coreg GUI rendering problem (OS X)

Dear list,

I'm an avid conda<https://docs.anaconda.com/anaconda/install/>-user for my python setups on various machines, and very pleased about the new ?experimental? 3.6-based mne-install<http://www.martinos.org/mne/stable/install_mne_python.html>.

Having installed based on the environment.yml<https://raw.githubusercontent.com/mne-tools/mne-python/master/environment.yml> I have an issue with getting the coregistration gui rendering properly: the view only uses 1/4 of the canvas (see attached screenshot). My system is

- OS X High Sierra (10.13.3)
- retina screen (running in default scaled mode)
- tried with both conda's pyqt=5.6.0 and pip?s pyqt5=5.10.1, no dice, cross-platform toolkit my ? :slight_smile:

Any hints on how to make this baby work are much appreciated!

/Chris

[cid:396A6DEC-76B7-4BA3-BA2A-B83C03DDCC17 at waoo.dk]

We need to use VTK8 for proper HiDPI (retina) support, which causes this
problem. VTK8 should actually now installable on OSX, Linux, and Windows
via `pip`:

https://github.com/enthought/mayavi/issues/84#issuecomment-361901157

So you could try removing the `vtk` conda package and your existing
`mayavi` package, and installing both via `pip`. I have been using this
setup on my OSX laptop for a few weeks with good results.

Eventually the experimental MNE 3.6 environment will probably be updated to
use this configuration.

Eric

Dear list,

I'm an avid conda <https://docs.anaconda.com/anaconda/install/&gt;\-user for
my python setups on various machines, and very pleased about the new ?experimental?
3.6-based mne-install
<http://www.martinos.org/mne/stable/install_mne_python.html&gt;\.

Having installed based on the environment.yml
<https://raw.githubusercontent.com/mne-tools/mne-python/master/environment.yml&gt;
I have an issue with getting the coregistration gui rendering properly: the
view only uses 1/4 of the canvas (see attached screenshot). My system is

- OS X High Sierra (10.13.3)
- retina screen (running in default scaled mode)
- tried with both conda's pyqt=5.6.0 and pip?s pyqt5=5.10.1, no dice,
cross-platform toolkit my ? :slight_smile:

Any hints on how to make this baby work are much appreciated!

/Chris

--
Christopher Bailey, MSc
MEG Engineer, MINDLab Core Experimental Facility
Center of Functionally Integrative Neuroscience (CFIN)
Aarhus University, Denmark

_______________________________________________
Mne_analysis mailing list
Mne_analysis at nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis

The information in this e-mail is intended only for the person to whom it
is
addressed. If you believe this e-mail was sent to you in error and the
e-mail
contains patient information, please contact the Partners Compliance
HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in
error
but does not contain patient information, please contact the sender and
properly
dispose of the e-mail.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20180228/07de6312/attachment-0001.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PastedGraphic-1.png
Type: image/png
Size: 63181 bytes
Desc: not available
Url : http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20180228/07de6312/attachment-0001.png

Hi Eric, thanks for the hints/links. I still had some residual problems with the ?standard? environment file, all obviously (py)Qt-related and stemming from the enthought-curated modules. I finally got a working coregistration GUI by:

- using git master for traits, pyface and traitsui, as well as mayavi (cf. this thread<https://github.com/enthought/mayavi/issues/595>)
- making sure the following are set (e.g. in ~/.bashrc):

export ETS_TOOLKIT=qt4 # actually not v4!! see here<https://github.com/enthought/traitsui/issues/407>
export QT_API=pyqt5 # v5 selected here?

I?ve attached the slightly modified env-file (NB: some testing-related modules not installed), though I realise it?s not very reliable in the long run: any changes to master on a number of modules might break the install :frowning: It seems that there is long-overdue activity on all of them right now, so perhaps things will stabilise soon.

Would this be a reason to have an mne conda-channel? To pin down working versions for a cleaner installation process? ?conda install mne? where all plots are pretty and publications high-impact :wink:

/Chris

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20180301/ccbdd5da/attachment-0001.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mne_py3env_notests.yml
Type: application/octet-stream
Size: 590 bytes
Desc: mne_py3env_notests.yml
Url : http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20180301/ccbdd5da/attachment-0001.obj
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20180301/ccbdd5da/attachment-0001.htm