RE: [Flightgear-devel] Teleportation

[Paul D. <pau...@ve...>]

Gents,

Thanks for the ideas.  I'm getting the lat/lon from an external bus using a
modified version of External.cxx, but have not changed the lat/lon
initialization.  I think I just need to get the initial lat/lon from the bus
in fgInitSubsystems() and then there will be no immediate teleportation.

Thanks,

Paul

> -----Original Message-----
> From: fli...@li...
> [mailto:fli...@li...]On Behalf Of Curtis
> L. Olson
> Sent: Friday, July 20, 2001 4:27 PM
> To: fli...@li...
> Subject: RE: [Flightgear-devel] Teleportation
>
>
> Norman Vine writes:
> > FWIW
> > This has always been a tricky issue with 'teleporting'.
>
> And it has gotten a lot tricker with the introduction of threaded tile
> loading.
>
> > In the previous incarnation of the tilemanager I implemented
> > a mechanism to halt loading of tiles then walk the 'ready' queue
> > freeing the 'loaded' tiles as I went, then emptied the 'ready' queue.
> >
> > I don't think that this is being done in the current scheme.
> >
> > The other approach is to somehow poll the 'ready' queue until
> > it is empty.  This will be a little tricky in that the 'loader' only
> > kicks in every X frames.  Therefore I would suspect some form
> > of my first version has to be written that will  work with the new
> > fgTileLoader < possibly threaded > mechanism.
> >
> > Otherwise just wait 'long enough' for the initial set of tiles to get
> > loaded. Decreasing your visibility will I believe decrease the number
> > of tiles that are scheduled to load.
> >
> > 'long enough' is determined empirically :-)
>
> You might also try increasing the relative tile cache size, although
> this might not save you if you teleport to many divergant locations in
> quick succession.
>
> Also, I have just checked in some changes that should make the loader
> more tolerant of a situation where the tile cache is full, but none of
> the entries qualify for removal (i.e. they are all at some stage in
> the loading process.)  It will just skip and try again the next frame
> until a cache entry finishes loading and can then be freed.
>
> Regards,
>
> Curt.
> --
> Curtis Olson   Human Factors Research Lab       FlightGear Project
> Twin Cities    cu...@hf...                cu...@fl...
> Minnesota      http://www.menet.umn.edu/~curt   http://www.flightgear.org
>
> _______________________________________________
> Flightgear-devel mailing list
> Fli...@li...
> http://lists.sourceforge.net/lists/listinfo/flightgear-devel
>