OK, I’ve got a few thousand audio clips, all with associated cycle markers, and I am exporting the markers as .CSV, editing the markers in Google Sheets, and then importing the edited markers back into Nuendo. I’m doing this so I can perform a cycle export of all the thousands of clips, each with the proper file name as reflected in the edited cycle marker list. Editing all this stuff is a lot easier in a spreadsheet.
But, when I re-import the edited cycle markers into Nuendo, they don’t land at the right spot. Look at the attached screen shot and you can see there’s about a 15 ms difference between the placement of the original marker and the re-imported marker.
The differences are not uniform. Some are good, others are pretty far off. But the placement is rarely exact.
Is this because the markers are exported in timecode format, not seconds, so my markers whose starting edge fall between frames just get dropped off at the closest frame?
If this is the case, is there an option to export markers using seconds instead of timecode? If so, I can’t find it!
This has been discussed before I think. No at this time csv only supports TC and is limited to frames in positional accuracy.
You should still be able to do what you need to do, but severe workarounds are required.
Is it worth it?
Probably not if what you do is a one time thing. If it is part of workflow then it might be.
I have not done this specific work myself but as I have done other hacks to solve issues I know parts of the workarounds.
Using Mac quickkeyes or a windows equivalent you can
Build automated hacked workflows.
Using a spreadsheet and a exported AES31 version or a track archive of your markers you can edit outside of Nuendo and using a series of automated search and replace to rename the markers in AES31 or directly in the track archive xml.
It will take time to figure out how to, it will only be worth it if it is really important.
Personally I would do it the other way around.
Rename the files according to the edited list after export using scripting.
Or perhaps the rename after list function in nuendo it self can help as well together with some unorthodox thinking.
But, as long as the csvs are limited to Tc there will be issues like this.
The easiest thing for to do is to simply snap each clip to the closest frame boundary, then drop the cycle markers, export the marker list, edit the list, and then re-import the edited markers before exporting the final clips with the correct names.
In the past, I’ve renamed the files after export, and this has worked fine, but this time around getting the file names correct right out of the gate would work better.
The clips are recorded piano samples and the final product doesn’t need any time reference, so it doesn’t matter if the start time for the clips is altered. And while the cycle markers’ end times will be incorrect, it doesn’t matter for export since the samples will have been faded out by then anyway. Inaccuracies of a few dozen ms one way or the other are OK for the cycle marker’s right boundary.
But it would be nice to have the option to be able to export markers in seconds. Mining the track archive or AES31 files to edit this data is a bit more work than I’d like to do!
Yep, those types of hacks are not for the faint of heart.
I built a rudimentary reconform tool before reconform was implemented as a test and have also built other workflow specific things a few times.
I built a rename event from marker to create a PT ADR session. It worked pretty well but as any such hacks they are limited to window configurations and personal preferences in how you want to work so are impossible to share. But if the need is big enough…
It really isn’t hugely difficult, but the time it takes to figure it out often just isn’t worth spending in most situations.
Yes, it’s a real adventure coaxing everything to be where you want it!
The piano sample projects involve many thousands of samples recorded at multiple velocities and mic perspectives, and the easiest way to get these things to drop where they’re supposed to go into the sampler is the name them correctly, often according to midi note number, key name, velocity level, round robin iteration, etc.
For me, it was a combination of Notepad++, Google Sheets and Winsome File Renamer to rename the clips, and a bit of Nyquist code written for Audacity that will assign RMS value to each of the clips to help in assigning correct midi velocity levels.
