I've been using MPLAB X and XC16 to build and program projects for a dsPIC 33EP128MC506 for about a year now. For the last two months, I've seen some new and unusual behavior on two separate machines. Both are running the latest MPLAB X and XC16. One PC is Windows 7, the other is Windows 8.
When I press "build and program" the project builds properly, then hangs. I sit there with a "Build successful message" and I can operate the IDE as normal, but it never proceeds to the program stage. The only way I can get the IDE to program my chip is to open the Windows task manager, kill the javaw.exe process, and reopen MPLAB X. I then get one good program before it starts exhibiting the same behavior.
In other words, every time I program a chip, I have to kill MPLAB and reopen it. That gets kind of old. The IDE is also constantly "Scanning projects", but I can't tell if that's related, because I can't make it stop.
I've opened a ticket with Microchip, but no response. I've recreated the project from scratch, no luck. I don't have this problem with dsPIC 30F6012a projects. Has anyone seen this? Is there a known solution?
Best Answer
Found this page and thought of sharing my experience. I searched on Google and found these two links:
http://microchip.wikidot.com/mplabx:jvm-configuration-options
https://performance.netbeans.org/howto/jvmswitches/index.html
These talk about tuning switches for the JVM so that Netbeans starts out well and uses a better GC, and Hotspot kicks in faster than it would normally.
I did what they said in here and now MPLAB-X is much more snappy. I noticed that my CPU usage and RAM usage goes up initially when the application starts and parses all the open projects, but then it settles down to a comfortable level and IDE performs much better.
Your Mileage May Vary.