So I knew about this but am rewatching the old X1 Advanced Workshop vids and at the very start where they show the timestretch cropping with the Smart Tool gave me some idea for an old live of the floor album that's been perplexing me for a while. Basically it's a live session recorded without a click and the tempo is all over the place. I've been dreading going in there to perform all the complicate Audiosnap procedures require to fix this ENTIRE album.
 
Really the main thing that needs to be done is for me to just get in there and make sure that when things start drifting I tighten up where the measures land on the timeline. I've been studying Audiosnap for this and there's a ton of useful info out there on the subject but now I'm thinking I could get a HUGE bulk of the work done much more quickly if I:
 
1) Figure out an approximate tempo
2) Set the start of the songs to an appropriate point on the timeline
3) Chop up all the tracks into sections each starting/ending on measures
4) Select ALL the clips for each section (as in all the instruments in the song) and use the Smart Tool Time Stretch cropping to line up the edge of the clips to the timeline allowing for a slight crossfade at each split
 
Is there some potential downside to this method? I figure that would get the bulk of the work done very quickly then I can deal with the smaller beat by beat correction using Audiosnap. Do I lose out on of all the offline render options doing it this way?
 
Anyway... just thought maybe someone has tried this and could provide some thoughts about how well it worked versus the more laborious Audiosnap methods.
 
Thanks.
 
 

When I tried fixing and/or changing the tempo of a project, I would first bounce all the tracks.  Next I change the tempo as you mentioned to match as closely as I can to the most consistent track (but usually focus on the drum/percussion track at first).  Then I would actually convert the tracks to groove clip.  Do a little stretching to match things up as close as possible.  Now I bounce the tracks again.  This was pre AS, so with it I would then do the tedious work of splitting where necessary and lining up transients to a preferable point in time.  Now at that level it was a matter of moving just the transient in question of moving some or all of the corresponding note transients in the clip.
 
That being said, going from your set 4, I am not confident that selecting all clips and doing a single stretch is going to bring things, that being the transients, into alignment, unless the performances are tight enough already.  I generally don't like using fades between clips if I can avoid it (which involves more detailed work when splitting at zero crossing).
 
Eh, my method/pov may not be helpful, but that's where I am at with the idea.
 
Kind regards,
 
Living Room Rocker

Hi, LRR. That's kind of how I had been viewing it all before (as in trying to squeeze transients across all tracks and/or screwing around with Groove Clips) but really the less minute changes to the original files the better.
 
For example (and I think this will keep everything in phase as it was originally recorded):
 
Measures 1-4 fit but measures 5-8 go off so the the start of measure 8 is a little late (the band slowed down slightly).
 
I would create a split across all the tracks at the REAL measure 8 on the original tracks, select all the tracks then hold the Alt key while cropping to get it in line with the project timeline.
 
This would leave either a gap or overlap (depending on whether the band sped up or slowed down). At which point I would but the edge of the next section (all clips) up against the end of measure 8 which is now in time.
 
I would extend/crop the start of the next section slightly into the previous section and create a slight crossfade then if the end of that next section is off I would stretch crop the end of it to fit the timeline and repeat the process.
 
I would do that incrementally until the end of the project and add in any need time sig and/or tempo changes as needed.
 
That SHOULD in theory preserve the original performance as much as possible.
 
I would then render the tracks hopefully being able to choose the best offline render settings (and perhaps I need to open the AS pallette to get those options but I thought that could be done via the Inspector on a track by track/clip by clip basis).
 
Now I've got (hopefully) a good, in time interpretation of the live session. From there I could do the actual Audiosnap transient editing to correct the drums, tighten up the bass, guits and vox to get everything super clean and render again.
 
Then I can go into overdub mode and add all the extra parts necessary like leads or double rhythms or whatever.
 
IDK... I guess I should just try it out. It may crush my system trying to stretch that many tracks at once but it isn't THAT many and I could crank my buffers through the roof while I do it.
 
Meh, I'm just kind of wondering about this method because all the schemes I've seen in regards to this seem highly labor intensive whereas this doesn't seem all that hard at all. I'm probably overlooking something.
 
Thanks and cheeeeers.