And with Mesa 10.1.5 on Fedora.
Anyone tested yet the workaround Paul just added, i.e.
(set-use-stroke-characters 1)
?
If it works for everyone I think in the CCP4 suite coot will have this
option on by default on Linux.
Marcin
On 23 September 2014 13:02, Florian Sauer
wrote:
> Hallo Jan,
>
> y
Hallo Jan,
yes the problem still exists with MESA 10.0.0 (on Ubuntu 13.10 / Intel
HD 4000).
OpenGL vendor string: Intel Open Source Technology Center
OpenGL renderer string: Mesa DRI Intel(R) Ivybridge Mobile
OpenGL core profile version string: 3.3 (Core Profile) Mesa 10.0.0
OpenGL core profile s
Hi Marcin,
is your COOT only slow when (atom) labels are shown or always?
All the others:
Has anyone tried a more recent MESA (10.x)?
Regards,
Jan
Am 23.09.2014 um 12:59 schrieb Marcin Wojdyr:
> Yes, the log says Ironlake (Arrandale), it's seems to be a bit older,
> from 2010, according to
> ht
Yes, the log says Ironlake (Arrandale), it's seems to be a bit older,
from 2010, according to
http://en.wikipedia.org/wiki/List_of_Intel_graphics_processing_units
In Diamond we also have RedHat 6 with the same Mesa. Computers with
Intel HD4000 and HD4600 run coot very slowly. No idea why exactly a
after posting, I additionally found out that the graphics are just called
Intel HD Graphics
but not HD3000 or HD4000.
Kay
On 09/20/2014 01:00 AM, COOT automatic digest system wrote:
It may all depend on the version of Mesa. As for not affected RHEL,
what's the output of
glxinfo | grep OpenGL
?
Marcin,
I can confirm that coot runs without problems when labels are on the
screen (no noticeable difference in the spe
Hi,
A temporary fix is to turn atom labelling upon recentering off
(set_label_on_recentre_flag(0) in python) - all the atom info is shown in the
bar below the graphics display anyway. Unfortunately there is no similar flag
that I am aware of to turn off atom labels after refinement or other mod
Le 19/09/14 17:40, Marcin Wojdyr a écrit :
> It's affected. I don't think it's a distro-specific problem.
Yep, same problem under Debian 8 (jessie) Sigh...
Miguel
I think coot on RHEL6 was running well with HD 3000 (not 4000 !)
graphics - a Toshiba Portege 13" notebook - will check/report next week.
Kay
Am 20.09.14 um 01:00 schrieb COOT automatic digest system:
>> Is this really the same problem (only slow with labels on)? Kay once
>> > reported that RHEL
On 18 September 2014 08:04, Jan Gebauer wrote:
> Am 17.09.2014 um 17:59 schrieb Marcin Wojdyr:
>> Some computers in Diamond running RHEL6 also have the slow-coot
>> problem. Our IT even informed RedHat about it:
>> Marcin
>>
>
> Is this really the same problem (only slow with labels on)? Kay once
On Wednesday, 17 September, 2014 16:59:31 Marcin Wojdyr wrote:
> On 17 September 2014 13:11, Paul Emsley wrote:
> > On 17/09/14 10:01, Jan Gebauer wrote:
> >>
> >> Hi,
> >>
> >> regarding the problem COOT has with Intel HD cards on Ubuntu
> >> (https://www.mail-archive.com/coot@jiscmail.ac.uk/msg0
On 17 September 2014 13:11, Paul Emsley wrote:
> On 17/09/14 10:01, Jan Gebauer wrote:
>>
>> Hi,
>>
>> regarding the problem COOT has with Intel HD cards on Ubuntu
>> (https://www.mail-archive.com/coot@jiscmail.ac.uk/msg03454.html) , I
>> wonder if there is a fix in the meantime.
>>
>
> Me too. Ha
have you guys tried compiling the latest mesa from source and installing it
into your system? I had to just do that for a bug I hit recently :
wget ftp://ftp.freedesktop.org/pub/mesa/10.2.7/MesaLib-10.2.7.tar.bz2
tar jxvf MesaLib
cd MesaLib
sudo apt-get install build-essential
sudo apt-get build-d
On 17/09/14 10:01, Jan Gebauer wrote:
Hi,
regarding the problem COOT has with Intel HD cards on Ubuntu
(https://www.mail-archive.com/coot@jiscmail.ac.uk/msg03454.html) , I
wonder if there is a fix in the meantime.
Me too. Has anyone with service contract with Canonical and reported
this prob
Hi,
regarding the problem COOT has with Intel HD cards on Ubuntu
(https://www.mail-archive.com/coot@jiscmail.ac.uk/msg03454.html) , I
wonder if there is a fix in the meantime.
Last year my students struggled with the "solution" of removing the
labels every time they clicked somewhere. Even with t
15 matches
Mail list logo