It would be helpful if you could a) specify your machine type b) specify if this happens with all plugins removed and only a single instance of a plugin working c) clear out your QC preferences and ensure that private preference options are disabled.
ie: Do you have MSAA enabled in private prefs or something? On Dec 28, 2011, at 12:49 PM, George Toledo wrote: > I've been trying to track down an issue, that's been a bit tricky, since it > may be GPU and OS version dependent. > > Basically, any code that would be expected to make a point or line actually > aliased, seems to fail in 10.6 and 10.7. I remember this working in 10.5, but > I was also using another machine than I am now (with a different GPU), more > often than not. I don't know if it ever worked in 10.6 or not. > > Whether or not one enables aliasing, it seems to be getting called in > QCPlugin, and GFPlugin. All v002 plugins (standard API), NI plugins (standard > API), Kineme plugins (GFPlugin), fail in this regard, on my machine. > > If you setup a conditional in your plugin code that enables and disables > smooth via a boolean input on a plugin, it fails to "not smooth" (e.g., draw > square points), when it's not False. That used to work fine. > > If you go the extra step of setting up "what happens" with an Else statement, > to make sure to dictate that no smoothing is going on, it still fails to draw > aliased points (or lines). The enabling/disabling/not calling nice hints, > alias stuff, or any permutation thereof, also make no difference. This also > used to work fine. When I say "used to", this isn't an issue specific to > 10.7, but 10.6 and 10.7 (with it having worked in 10.5, at least on one > machine!). > > The one patch I can find that successfully draws aliased (square) points, is > the QCMesh Render patch. This makes me feel like the issue isn't GPU > dependent, but some obscure call that needs to be made, or not made, now. I'm > curious about this, to test performance related issues. I wish to see if > rendering of point clouds is quicker when drawing aliased points. Also, there > seems to be something slightly non-standard or highly optimized going on with > the texturing of the QCMesh gl points that allows it to perform much more > quickly than alternatives. Any insight into that would be great. > > Best regards, > George Toledo > > > _______________________________________________ > Do not post admin requests to the list. They will be ignored. > Quartzcomposer-dev mailing list (Quartzcomposer-dev@lists.apple.com) > Help/Unsubscribe/Update your Subscription: > http://lists.apple.com/mailman/options/quartzcomposer-dev/doktorp%40mac.com > > This email sent to dokt...@mac.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Quartzcomposer-dev mailing list (Quartzcomposer-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/quartzcomposer-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
