Just had a listen on the cans. I'm with Jeff that the mixes are too different to compare ... sonic character is so radically different that I cannot focus on subtleties. I'm torn between the mixes liking one element better here, another in the other mix...
Actually I tried something similar in the past few days while trying to get a grip on Mixbus 3.3. I did use only 5 tracks (1x lead vox, 1x piano, 3x back vox), which leaves a lot of space in between to listen deep. And I did limit myself to using the same reverbs in either DAW, then iteratively working either mix. Sonar first, Mixbus next. Bounce. A/B compare next day. Make notes. Work both mixes again ... currently about to start the 3rd iteration; about to do some more listening/making notes tonight ...
While I can't put these mixes out (as they originate from a rough demo tracking session of material that is not to be released for another half year), this is great learning experience.
Round #1 (due to inexperience with Mixbus) definitely ended 1:0 for Sonar (the first Mixbus mix couldn't keep up)
Round#2 had Mixbus to tie with Sonar. Mixes got closer together sonically, and I did start to like the Mixbus version ... and I had far fewer plugs in the Mixbus project ...
I did realize quite quickly that both DAWs will make you go different ways. So the workflow has an impact. Even if you have a vision that you work towards. Since raw tracks were just quick scratch tracks from a demo session, mic/preamp settings did not properly suit the voice. Somehow Mixbus did a better job in smoothing out some of the harshness of the lead vocals, but in neither DAW I could polish the turd far enough ...
I'll definitely do a round #3 because I want to dig deeper in Mixbus ... and I'm not sure where it ends. While Mixbus definitely has some nicely tuned EQ bands at hand, it may fall behind when automation gets into the picture (first automation attempts in Mixbus were neither stable nor really working that well) ...