When I was getting ready upgrade my system, I borrowed a couple different charge controllers and had a couple friends come over to help me play with and decide what I wanted to do, and consume a few beers of course. With the mppt controllers we had to use for series work, and using the same controller, just different panel configurations we didn't see any significant difference in output between the two configurations unless you shaded a panel in parallel configuration. The controllers we tried really didn't like series parallel of any sort with shading. Pretty much have to use a separate controller for each sting.
Using a PWM controller, parallel config of course, shading of panels resulted in predictable results. We couldn't of course play with series setups with these controllers.
As for reliable information on internet forums, one of the first things I learned when I first got into testing and evaluations decades ago was to trust the interments. People tend to see what the expect to see, or want to see. Unfortunately, it's often human nature to set up data monitoring, logging and interpretation to produce the expected results. Just human nature.
Myself, I tend to want to see for myself as much as possible, but just end up going with what I know will work under the circumstances I'll be working in and then not worry about it. The system I went with breaks all the so called rules. Wet cell batteries, 10 panels in parallel, PWM controllers for a 1,000 watt system etc. Going against all common knowledge or not, seems to meet our needs really well in just about any conditions.
