[Joe Huggins]

Bouke,

We do have your wonderful reconnect software, but it will not work in this case. We need to reverse assemble a camera roll – 10min – using the ALE file to show clip duration. It’s not a matter of file name from offline vs TC stamps.

I also posted something similar on FCP’s site at this thread if you have any additional insight on reverse assemble and altering CTDB to ALEs.

https://forums.creativecow.net/thread/8/1117297

Joe

Thanks, Film Joe

[Joe Huggins]

Job,

Good news, we did make the reverse assembly work. Turns out Avid does cut the subclip based on the TC and not the first frame of the captured video.

Follow up. Do you work at all with FCP? We are looking for a similar technique that will allow the ALE imported to CinemaTools to reverse assemble a FCP 10 minute clip. Unless it sees a clip the correct size of the ALE take, it can not match the cuts and KK to any clip.

Thanks again for confirming our test findings.

Thanks, Film Joe

[Joe Huggins]

Job,

Good feedback. I know Avid stopped supporting MediaStation long ago (ex Avid salesman here). But VRT control is only for playback, Avid can not be used as a Record deck and make an ‘edit’ and neither can AJA KiPro, only slam record or RS422 playback control.

Deluxe has a very expensive MTI control dailies system that is like a Bones or DVS Clipster, after it inputs a telecine file under frame accurate TC, it can make Avid or ProRes and match the ALE there as the frame count is the same. This is what I want to acheive with a manual record to Avid, only I can’t make the first frame of the file a specific 1 hr 00 (or similar) TC. Avid uses the first frame of video to count frames to match TC and KK, so we are looking for a way to trim that back or offset the ALE and make subclips.

Seems no other telcine house is trying this but we are looking for a tapeless workflow to be similar to Control Dailies without that high cost. Thanks for any other ideas. Joe

Thanks, Film Joe

[Joe Huggins]

Job,

I guess the concern is that the captured file is longer, from the head, than the TC stamp on the ALE file. Our tech said he did not know of a technique to take the ALE and break up the continuous camera roll into subclips. Maybe your reply will clear that up. Yes, 23.98 format.

The goal is to create an accurate ALE without going to tape, then digitizing each clip in from the tape. If I can make an Avid HD file on the fly, I save a digitize step, the use of an HDCam deck, and tape stock. I’m looking to save costs on film transfers for features and provide an accurate ALE file. We sync the audio after telecine in the Avid and that reduces time and costs too then make subclips from there. The start of a file, unlike tape, can not match to an exact frame TC where we pull an edit on tapes. Does this make more sense?

Joe

Thanks, Film Joe

[Joe Huggins]

Doug, Sorry for the delay I was out of the office.

Yes, we can burn in code and manually conform. That is too time costly. I need an EDL file that will auto conform to source TC after an FCP edit. Did I miss something in your reply? How do we get FCP to take source code from an ALE made in Scratch when we output in realtime to ProRes for offline?

Thanks, Film Joe

[Joe Huggins]

Job, thanks. As we are more FCP proficient than Avid, where is the Transcode menu located? Can the transcode take place in the background while we work to create circle takes? And is the transcode render usually faster than real time or more time consuming on most 8 core MacPros?

Thanks, Film Joe

[Joe Huggins]

Thanks, we tried that with no luck. Adding a house ref signal made no difference. If we use the AJA VTR Xchange program the capture is perfect every time with a -1 Capture Offset. (by ‘perfect’ I mean starts at the correct frame of video with the correct timecode stamp … the out point is ‘sloppy’ and goes past my programed ‘out’ timecode … but the actual part of the clip that I wanted is correctly captured at 23.98.) So the Cinema Tools database is corrupt using AJA program since the head is right, but the tail bad.

That points to the Kona being OK, and FCP not being accurate. This is hard to believe that no one has had problems like this in FCP and a 23.98 capture. This is basic film 101 concerns right, back before we had HD dailies with no 2/3 pulldown at all.

Where are we going wrong? BTW AJA is moving with no phone support until Tuesday. – bad time to start a feature.

Thanks for any extra ideas.

Thanks, Film Joe

[Joe Huggins]

Thanks. We do calibrate to the Device Control Menu. Since it is on the Aframe on some takes and off the A frame on others, there is no consistent offset to correct this. If the takes were consistently off, that menu setting might help.

So I think the issue is still open. Any other thoughts from anyone?

Joe

Thanks, Film Joe

[Joe Huggins]

Thanks for the feedback. Here is some more detail.

We are the lab and did the telecine also.
Tape is a Telecine Master @29.97 and we are digitizing from the Telecine ALE file.

Tape is a single edit pulled at 1:00:00:00 till the end of the film roll (so no cadence change) all in points are on A-frames (and confirmed against the tape) on either a :00, :05, :10, :15, :20, or :25.

About half the takes come in on the A-frame inpoint … about half of them come in on the B-Frame video but show the A-Frame timecode ….
Any take that comes in as a B-Frame that is offlined and re-captured is about 50% likely to come in as a perfect A the next time. (Again, the tape has been double checked and is fine … and the tape will come into an Avid system with zero problems.)

Capture hardware is a Kona 3 card …. Compression is ProRes(HQ) …
system is a MacPro Quad-core. Problem is the same when digitizing to a fiber raid array or a FW800 LACiE so drive specs do not seem to be any relation to the issue.

Any new thoughts based on this input? Seems odd that the older SD dailies process has a ‘hit or miss’ error on A frames. We must have a setting wrong?

Thanks, Film Joe

[Joe Huggins]

David,

Wow, that’s a great response to an old post. You found it! I guess they have not updated that to current files like ProRes, but that is the exact page. Thanks again.

Thanks, Film Joe