I own a number of engine simulation programs ranging from early Desktop Dyno to the top-of-the-line Engine Analyzer Pro. EA Pro gave me sort-of-accurate results AFTER I gamed a bunch of the parameters to try and get it to replicate actual engine dyno results for known BBM combinations. Plugging in the actual data prior to that gave questionable-at-best results, which is I why I felt the need to jack around w/ the features to try and get more realistic results.
After I had gamed my "sort of accurate" EA Pro baseline, I provided the actual inputs for someone else to run in the latest version (at the time) of Dynomation. Not only were the results very different from each other, neither were that close in terms of outputs & RPMs to the actual data from my engine dyno session(s). The Performance Trends Engine Analyzer programs -- I have (had?) -- showed more realistic TRENDS from certain component spec changes than the Dynomation program did, but weren't necessarily accurate from a predictive standpoint.
I gave up on engine simulation programs after that and consider them all toys, or at least the ones I can afford to purchase. If I ever got the itch to buy another program (NOT likely), I'd look more seriously at the latest PipeMax release (4.?). I have an older 3.9x version which is far less robust in features, but at least provided some usable outputs for what I was doing.
EDIT: This is my "Last Hurrah!" on the subject: If you're a Speed-Talk member and can access the Advanced Engine Tech forum, you might want to read through this:
https://www.speed-talk.com/forum/viewtopic.php?f=15&t=49826&hilit=mach