Jeff Evans
I did an experiment some time ago summing an exact mix into 4 DAW's. Logic, ProTools, Sonar and Studio One.  No plugins used though just level and panning. (Panning was also only L,C,R as well.) I got identical result from all 4.  I could null any two of them.  A room full of experts could not hear it either.  The moment you start using plugins in any of them, then all bets are off.  There will always be differences then.  It is not the summing engines that sound different.  It is everything else that sounds different. 
 
I also did another experiment earlier too. I had a turntable with the finest pickup known to man playing back a very high quality vinyl recording (Sheffield Lab stuff) The pickup had frequency response to 50kHz as well. (Ortofon SL15Q) Used a $1000 RIAA equaliser preamp too.  This signal went to one side of an A/B switcher.  I also fed the signal into A to D and D to A converter at 44.1K and 16 bit.  And put that on the other side of the switch box.  So in one position you were hearing the turntable direct and in the other the A-D D-A version of that sound at 44.1K 16 bit.  Perfectly level matched of course.
 
Room full of engineers with amazing ears could not pick either.  So to papacucko I say horse pucky.  You would not have a hope in hell in a blind test like this.
 
On this sampling rate issue you might want to read this:
 
https://www.mixonline.com/recording/emperors-new-sampling-rate-365968
 
 

We are on the same page! 

Jeff Evans
On this sampling rate issue you might want to read this:
 
https://www.mixonline.com/recording/emperors-new-sampling-rate-365968

 
I agree that playback at 96 kHz is a non-starter, I never met anyone who could tell the difference between a 96 kHz master and the same master sample-rate-converted to 44.1 kHz. BUT do remember that if you're recording inside the box, 96 kHz can make a substantial difference in sound quality for reasons I've explained many times over. Fortunately SONAR's upsampling can give equivalent results but AFAIK, it's the only DAW that offers this option.
 
 

Jeff Evans
I did an experiment some time ago summing an exact mix into 4 DAW's. Logic, ProTools, Sonar and Studio One.  No plugins used though just level and panning. (Panning was also only L,C,R as well.) I got identical result from all 4.  I could null any two of them.  A room full of experts could not hear it either.  The moment you start using plugins in any of them, then all bets are off.  There will always be differences then.  It is not the summing engines that sound different.  It is everything else that sounds different. 

 
Does anyone mix songs without using plugs?
 
If "all bets are off" when plugs are used, then yes, all daws sound different when used to their proper potential.
 
 
 

 
 

A Horses arse statement if ever I heard one.
One can bet he has never even tried Sonar.
He can be forgiven as we know he is blind for not looking and not take too much offence to his uninformed blindness.
 
 

rabeach
Sonar's lowpass filter's passband frequency response in it's sample rate converters (SRCs) is more narrow than protools, logic, cubase, reaper, mixbus, samplitude, and fl studio. The design parameter is that it should be less than one-half of the sampling frequency. So it complies with that but I think cakewalk would have to address why they opted to design it this way.  http://src.infinitewave.ca/   The data was collected from X3.
This might impart some difference in sound that possibly could be heard by discerning ears, certainly it could be measured.

That's interesting, but worth noting for the record that it's a bit of a fringe factor. I work in Sonar all day every day, and I reckon I do sample rate conversions maybe twice a year, tops.

Jeff Evans
I did an experiment some time ago summing an exact mix into 4 DAW's... I got identical result from all 4.  I could null any two of them.  A room full of experts could not hear it either.  

I sure hope a room full of experts could not hear a difference between files that null with each other! Otherwise, wait, what were they experts in again? 
 

rabeach
Sonar's lowpass filter's passband frequency response in it's sample rate converters (SRCs) is more narrow than protools, logic, cubase, reaper, mixbus, samplitude, and fl studio. The design parameter is that it should be less than one-half of the sampling frequency. So it complies with that but I think cakewalk would have to address why they opted to design it this way.  http://src.infinitewave.ca/   The data was collected from X3.
This might impart some difference in sound that possibly could be heard by discerning ears, certainly it could be measured.

I'd forgotten about that site, good to see it's still there and relatively up to date. These measurements have to do with SRC from 96 to 44.1, so if you're not doing that, it obviously doesn't apply. Still, it's good to be reminded... if absolute pristine quality is of the utmost importance, maybe it's better to export a mix at 96 (if it was recorded there) and downsample it as needed with something else... I like the look of the iZotope curves (and have used them for years) but in fact I seriously doubt that in most listening environments it would make a perceptible difference. Maybe even to a room full of experts! (see above )
 

Hey that was real funny it made me laugh!  Yes all the experts were there listening to the nulls and we could all hear the silence so well!   I have to be careful how I phrase things!
 
Well doing a mix with no plugins is rare for sure.  But these raw tracks were just amazing.  It was at the time I was working for Roland Australia selling the V700 system.  They employed this great engineer to record a great live band. Very classy indeed.  The tracks were so well recorded all you had to do was push up the faders and get a balance, some panning and it was a perfect pristine mix!  (Masterclass in mic choices and placement)  This is rare I do agree.  But ideal for comparing summing busses though. (mainly because the end mix sounded so good with no processing anywhere)
 
I guess the difference might be in the stock plugins and of course people using different third party plug-ins set differently in each DAW.  If you used the same third party plugins in all 4 DAW's and applied exact settings to each I would imagine those nulls would also happen.  I tested the nulls at home before the listening test with the engineers of course but they did null rather well in fact.  So it does sort of prove at least summing engines are very similar.  The listening test was also pretty interesting.  No one had any idea what DAW they were listening to.
 
By the way I did a similar test recently comparing Studio One to Mixbus.  I recorded my son playing drums with a multi track session.  Set up exactly the same mix on both systems. i.e. no processing in Studio One.  However with Mixbus I also urned off all processing and used no saturation etc.. Also got a complete null as well.  So there is no magic Mixbus sound buried in the summing engine at all in fact.  The Mixbus sound (and it is good too) comes from elsewhere. 
 
I agree with Craig re using higher sampling rates during the production though.  Native Instruments Prism can sound pretty different at 96K compared to 44.1K for sure.  (for some sounds, not all)  But the 96K sound can be downsampled back to 44.1K and the 96K sound is still there.  So yes, higher sampling rates (don't need 192 though) can be useful during the production.  Not so needy at the end though. 

It's not unusual for specific converters to do better at one sample rate over another, but that's not due to some intrinsic superiority of a particular rate. Rather, it's because the engineers (talking real engineers, not us knob-twisters) made calculated design decisions based on how they presumed their products would be used. Prism, Apollo, Lavry, etc. are optimized for 96 KHz.

All good points.  Just FYI, I am talking about mixing several VSTI and audio tracks in sonar at 96 as compared to 44.   If you don't hear a difference in a mix or of audio at 44 and 192.   I doubt I can discern an individual track.  With VSTI resolution and upsampling if you have not tried this and your processor can handle it.  Amazing.  (I should mention I still use an RME multiface2).  I do think there is no reason to doubt that audio is just as subjective if not more than other things.  Have you ever switched back and forth between Fox news and CNN? Or paid attention to the officiating in professional sports even with instant replay?  Sometimes you find yourself looking at your friend in amazement like "are we looking at the same thing?"  Funny.  I downloaded pro tools first on my rig last night.  It hosed my ilok manager installation and would never actually start just kept spinning. tsk tsk.  Silly wabbit.