I'm bumping this tread because it really seems that timestretch has been downgraded in X2.  IE, it's only accurate to one decimal place instead of 3 decimal places?  Can someone from Cakewalk please confirm or deny this?
 
Basically it seems timestretch is no longer useful for large files?  Are there any plans to restore it's functionality?
 
In Sonar 8.5 you can time stretch to an accuracy of 3 decimal places.  Eg 110.867%.  In X2 you can only time strecth to an accuracy of 1 decimal place.  Eg 110.9% This makes time strectches to target tempos and target lenghts for big files just about worthless.  What's up with this?
 
Another big AARGH....
 
Am I missing something?  The Izotope timestretch algorithm in S8.5 is still pretty awesome.  VERY little artifact.  Is this some deal they made with Izotope to limit the usefulness in X2?

There are better tools for time stretch. Why not use your mouse to stretch it? If that isn't good enough for you, then you should use Audio Snap or a non native program like Melodyne.

I'm talking about audio snap. In X2 the stretch percentage in audio snap is only accurate to 1 decimal point. In S8.5 it's accurate to 3 decimal points! 
 



>>>> you should use.... a non native program like Melodyne.
 
That's the thing.  Sonar's time stretch algorithm is GREAT.  It's actually an Izotope algorithm.  My beef is that they seem to have crippled it in X2 compared to S8.5.
 
>>>>There are better tools for time stretch.
 
I'm not sure there are in Sonar's price range.  I've used the radius advanced algorithm to strech vocal, piano, bass and instrumental tracks VERY convincingly.  It's not perfect for stretching mixes, but for individual tracks it's still pretty amazing.  Very sad that S8.5 is superior to X2 for time stretching!
 
Peace,
 
Ian
 

I agree with Ian, I never understood that change, those extra decimal places made a big difference for time stretching accuracy.

  You can actually enter additional decimal places and they have the expected result on the New duration, it's just that the displayed % is rounded to the nearest 0.1. You can also manipulate the Duration at the 1/1000th of a second level. 

Bigdogs - I don't think that true.  If I timestretch to a target tempo in S8.5, at the end of the song it's still clicking along w/ the metrinome.  In X2 if I do the same thing, it's way off by the end of the song.  I think the displayed decimal points are what you get in X2.

I see that if you type in a value of 3 decimal points, it changes the length.  But when you play around with this, X2 does erratic computations-IE, it doesn't work.  It doesn't seem possible to stretch accurately to a target tempo in X2, even with trial and error.  I really think they've made timestretch less functional in X2!

I'm bumping this tread because it really seems that timestretch has been downgraded in X2. IE, it's only accurate to one decimal place instead of 3 decimal places? Can someone from Cakewalk please confirm or deny this?

Basically it seems timestretch is no longer useful for large files? Are there any plans to restore it's functionality?

Can someone from Cakewalk please confirm or deny this?

As you probably know, the Bakers don't watch the forum that closely. If you really want this fixed, you'll want to file a formal problem report. Even though one decimal point is probably "as intended" for this release, I suspect they would probably get it fixed sooner rather than later, since it was better in earlier versions.

I can check it, but it seems pretty clear you've confirmed it won't do 3-decimal place precision any more.

A workaround might be calculate the new end point yourself, and drag-stretch the clip it to that point.


EDIT: I checked it out. It seems Bigdogs is right... and then some. Not only can you enter a precision of three decimal places, you can enter as many as needed to get 1-sample precision, and it will stretch to that precision, even though it will only display 1 decimal place in the inspector when it's done. Slipstretching also works.

If you have an example of a particular  stretch scenario that didn't give you what you wanted, please post exact numbers (e.g. start end, new end, expected stretch percentage), and I'll try it.

Whenever I've had odd length results from audio snap, I've found it's down to the "average tempo" setting being incorrect. Sometimes Sonar's guesses at the original tempo are a bit dodgy, and it's always worth checking this setting on a per-clip basis. It's also not obvious to me why this would affect percentage-based stretches, but it does seem to.

John and brundlefly.

I'm playing around with the timestretch in a different song than before and it seems to work OK in this song.  So I think you both may be right.  X2 may have made some incorrect guesses before.  Or it might be a programming thing with the new release.  It definitely doesn't work correctly in the previous song though.  So now it's an intermittent problem... the worst kind.  But at least CW intends for it to work!

Thanks guys.  I'll update if I learn something more.

Ian