Jump to content

16mm w/ DAT (no timecode)


Josh Hill

Recommended Posts

  • Premium Member

So I'm a little hazy on the whole film/sound thing. I'm coming up on my first 16mm sync sound project (CP16R and HHB PortraDAT) which will be telecined for editing (unlikely that we will make a film print, but that all depends on how the short turns out).

 

Since the DAT is not timecode, we're going to be syncing the old fashioned way with a clapper slate (a little less old fashioned because we are going to sync our own sound on the computer). We will be shooting 24fps for transfer to 30fps.

 

Are we going to need to adjust the audio speed in the computer? This is the issue that I'm confused with. I was under the impression that, since we are not using a timecode DAT, we will end up with the same number of second of film/video with the same number of seconds of audio and we simply have to sync the DAT to the clapper and we are good to go.

 

What is the procedure for syncing telecined video and non-timecode audio on a computer? What adjustments have to be made?

 

Thanks,

 

Josh

Link to comment
Share on other sites

  • Premium Member
So I'm a little hazy on the whole film/sound thing. I'm coming up on my first 16mm sync sound project (CP16R and HHB PortraDAT) which will be telecined for editing (unlikely that we will make a film print, but that all depends on how the short turns out).

 

Since the DAT is not timecode, we're going to be syncing the old fashioned way with a clapper slate (a little less old fashioned because we are going to sync our own sound on the computer). We will be shooting 24fps for transfer to 30fps.

 

Are we going to need to adjust the audio speed in the computer? This is the issue that I'm confused with. I was under the impression that, since we are not using a timecode DAT, we will end up with the same number of second of film/video with the same number of seconds of audio and we simply have to sync the DAT to the clapper and we are good to go.

 

What is the procedure for syncing telecined video and non-timecode audio on a computer? What adjustments have to be made?

 

Thanks,

 

Josh

 

Are you shooting 24fps second and doing a 2/3 pulldown for a xfer to video which is at 30 fps?

 

 

If your are shooting 24fps you xfer at 24fps.

 

or shoot at 30fps and xfer at 30fps is a very good option if you are doing a video SD/HD only project.

 

In any case the DAT will sync if your takes are not to long--- beyond 5 ot 10 minutes per take/beyon that it may drift but nothing which can's be adjusted.

 

You can Shoot and Xfer at 7.5 fps second and the DAT will sync.

 

Have the person handling the slate call out the Scene and take number after the clap, so you don't cut off the reference when you sync.

 

You can use the waveform of the audio for an obvious match to the clapper sticks coming together. You will figure out there is a 1 or 2 frame offset for dead sync if your clapper person has not had to much espresso. Shooting and Xfer at 30fps gives you a little sharper image because of a faster shutter speed (1/60th as opposed to 1/48th) and you will not have to remove the 2/3 pulldown if you composite. Shooting 30fps is my SOP for shooting commercials. 1 frame of film equals 1 frame of video-elegant when xfering to progressive video formats.

 

Since film looks like film you don't have to shoot at 24fps. And going thru elaborate of 24 fps video schemes and post production processing plug-ins and all the stuff you have to do and make excuses for is simply not necessary. Film just is what it is!

Link to comment
Share on other sites

  • Premium Member

By transferring to 30fps I mean that I'm going to be transferring to NTSC video (if that was unclear). I would shoot 30fps if I can, but I don't have the adapter gears for the CP to make it shoot 30fps (nor would I want to open up my camera and start tinkering around with it if I did).

Link to comment
Share on other sites

Shooting and Xfer at 30fps gives you a little sharper image because of a faster shutter speed (1/60th as opposed to 1/48th) and you will not have to remove the 2/3 pulldown if you composite. Shooting 30fps is my SOP for shooting commercials. 1 frame of film equals 1 frame of video-elegant when xfering to progressive video formats.

You can do the same thing even if you shoot at 24fps. You can telecine the footage at 30fps, and then have the NLE capture it at 24fps. This way you avoid the pulldown and get to work with 24 frame progressive material WITHOUT recompression.

 

You also need to be aware of the drop frame issue, if this is going to NTSC. If you shoot at 24.000fps, this will need to be slowed to 23.976fps for NTSC at some point. That means the audio must also be slowed down to match. This can be done in audio post after you've got a locked edit. The same rule applies if you shoot at 30.000fps. It will eventually need to be 29.970fps, and again, the audio needs to be slowed down to match.

Link to comment
Share on other sites

  • Premium Member
You can do the same thing even if you shoot at 24fps. You can telecine the footage at 30fps, and then have the NLE capture it at 24fps. This way you avoid the pulldown and get to work with 24 frame progressive material WITHOUT recompression.

 

You also need to be aware of the drop frame issue, if this is going to NTSC. If you shoot at 24.000fps, this will need to be slowed to 23.976fps for NTSC at some point. That means the audio must also be slowed down to match. This can be done in audio post after you've got a locked edit. The same rule applies if you shoot at 30.000fps. It will eventually need to be 29.970fps, and again, the audio needs to be slowed down to match.

 

The key issue is whether the camera and DAT run at a consistent speed. You can shoot at 24/23.970/30/29,970/48/64 or what ever it does not matter. Since the DAT does not record at 24 or 30 it does not matter. You capture your audio at the frame rate you will edit, that's what matters. It will sync up fine.

 

The difference between 24 and 23.97 and 30 and 29.970 is so inconsequential it hardly makes any difference except for very, very long takes.

 

Since film is film and does not require the gimmick of 24fps to look like what it is; xfer at 30fps progressive gives you the advantage of (non interlace) progressive recording.

 

 

 

If you shoot at 24 then xfer at 30fps the image will be speed up. Capturing it at 24 fps will not remove that increase in speed. You will have to process it using a compositing program or temporal manipulation program-not recommended.

Link to comment
Share on other sites

I.m confused by some of the things in this post...assuming he shoots 24fps and then telecines to 29.97 why can.t he just take the audio files and slow them down by .1% the same as the film and have perfect synch? What is this stuff about this not working for long takes? I.m curious because I.m planning to shoot in a few weeks and use this approach.

 

Mark.

Link to comment
Share on other sites

  • Premium Member
I.m confused by some of the things in this post...assuming he shoots 24fps and then telecines to 29.97 why can.t he just take the audio files and slow them down by .1% the same as the film and have perfect synch? What is this stuff about this not working for long takes?

 

I.m curious because I.m planning to shoot in a few weeks and use this approach.

 

Used a non-timecode DAT for years with no problems

 

Mark.

There are three issues we are taking about.

 

1) If you shoot film at 24 frames per second and transfer at 30fps(29.970) the picture will appear sped up

 

2) If you shoot at 24 fps and do the normal transfer to video; the film transfer device (Cintel/Spirit) will run at 24fps and deliver a 2:3 pulldown which is recorded by a video tape deck at the video standard of 30fps. Please look at the archives of CML or this site for a detailed explanation of "2:3 PullDown"

 

3) Since the time base of the DAT Audio format is unique (it is not 24 or 30fps)to the DAT when you capture the audio to your editing system it is converted to the time base you are working with.

 

As long as the DAT has an accurate interal clock and the motor on the camera has an accurate crystal control the film and audio will sync up.

 

You will have to manually sync your picture to audio as you will not be able to use automation to line up picture and sound.

 

Since you are not using a timecode DAT, the sync will "drift" slightly over time. Like one or two frames every five or six minutes.

 

This in practice is not a major problem as most folks rarely have takes that long. Attempting to apply a percentage speed up or slowdown to the entire clip just compounds the sync error. Shifting of a frame or two in the edit is a better solution, you may have to cut the audio track up into a couple of chunks to get everything to sync ---this is worst case-you may not have much error at all-but if you do!-You know what to do.

Link to comment
Share on other sites

If you shoot at 24 then xfer at 30fps the image will be speed up. Capturing it at 24 fps will not remove that increase in speed. You will have to process it using a compositing program or temporal manipulation program-not recommended.

This is simply not true. The actual frames of the picture and the playback frame rate are independent of one another. Simple proof of this is the way the Panasonic DVX100 records in 24pA mode. The video is recorded at 29.97fps, but then transferred to the NLE at 24fps after discarding redundant frames. Avid, for example, is quite capable of capturing 30fps material at 24fps, after which it will play back and edit at 24fps (as originally shot). All of this happens without recompression, as opposed to doing a "normal" telecine at 29.97fps with 3:2 pulldown, and then performing a "reverse telecine" in the NLE to recover the original 24 frames.

 

As long as the DAT has an accurate interal clock and the motor on the camera has an accurate crystal control the film and audio will sysnc up.

This is absolutely correct -- however, and NTSC signal does not necessarily play back material at the same rate it was recorded. If you shoot material at 24.000fps, there is no video medium that will ever play it back at that same speed. The closest you will ever get is 23.976fps, since NTSC is drop-frame. Even with PAL, the proper way to transfer 24fps material is to speed it up by approximately 4%. This is why the audio must be sped up or slowed down to match. It doesn't matter that both the film and tape were in sync on the day of the shoot. The problem now is that you do not have the ability to play back the picture at the original rate, hence the audio drift.

 

However since you are not using a timecode DAT, the sync will "drift" slightly over time. Like one or two frames every five or six minutes.

Whether the DAT uses timecode or not is irrelevant. The phenomenon you are witnessing is a direct result of what I am talking about. It's not because the DAT doesn't use timecode. It's because the audio is not, in fact, playing at the same speed as the picture. The only way your audio would ever match your picture perfectly is if the film camera was shooting at 23.976fps and the DAT was recording at 29.97fps. I believe there are cameras that can shoot at 23.976fps, but most shoot at 24.000. So, be prepared to slow down your audio if you want it to match perfectly.

 

I.m confused by some of the things in this post...assuming he shoots 24fps and then telecines to 29.97 why can.t he just take the audio files and slow them down by .1% the same as the film and have perfect synch? What is this stuff about this not working for long takes? I.m curious because I.m planning to shoot in a few weeks and use this approach.

You can, and that is the proper way to do it. I would highly recommend against transferring with 3:2 pulldown if you are going to sync the audio in post anyway. The advantages to 1:1 telecine are higher picture quality, and less storage space on the NLE since you are storing 24p material as opposed to 29.97 with 3:2 pulldown.

 

All of what I am talking about is not theory. I have done it myself, and it works. I urge others to use this technique whenever possible.

Link to comment
Share on other sites

  • Premium Member
This is simply not true. The actual frames of the picture and the playback frame rate are independent of one another. Simple proof of this is the way the Panasonic DVX100 records in 24pA mode. The video is recorded at 29.97fps, but then transferred to the NLE at 24fps after discarding redundant frames. Avid, for example, is quite capable of capturing 30fps material at 24fps, after which it will play back and edit at 24fps (as originally shot). All of this happens without recompression, as opposed to doing a "normal" telecine at 29.97fps with 3:2 pulldown, and then performing a "reverse telecine" in the NLE to recover the original 24 frames.

 

 

If you shoot 16mm film (note the title of the thread) at 24 frames per second and the xfer device Cintel or Spirit is set to poerate at 30 frames per second, the footage will look sped up.

Fact of life, good in all time zones.

 

I sure that everything you say works. But I don't think it would be considered a SOP. Your technique sounds really smart. I don't work in the formats you mentioned of course so it won't cost me a dime if it doesn't

Link to comment
Share on other sites

... The only way your audio would ever match your picture perfectly is if the film camera was shooting at 23.976fps and the DAT was recording at 29.97fps. I believe there are cameras that can shoot at 23.976fps, but most shoot at 24.000. ...

 

Almost any camera with an accessory speed control socket can film at 23.976 or 29.970 to match the actual running speed of a Rank etc. if you add one of our famous Tobin Videoframe Controllers. This inexpensive device was intended for just this purpose, eliminating having to adjust the speed of the audio in post. Also it has a Phase button for filming from a video monitor. Just the thing if you don't need the full 50,000+ speeds of a Milliframe Controller. See at http://www.tobincinemasystems.com/page12.html .

 

Also our latest TXM-27 crystal control for the Arri 16-BL has these two video speeds and a phase button.

Link to comment
Share on other sites

Are these wondrous tools available for the CP16R?

 

We have made little crystal boards for the CP-16 for running at 29.97, 30, or 23.976 speeds which we used to sell to Whitehouse for installing in the cameras. We haven't sold any for a while. I don't know if Whitehouse could still install them with the new ownership, or if Visual Products could take over the installation. We could still make them although they are not listed on the website: www.tobincinemasystems.com .

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...