[FieldTrip] Sourceplots suddenly have huge file size

[Erik te Woerd]

Hi Jens,

The problem might be that the cfg.previous-structure is becoming quite
large. Try to clear this before plotting the data, like:
data.cfg.previous = [];

This is especially helpful when plotting grandaverage-data. (I've seen
cases in which the size went from 13 GB to 150 MB just by clearing the
cfg.previous)

Best,

Erik



2014-10-09 11:51 GMT+02:00 "Jens Klinzing, Universität Tübingen" <
jens.klinzing at uni-tuebingen.de>:

> Dear fieldtrip community,
> I encountered a weird problem. Lately, my source plots started to have
> huge file sizes when I save them as figures (.fig). This applies to both
> slice and surface plots. Surface plots used to have around 3-7 MB, now
> they have around 220 MB or even > 4 GB (ie. almost the size of the
> plotted data set). They also look slightly different now, with the
> surface being way darker than before. Slice plots look the same but are
> also larger than before.
>
> The problem seems to be in the plotting process itself (not the
> beamforming). When I plot older source localizations (which by the way
> have the same file size, fields, and basic properties as newer ones) the
> problem is still there.
>
> The code I am using for plotting did (to the best of my knowledge) not
> change at all:
> cfg                                     = [];
> cfg.method                      = 'surface';
> cfg.funparameter            = 'avg.pow';
> cfg.maskparameter         = cfg.funparameter;
> cfg.funcolormap              = 'jet';
> cfg.projmethod                = 'nearest';
> cfg.surfdownsample        = 5;
> ft_sourceplot(cfg,data)
>
> I thought it may be a change in fieldtrip so I reverted my ft_sourceplot
> to an older version but that did not help. I also used an older Matlab
> version (currently I use 2013b but I also tried 2013a) and another
> computer (Windows and Unix). In all cases I used opengl as the renderer.
>
> I am sure it is something very simple and I'm just too blind to see it.
> Any suggestions?
>
> Thanks a lot for your help,
> Jens
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20141009/2d0d1712/attachment.html>