OpenGL bug on Mac, shows up in schnakenberg example

[sebjameswml]

Looks like there's a null vertex that shows up when the schnakenberg example runs on a Mac.

I suspect a bug in the setting of vertices or normals in HexGridVisual.

Perhaps @stuartwilson would like to buy that Macbook for me to debug it ;)

[sebjameswml]

I think this bug is visible on Linux, too, though with a slightly different "look" - a black hex in the middle of one of the maps.

[jmartinbrooke]

Hi Seb Yes I have seen that on linux. John

…

On Thu, 28 May 2020 at 10:46, Seb James ***@***.***> wrote: I think this bug is visible on Linux, too, though with a slightly different "look" - a black hex in the middle of one of the maps. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#19 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AANK24XGGGLY276F4L2CZILRTYXI7ANCNFSM4MNRCJLQ> .

[sebjameswml]

I think this occurs when there is a bad value for a hex. On Linux, this shows as a black hex. As it happens, on the new Macbook Air, when the GL is linked to Cocoa, the horrible output that Stuart was seeing has gone. Even the black hexes don't show. The new build process in features/cmake_update improves the linking, and the build process for morphologica in general.

However, I need to look at the hexgrid visual code, to see what it does if one of the data variables is nan and make sure it falls back gracefully as far as creating the model vertices.

[sebjameswml]

2e03538 fixes the black hexes which were showing up.

[stuartwilson]

Yey!!

On Tue, 15 Sep 2020 at 17:22, Seb James ***@***.***> wrote: 2e03538 <2e03538> fixes the black hexes which were showing up. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#19 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABJQ3OB2JUTJNRV4JTPMKP3SF6IEBANCNFSM4MNRCJLQ> . --

Best wishes, Stuart Dr. Stuart P Wilson (Lecturer in Cognitive Neuroscience) Department of Psychology University of Sheffield Sheffield, S10 2TP, UK. *Tel:* +44 (0) 114 222 6595 *Fax:* +44 (0) 114 276 6515 *Email:* S.P.Wilson@Sheffield.ac.uk *Group: abrg.group.shef.ac.uk/people/stuart/ <http://www.abrg.group.shef.ac.uk/people/stuart/>* *Personal: spwilson.staff.shef.ac.uk <http://spwilson.staff.shef.ac.uk>*

[sebjameswml]

After some further changes to CMakeLists, this seems to have gone away for Stuart's Mac! See 6411246 for careful binding and unbinding code, which may have been the fix. Also fbc8a92 for incorporating OpenMP correctly - the GL rendering worked for Stuart when programs were compiled single-threaded. Maybe the rendering worked 'by accident'?

[sebjameswml]

Closing, as I believe this has gone now.