Sampling Rate...Sync Sound...And A Headache.

[Mark Smith]

Okay, let me start off this way....I know this post should go in the sound forum...but i took a look there...no one.s posted in over a month, so I thought I.d try here. I.m a writer/director of a new film that I plan to shoot this summer...i go to a film school but synch sound is not covered till third year and I want to get a start on it now. Trial and error learning. I am going to be shooting either 16mm or super 16mm film and then telecine it down to 29.97 for editing in FCP...it will not go back out to film and will end up on DVCAM. What I need to do is buy a recording solution. I.ve looked at DAT recorders and CD-R recorders and I understand that they record via sampling rates [44.1 etc.] but what I DO NOT understand, is how that rate applies to film...say I have a 24fps scene [i will be doing no slo/fast motion] but then i telecine it to 29.97....well, what am I supposed to set the recorder to? And is there a difference between what I would set it to for 24fps and 29.97? Or is sampling rate more about the quality than a speed difference? I plan to synch the audio by mearly using a clapper and identify the marks in FCP...i understand that this will degenerate my synch over time? ... is this truly the case? And if so, how is it compensated?

Lastly, I am looking for microphones. I live in New York, and will be shooting on the streets of the city...I know I.ll end up having to ADR a ton of dialogue anyway, but what sort of mic. [type and brand/model] would be recommended by you all?

 

A million questions, I know and apologize.

 

But thank you for any and all responses,

Mark.

[Chris Keth]

Okay, let me start off this way....I know this post should go in the sound forum...but i took a look there...no one.s posted in over a month, so I thought I.d try here.  I.m a writer/director of a new film that I plan to shoot this summer...i go to a film school but synch sound is not covered till third year and I want to get a start on it now.  Trial and error learning.  I am going to be shooting either 16mm or super 16mm film and then telecine it down to 29.97 for editing in FCP...it will not go back out to film and will end up on DVCAM.  What I need to do is buy a recording solution.  I.ve looked at DAT recorders and CD-R recorders and I understand that they record via sampling rates [44.1 etc.] but what I DO NOT understand, is how that rate applies to film...say I have a 24fps scene [i will be doing no slo/fast motion] but then i telecine it to 29.97....well, what am I supposed to set the recorder to?  And is there a difference between what I would set it to for 24fps and 29.97?  Or is sampling rate more about the quality than a speed difference?  I plan to synch the audio by mearly using a clapper and identify the marks in FCP...i understand that this will degenerate my synch over time?  ... is this truly the case?  And if so, how is it compensated?

Lastly, I am looking for microphones.  I live in New York, and will be shooting on the streets of the city...I know I.ll end up having to ADR a ton of dialogue anyway, but what sort of mic. [type and brand/model] would be recommended by you all?

 

A million questions, I know and apologize.

 

But thank you for any and all responses,

Mark.

<{POST_SNAPBACK}>

 

 

Well, the thing is that sampling rates don't apply to film frame rates at all. That just amounts to the quality of the audio. If you're going to final cut, I would just shoot at 30 fps (or 29.97 if your camera has the capability, I'm sure some do) and record the audio any way you want to. Slates would help sync it in final cut. Assuming a quality, accurate audio recorder (that will keep speed absolutely constant) the sync shouldn't creep over time, but you could definately sync every shot and it should be rock solid.

[Mark Smith]

I actually can.t shoot in 29.97...and i.d prefer to shoot on 24 in the event I ever do master it back out to film...[i know i.d have to speed up the audio again]...

 

I.ve been doing some research on the net, and sort of found out that I can do this by slowing down [pulling down?] the audio by 1% ... the only tutorial i.ve found on this is for pro tools though...Does anyone know how to take say a 48khz file and match the speed to the telecined 29.97 video strictly with final cut [compressor???]

 

Thanks again,

Mark.

[James Visiano]

If you shoot at 30 fps, then you would be shooting in slow motion!

[jasarsenault]

I did this exact thing this summer. I shot 16mm telecined to dvcam with my sound recorded on Dat. I shot at 24 frames per second on my cam. I asked the lab about the telecine and the sound question. They recomended I transfer my footage and record sound on dat at 29.97 ndf. They said this would give best results. While editing, I found that after syncing my clapper every couple seconds there was a drift in the sync. It was a pain but I had everything in sync within a few hours and it looks fine.

 

So, shoot at 24 frames. Record sound at 29.97 ndf. Transfer your footage from film to video 29.97 ndf.

Edited January 13, 2005 by jasarsenault

[Mark Smith]

Well I.ve been reading about the drift that happens to the audio...supposedly you.re supposed to change the sampling rate...which I have no idea how to do...

 

Anyone?

 

Mark.

[jasarsenault]

Sampling rate is quality. Check my above post. The way I fixed the drift in fcp was every couple of seconds I would cut the audio in a silent part of the film when I knew it was falling out of sync. I would then move the audio forward a few frames to resync and lock down audio. I would then continue until I noticed the falling out of sync and do the same thing.

Edited January 13, 2005 by jasarsenault

[Guest 22west]

Why telecine to 29.97 for editing in FCP?

I don't think you have to do this.

[jasarsenault]

Well video runs at a different frame rate then film. Film runs normal speed at 24 frames while video at 29.97. you can go non drop frame or drop frame. Look at the speed on your Dat timecode adapter. It will give you the options of recording at 24 f, 29.97 ndf, or 29.97 drop frame. Then go check final cut and see what your options are for playback at normal speed. It will be 29.97 ndf or drop frame. This is what was recomended by the lab I used for processing and telecine (Alphacine) and my film to video project is completly in sync!

Edited January 13, 2005 by jasarsenault

[Chris Keth]

If you shoot at 30 fps, then you would be shooting in slow motion!

<{POST_SNAPBACK}>

 

 

Only if he projects it. I meant that he should do that for transfer to video, then he would have a more exact frame-to-frame correlation.

[jasarsenault]

Essentially your right bunnies. When played back on fcp it will be at 30 frames/sec or 29.97 ndf or drop frame. Shoot your film at 24 frames. Transfer to video. It will then be played at 29.97 ndf or drop frame depending on your choice. Set your Dat to record at one one of those speeds. But bunnies is right again. The speed option on your dat doesnt really matter too much if you are syncing completly by eye. I shot at 24 , telecined to 29.97, and had my dat set to 29.97 ndf... but when I started I had my dat set to 24 frames. You can still sync it by eye. If you shoot 24, it will look normal when played back on video at 29.97.

Edited January 13, 2005 by jasarsenault

[Mark Smith]

Well, here.s the question...I.ve been going around to a lot of different music supply shops lately, and EVERYONE is telling me to screw the DAT format. They tell me the best solution - since I am not having the sound added via telecine - is to use a product like digidesigns Mbox or 002 and capture the audio into a laptop computer ... that way it is already in the correct file format [aiff]...my question then becomes again, I am shooting at 24fps...I am then transferring it to 29.97 a 1% slow down [i think?]...so i have to slow the audio down...right?...to match? How can I take a file that is matched [in time] to the film at 44.1khz and slow it down to match with the telecined film? Can i do that in FCP? Logic? Pro Tools [would prefer the answer not be pro tools...i work a side job at apple, so logic I can get for cheaper].

 

Thanks again,

Mark.

[Chris Keth]

24 to 29.97 is more like a 12% decrease in speed. It would be noticeable. What I'm saying is just shoot at 30fps. Then transfer to video it will essentially be a 1 to 1 correlation ebtween film frames and video frames (I know it's interlaced, but FCP still tells you in frames). Then audio recording is continuous, as far as I know. You don't need to deal with frame rates or anything, just with making sure you're recording at a constant speed. Since you're going right into FCP. just do it on your DAT or straight into the laptop in .aiff format. Then once you have your film transferred to video, you can import the audio and sync it with the video frames by eye. Very simple to do.

Edited January 13, 2005 by Mr. Bunnies

[Mark Smith]

Right, but from what I understand, if I did what you were saying and then my final project was projected back out to film and the audio was played, it would be in sync...but because i have to add frames to it in the telecine process, it pulls it down 1% and so I have to make those adjustments in the audio as well, or it will drift...

 

Mark.

 

I just don.t know how to make the adjustments...i don.t use music software.

[Chris Keth]

See my edited post. Don't shoot at 24fps if you only have video transfer in mind. You only have to worry about pulldonws and whatnot when you want to transfer to video AND project it (which must be done at 24fps)

[jasarsenault]

I finished in video. I did not go back to film. So it is quite possible that there is a different process for that. Why not call the lab that will be doing the telecine? That's what I did for my project.

[Mark Smith]

Well the thing is, there is no lab doing the telecine. I have access to the telecine room at my school and will be doing it myself...Also, the camera I am shooting at is only varriable from 24fps to like 48 or something. I do want to shoot in 24 so I could have the ability to go back out to film at some future point, but right now i belive it will end on video.

 

Mark.

[jasarsenault]

This post is giving me a headache. :D

[Mark Lyon IMPOSTOR]

While there are some people who shoot film at 30 fps for video transfer without pulldown, I think for most projects it's better to shoot 24 fps, so you get more of the feel of film. Because people watch telecined film all the time on TV, the 3:2 pulldown used to get the 24 fps film to 30 fps video is part of the look. Then you can simply set your DAT to whatever the telecine is set to (we always use 30NDF), and it lines up fine.

 

Good luck,

 

Mark Lyon

Mighty Max Films

[Mark Smith]

Hate to give everyone a headache, but it is stated in the subject! :)

 

I think after talking with a lot of people in the audio world though, I am leaning away from the DAT recorders and being able to set 30 etc...

 

 

I am looking at a product by Tascam which is basically a pre-amp audio device that lets you record directly to a harddrive on a laptop. This seems to be a good rode. I am just wondering if I will have a sync problem this way. I imagine I still will because of the pull down. What exactly does that do, does it speed up the film 1% or slow it down 1%...or? Because then I need to match my audio accorrdingly??

 

Sorry again,

Mark.

[Mark Smith]

Oh, and one more thing I might add, my scenes are not simple 20 second dialogue scenes before cuts, I have some entire scenes [5-9 mins] where the camera does not cut on the dialogue [think of a woody allen film for the effect i.m talking about]...just so you know, so if i even used a DAT recorder at 30ndf and the video was 29.97 i think it would get off mid-way into my scene...these aren.t short cuts.

 

Mark.

[jasarsenault]

I shot a doc rolling sometimes full 400 ft rolls. When my sync drifted I would cut the audio line where there was a silent space in the dialogue. I would then move the audio ahead or back a few frames to re-sync, then locking down audio. Then I would keep playing the film until it was out of sync and repeat the process. I had full ten minute interviews where the audio had been cut in between silence numerous times to re sync, yet you could not tell during the interview. I was kidding about the headache!

Edited January 13, 2005 by jasarsenault

[jasarsenault]

All I can say is that if you are going to stay on video, I shot at 24 frames/sec telecined to 29.97 ndf. I recorded sound on Dat. My interviews sequences were long. It did drift on fcp, but I just kept re-syncing in audio pauses. It worked.

Edited January 13, 2005 by jasarsenault

[Mark Smith]

I could always do that I guess...synch in the audio pauses, it was something I had given though too...But I was hoping to avoid it.

 

I want to quote a website I.ve been reading, about the pulldown and transfers, to show you what I.m coming from in my understanding....

 

If you're shooting on digital video none of this probably applies, because you have single-system sound with picture and sound already in sync and both are already compatible with video frame rates. But if you're shooting film, digitizing the film via telecine, and completing your post on a non-linear system, the film you shot on the set -- the "real time" that you recorded -- is going to get slowed down slightly during the telecine process. If you filmed for exactly ten minutes on the set, that footage, after being telecined, will play back slightly slower and last 10 minutes and 6/10th of a second. The difference is enough to make dialog obviously out of sync -- if you sync up the clapper at the start of the take, the sound will drift noticeably out of sync after just a minute or two. To avoid this problem, you need to "stretch" the Nagra recording, slowing it down by a corresponding amount, which happens to be .1%. This is the difference between a theoretical video frame rate of 30 fps and the actual NTSC frame rate of 29.97 fps. Likewise, it's the difference between the Nagra's original sync pulse reference of 60 Hz and 59.94 Hz.

 

During playback, the Nagra "expects" to see an external reference of 60 Hz to compare to its pre-recorded reference of 60 Hz. It will then make any tiny adjustments in playback speed that are needed to stay "in perfect pitch" with this external reference. When we use the 59.94 reference box instead, the Nagra will seem to be playing back a sync pulse that is subtly "sharper" in pitch than is desired so the correction circuitry will slow the Nagra playback down very slightly so that it is once again playing "in perfect pitch."

 

This is best done during the digitizing itself so that the resulting AIFF files will already be "stretched" to the correct length.

 

Remember that none of this resolving takes place unless the Nagra is playing back with the Selector Switch lever all the way down in the "speaker icon" mode. (The Nagra will play back audio in the "first click" position but will not resolve in that mode.) Also note that resolving and pulldown are not the same thing. If you failed to resolve a Nagra transfer, doing a pulldown through file conversion tricks could still result in sound that drifts in sync because the Nagra was essentially "freewheeling" during playback and not doing any subtle ongoing speed corrections. The pulldown is an overall speed adjustment that affects an entire take; resolving involves internal incremental adjustments during a take.

 

 

 

About achieving pulldown through file conversion.

 

Maybe you goofed and digitized a number of Nagra tapes while resolving to 60 Hz instead of 59.94. Or maybe you recorded sync sound on a standard DAT machine and you loaded it into Pro Tools through a digital-to-digital S/PDIF connection. In either case, you need to stretch those files so that they correspond to the "video speed" of the telecined picture.

 

If you were strictly working within Pro Tools, this would be relatively simple. Pro Tools defaults to using the Mac's clock reference to determine the playback speed of all audio files. That is, the session itself "assumes" a rigid speed based on the sample rate you choose for your session -- in this case, 48kHz. For instance, if you bring a 44.1 kHz file into the session without doing a file conversion, it will play back at the wrong speed -- it will occupy less space on the Pro Tools timeline, and play back faster than its original speed. So one way to achieve a pulldown would be to select a file and go to the Audio Menu of the Regions List, choose Export Selected as Files >Sample Rate>48kHz (Pull Up/Down)>48.048. You would then save these "sample rate tweaked" versions of the files into a different folder.

 

Now if these files were played back at their correct sample rate of 48.048 they would be exactly the same length as before. (And in fact Quicktime Player would do this.) But if you import these files directly into Pro Tools without converting them to "48kHz exactly" during the import, then they will end up sounding subtly slower and taking up more space on the timeline -- in other words, they'll be stretched to the proper pulled down length.

 

However, Final Cut takes more of a Quicktime Player approach to sound playback, essentially looking at the file and saying, "So -- you're a 48.048 file and if played at that speed should last exactly ten minutes. Okay -- we'll adjust accordingly so that you end up lasting exactly ten minutes." So FCP expects to deal with files of different sample rates in the same session and adjusts on the fly. Impressive, but in this case it complicates matters since we probably want files that will be stretched to work in FCP.

 

There are two solutions:

 

1. We can trick FCP into thinking our 48.048 file is actually 48kHz exactly. We can do this by using a sound utility to change the header information of each audio file to read "48 kHz." The header information is what tells Quicktime/FCP how to approach the sample rate. Disadvantage: While this method works, it is tedious since I haven't found a program that will change this header info in batches rather than one file at a time.

 

2. We can create true 48 kHz files that have already been stretched to be longer. Let's go back to our Pro Tools session where we created 48.048 kHz versions of our files and brought them into the timeline. They are now stretched in time. We simply need to create new copies that are converted to 48 kHz. One method is to highlight individual files and select Edit>Consolidate. But it's easier to drag a succession of files into the timeline, select them all, and then go to Audiosuite>Duplicate, making sure that we choose the option to "create individual files" rather than "create continuous file." This will result in a series of files with the same file names but with "DUPL-01, etc" appended. These will default to going into our Session Folder's Audio Files folder and we can later copy or move them for import into FCP.

 

If anyone knows how to do this for say 48khz 24bit files [like an mBox or tascam us-122 would be recording] that I think would fix this problem of having to readjust during pauses in dialogue, because for one, my dialogue is VERY rapid fire, pauses don.t come along too much, and the synch may drift in the middle of a character rant etc. I.d rather have the time ironed out.

 

???

 

Mark.

 

Sheesh, you may have been kidding, but I am getting a headache :)

[Mark Smith]

Okay, well good news. That article I posted, I contacted the author, and he actually replied! He was very nice and answered all of my questions [he.s an actual sound editor on some big budget hollywood films - flight of the phoenix is the latest]...so we worked out the problems, I do have to do a 1% conversion...use compressor to take my 48khz file and make it 48.048 But then I have to fool Final Cut into believing that this is still a 48khz file...so he gave me a link to some freeware on the web that will let me edit the header information. And this pretty much solved my problem! So thanks to everyone on the board. I only found this site because of this question of mine, but It.s such a great site, i imagine i.ll be here often.

 

Thanks again,

Mark.