WGL_NV_SWAP_GROUP

Does anyone have any example code for using this extension?
The particular scenario I’m after is synchronising buffer swaps on two separate machines, using the SLI gsync cards.
I’m a bit puzzled as to what a swap barrier is (in terms of this extension). Is it the vsync swap interval for that device context?
Much obliged.

P.S. Not being funny but, please only answer this if you’ve got facts and/or experience with this particular extension, I’m not interested in guessing games, or I3D experience.

Nice joke.
WGL_NV_SWAP_FORUM anyone?

I’m afraid that was over my head, mjkrol.
I should have added “no jokers” to my P.S., but didn’t think I’d need to.

Swap groups are for multiple windows on a single system. To synchronize across several systems you have to define a barrier and bind the groups to that.
(theoretically, but not a guess, just quoting documentation.)

I hope I am not violating knackered new imposed rules on this forum who is allowed to answer his posts… :wink:

Ok, without a specification for this extension, I’m going to use the SGI equivalents as a reference, and hope it’s a one-to-one match.
http://oss.sgi.com/projects/ogl-sample/registry/SGIX/swap_group.txt
http://oss.sgi.com/projects/ogl-sample/registry/SGIX/swap_barrier.txt

Google for “nvidia frame lock users guide” there is a preliminary pdf floating arround. It seems to go into more depth.
What hardware setup are you working with? Our customers mostly refrain from using gsync cards due to budget reasons. (talking about maybe 25 systems equipped with QuadroFX4500s and G-sync modules…) :eek:

I’ve read the user guide, but can’t find any specifics about the various scenarios they can be used in. Examples is what I need, so I can directly compare what I’ve done with some standard way of using them - for troubleshooting.
I’ve used the extension successfully before now, just by issuing the calls randomly, but I need some specifics to diagnose a problem I’m having, but nvidia don’t provide them. They don’t even provide a test program with their genlocking hardware, for gods sake.

Originally posted by knackered:
They don’t even provide a test program with their genlocking hardware, for gods sake.
C’mon, they can’t be that bad? Can they? I mean, that’d be like telling you “Don’t use any of the two kernel platforms we claim to “support”, use something more suitable (though, we don’t “support” it so good luck getting any documentation on our hardware we sold you from us!”). ;->

But seriously, they can’t be that evil/incompetent/ignorant/moronic/<whatever> (I didn’t even know they could spell “genlock” - I’m not sure it’s still not April 1:st).

Has the world gone mad?
Tamlin, sod off if you’ve nothing useful to say - go and spout your insane gibberish to your land-lady, if you need someone to listen.

Any NVidia people reading this?
This is an ideal opportunity for you to supplement your inadequate documentation with some detail, here on this forum - so people can search for it in the future.