Note, I am using some options which appear conflicting.
This config in theory makes the IDE run /slower/ for the first 60,000 somethings[commands executed?].
Standard client config sets this to 1000
Server sets it to 20000
However, in practice it makes the user experience much much faster. Basically the IDE loads faster and opens the project faster because it is not trying to compile the code at the same time as opening.
And then I do the opposite with:
I assume this is the same thing as PHPStorms "background indexing" - ie performing the task at a lower priority to avoid interfering with the UI. Wheras I don't want the compile dragging on forever, I want it to complete as soon as it can. I find ongoing performance to be much better with this setting.
# No clue but my system is running faster
# Make sure compile threads equals cpu cores/threads
# --client sets this to 1000
# --server sets this to 20000
# As I understand it, this is the number of commands
# Java will execute in interpretted mode before starting
# to compile. By upping it to 60k startup time
# is a LOT faster, and I make sure to open a project
# automatically at start since the goal of delaying
# compile is to give java a chance to see what is
# being executed in order to optimize compile
# Sometimes fun to watch
# Compile as quickly as possible, don't do it
# in the background. Results in much zippier performance