Robert, Jason, Luigi, Gert, and all
Luigi's right: when "offset" is forced to be 0, textures are applying
OK.
I don't quite understand everything going in DataSet.cpp, but I do
know that these calculated offset are in *degrees*, a unit I rarely
seen in the datasets I'm used to. As to why "d_bb.xMin()" can be
smaller than "s_bb.xMin()", that doesn't seem to have nothing to do
with the data I pipe everyday to osgdem.
Maybe there should be some check as to whether the dataset is in
degrees before setting the said "offset"? I see that later "unsingned
int numXChecks" is set to 2 if d_bb.is_Geographic is true. Is that a
condition we should add to the "offset" thing? Does
GeospatialExtens::is_Geographic means the data is in degrees?
Hum. I'll pass the puck to the ones who know.
Lost in this thread is another problem which now seems not to be
related, ie:
Also, while trying to recreate the problem with a smaller dataset,
another problem appeared, ie a clear white line in the middle of
nowhere. See http://graphsynergie.com/osg/whiteline.jpg. That
white line does'nt correspond to an edge as seen in the
corresponding mathing wireframe of the same image: http://
graphsynergie.com/osg/whiteline_wireframe.jpg.
Any ideas?
Regards,
Paul
On 6-Dec-06, at 11:53 AM, Jason Beverage wrote:
Robert is probably right, the fix that I applied probably wasn't
general
enough. I've got no idea right now what could be causing the issue.
I'm kind of swamped with work until maybe early next week or this
weekend, so I won't be able to look into this until then. If you guys
can come up with anything, that would be great.
Thanks,
Jason
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Paul de
Repentigny
Sent: Wednesday, December 06, 2006 11:37 A
To: osg users
Subject: Re: [osg-users] osgdem: offset in texture pages
Jason, Luigi,
The same problem occurs with the latest Nightly Tarballs
( 2006-12-05 ), cleanly built on Linux FC4.
Ciao,
Paul
On 6-Dec-06, at 11:17 AM, Luigi Calori wrote:
I' ll chechout as soon as possible and I' ll let you know the result.
Is there a parameter to set in order to disable the shifting or is
it infering from the dataset?
Jason Beverage wrote:
Hi Luigi and Paul,
Just to clarify, are you guys using the latest CVS or a previous
version. As Robert and Gert mentioned, that line is changed in
the CVS
to fix the problem that Gert found. If you're not using the latest,
could you do a CVS update and see if you have the same problems?
Thanks,
Jason
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Robert
Osfield
Sent: Wednesday, December 06, 2006 9:08 A
To: osg users
Subject: Re: [osg-users] osgdem: offset in texture pages
Hi Luigi and Paul,
On 12/6/06, Luigi Calori <[EMAIL PROTECTED]> wrote:
Using your data, I have the same problems (shift in image wehen lod
sawp...i think)
This problem seems related to the line
double xoffset = d_bb.xMin() < s_bb.xMin() ? -360.0 : 0.0;
in osgTerrain/DataSet.cpp, line 520
With this deta, this value is getting set (incorretly? ) to -360.0
If you reset it to 0.0, the problem seems to go away
Hope it helps in diagnosis
The above line is part of fix to handle geocentric datasets that
cross
over the dateline that I applied a couple of months back. Since
then
Jason has done a fix to it so that it isn't applied
inappropriately -
as some dataset you don't want to have the offset added. Perhaps
Jason's fix isn't general enough.
I'm currently focused on completely some other work so can't
dedicate
myself to looking into this right now so if others can jump in an
spot
a fix then please do.
Robert.
_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/
_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/
_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/
