links in this regard, all suggestions are welcome.

The goal is to be fully compatible with this guy's 24p equipment, so we could bring in one of these SONY HD decks in here, and I could layback right onto his HD master directly from my Mix Plus with 02R setup.

Actually, his 24x project as it originally came in here DID play back fine on my Protools rig, with a somewhat jerky, small, widescreen video image in a window on my dual screen Mac. Playback performance seemed a bit too jerky to make good judgement calls on issues of lip sync. And my client expected to see the project mix on a big monitor. Thus the somewhat fellatious decision to mix at 29.97. Sizes for 24x quicktime files varied from 660megs for a 20 minute reel to 2.2 gigs for a 20 min reel.

-J