Sorry for the late replay.
1) Yes, PC is the source - there's ways of implementing it without a PC - MiniDSP has released Ambio units which will get you pan ambio - no convolution, but if you reduce the volume of the back channels it will expand the width a slight bit further according to some of the pages, just dial in where it sounds good and leave it.
2) Complexity and cost most likely. Depending on what you're intent is a simple stereo setup can be done with a single VST. When you involve convolution, panambio, etc. cost and complexity goes up. I'll be happy to answer any setup questions I can as to what's the best solution IME. You also need a a somewhat faster PC if you plan on doing anything on the PC while it's doing its magic - I use a i5 2500K and hit roughly 5.5% CPU usage in Audiomulch. The latency isn't too high but faster is always better in this case. This is of course if doing anything beyond simple stereo, if only doing stereo and music it's unlikely even somewhat older PCs will have much an issue.
3) Your aversion is somewhat justified - there is a lot of software chaining going on in my config. You have to do some research and work to get it to function how you want. Then again, with someone that has already done it the setup should be substantially easier . . .
If you want to just do a simple stereo test the VST alone is $10 at electro-music. You pay, and they eventually email you a link. Turn around and make sure you get version 1.2. From there put your speakers right next to each other, sit so you have a total of 30 degree angle (15 per side) max. Use a plugin to use the VST if using FB2K or use it natively with jriver. Pop up the VST and lower the delay to 45ms, up recursion to 50%, volume max, zentum up a bit if you want the center to be a bit forward in the stage (I set it to one), and adjust space till it sounds reasonable (I have it at 3.0 myself). That will give you the effect with a very straight forward sound, possibly in the studio with the band is how I'd describe it. To get the "live" type sound takes a lot more work, and the more realistic you want the more work/equipment it needs. I think mine is a nice happy medium.