Andy Ross writes: > David Megginson wrote: > > Andy Ross writes: > > > + It's so close to the near clip plane that the precision errors cause > > > nasty jittering; this happens to the rest of the cockpit geometry > > > too. Honestly, I think this is probably unavoidable. We should > > > really look at solutions for drawing the cockpit into a separate > > > depth range from the terrain. > > > > We already do -- when you are inside the cockpit, the aircraft is in > > its own scene graph with its own clip plane; it's only when you're > > outside the the aircraft model has the same clip plane. > > Hrm, really? What are the clip plane settings? Something is causing > a really nasty 1-2 pixel jitter of the near-field geometry on my > Geforce. Maybe we are hitting a precision limit somewhere in software > that's making its way into the modelview matrix instead? > > Maybe this is specific to my box -- does no one else notice the > significant jitter of the A-4 cockpit?
I see the jitter too ... it appears to be the vertices that are jittering and not the texture coordinates. (You can get texture jitter if you use really large texture coordinates ... i.e. > 4, 8, 16, 32 range depending on your video card. (Voodoo 1/2/3 were a lot worse with this than the GeForce cards.) Normall to map a whole texture onto a face you would use texture coordinates in the 0.0 - 1.0 range but you can use larger numbers to get the texture to repeat/tile. But I digress ... that doesn't seem to be the problem anyway. Here's what I would suggest. OpenGL uses floats internally so think about the values of the vertices, or the values that could be generated as the object is transformed/rotated to it's final location. If the verticies or the intermediate values ever are greater than 100,000.0 - 1,000,000.0 range, the result will get rounded to fit inside the space of a float. Because the rounding could go up or down from frame to frame as view points and transforms change, this can lead to visible jitter. For instance if you create a road and plop it down at around x = 800,000, y = 300,000 and then drive along it, you may see the lines jumping side to side or back and forth by a meter or so. I haven't looked at the specific implimentation of this instrument, but when you are trying to track down jitter, think about what floating point values you are using and if you are losing enough bits somewhere along the way to account for this much jitter. 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
