IF you'll read the AEM thread - yes it can be changed on the fly. That was one of the last questions Jason answered. As for doing it "right" - look at it this way...
The stock ECU is a black box. It's timing curves, it's fuel curves, it's temp compensations, and it's altitude compensations are all pretty much UNKNOWN. What you're doing with those add-on boxes is fiddling with the factory sensors - outputs and inputs. You're changing what the ECU sees so that the ECU will do what you want it to do - or at least you hope it will. Since you don't know everything that's in the factory ECU's programming you cannot know for sure what the results will be!
Here's a good example - look at the Mustang guys with their MAFs calibrated for bigger injectors. Do they realize that they hose their spark tables switching to those units? The MAF allows a different injector to be run by modifying the MAF's output voltage curve. To get a bigger injector to idle it needs a shorter voltage P/W - to get this they setup the MAF so that it puts out s signal telling the ECU that the amount of air going through it is actually LOWER than what's actually occuring. That means that the engine thinks it's seeing lowspeed idle airflow when it's actually receiving much more going down the road, as a result it outputs spark designed for IDLE. Changing the MAF changes things across the board! The correct way to have done this would've been to modify the variable in the ECU to tell it that it was running larger injectors.
So, how does that effect you? What do YOU change when you goto bigger injectors? What does the VPC output when you remove the MAF? Does it mirror what the original MAF output for various airflows? It might be close but MAP infers airflow over MAF's actual measuring of airflow. If the lookup tables aren't just right it will infer incorrectly and send the wrong signal. That signal might still let the car run but it could be slightly rich or slightly lean - you'll never know.
Fiddling with the outputs of the ECU is probably smarter. Take direct control of the engine this way and there's less chance of errors occuring. Except now you've got to tune those boxes. Using more than one? Have fun tuning them and getting everything to interact correctly. I'm betting you've not got nearly the tuning range with those boxes either. You'll have a few points in the RPM curve where you can adjust up and down what the ECU is sending out or maybe take over completely. But what about the RPMs in between those points? The box will interpolate, hopefully smoothly, what it does between them. Except that too might not be right.
In the end it's a compromise. Yeah it obviously can work as lot's of people are using these things but do you really think they won't pick up power going to a full on standalone that can adjust with MUCH more granularity? IF this box is everything it's supposed to be I'm betting that people who make the switch pick up power, hopefully lot's of it. Maybe not peak power but power for sure.
So, $700 to raise the rev limit? Or $1300 to be able to do anything you want. It's up to you but I was simply suggesting which might be a better more flexible way to go. Hrm, so what questions didn't Jason answer that you asked?