Glued items should look identical to old items, not...?

[Fabian]

I have a bunch of items that I have chopped up and adjusted play rate to fit the grid. Crappy guitar player (that's me!). Then I copied the whole bunch to a new track and glued the items there. Now that I look at the two tracks, they do not look identical, not at all. See here:



On top, the original chopped items with different play rates. Bottom the glued version. Obviously, they do not look identical.

Strange thing is.. they do null. But since they look so obviously non-identical, there's gotta be something not right with the waveform drawing... or?

[timlloyd]

odd ... what if you zoom in?

edit:
Although, actually it's not that odd, because surely one would expect the real-time graphical resampling/stretching to be of lesser quality than the actual audio processing, and therefore probably look different.

[Fabian]

Originally Posted by timlloyd

odd ... what if you zoom in?

edit:
Although, actually it's not that odd, because surely one would expect the real-time graphical resampling/stretching to be of lesser quality than the actual audio processing, and therefore probably look different.

Hmm... maybe you got a point there...

Zooming in reveals yet more non-identicality. But I guess it makes sense that the graphical representation of the time-stretched/resampled items is not as accurate as the graphics representing their glued counterpart. Still looks strange though...

[AdamWathan]

The graphic representing the time stretched wave form is not accurate AT ALL, it just stretches the graphic. If you actually played what the graphic is showing when you adjust the rate slower, it would play back at a lower pitch. I imagine it would be a lot of extra work on the CPU to redraw the peaks on the fly for every pitch adjusted item.

This is the same in all other DAWs btw... Whenever you use Flex Time or Elastic audio it just visually stretches what's there. If you bounce it down, it won't look like that at all. The bounced version is the actual waveform created using the pitch shifting algorithm.

[Xenakios]

Originally Posted by AdamWathan

it just stretches the graphic

Or the peaks data of the original audio is read at a different rate.

[KevinW]

Sorry if this is a bit unrelated, but when you glued the items, did you do it all in one step? As in, did the items have varying playrates? Cause when I've tried to do that, I get some weird results. So what I do is "glue" each item to itself, so it's playrate goes to 1.0, then glue them all together as a separate step. I've never noticed the waveform looking different, btw.

[Fabian]

Originally Posted by KevinW

Sorry if this is a bit unrelated, but when you glued the items, did you do it all in one step? As in, did the items have varying playrates? Cause when I've tried to do that, I get some weird results. So what I do is "glue" each item to itself, so it's playrate goes to 1.0, then glue them all together as a separate step. I've never noticed the waveform looking different, btw.

Yes, they had varying play-rates, and I glued them all at once. I hear no strangeness, and the two versions do null, so I guess these are only drawing artifacts.

And, no, I've never noticed different looking waveforms either, earlier, but I've also never compared them side by side like this before...