2017-06-21 12:28 GMT+02:00 Viktor Engelmann <viktor.engelm...@qt.io>: > Hello Elvis, > > Haswell and Ivy Bridge indeed ring a bell here. We have a lot of > WebEngine bug reports that affect these GPUs only. > Can you create a bug report please?
Alright, good to know that I'm not completely alone. However, I was hesitant to create a Qt bug report as I'm not at all sure yet that it's a Qt problem. There's still a possibility that it's a bug in the VTK rendering engine (though it does work when VTK is configured to use GLEW). Perhaps I should try to make a pure Qt test case first that shows the problem. If I fail, then maybe that would be an indication that it's something in VTKs rendering engine that is at fault. > > OpenGL 3.2 context raises an eyebrow here, because just yesterday I have > seen a qtdiag output from a Debian 7 machine (2013 isn't THAT long ago) > with mesa 8.0.5, which only supports OpenGL 3.0. OK. The machine that I'm testing on is a Lenovo Thinkpad E540 with a HD 4600 GPU. The GPU is supposed to have 100% support for 3.2 core profile (verified this using "OpenGL Extension Viewer", http://realtech-vr.com/admin/glview), and I've verified that VTK does indeed get a correctly versioned context from Qt. The problem only seem to occur on Windows. It's working fine on my own Kubuntu 16.04 laptop (HD 4400 GPU). Elvis > > Kind regards > > > On 21.06.2017 11:38, Elvis Stansvik wrote: >> This may or may not be a Qt issue. >> >> I'm trying to help the VTK folks narrow down a problem with their new >> QVTKOpenGLWidget, which is a newly-introduced QOpenGLWidget-based >> class for letting VTK render into a QOpenGLWidget. >> >> The problem manifests itself as volume renderings not showing up on >> Windows machines with Ivy Bridge (e.g HD4000) and Haswell (e.g. >> HD4600) Intel GPUs when using the new class. VTKs volume rendering is >> done using float textures. >> >> Volume renderings works fine on the same machines when using "plain" >> VTK. This is when VTK itself sets up the OpenGL context (using GLEW I >> believe), but apart from that, the rendering pipeline is exactly the >> same I think. This is what leads me to believe that the issue is >> something specific to QOpenGLWidget, or to how VTK uses it. >> >> VTK requests a OpenGL 3.2 core context, and through some debugging, >> I've verified that VTK is actually getting the context that it is >> requesting. >> >> The issue is reported by me, here at VTKs GitLab: >> >> https://gitlab.kitware.com/vtk/vtk/issues/17058 >> >> The report contains some more information, such as the generated >> fragment shader that VTKs volume rendering backend is using, and a GL >> report from the problematic machine. >> >> Does this specific issue (float texture rendering not showing up on >> these semi-old Intel GPUs) ring a bell to anyone else? >> >> Many thanks for any tips/pointers/leads on how to debug this. >> >> I'm really interested in getting this issue in VTK fixed, since >> laptops with these GPUs are quite common. It's also affecting recent >> versions of ParaView, which has (just like me) recently ported to the >> new QVTKOpenGLWidget. >> >> Elvis >> _______________________________________________ >> Interest mailing list >> Interest@qt-project.org >> http://lists.qt-project.org/mailman/listinfo/interest > > -- > > Viktor Engelmann > Software Engineer > > The Qt Company GmbH > Rudower Chaussee 13 > D-12489 Berlin > viktor.engelm...@qt.io > +49 151 26784521 > http://qt.io > > Geschäftsführer: Mika Pälsi, > Juha Varelius, Mika Harjuaho > Sitz der Gesellschaft: Berlin, > Registergericht: Amtsgericht > Charlottenburg, HRB 144331 B > > The Future is Written with Qt > www.qtworldsummit.com > > > _______________________________________________ > Interest mailing list > Interest@qt-project.org > http://lists.qt-project.org/mailman/listinfo/interest _______________________________________________ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
