Issue with Changing Sample Rates in X3

Any help on this one? Please, I'm desperate!
 
Here's an illustration I made that explains my issue in a nutshell. Anyone have an ideas how to wake Sonar up and rescan the wav files so they get properly rendered?:

Sounds like more info is needed.

SONAR does Sample Rate Conversion automatically when a file is imported to the project setting, so if you export tracks from the 44.1K project and import them into the 48K project they will be correct.
 
I am not sure what is occurring in your situation above, but it appears that the wav file may have a header on it that SONAR does not recognize... like it was NOT converted, but the 3rd party program told it to "play back at 48K." It looks like SONAR simply imported the samples (so is probably 10% shorter?). Has the audio that does play been pitch shifted up?

So I played around with this a bit last night looking specifically at the header trail and couldn't make it work trying different command line parameters with SOX. I checked the pitch on the resulting clips and it's fine.
 
Here's another possible piece of the puzzle though: going from 44.1k to 48k the resulting 48k will have more samples for the same time duration. What's probably happening is that although Sonar is recognizing the higher 48k sample rate there's probably a clip parameter under the hood that represents the total sample count and perhaps the 44.1k sample count is still associated with the clip instead of the 48k sample count. This is consistent with the symptom I'm experiencing in which the last 10% or so of the audio is missing in the clip.
 
Does this spur any more ideas with anyone?
 
Just to reiterate, if I import one of the underlying wav files into Sonar in a new location there is no problem; Sonar recognizes 100% of the audio in the clip at the correct sample rate. This also supports the idea that the wav files are not the problem, but it's more likely the glue in Sonar that binds the wav files to the Sonar clips.
 
More ideas anyone?

chuckebaby
Why not just as a test, try another sample converter

This one is free and I don't believe there are any limitations on this one. it is not as good as the pro version but it should work.
 
This will take the guess work out of "Is it sonar or is it SOX".

r8brain is where I started, so that's why I was trying SOX. But for sh!ts and giggles I tried r8brain again and still get the same result. As you can see I got identical results between r8brain and SOX. You cannot rule out that r8brain is using the SOX library. If I were going to write r8brain and there was already an open utility library, I sure wouldn't reinvent the wheel!
 
ORGINAL WAV FILE
Input File : 'Shes_Got_You, Barb vocal , Mix (587)BAK.wav'
Channels : 1
Sample Rate : 44100
Precision : 25-bit
Duration : 00:02:24.21 = 6359591 samples = 10815.6 CDDA sectors
File Size : 25.4M
Bit Rate : 1.41M
Sample Encoding: 32-bit Floating Point PCM
 
CONVERTED TO 48k WITH SOX
C:\......\Audio>sox --i "Shes_Got_You, Barb vocal , Mix (587).wav"
Input File : 'Shes_Got_You, Barb vocal , Mix (587).wav'
Channels : 1
Sample Rate : 48000
Precision : 25-bit
Duration : 00:02:24.21 = 6922004 samples ~ 10815.6 CDDA sectors
File Size : 27.7M
Bit Rate : 1.54M
Sample Encoding: 32-bit Floating Point PCM
 
CONVERTED TO 48k WITH r8brain
C:\.......\Audio>sox --i "Shes_Got_You, Barb vocal , Mix (587).wav"
sox WARN wav: wave header missing extended part of fmt chunk
Input File : 'Shes_Got_You, Barb vocal , Mix (587).wav'
Channels : 1
Sample Rate : 48000
Precision : 25-bit
Duration : 00:02:24.21 = 6922004 samples ~ 10815.6 CDDA sectors
File Size : 27.7M
Bit Rate : 1.54M
Sample Encoding: 32-bit Floating Point PCM

Well, I gave up and just did the least painful grunt work way as was suggested by Milton on a different thread:

Milton
Solved. I had to save my project as a "Template" in the templates folder (without deleting any audio clips!). I then started a 'new' project (with a new project name) using the just saved template but setting the new sample rate to what I needed e.g. 96KHz to 48KHz etc. Then after opening the newly created project I deleted all audio clips and then saved it again to the same "Templates" folder as previously in order to be able to use this new template at the new sample rate for future projects. Seems like a long solution, but it worked. This allowed me to keep all my tracks, buses, project views, etc. without having to re-build my project from scratch. Sure hope Cakewalk makes this task easier in a future update. They need to see how Cubase Pro handles this (much easier and quicker).

I agree with Milton that Cakewalk needs to address this. It looks like to me that there is some parameter attached to Sonar clips that contains the total number of samples and this parameter does not get regenerated if the sample rate is changed in the project.