Well, thanks to my friend Jerry C., I've got an older NCE Powerhouse Pro system in my hands to try out. I'll share today what I've learned about setting up and using it. As part of that setup, I have also added JMRI via an old notebook that I had and have even used my Samsung S3 phone as a remote throttle :0).
While the NCE system that Jerry has is an older unit, upon study of the online documentation for the latest equipment, I've not found much in the way of functional differences. It is clear that they simplified the wiring hookup and I'm guessing that they have made performance improvements. I'm also guessing that the price is more now ... this system is of the pre-2004 variety.
In summary, my experience has been reasonably good with installing and operating the NCE. Jerry was kind enough to let me borrow two AZL GP38-2 locos with Digitrax DZ123Z0 decoders installed so I didn't have to mess with that.
Starting with using just the NCE Powerhouse and a loop of test track, I found the hook up to the track was relatively simple. The interesting thing about this version of the system is that within the box is actually two physically separate products, one being the DCC control station and the other being a 5 Amp booster. The two are connected externally with a pair of wires - I note that the newer system has eliminated these wires so the two sides of the system have obviously been connected internally (or even combined on a common PCB?).
Using the supplied Pro Cab, was able to almost immediately run each of the test locos using the default short address of 003. I also was able to fairly quickly, using the instructions in the manual, reprogram one of the locos to a long address matching the loco running board number (3108) and again run the engine successfully independent of the other loco. To accomplish the programming, I did have to switch which side of the system box that the track power is hooked up to ... I did this simply in this trial by swapping the wiring plugs so no need to remove or reattach the wires with a screwdriver. I would expect this to be done with a DPDT switch in a more permanent setup.
Moving on to JMRI ... I picked up a serial-to-USB adapter cable and in the end, it was installing this cable on my older notebook running Windows Vista that took the longest time for the rest of the setup and testing :o|.
Installing JMRI 3.10.1 as a download from SourceForge was painless. Since I initially just wanted to run locos, I started up Panel Pro and was easily able to locate my USB serial port (COM6 in my case) and then add a throttle configured for Loco #3108. Immediately I was able to control speed, direction and lighting through the NCE system. BTW, I still had the NCE Pro Cab attached at the time and while the two throttles didn't "know" about each other, the loco responded to either throttles. At this point, I disconnected the NCE Pro Cab and set it aside.
One final step, running off to get my cell phone (a Samsung S3 running Android Kit Kat), I return with Engine Driver installed and then launch WiThrottle on the JMRI software. It's a matter of a couple of minutes from app install to running locos as the app readily discovered my JMRI setup on the WiFi and just as easily I found Loco #3108 available to run with the app throttle. This time, I note that the JMRI throttle and the Engine Driver throttle are coordinated and any changes made with one are relatively soon reflected on the screen of the other ... nice!
As a side note, as you may have noticed, I removed the cover from the NCE system unit in order to see inside and almost immediately went straight to the track voltage adjustment. Using a multi-meter, I easily adjusted the voltage down from the 14+ volts that NCE presets the unit at to 10 volts. Although I've heard that many operate DCC with Z-scale without making such an adjustment, I chose to stay on the safe side by making this adjustment. After noting that the locos were running a bit slower than I expected, I adjusted the voltage up to 12 volts and that seemed to help a bit. While I can't accurately measure the track voltage across the rails, I'm suspecting a bit of a drop from the 12 volts measured inside of the NCE system box.