OMF/FCP import/export mess: Why is DP mangling OMFs?

[dix]

I'm normally I'm a composer, but offered to mix a friend's short film. The OMF created from FCP that he sent was an unusable mess. I've mixed a few films before and the Embedded OMFs that showed up were usually fine, with pans, fades etc. intact. This one had nothing to do with the picture (all kinds of garbage on it - multiple scene takes, slates etc). I had sent him the spec sheet that I've used before (provided by audios - thx!), but clearly there's something more he needs to know. I'm not familiar enough with FCP to know what to tell him. Does anyone know what I could tell the editor to get him on the right track?

Here's audios's spec sheet, but I have a feeling there's some other FCP operation that needs to be done before doing the OMF export.?

Thanks!
-------------------

For OMF exports

Picture Elements

Quicktime movie in H.264 that includes dialog only audio in 16/48K AIFF MONO.

- It should have a head and tail 2-pop
- Program should have 5 seconds of black prior to first audio or first picture.
- First video or “black” should start on the hour @ 01:00:00:00
- Time code window should start running at 1st frame on the QT movie.
Timecode can be drop frame or non-drop frame as your project stands now. Just indicate the frame rate on the packaging or email communications. Ideally the visual burn-in of timecode is located at the very top of “Title safe”.

Audio Elements

- OMF type 2 (Embedded)
- 300+ frame handles

[MIDI Life Crisis]

If it were me, the first thing I'd do is ask him to make sure of the type of OMF (it seems you have covered that as type 2 but you want to make sure he knows what he's doing.)

In the event he is totally in the dark, or there is a system incompatibility of some sort, you might request a 'checker boarded' set of tracks to work from. I've used this method on high end projects in the past when the systems just did not want to talk to each other.

Basically, he will provide audio tracks that are on separate tracks (usually 3 or 4). Then you can mix is and deliver a stereo aiff and with the 2 pop and end pop as reference, he should be good to go.

Why is it that 'professional favors' ALWAYS end up being the most difficult tasks to accomplish? Never fails. And it usually ends up that the people who get you to do the favor for zero dollars are the most demanding in the end and second guess everything you do? OK, that's off topic, for sure, but is it really? I'm not so sure. It seems to me that people who pay you because they respect you as a professional extend that respect (for the most part) into the collaborative process. Not always, but more often than those who don't want to pay you seem to be quick to try and beat you down for the stupidest things.

Yeah, I've got a few axes to grind and will never work for free again. EVER! The only exception is that I will 'barter' with fellow artists who have skills I can use or collaborate where I am a producer with a real back end or something of value to gain. But favors such as these are a thing of the past. It took a long time to realize just how much time and energy they waste, and frankly, if I am going to waste time at all, it will be with my buds and not with the leeches. IOW: it's either a pro setting or I'm not playing.

[barreltone]

Doing the OMF export from FCP is a pretty simple process, so I would be surprised if he is missing a step. He may just have a lot of garbage in the timeline.
Is it possible that the "handles" (300+ frames is like 10 seconds on either side, of course) he provided just made for some junk within the OMF?

Or is it possible that he had a different timeline open and gave you an OMF of the wrong one?

[zaratero]

My biggest issues lately have been related to fade ins and disabled clips in the FCP timeline. Some systems had to reactivate (or delete) all disabled clips that contained a fade. Those fade times would add up making the the sync and the OMF useless after a few of them delaying everything.

[dix]

Thanks. Clearly, the editor is newer than even I am to all this, so these responses might help him find a clue.

barreltone, I don't think it's the handles that are the problem. I'm hearing entire takes sequentially on the tracks ("scene 1 - take 1, scene 1 take 2…"). It's much more than a sync/drift issue. More likely a different timeline was exported. I was thinking there might be a Consolidate Project step that he left out. ..dunno.

I'll probably take another stab at an OMF and also have him give me the tracks rendered with head and tail pops - what MLC calls checkerboard tracks I think (although I'm not sure why it's called that). However, one of the reasons I took on the favor is that I'd like to get more serious about film mixing and reckoned this'd be a good guinea pig, so I'd like to do it the "right" way as much as possible, using conventional standard procedures. That was the idea anyway!

Thanks again.

[MIDI Life Crisis]

It's called checker boarding because the cues alternate like the red and black of a checkerboard. When a cue in track A ends the cue in track B starts, etc.

Cal your guy export an AAF? I've has good results with those in DP.

[dix]

I've never worked with AAFs. Is there an advantage over OMFs? Is it any simpler on the export end? Thx!

[MIDI Life Crisis]

My understanding is that AAF files compute fades and other data but it is more than that. The OMFs I get also seem to comoute fades so it is a little confusing. From here:

http://www.joyoffilmediting.com/?page_id=329" onclick="window.open(this.href);return false;" onclick="window.open(this.href);return false;" onclick="window.open(this.href);return false;

AAF. Advanced Authoring Format created by the AAF Association, a trade organization, and beginning to be used by editors working on digital systems. An expanded OMF, an AAF contains EDL information plus audio and video media as well as metadata and data on an increased number of complex effects and titles and can operate with a variety of different media types such as QuickTime and MXF. Avid, Quantel, and other manufacturers’ proprietary information can be encoded in the metadata for exchange only within the manufacturers’ equipment.

From the same site (before you ask... ☺ ):

EDL. Edit Decision List. The EDL contains the reel number and time code numbers for each edit in the final offline cut, the EDL is used to recreate the offline cut during online. It is used by the online editor to conform the show.

[dix]

Okay. This appears to be a DP issue.

I reexported the OMF from FCP and got the same awful results when I imported the file into DP (I even tried importing into DP6 and DP5), however the OMF opens in Logic just fine. From Logic I can render discreet tracks I suppose, but I'd lose the handles.

Does anyone have any further ideas as to what the problem could be? It is a rather large OMF. 1.9gb with a ton of audio files. DP seems to be not recognizing the audio file names correctly. Files with completely different names from different parts of the project often sound identical. A soundbite marked "debbie dia" and "ocean fx" sound exactly the same for example. Why is DP goofing this up so badly?

Thanks!

[dix]

Bumping this. I'm still hoping to iron this out in DP. Has anyone else had DP mess up OMFs on import like this?

Thanks.

[barreltone]

Sorry, dix, but I've never had this happen with an OMF. And I've pretty much only opened OMF's from FCP. Admittedly, I just upgraded to DP7 pretty recently and I haven't tried with that, but every OMF worked flawlessly for me in DP6.

Did you export it yourself from FCP, or did your friend?

[dix]

Thanks.

I didn't export this second OMF myself, but I was present when it was output, so I'm sure it was the correct timeline.

It's a PIA, but the workaround is to open the OMF in Logic and then output another OMF (as mentioned above, OMFs from the project open in Logic np). The Logic-generated opens in DP correctly.

Still no idea what's going on here. Could it be that DP doesn't like OMFs that are so close to the 2gb limit?