OK, I *think* I have nailed the platform-dependant YASim solution failures. What I found was that the solution heuristics hid a bunch of pseudo-chaotic instabilities that were introduced when the approach elevator trim feature went in a few weeks back. This was deterministic, but weird -- increasing the value of a parameter by a tiny amount could throw the solution into an oscillation.
Since tiny differences in calculation results ncould (I suppose...) be due to compiler differences, I'm willing to believe this was the culprit; the older gcc had a rounding mode bug or somesuch that caused the solver to pick up on a different oscillation mode. I validated with a command line YASim compiler that the results generated by 2.95.2 were identical to those from 2.96. Unfortunately, my FlightGear 2.95.2 build has begun crashing while loading tiles, so I can't test that until tomorrow. Anyway, try the new code and see if that works for you. You'll also want new planes, as some of the old ones went pretty wacky once the fix went in: http://www.plausible.org/andy/yasim-aircraft-052902.tar Hopefully we can bury this one. Andy -- Andrew J. Ross NextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) _______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
