Ok guys, I may get laughed out of the room on this one, but I'm going to put it out there for those that like to investigate performance related issues.
In my BB1 I'm running an NXT1 (blue board with the Intel Celeron CPU and 512 megs of RAM). The game that I'm running (currently) is Zeus II on a 13.x OS. I haven't tested this on 9.x. I have the game set up with multidenom of .01/.02/.05/.10.
I was running with a Louisiana JUR chip which consequently ran the annoying clicking meters. Today I changed the JUR chip to a Mississippi JUR chip which disables the meters. Here's where the fun starts...
The entire time I ran the Louisiana JUR chip, I noticed that when the meters were clicking away during gameplay, the performance was a bit sluggish with the reel spins until the clicking stopped. Usually, this would mean when I pressed the spin button on a max bet reels 1 and 2 would start off slower with the meters clicking, and then the performance would get better for the rest of the reels when the meters stopped clicking. This was on .10 denom which would create more meter activity compared to .01 cent.
With .01, I didn't notice as much of a performance hit with the reels spinning because there wasn't as much clicking compared to .10 when doing a max bet spin.
Now that I've replaced the Louisiana JUR chip with the Mississippi JUR chip, I don't notice the performance hit with reel spins on any denoms, and they all perform the same (better).
So I guess the question is...
Do active meters tax the CPU with additional cycles that impact performance in some fashion compared to inactive meters? The financial data is still being written and logged in the system, but without the meters spinning away. I know many think I'm somewhat insane, and that is true to some degree, but who isn't right?
Cheers!
Jason