Where did these XML "update" times come from?

[Guest HC001ST - Stick]

Where did this data come from?This "Update Frequency" determines the interval with which FS schedules the gauge (i.e the refresh rate)If line absent, value=0 or value>9: 0.055 secValue=1: 1.00 secValue=2: 0.50 secValue=3: 0.33 secValue=4: 0.22 secValue=5: 0.17 secValue=6: 0.17 secValue=7: 0.11 secValue=8: 0.11 secValue=9: 0.11 secWas this emperically determined, or is this in the SDK somewhere and I missed it?Thanks.

[rcbarend 37]

If you copied that from my post:http://forums.avsim.net/dcboard.php?az=sho...ing_type=searchYes, this was indeed emperically determined; at least I wouldn't know of any official source that specifies this :-)I don't know why this is such a strange valuelist (e.g. why 5/6 and 7/8/9 are the same), but I get the same results under all circumstances I can think of.Cheers. Rob Barendregt

[Guest HC001ST - Stick]

The strangeness might be due to the uncertainty (error variance) with which they were "measured".I beleive that all the values would be directly proportional to 1/18th a second. Perhaps like this:Ordinal Ticks Refresh Interval1 1/2^0 1.000 sec2 1/2^1 0.500 sec3 1/2^2 0.250 sec4 1/2^3 0.125 sec5 1/2^4 0.063 sec6 1/2^5 0.031 sec*7 1/2^6 0.016 sec*8 1/2^7 0.008 sec*9 1/2^8 0.004 sec*except that the "tick" is the limiting factor, and hence anything 6 or up would really be 0.055 secs, and updating 18 times per second all things being equal.This might explain some other strange behaviors as well.But, just a theory.

[rcbarend 37]

Hi Stick,I'm not sure what you mean by "error variance"The times I mentioned have nothing uncertain about it.Example:If I let a gauge run with UpdateRate=7, it is scheduled 545 times per minute (= once per 110 msec).This can easy verified with a counter that increments every time the gauge is scheduled.And with UpdateRate=8 and 9, I get exactly the same result.Cheers, Rob

[Guest HC001ST - Stick]

Rob,"error variance" is a science thing. Your measurment has built in "variance" that is not due to the independent variable you are modifying (the updaterate). Anyway, it's not important, because the rough estimates are more than close enough for what we all need. I was just curious to go deeper for other reasons having to do with a current development project.