Be Aware of "Orphaned" Slivers

I couldn't find the thread where this was originally mentioned, but to summarize, someone doing narration was hearing occasional, quiet clicks after rendering but not during playback. I also don't recall that he got an answer, but I had something similar happen last night. It was driving me nuts because I would keep hearing a click after rendering where seemingly, no audio material existed.
 
It turned out the clip was so short it was no wider than a grid line (!). It was not visible and the only way I stumbled on it was passing the mouse over a (supposedly) blank track, and the slip-editing tool appeared. As to why I didn't hear it during playback, the only theory I can come up with is that it was so short it blended in with the tracks, but when rendering a finished track which included a limiter in the output bus, it "tripped" the limiter. 
 
I found a couple other tiny slivers like this. As to why they existed, I do a lot of cut/paste/move etc.  Sonar defaults to snapping to the nearest audio crossing and usually, I don't want that. However, I don't always remember to uncheck it until after I've worked with the program for a while. Maybe I somehow ended up cutting in such a way that a little sliver of sound was left over after cutting at a snap point as part of a mass cut instead of at a zero crossing.
 
In any event, after I tracked these down and deleted them, the problem went away.

One possible source of the slivers could be the comping mode, I sometimes will split a clip in a take lane so I can edit out a string noise that may be present in all three takes. If you're not careful when trying to slip edit you can create additional take clips, (not in new lanes).
 
And further more they seem to be underneath the clip you've slip edited.  It's still all a little new to me and I can't confirm a recipe nor that these clips cause a clicking. Simply an observation of something that has frustrated me.

Just to follow up - it does seem related to the limiter in the output bus being "slammed" rather than the shortness of the clip. I should probably delete this thread so no one gets let down the wrong path.

Good point. Instead of deleting, I took out the material that could have led someone down the wrong path, changed the title, and left the useful stuff in.

I've had these slivers appear from time to time. Have never worked out where they come from.

John
My concern is that the idea that we can't be sure about playback absolutely mirroring render is a huge problem. Clearly I don't think Craig had that in mind when he posted this thread. I think he was trying to solve a problem with clicks and pops that some few have reported. 
 
However the ramifications of saying that playback can miss things and render may not is extremely troublesome. 
 
Therefore this is a very important thread that needs to stay put and hopefully a good resolution will come out of it.
 
I'm hoping to hear from CW and members that can shed light on this.  

There are a few ways in which playback and render could differ, just down to settings. Like whether or not 64 bit processing is on for one and not the other. Also, anything in a "live" audiosnap clip will always be different for playback and render to some degree. That's off the top of my head, I'm sure there are more examples.

I'm not sure I've ever found a sliver, but I could see that it could occur and that it could cause a pop/click when it was truncated (where the wave wasn't at a 0 crossing as the clip abruptly ends).
 
Good stuff.

John T
 
There are a few ways in which playback and render could differ, just down to settings. Like whether or not 64 bit processing is on for one and not the other. Also, anything in a "live" audiosnap clip will always be different for playback and render to some degree. That's off the top of my head, I'm sure there are more examples.

Yes, that's why I deliberately threw in the "All other things being equal" caveat. Barring differences in render settings from real-time settings and plugin randomness or response to high data rates, SONAR should (and generally does), produce the same output on render that you hear on playback. And something as problematic as pops vs. no pops should not happen in any case (with the possible exception of stretching artifacts from differing Audiosnap algorithms). Even that difference can be eliminated by the "Same as Online" rendering option for the Groove Clip and Percussion algorithms.
 
Like john, I just want want to be clear that the audio engine itself does not behave differently on Fast Bounce or Export than during real-time playback, other than in the rate at which buffers are processed and that the buffers of processed audio get written to file instead of being handed off to the audio driver.
 
 
 
 

Anderton
Good point. Instead of deleting, I took out the material that could have led someone down the wrong path, changed the title, and left the useful stuff in.

Too late. I'm already half way down the wrong path and since you deleted that material I can't seem to find my way back. Can you please reinsert it?
 
 
Just kidding, of course.