There are many ways to use a modeler (or profiler). Some approaches, like going modeler (or profiler)>tube power amp>guitar cab, will limit flexibility (in that configuration), but result in a sound closer to that of an amp into a guitar cab. Some users are after that, and it works for them. Others (like myself) want the full flexibility of the system, which requires an accurate power amp and cab (Full Range, Flat Response). Using this approach, you will not get the "amp in the room" result, but you will get a result which accurately represents the amp/cab models. With the latter approach, you are after the "mic'd cab sound", not the "amp in the room" sound. They are different, one just has to go the direction that makes them happy. I've hooked my Fractal up with those 2 approaches running in parallel (SS power amp driving 2 4x12's, in tandem with direct to 2 powered FRFR cabs). The signal driving the 4x12's had power amp modeling turned on, but cab modeling bypassed. The signal driving the powered FRFR cabs had both power amp and cab modeling enabled. While that setup sounded huge, it was primarily for self indulgence, and would not be practical to take back and forth to practice. The modeler>powered FRFR 1x12 cab gives me the full flexibility of the system, in a small footprint, which is easy to transport. That small setup, with it's wide dispersion cab, will out perform a 100 watt halfstack (I've used both).