29.97 to 30 problem

[steve mag]

I accidently loaded 30fps field DAT's (from a film shoot) into Pro Tools with the DAT player playing at 29.97 fps. Pro Tools chased the DAT's LTC using USD. Obviously all my timelines are off significantly. Can I do a 30fps Pull Up to correct the timeline? If so can I export the corrected audio files to another Pro Tools session with the audio files keeping the corrected time code? Will there then be any pitch issues when those audio files are pulled down to 29.97 to sync to video for the offline sound edit in the other pro tools session? My plan was to re-create all field DATS as PT sessions so I'd have all recored material available for future use. I'd then start a new PT session being the actual film edit/mix. I'd import audio from the field DAT sessions and place in the film's session. Thus the field DATS' timelines must accuratelt reflect the 30fps timeline they were recorded in so when I import them into the film's session they are the correct pieces. Now recorded at 29.97 by mistake they are obviously completely on a wrong timeline. BTW there are 15 hours of material on 28 DATs so re-doing it is possible but takes about 50 hours to do with PT chasing the DAT's LTC via a USD.
Thanks greatly for any advice!

[Kasper]

I'm a bit confused by the way you have presented your problem but I'll give it a shot.

If you put PT in pull-up mode in the session set up pref. that should adjust the display of the timeline for 30FPS.

If your sample clock is set for 30FPS, a pull-up if you went in at 29.97, then the audio should play back at film speed and the timecode should be the same as the orig. If this method works You should be able to use that session as Your ref. for the orig. TC without any need to create a new session.

I don't know how PT deals with the timecode stamp or the sample rate issue when exporting files. I would just do a test. Go way downstream in the session and export a file and see if it matches the orig. 30FPS TC. If not, time stamp all the regions w/the new code where they sit in the corrected session.

How long are the takes from the DATs ?
The reason i ask is that i would assume at each break in code the USD starts generating new code from the numbers it sees comming in, right ? So, unless the takes are really long without a break in the code you shouldn't be that far off from 29.97 to 30 FPS. Am I missing something ?

kasper

[Noiz2]

Did you load digitaly? w/ digital sync? If so then you have files that will play at normal speed in a 30 session (and are actualy playing slightly slow right now. If this were so you could put user time stamps at there curent locations , make a new 30 session import the files and spot to user timestamp and be home free.
SK

[steve mag]

Originally posted by Kasper:
I'm a bit confused by the way you have presented your problem but I'll give it a shot.

If you put PT in pull-up mode in the session set up pref. that should adjust the display of the timeline for 30FPS.

If your sample clock is set for 30FPS, a pull-up if you went in at 29.97, then the audio should play back at film speed and the timecode should be the same as the orig. If this method works You should be able to use that session as Your ref. for the orig. TC without any need to create a new session.

I don't know how PT deals with the timecode stamp or the sample rate issue when exporting files. I would just do a test. Go way downstream in the session and export a file and see if it matches the orig. 30FPS TC. If not, time stamp all the regions w/the new code where they sit in the corrected session.

How long are the takes from the DATs ?
The reason i ask is that i would assume at each break in code the USD starts generating new code from the numbers it sees comming in, right ? So, unless the takes are really long without a break in the code you shouldn't be that far off from 29.97 to 30 FPS. Am I missing something ?

kasper

Yes Kasper the DATS were all recorded with the time code generated in relation to time of day. So even though each DAT only contains about a 30 minutes of material, the time code actually spans around 8-10 hourse per DAT so the code error extrapolates a great deal over that time.

"If your sample clock is set for 30FPS, a pull-up if you went in at 29.97, then the audio should play back at film speed and the timecode should be the same as the orig"
the same original meaning the original 30fps they were rcorded at or 29.97 I accidentally loaded them at? Thanks!!!

[steve mag]

unfortunately I had to fly them in analog, so that won't ork for me, I guess. Thanks and any other ideas would sure be appreciated.

[Kasper]

Originally posted by steve mag:
Yes Kasper the DATS were all recorded with the time code generated in relation to time of day. So even though each DAT only contains about a 30 minutes of material, the time code actually spans around 8-10 hourse per DAT so the code error extrapolates a great deal over that time.

"If your sample clock is set for 30FPS, a pull-up if you went in at 29.97, then the audio should play back at film speed and the timecode should be the same as the orig"
the same original meaning the original 30fps they were rcorded at or 29.97 I accidentally loaded them at? Thanks!!!

OK, bear with me steve, just want to get this right.

The DATs were recorded at 30FPS free-running time of day.
(standard procedure)

Every time a new take comes along, when the mixer put the DAT machine into record, the TC jumps to a new number, corresponding to however long it has been since the end of the last take. Yes ?

If the USD was generating new code at every change in code then you should only be off by the distance at the start of each code break and not the distance from the start of each tape. Is this the case ? If so, then the dif. will be so small it should not be a real problem, unless you are using some sort of auto conforming software to place the takes. Do you have audible slates on each take to identify them ? The dif. between 30 and 29.97 is only 3 sec. and 18 frames, over a one hour continuous run.

> the same original meaning the original 30fps they were > rcorded at or 29.97 I accidentally loaded them at? Thanks!!![/B][/QUOTE]

Yes. The orig 30FPS. If you went in at 29.97 FPS TC rate and the sample clock was set for standard 44.1 or 48 then changing the session set up rate to pull-up will adjust the TC display on the time line to 30FPS. Set the sample clock to pull-up will change the speed of the audio to film speed. You should be back to the original TC and pitch the audio was recorded at in the field.

kasper

[steve mag]

Kasper,
You mention when the USD generates time code... I actually had the USD chasing the DAT machine (I assume you know that, but just to be clear). So the DAT machine is generating code at 29.97 but the tape was actually recored at 30 fps, time of day, standard as you said, yes. Pro Tools session set up is set to 30 fps no pull up or don ( I thought this was the time code speed being read by and generated to USD by the Sony 7030 machine). I tried a 30 fps pull up but it didn't seem to correct the problem. Does this make the problem clearer? Thanks yet again!

[steve mag]

Kasper,
You mention when the USD generates time code... I actually had the USD chasing the DAT machine (I assume you know that, but just to be clear). So the DAT machine is generating code at 29.97 but the tape was actually recored at 30 fps, time of day, standard as you said, yes. Pro Tools session set up is set to 30 fps no pull up or don ( I thought this was the time code speed being read by and generated to USD by the Sony 7030 machine). I tried a 30 fps pull up but it didn't seem to correct the problem. Does this make the problem clearer? Thanks yet again!

[Kasper]

Hi steve,

Gonna get to the bottom of this, each time i get more info i think i get a better understanding. If I can't help you here perhaps email contact directly between us would help you get over this a bit sooner.

Where were we...

The session was originally set up for 30FPS when you recorded in, w/the 7030 running at 29.97. OK. This is a problem because i think you would be pulling the wrong way.
Still, when you change the frame rate in the session set up window your TC display should change. Even if it goes the wrong way it has to change. Select a region that is near the end of your session then change the TC rate in the session set up window. The start time has to change. Make sure a pull-up or down is not checked when you change.
I would create a new session at 29.97 import all your files spot to orig. TC and then change the session to 30 and pull-up the sample clock but don't check the pull-up in the session set up window or you'll be back where you started.

When you recorded the DATs in to PT the USD was re-genning the code from the 7030. Did it not change when the code on the 7030 changed ? (at the record in point of each take on the DAT)
I assume you have many regions in the session spaced apart by the distance between record in points on the DAT.

One last note. How were you locking the 7030 to 29.97 ? This has to be done a very specific way or it willl not work. I have not attempted it in a few years but unless Sony updated the software I seem to remember the manual stating some quirks with doing a pull-down.

Helping at all ?

kasper