For what it's worth, I did a high speed taxi down the same runway in
the JSBSim C172.  At the same point on the runway, the C172 takes a
hard bounce, goes almost completely nose down vertical, but recovers
without "crashing".  I don't see any visual discontinuity, so I assume
this must be a problem with the code that reports ground elevation.

It appears to be happening as we cross tile boundaries, so mostly
likely that first elevation change as we cross is bad.  We had a
potentially similar problem visually when crossing tile
boundaries where the visuals would jump way out of position for just
one frame.

I dumped out the ground elevation as I went over the anomoly and
here's what I get:

elev = 164.043
elev = 164.041
elev = 164.037
elev = 164.032
elev = 166.037  <!--
elev = 164.023
elev = 164.023

So at the point of the crash, the scenery system is returning a
terrain height of 2 meters higher than it should be.  The subsequent
frames are correct again.  This corresponds the the first hit returned
after crossing a tile boundary.

This is really wierd, it's a bug that crept in at some unkown point
and it seems that the a4-yasim is the most sensitive to it, althoug
hit also affects JSBSim aircraft too.

Could the hitlist code be overly aggressive on what it's caching
internally?

This is kind of strange (?)

Regards,

Curt.
-- 
Curtis Olson   IVLab / HumanFIRST Program       FlightGear Project
Twin Cities    [EMAIL PROTECTED]                  [EMAIL PROTECTED]
Minnesota      http://www.menet.umn.edu/~curt   http://www.flightgear.org

_______________________________________________
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel

Reply via email to