[osg-users] OpenSceneGraph-3.6-TexStorage branch created, needs community testing :-)

[Robert Osfield]

HI All,

I have created a OpenSceneGraph-3.6-TexStorage branch to allow
glTexStorage support that was just in the master branch to be ported
back onto 3.6. There isn't any public API changes so it should drop
in OK compile wise, runtime wise it's a different matter though.

We need testing across a wide range of platforms/hardware and
application usage cases to know whether this change is safe. If it
all looks OK then I'll merge 3.6-TexStorage into the main 3.6 branch
and then we can start working towards a 3.6.3 release in September.

So please test if you can. Many thanks,
Robert.

[michael kapelko]

Hi. Who's the audience of the change? (i.e. should I care?)

Post by Robert Osfield
HI All,
I have created a OpenSceneGraph-3.6-TexStorage branch to allow
glTexStorage support that was just in the master branch to be ported
back onto 3.6. There isn't any public API changes so it should drop
in OK compile wise, runtime wise it's a different matter though.
We need testing across a wide range of platforms/hardware and
application usage cases to know whether this change is safe. If it
all looks OK then I'll merge 3.6-TexStorage into the main 3.6 branch
and then we can start working towards a 3.6.3 release in September.
So please test if you can. Many thanks,
Robert.
_______________________________________________
osg-users mailing list
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

[Robert Osfield]

Post by michael kapelko
Hi. Who's the audience of the change? (i.e. should I care?)

With my focus on new feature dev to the VSG project for the next year
the next major OSG release (4.0) is pushed a bit further out, so
availability of new features that are in master in a stble release
will take longer, so if there are features in master that can be
pulled into the 3.6 branch without affect the public API and without
risking stability then I'm open to do this.

The glTexStorage functionality is a possible candidate for this as it
only changes the implementation, it doesn't affect the API. However,
the "risking stability" is an open question, this can only be properly
answered by wider testing in the community.

Robert.