OMF export from FCP3 clips on import into PTLE6.1

[brinaddison]

OMF, Great Idea!

Acctually I've never been 100% successfull at this Protools OMF import thing. Sure the files always show up in the right places on the timeline and in sync with the movie, but inevitably, there will be a lot of "Clipping spikes" all over the place. Sometimes I can draw them out but all too often they're in abundence and no one could pay me enough to draw them all out.

The only way I've been able to get around this is to export individual audio tracks (essentially consolidating them) from Final Cut Pro 3 and then importing them into PT LE as audio tracks. Now if this were the way to do it, what's the point of OMF? Well I guess OMF is the right way to do it.

So, don't get me wrong, I like the OMF feature, it makes a lot of sence, especially when dealling with long movies. It puts things right where they need to be. However if the audio files get corrupted along the way, it introduces a whole other problem.

My initial thought is Levels too hot on EXPORT. unfortunately this is not adjustable in FCP. All levels maintain their original captured values. Now I know DV camcorders are notorious for producing HOT levels and unless whomever did the filming can set the audio levels manualy, this is unavoidable.

They play just fine in FCP3

But you've got to wonder?........After working on a film for about a year in FCP with no problems and finaly exporting the OMF file for AUDIO MIX in Protools......Why?.....why does it have to clip and distort?

I'm talking obnoxious distortion created my milions of little spikes congregating on my audio files like raging Lucifer devil teeth.

What they look like? ......LIKE DEVIL TEETH (said in a deep Hollywood voice)

Well it's as if the top of the waveform has been chopped off one after the other in a cluster. Before you get down to the sample level while zooming, it is evident that there is a problem with the way the waveform looks.

There are no posts on this issue so I thought I'd be specific.

Am I the only one with the "curse of the OMF import"?

Anyone tryed this on a 001? [img]images/icons/shocked.gif[/img]

[hdace]

You should have a look at this link:
http://www.automaticduck.com/products/index.php

The workaround could be to use duck to output from fcp to avid, find a downtime avid friend somewhere to let you use one for an hour, import into avid media composer, then export to pro tools. The options are complicated, so read the avid manual carefully.

Another solution would be to upgrade to fcp4 and see if problem persists.

I've been an Avid editor since '94 and fcping since 2000. I've never used omf from fcp. But even Avid omfing to pro tools has always been problematic. About one in three times there's problems, but it always works in the end. If Avid can't even get their own digi colleagues to get omf to work properly, how can you expect apple to get it to work? But maybe it's been ironed out with fcp4.

Sorry I don't have a more direct solution, but just to let you know that omf has always been a great idea not very well executed. Good luck.

[hdace]

also, have you tried importing to ptle 5.2.1?

[CTedone]

Hmmm, I had this problem once, importing OMF files from an Avid system into Pro Tools TDM. The audio was far too hot , and it was distorting for me, but sounded fine on the avid. I guessed it was a reference level problem , and it was. The Avid was set at -20dB, while we always run our protools at the standard -18dB. Adjusted ref level in the software, and calibrated the Audio interface, and it worked. Problem solved.
So, this might be the issue, though I don't know Final Cut well at all, so have no idea if you can even adjust the audio reference level.(My initial thought is Levels too hot on EXPORT. unfortunately this is not adjustable in FCP. All levels maintain their original captured values-comment by you...)Not sure if you mean actual levels, or reference levels.
Seems to me that if it sounds fine in FCP, but distorted in ProTools, this is the problem.
If this does not work, put your FCP reference level at
-16dB, then maybe add the extra gain in protools, once you have a clean version (this is assuming you have control over this in final cut). Try that, and let me know how it goes. Might be able to help further... if it does not.
An all digital transfer should never have 'overs' or distortion if it always remains digital at an equal reference level. You would hear it in FCP as well.
I have spent many hours re-drawing waveforms, and i understand your pain. My buddy at work, had to spend 2 months straight, nearly seven days a week, on a 48 track live show, removing digital artifacts and clicks. It was the worst project of his life, as far as he is concerned, and he was paid well for it. He still has nightmares of little waveforms with sharp little clusters.

-Chris

[hdace]

FCP does not have "reference level" control. I don't know why. In Avid, it always seems crucial to get that right (always, always, -18dB). In FCP I always feel like I'm working blind.

CTedone, thanks for reminding me about that. I still think my solution will work, and the trick would be to try playing with the reference levels in Avid.

Another way would be to globally lower the volume level of each track in Avid before export. Avid is not limited, like FCP, to exporting the original recording levels. Globally lowering the level of each track by about 4dB should do the trick.

However, since Avid uses 888's too, and all the same protocols as Pro Tools, there's a chance that the same spikes will arrive upon opening up in Avid. Ugrading to FCP4 might be best solution, but check out the specs on the apple website first to see if they've improved omf.

[brinaddison]

This is good info.

And I'm feeling more confidend now that this is not a problem that only I'm experiencing. I'm going to check if there is a way to set the referrence levels in FCP. If only there was a way to adjust the audio levels on capture from DV. This seems to be an easy sollution to the hot audio that is set on DV and miniDV camcorders.

Or mabe we can petition all the manufacturors to bring the audio "auto level" down by 2db's in all of their products. Haha, ,,,,,,that way we wouldn't have to deal with Apple........oh...my bad, cheap shot! sorry it just slipped out. (They're just like dealing with a foreign Iron-curtain government, at times)

Anyway thanks for the advice, implementation is another ball game.

[brinaddison]

Oh, I forgot to mention, I've tryed this in PT 5.1.1 and 5.2.1 with the same results.

Acctually I did another test last night. I found one of the problematic waveforms and went back to the source file and imported that with import audio to track. I then set the two side by side and compared the shapes. Well there was an obvious correlation between the spikes and clips however the source file was tollerable and did not show the jaggered spikes the way the OMF file did.

So my conclusion is that if the audio were too hot to begin with then it would create these spikes during the OMF file export.
Now what I don't understand is why if the OMF export just rewrites the digital information and includes the metafile for correct placement in the timeline, how does it corrupt the audio if it is a straight digital copy of the source?

I mean I can copy a file a hundred times and it will be the same everytime, but when the OMF export is used, It exaggurates the short-commings of the file.

Now that's one for the experts.

[brinaddison]

Oh, I forgot to mention, I've tryed this in PT 5.1.1 and 5.2.1 with the same results.

Acctually I did another test last night. I found one of the problematic waveforms and went back to the source file and imported that with import audio to track. I then set the two side by side and compared the shapes. Well there was an obvious correlation between the spikes and clips however the source file was tollerable and did not show the jaggered spikes the way the OMF file did.

So my conclusion is that if the audio were too hot to begin with then it would create these spikes during the OMF file export.
Now what I don't understand is why if the OMF export just rewrites the digital information and includes the metafile for correct placement in the timeline, how does it corrupt the audio if it is a straight digital copy of the source?

I mean I can copy a file a hundred times and it will be the same everytime, but when the OMF export is used, It exaggurates the short-commings of the file.

Now that's one for the experts.

[hdace]

Your test last night is fascinating. As I understand it, different digital audio manufacturers offer different amounts of hidden "headroom" above 0dB to help avoid clipping. Clearly Digidesign give more head than Apple! Or maybe it's the other way 'round. I'm gettting confused.

[brinaddison]

Hmmm, It seems like the headroom in Protools while recording is pretty good 'cause even when things go over (into the red) there is no evidence of clipping in the audio waveform. One really needs to be excessive for clipping to be a problem.

Apple or FCP can handle loud playback but as far as recording to it for voice over, I've never tried it. I always do the audio in PT.

I'm sure there is a difference from manufacturer to manufacturer but I'm sure not by much.

I think it's a OMF rewrite issue. I should try bringing the OMF file back into FCP to see what the results are. I have a feeling it will produce the same corrupted audio upon import.

We'll see.