Well, I've just tested with SwapGroups and got no joy.  I'm getting Xprotocol errors and Ican't seem to figure out why as the values it is complaining about are valid ones.

Anyway, however, I have my doubts about Swapgroups helping here.  These calls are designed to synchronize swaps at the GPU driver level.  Our concern is the blocking of the CPU, not the GPU.  Swap ready type synchronization will just impose another condition that SwapBuffers will need to wait for besides Vsync before executing.

BUt in the CPU we are doing :

             SwapBuffers( drawable1 );
             SwapBuffers( drawable2);

If SwapBuffers( drawable1) is blocked until next Vsync, then SwapBuffers(drawable2) won't happen until 16 milliseconds later.  And if both of these belong to the same swapgroup, then it will never happen and the application will wedge.

-don

On 10/18/06, Garry Keltie <[EMAIL PROTECTED]> wrote:

This document is (NVidia specific obviously) aimed at g-sync card users,
although multiple windows can reside on the same system (I take this to
be graphics head, single or SLI).

From the document: "NVFrameSync_UG_20.pdf"

Most of the document is about frame lock/swap sync and genlocking but I
wonder if the general principle of groups and barriers holds. I also
don't know if you need g-sync hardware in order to use the functions for
the extensions (ie if your not externally syncing via hardware to
another 'system' and need the physical card).

/snip

• Swap Groups
If buffer swaps must be synchronized across several windows on a single
system, you can define a "group" that consists of the specified windows.

• Swap Barriers
If buffer swaps must be synchronized across several systems, you can
define a "barrier" that consists of the specified groups.

/endsnip

Garry

----- Original Message -----
Colin Dunlop wrote:
...
> did this used to work on SGI's from the mid 90's?
> http://www.opengl.org/registry/specs/SGIX/swap_group.txt
>> This extension provides the capability to synchronize the buffer swaps
>> of a group of GLX drawables. A swap group is created, and drawables are
>> added as members to the swap group. Buffer swaps to members of the swap
>> group will then take place concurrently.
>
> Perhaps the chaps who authored the above work for NVidia now? ;-)
...

--
Garry Keltie

garrydotkeltieATrmit.edu.au
VR Centre - Research & Innovation
RMIT University

/ ... .... .. - /  .... .- .--. .--. . -. . -..
_______________________________________________
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/

Reply via email to