Sanderxpander
I'll be interested to hear how this plays out. I would normally first doubt your perception but since you say you are aware of this effect and it was very clear beyond this, my guess goes towards some unintended setting. I have a really really hard time believing that DAWs would have a different sound from regular stereo WAV playback. Even the voodoo differences people normally talk about (DAW X sounds better than Y) tend to focus on things like summing.

We are on the same page. I have posted often that the sound of a DAW should be the same as any other DAW except the few that have DSP going on all the time; Mixbus for example.
 
 

Did you compare you mixdown WAV in Samplitude to your running project in Sonar or did you load the WAV into empty projects in both DAWs?

EDIT: sorry, just read your OP again - same file. Curiouser and curiouser.

However, it seems to me that if the DAW imparts a sound that should hold through mixdown. So perhaps you could bounce the Samplitude file and upload it together with the Sonar-created original. Perhaps one of us can find a clue what caused it, or at least admit hearing a difference.

My first instinct upon reading the opening post was that this might just be the result of your unfamiliarity with Samplitude. That would my assumption if it was me - that I was committing some noob mistake.
 
As you're well aware, loading a stereo wave and playing it back in its original file format without any processing (not even volume adjustments) should yield exactly the same results regardless of which DAW you're listening to. Pan laws do not come into play.
 
You could export the wave file from each DAW and perform a null test. If they null out perfectly (as they should) then you'll know with certainty that it's expectation bias at work. If they don't null, then there's been some kind of inadvertent processing applied, e.g. wordlength reduction or accidental application of dither. 

bitflipper
My first instinct upon reading the opening post was that this might just be the result of your unfamiliarity with Samplitude. That would my assumption if it was me - that I was committing some noob mistake.
 
As you're well aware, loading a stereo wave and playing it back in its original file format without any processing (not even volume adjustments) should yield exactly the same results regardless of which DAW you're listening to. Pan laws do not come into play.
 
You could export the wave file from each DAW and perform a null test. If they null out perfectly (as they should) then you'll know with certainty that it's expectation bias at work. If they don't null, then there's been some kind of inadvertent processing applied, e.g. wordlength reduction or accidental application of dither. 

Dave I thought of that too. I wouldn't expect them not to null. No processing is going on therefor no change to the file. 
 
I was mistaken on one issue. Samplitude Pro X2 has its own ASIO driver and that was engaged when I listened. In Sonar Platinum I use WDM/KS. I had forgotten I had changed this in Sam for various reasons.
 
One thing to be aware of here.  I am no noob. I know Sam well enough. Also, even if I didn't playing back a wav file with nothing else going is not rocket science. 

Sooo... Don't leave us hanging - does it sound the same when they're both using the same WDM driver?

Yes they sound very very close. This is using the same driver (WDM) in both.
 
Keep in mind I thought they were the same. But now I recall switching to Magix's  ASIO low latency driver because certain things within Samplitude uses it and are unavailable with WDM. I wont make that mistake again. 
 
I wouldn't have noticed this if I was just doing a recording or creating a song with Sam. It was only because I knew how this song was supposed to sound that I got curious. Nor did it occur to me that a third party driver could impact the sound so much in such a bad way. 
 
The mystery has been solved. 

I wouldn't have expected that either. Thanks for the heads up.