pymol-users Mailing List for PyMOL Molecular Graphics System

Hi Thomas,

Thank you for the email. Yes, it does seem to be the problem of
OpenGL. Here is the engine:
 Detected OpenGL version 4.5. Shaders available.
 Detected GLSL version 4.50.
 OpenGL graphics engine:
  GL_VENDOR:   Intel
  GL_RENDERER: Intel(R) UHD Graphics 630
  GL_VERSION:  4.5.0 - Build 24.20.100.6227

Fixed by setting precomputed_lighting.
I found setting use_shaders, 0 seems also a workaround, not sure if it
is the right way though.

Yufeng
On Wed, Oct 31, 2018 at 2:58 AM Thomas Holder
<tho...@sc...> wrote:
>
> Hi Yufeng,
>
> This is a problem with a recent graphics driver update from Intel on
> Windows 10. So far it appears that the build series "4.5.0 - Build
> 24.20.100.xxxx" is affected. Can you please tell us what PyMOL reports
> as "OpenGL graphics engine" (3 lines printed in log window on
> startup)?
>
> Luckily, there is a simple workaround, PyMOL has two different
> implementations of the lighting model and only one is affected. Type
> this into the PyMOL command line or put it in your pymolrc file:
>
>     set precomputed_lighting
>
> Cheers,
>   Thomas
> On Tue, Oct 30, 2018 at 11:48 PM Yufeng Tong <yuf...@gm...> wrote:
> >
> > Hi there,
> >
> > I installed PyMol on a new laptop (Windows 10, within Anaconda3) and
> > run into a display problem, which does not exist on a Windows 7
> > laptop.
> >
> > Basically, the displayed molecules does not have the 3D visual effect
> > and the colors are very dark. Only after I apply "ray", it shows
> > normal color but then of course I cannot rotate the molecules without
> > loosing the normal view. The same problem occurs to PyMol version 2.3
> > or 2.2 from the Schrödinger channel or the Gohlke's unofficial
> > packages. Tried to change color space, light position, and all options
> > under Display menu and none seems to change the results. Installed
> > PyQt and the problem persists.
> >
> > I tried the same installation process on an older Windows 7 laptop and
> > all looks normal. Attached are two screen copy of the display before
> > and after applying "ray", they can also be viewed from
> > https://1drv.ms/f/s!Ao4PGr4qLLsZjQiafUTIaUFTyxNn
> >
> > I guess it is a problem of the GUI. Any hints are greatly appreciated.
> >
> > Yufeng
> > _______________________________________________
> > PyMOL-users mailing list
> > Archives: http://www.mail-archive.com/pym...@li...
> > Unsubscribe: https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe

Hi Yufeng,

This is a problem with a recent graphics driver update from Intel on
Windows 10. So far it appears that the build series "4.5.0 - Build
24.20.100.xxxx" is affected. Can you please tell us what PyMOL reports
as "OpenGL graphics engine" (3 lines printed in log window on
startup)?

Luckily, there is a simple workaround, PyMOL has two different
implementations of the lighting model and only one is affected. Type
this into the PyMOL command line or put it in your pymolrc file:

    set precomputed_lighting

Cheers,
  Thomas
On Tue, Oct 30, 2018 at 11:48 PM Yufeng Tong <yuf...@gm...> wrote:
>
> Hi there,
>
> I installed PyMol on a new laptop (Windows 10, within Anaconda3) and
> run into a display problem, which does not exist on a Windows 7
> laptop.
>
> Basically, the displayed molecules does not have the 3D visual effect
> and the colors are very dark. Only after I apply "ray", it shows
> normal color but then of course I cannot rotate the molecules without
> loosing the normal view. The same problem occurs to PyMol version 2.3
> or 2.2 from the Schrödinger channel or the Gohlke's unofficial
> packages. Tried to change color space, light position, and all options
> under Display menu and none seems to change the results. Installed
> PyQt and the problem persists.
>
> I tried the same installation process on an older Windows 7 laptop and
> all looks normal. Attached are two screen copy of the display before
> and after applying "ray", they can also be viewed from
> https://1drv.ms/f/s!Ao4PGr4qLLsZjQiafUTIaUFTyxNn
>
> I guess it is a problem of the GUI. Any hints are greatly appreciated.
>
> Yufeng
> _______________________________________________
> PyMOL-users mailing list
> Archives: http://www.mail-archive.com/pym...@li...
> Unsubscribe: https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe

Hi there,

I installed PyMol on a new laptop (Windows 10, within Anaconda3) and
run into a display problem, which does not exist on a Windows 7
laptop.

Basically, the displayed molecules does not have the 3D visual effect
and the colors are very dark. Only after I apply "ray", it shows
normal color but then of course I cannot rotate the molecules without
loosing the normal view. The same problem occurs to PyMol version 2.3
or 2.2 from the Schrödinger channel or the Gohlke's unofficial
packages. Tried to change color space, light position, and all options
under Display menu and none seems to change the results. Installed
PyQt and the problem persists.

I tried the same installation process on an older Windows 7 laptop and
all looks normal. Attached are two screen copy of the display before
and after applying "ray", they can also be viewed from
https://1drv.ms/f/s!Ao4PGr4qLLsZjQiafUTIaUFTyxNn

I guess it is a problem of the GUI. Any hints are greatly appreciated.

Yufeng

Thanks for the tip!

On Tue, Oct 30, 2018 at 5:10 AM Thomas Holder <tho...@sc...>
wrote:

> Hi Steve and Shintaro,
>
> Shintaro is right, PyMOL's ray tracing is CPU only. It runs parallel
> on multiple CPU cores.
>
> Our graphics development over the last years was focused on getting
> the real-time OpenGL rendering (which uses the GPU heavily) as close
> to the ray traced images as possible. Lighting, antialiasing, and
> pixel-perfect sticks and spheres should be identical with "draw" and
> "ray". There are still some gaps, the ray tracer has shadows, better
> multi-layer transparency, and outline modes
> (https://pymolwiki.org/index.php/Ray#Modes ) that we don't have in
> real-time yet.
>
> For movies, I would certainly consider "draw" rendering a good option,
> unless you need shadows or better transparency.
>
> # write out PNG images
> mpng prefixdraw, mode=1, width=1280, height=720
> mpng prefixray, mode=2, width=1280, height=720
>
> # write out MPEG 4 (requires ffmpeg)
> movie.produce draw.mp4, mode=draw, width=1280, height=720
> movie.produce ray.mp4, mode=ray, width=1280, height=720
>
> Cheers,
>   Thomas
>
> On Mon, Oct 29, 2018 at 1:48 PM Shintaro Aibara
> <shi...@gm...> wrote:
> >
> > I am under the impression that ray tracing is an entirely CPU process in
> pymol. GPU accelerated ray tracing is not implemented in pymol, beyond the
> command "draw" which is not the same anyway. The software developers need
> to decide whether ray tracing in real-time is widely enough adopted
> hardware and software to decide to implement code that will use these
> (think about AMD card users, for example).
> >
> > Currently you are probably better off with upgrading the CPU to
> something like the 2990WX
> >
> > Best wishes
> > Shintaro
> >
> >
> >
> > On Mon, Oct 29, 2018 at 2:08 AM Stephen Gravina <
> ste...@gm...> wrote:
> >>
> >> Just installed a NVIDIA RTX 2080 for the Ray Tracing capabilities in
> PyMol. Just tested a GTX 1080 and the RTX 2080 and no difference. I did
> install the new Win10 patch and drivers. Update Bios.
> >>
> >> The GTX 1080 takes an hour to render the 30 frames for a second of a
> ray-traced movie of my protein a C class GPCR. I need 30 seconds, thus 30
> hours.
> >> .Only to find out a messed it up!
> >>
> >> Any suggestions?
> >>
> >> Thanks
> >>
> >> Steve Gravina, Ph.D.
> >> _______________________________________________
> >> PyMOL-users mailing list
> >> Archives: http://www.mail-archive.com/pym...@li...
> >> Unsubscribe:
> https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe
> >
> >
> >
> > --
> > Yours Sincerely,
> > Shintaro Aibara
> > _______________________________________________
> > PyMOL-users mailing list
> > Archives: http://www.mail-archive.com/pym...@li...
> > Unsubscribe:
> https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe
>

Hi Steve and Shintaro,

Shintaro is right, PyMOL's ray tracing is CPU only. It runs parallel
on multiple CPU cores.

Our graphics development over the last years was focused on getting
the real-time OpenGL rendering (which uses the GPU heavily) as close
to the ray traced images as possible. Lighting, antialiasing, and
pixel-perfect sticks and spheres should be identical with "draw" and
"ray". There are still some gaps, the ray tracer has shadows, better
multi-layer transparency, and outline modes
(https://pymolwiki.org/index.php/Ray#Modes ) that we don't have in
real-time yet.

For movies, I would certainly consider "draw" rendering a good option,
unless you need shadows or better transparency.

# write out PNG images
mpng prefixdraw, mode=1, width=1280, height=720
mpng prefixray, mode=2, width=1280, height=720

# write out MPEG 4 (requires ffmpeg)
movie.produce draw.mp4, mode=draw, width=1280, height=720
movie.produce ray.mp4, mode=ray, width=1280, height=720

Cheers,
  Thomas

On Mon, Oct 29, 2018 at 1:48 PM Shintaro Aibara
<shi...@gm...> wrote:
>
> I am under the impression that ray tracing is an entirely CPU process in pymol. GPU accelerated ray tracing is not implemented in pymol, beyond the command "draw" which is not the same anyway. The software developers need to decide whether ray tracing in real-time is widely enough adopted hardware and software to decide to implement code that will use these (think about AMD card users, for example).
>
> Currently you are probably better off with upgrading the CPU to something like the 2990WX
>
> Best wishes
> Shintaro
>
>
>
> On Mon, Oct 29, 2018 at 2:08 AM Stephen Gravina <ste...@gm...> wrote:
>>
>> Just installed a NVIDIA RTX 2080 for the Ray Tracing capabilities in PyMol. Just tested a GTX 1080 and the RTX 2080 and no difference. I did install the new Win10 patch and drivers. Update Bios.
>>
>> The GTX 1080 takes an hour to render the 30 frames for a second of a ray-traced movie of my protein a C class GPCR. I need 30 seconds, thus 30 hours.
>> .Only to find out a messed it up!
>>
>> Any suggestions?
>>
>> Thanks
>>
>> Steve Gravina, Ph.D.
>> _______________________________________________
>> PyMOL-users mailing list
>> Archives: http://www.mail-archive.com/pym...@li...
>> Unsubscribe: https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe
>
>
>
> --
> Yours Sincerely,
> Shintaro Aibara
> _______________________________________________
> PyMOL-users mailing list
> Archives: http://www.mail-archive.com/pym...@li...
> Unsubscribe: https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe

I am under the impression that ray tracing is an entirely CPU process in
pymol. GPU accelerated ray tracing is not implemented in pymol, beyond the
command "draw" which is not the same anyway. The software developers need
to decide whether ray tracing in real-time is widely enough adopted
hardware and software to decide to implement code that will use these
(think about AMD card users, for example).

Currently you are probably better off with upgrading the CPU to something
like the 2990WX

Best wishes
Shintaro



On Mon, Oct 29, 2018 at 2:08 AM Stephen Gravina <ste...@gm...>
wrote:

> Just installed a NVIDIA RTX 2080 for the Ray Tracing capabilities in
> PyMol. Just tested a GTX 1080 and the RTX 2080 and no difference. I did
> install the new Win10 patch and drivers. Update Bios.
>
> The GTX 1080 takes an hour to render the 30 frames for a second of a
> ray-traced movie of my protein a C class GPCR. I need 30 seconds, thus 30
> hours.
> .Only to find out a messed it up!
>
> Any suggestions?
>
> Thanks
>
> Steve Gravina, Ph.D.
> _______________________________________________
> PyMOL-users mailing list
> Archives: http://www.mail-archive.com/pym...@li...
> Unsubscribe:
> https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe



-- 
Yours Sincerely,
Shintaro Aibara

Just installed a NVIDIA RTX 2080 for the Ray Tracing capabilities in PyMol.
Just tested a GTX 1080 and the RTX 2080 and no difference. I did install
the new Win10 patch and drivers. Update Bios.

The GTX 1080 takes an hour to render the 30 frames for a second of a
ray-traced movie of my protein a C class GPCR. I need 30 seconds, thus 30
hours.
.Only to find out a messed it up!

Any suggestions?

Thanks

Steve Gravina, Ph.D.

Hi Thomas,


Thanks for clarifying this for me!


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University
________________________________
From: Thomas Holder <tho...@sc...>
Sent: Friday, 26 October 2018 3:41:40 PM
To: Joseph Brock
Cc: Jarrett Johnson; Pymol User list
Subject: Re: [PyMOL] distance

Hi Joseph,

Thanks for the example, I can reproduce it! Not sure if this should be considered a bug, or just a misleading user interface. "involving side chains" really means "side chains within selection". There are none if you only select the ligand. If I instead click on "to other atoms in object" I'll get the expected result.

You might also find one of these command line options useful:

distance polar1, (resn GSH), not (resn GSH), mode=2
distance polar2, (resn GSH), (sidechain), mode=2
distance polar3, (resn GSH), not (solvent), mode=2

Cheers,
  Thomas

> On Oct 26, 2018, at 2:34 PM, Joseph Brock <jos...@DB...> wrote:
>
> Hi again Jarrett,
>
> I have built the latest version from git (PyMOL 2.3.0a0 Open-Source) on my system.
>
> The behaviour is still there, but I realise that this it is now only when the selection is a ligand. For instance trying to find polar contacts between a ligand and sourrounding side chains. If I make a selection involving both the ligand and side chains of interest first however, it works fine.
>
> It is an easy workaround, but i would be curious to know if you can replicate it on your system. For instance if you fetch pdb id: 1EEM and make a selection from resname GSH, then try to find polar contacts involving side chains from the action menu, you should see what I mean.
>
> Cheers,
> Joseph.
>
>
> Joseph S. Brock | PhD
> Researcher
> Drew Lab
> Department of Biochemistry and Biophysics
> Stockholm University
> From: Joseph Brock <jos...@db...>
> Sent: Friday, 26 October 2018 8:57:44 AM
> To: Jarrett Johnson
> Cc: Pymol User list
> Subject: Re: [PyMOL] distance
>
> Thanks Jarrett! I will try that!
>
> Cheers,
> Joseph.
>
>
> Joseph S. Brock | PhD
> Researcher
> Drew Lab
> Department of Biochemistry and Biophysics
> Stockholm University
> From: Jarrett Johnson <jar...@sc...>
> Sent: Friday, 26 October 2018 12:23:46 AM
> To: Joseph Brock
> Cc: Pymol User list
> Subject: Re: [PyMOL] distance
>
> Built from source (github repo).
>
> On Thu, Oct 25, 2018 at 5:56 PM Joseph Brock <jos...@db...> wrote:
> Hi Jarret,
>
> Thanks for the reply!
>
> Unfortunately so ;(
>
> Did you build from source or install via apt-get?
> https://pymolwiki.org/index.php/Linux_Install
>
> Cheers,
> Joseph.
>
>
> Joseph S. Brock | PhD
> Researcher
> Drew Lab
> Department of Biochemistry and Biophysics
> Stockholm University
> From: Jarrett Johnson <jar...@sc...>
> Sent: Thursday, 25 October 2018 11:34:28 PM
> To: Joseph Brock
> Cc: Pymol User list
> Subject: Re: [PyMOL] distance
>
> Hi Joseph,
>
> I'm not able to replicate that on my end using open-source PyMOL 2.2 on Ubuntu 16.04. Does this behavior also occur for molecules fetched straight from the PDB?
>
> Jarrett J.
>
> On Thu, Oct 25, 2018 at 4:44 PM Joseph Brock <jos...@db...> wrote:
> Hi pymol users,
>
> I've observed a weird behaviour with the open source pymol (v 1.7.2.1-2.2 running on ubuntu 16.04).
>
> When I try to create some distance objects either via the command line of the actions menu, an object is created but no dashed lines appear. When I "zoom" this object it is very far in space from the atoms I selected to make the distances between, and there is still nothing I can do to make the dashed lines appear.
>
> If anyone can let me know what I'm doing wrong I would appreciate it!
>
> Cheers,
> Joseph.
>
>
> Joseph S. Brock | PhD
> Researcher
> Drew Lab
> Department of Biochemistry and Biophysics
> Stockholm University
>
> --
> Jarrett Johnson
> PyMOL Developer
> Schrödinger, Inc.

--
Thomas Holder
PyMOL Principal Developer
Schrödinger, Inc.

Hi Joseph,

Thanks for the example, I can reproduce it! Not sure if this should be considered a bug, or just a misleading user interface. "involving side chains" really means "side chains within selection". There are none if you only select the ligand. If I instead click on "to other atoms in object" I'll get the expected result.

You might also find one of these command line options useful:

distance polar1, (resn GSH), not (resn GSH), mode=2
distance polar2, (resn GSH), (sidechain), mode=2
distance polar3, (resn GSH), not (solvent), mode=2

Cheers,
  Thomas

> On Oct 26, 2018, at 2:34 PM, Joseph Brock <jos...@DB...> wrote:
> 
> Hi again Jarrett,
> 
> I have built the latest version from git (PyMOL 2.3.0a0 Open-Source) on my system.
> 
> The behaviour is still there, but I realise that this it is now only when the selection is a ligand. For instance trying to find polar contacts between a ligand and sourrounding side chains. If I make a selection involving both the ligand and side chains of interest first however, it works fine.
> 
> It is an easy workaround, but i would be curious to know if you can replicate it on your system. For instance if you fetch pdb id: 1EEM and make a selection from resname GSH, then try to find polar contacts involving side chains from the action menu, you should see what I mean.
> 
> Cheers,
> Joseph.
> 
> 
> Joseph S. Brock | PhD
> Researcher
> Drew Lab
> Department of Biochemistry and Biophysics
> Stockholm University
> From: Joseph Brock <jos...@db...>
> Sent: Friday, 26 October 2018 8:57:44 AM
> To: Jarrett Johnson
> Cc: Pymol User list
> Subject: Re: [PyMOL] distance
>  
> Thanks Jarrett! I will try that!
> 
> Cheers,
> Joseph.
> 
> 
> Joseph S. Brock | PhD
> Researcher
> Drew Lab
> Department of Biochemistry and Biophysics
> Stockholm University
> From: Jarrett Johnson <jar...@sc...>
> Sent: Friday, 26 October 2018 12:23:46 AM
> To: Joseph Brock
> Cc: Pymol User list
> Subject: Re: [PyMOL] distance
>  
> Built from source (github repo).
> 
> On Thu, Oct 25, 2018 at 5:56 PM Joseph Brock <jos...@db...> wrote:
> Hi Jarret,
> 
> Thanks for the reply!
> 
> Unfortunately so ;( 
> 
> Did you build from source or install via apt-get?
> https://pymolwiki.org/index.php/Linux_Install
> 
> Cheers,
> Joseph.
> 
> 
> Joseph S. Brock | PhD
> Researcher
> Drew Lab
> Department of Biochemistry and Biophysics
> Stockholm University
> From: Jarrett Johnson <jar...@sc...>
> Sent: Thursday, 25 October 2018 11:34:28 PM
> To: Joseph Brock
> Cc: Pymol User list
> Subject: Re: [PyMOL] distance
>  
> Hi Joseph,
> 
> I'm not able to replicate that on my end using open-source PyMOL 2.2 on Ubuntu 16.04. Does this behavior also occur for molecules fetched straight from the PDB?
> 
> Jarrett J.
> 
> On Thu, Oct 25, 2018 at 4:44 PM Joseph Brock <jos...@db...> wrote:
> Hi pymol users,
> 
> I've observed a weird behaviour with the open source pymol (v 1.7.2.1-2.2 running on ubuntu 16.04).
> 
> When I try to create some distance objects either via the command line of the actions menu, an object is created but no dashed lines appear. When I "zoom" this object it is very far in space from the atoms I selected to make the distances between, and there is still nothing I can do to make the dashed lines appear.
> 
> If anyone can let me know what I'm doing wrong I would appreciate it!
> 
> Cheers,
> Joseph.
> 
> 
> Joseph S. Brock | PhD
> Researcher
> Drew Lab
> Department of Biochemistry and Biophysics
> Stockholm University
> 
> -- 
> Jarrett Johnson
> PyMOL Developer
> Schrödinger, Inc.

--
Thomas Holder
PyMOL Principal Developer
Schrödinger, Inc.

Hi again Jarrett,


I have built the latest version from git (PyMOL 2.3.0a0 Open-Source) on my system.


The behaviour is still there, but I realise that this it is now only when the selection is a ligand. For instance trying to find polar contacts between a ligand and sourrounding side chains. If I make a selection involving both the ligand and side chains of interest first however, it works fine.


It is an easy workaround, but i would be curious to know if you can replicate it on your system. For instance if you fetch pdb id: 1EEM and make a selection from resname GSH, then try to find polar contacts involving side chains from the action menu, you should see what I mean.


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University
________________________________
From: Joseph Brock <jos...@db...>
Sent: Friday, 26 October 2018 8:57:44 AM
To: Jarrett Johnson
Cc: Pymol User list
Subject: Re: [PyMOL] distance


Thanks Jarrett! I will try that!


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University
________________________________
From: Jarrett Johnson <jar...@sc...>
Sent: Friday, 26 October 2018 12:23:46 AM
To: Joseph Brock
Cc: Pymol User list
Subject: Re: [PyMOL] distance

Built from source (github repo).

On Thu, Oct 25, 2018 at 5:56 PM Joseph Brock <jos...@db...<mailto:jos...@db...>> wrote:

Hi Jarret,


Thanks for the reply!


Unfortunately so ;(


Did you build from source or install via apt-get?

https://pymolwiki.org/index.php/Linux_Install


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University
________________________________
From: Jarrett Johnson <jar...@sc...<mailto:jar...@sc...>>
Sent: Thursday, 25 October 2018 11:34:28 PM
To: Joseph Brock
Cc: Pymol User list
Subject: Re: [PyMOL] distance

Hi Joseph,

I'm not able to replicate that on my end using open-source PyMOL 2.2 on Ubuntu 16.04. Does this behavior also occur for molecules fetched straight from the PDB?

Jarrett J.

On Thu, Oct 25, 2018 at 4:44 PM Joseph Brock <jos...@db...<mailto:jos...@db...>> wrote:

Hi pymol users,


I've observed a weird behaviour with the open source pymol (v 1.7.2.1-2.2 running on ubuntu 16.04).


When I try to create some distance objects either via the command line of the actions menu, an object is created but no dashed lines appear. When I "zoom" this object it is very far in space from the atoms I selected to make the distances between, and there is still nothing I can do to make the dashed lines appear.


If anyone can let me know what I'm doing wrong I would appreciate it!


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University
_______________________________________________
PyMOL-users mailing list
Archives: http://www.mail-archive.com/pym...@li...
Unsubscribe: https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe


--
Jarrett Johnson
PyMOL Developer
Schrödinger, Inc.


--
Jarrett Johnson
PyMOL Developer
Schrödinger, Inc.

Thanks Jarrett! I will try that!


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University
________________________________
From: Jarrett Johnson <jar...@sc...>
Sent: Friday, 26 October 2018 12:23:46 AM
To: Joseph Brock
Cc: Pymol User list
Subject: Re: [PyMOL] distance

Built from source (github repo).

On Thu, Oct 25, 2018 at 5:56 PM Joseph Brock <jos...@db...<mailto:jos...@db...>> wrote:

Hi Jarret,


Thanks for the reply!


Unfortunately so ;(


Did you build from source or install via apt-get?

https://pymolwiki.org/index.php/Linux_Install


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University
________________________________
From: Jarrett Johnson <jar...@sc...<mailto:jar...@sc...>>
Sent: Thursday, 25 October 2018 11:34:28 PM
To: Joseph Brock
Cc: Pymol User list
Subject: Re: [PyMOL] distance

Hi Joseph,

I'm not able to replicate that on my end using open-source PyMOL 2.2 on Ubuntu 16.04. Does this behavior also occur for molecules fetched straight from the PDB?

Jarrett J.

On Thu, Oct 25, 2018 at 4:44 PM Joseph Brock <jos...@db...<mailto:jos...@db...>> wrote:

Hi pymol users,


I've observed a weird behaviour with the open source pymol (v 1.7.2.1-2.2 running on ubuntu 16.04).


When I try to create some distance objects either via the command line of the actions menu, an object is created but no dashed lines appear. When I "zoom" this object it is very far in space from the atoms I selected to make the distances between, and there is still nothing I can do to make the dashed lines appear.


If anyone can let me know what I'm doing wrong I would appreciate it!


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University
_______________________________________________
PyMOL-users mailing list
Archives: http://www.mail-archive.com/pym...@li...
Unsubscribe: https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe


--
Jarrett Johnson
PyMOL Developer
Schrödinger, Inc.


--
Jarrett Johnson
PyMOL Developer
Schrödinger, Inc.

Built from source (github repo).

On Thu, Oct 25, 2018 at 5:56 PM Joseph Brock <jos...@db...> wrote:

> Hi Jarret,
>
>
> Thanks for the reply!
>
>
> Unfortunately so ;(
>
>
> Did you build from source or install via apt-get?
>
> https://pymolwiki.org/index.php/Linux_Install
>
>
> Cheers,
>
> Joseph.
>
>
>
> Joseph S. Brock | PhD
> Researcher
> Drew Lab
> Department of Biochemistry and Biophysics
> Stockholm University
> ------------------------------
> *From:* Jarrett Johnson <jar...@sc...>
> *Sent:* Thursday, 25 October 2018 11:34:28 PM
> *To:* Joseph Brock
> *Cc:* Pymol User list
> *Subject:* Re: [PyMOL] distance
>
> Hi Joseph,
>
> I'm not able to replicate that on my end using open-source PyMOL 2.2 on
> Ubuntu 16.04. Does this behavior also occur for molecules fetched straight
> from the PDB?
>
> Jarrett J.
>
> On Thu, Oct 25, 2018 at 4:44 PM Joseph Brock <jos...@db...>
> wrote:
>
>> Hi pymol users,
>>
>>
>> I've observed a weird behaviour with the open source pymol (v
>> 1.7.2.1-2.2 running on ubuntu 16.04).
>>
>>
>> When I try to create some distance objects either via the command line of
>> the actions menu, an object is created but no dashed lines appear. When I
>> "zoom" this object it is very far in space from the atoms I selected to
>> make the distances between, and there is still nothing I can do to make the
>> dashed lines appear.
>>
>>
>> If anyone can let me know what I'm doing wrong I would appreciate it!
>>
>>
>> Cheers,
>>
>> Joseph.
>>
>>
>>
>> Joseph S. Brock | PhD
>> Researcher
>> Drew Lab
>> Department of Biochemistry and Biophysics
>> Stockholm University
>> _______________________________________________
>> PyMOL-users mailing list
>> Archives: http://www.mail-archive.com/pym...@li...
>> Unsubscribe:
>> https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe
>
>
>
> --
> Jarrett Johnson
> PyMOL Developer
> Schrödinger, Inc.
>


-- 
Jarrett Johnson
PyMOL Developer
Schrödinger, Inc.

Hi Jarret,


Thanks for the reply!


Unfortunately so ;(


Did you build from source or install via apt-get?

https://pymolwiki.org/index.php/Linux_Install


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University
________________________________
From: Jarrett Johnson <jar...@sc...>
Sent: Thursday, 25 October 2018 11:34:28 PM
To: Joseph Brock
Cc: Pymol User list
Subject: Re: [PyMOL] distance

Hi Joseph,

I'm not able to replicate that on my end using open-source PyMOL 2.2 on Ubuntu 16.04. Does this behavior also occur for molecules fetched straight from the PDB?

Jarrett J.

On Thu, Oct 25, 2018 at 4:44 PM Joseph Brock <jos...@db...<mailto:jos...@db...>> wrote:

Hi pymol users,


I've observed a weird behaviour with the open source pymol (v 1.7.2.1-2.2 running on ubuntu 16.04).


When I try to create some distance objects either via the command line of the actions menu, an object is created but no dashed lines appear. When I "zoom" this object it is very far in space from the atoms I selected to make the distances between, and there is still nothing I can do to make the dashed lines appear.


If anyone can let me know what I'm doing wrong I would appreciate it!


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University
_______________________________________________
PyMOL-users mailing list
Archives: http://www.mail-archive.com/pym...@li...
Unsubscribe: https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe


--
Jarrett Johnson
PyMOL Developer
Schrödinger, Inc.

Hi Joseph,

I'm not able to replicate that on my end using open-source PyMOL 2.2 on
Ubuntu 16.04. Does this behavior also occur for molecules fetched straight
from the PDB?

Jarrett J.

On Thu, Oct 25, 2018 at 4:44 PM Joseph Brock <jos...@db...> wrote:

> Hi pymol users,
>
>
> I've observed a weird behaviour with the open source pymol (v 1.7.2.1-2.2
> running on ubuntu 16.04).
>
>
> When I try to create some distance objects either via the command line of
> the actions menu, an object is created but no dashed lines appear. When I
> "zoom" this object it is very far in space from the atoms I selected to
> make the distances between, and there is still nothing I can do to make the
> dashed lines appear.
>
>
> If anyone can let me know what I'm doing wrong I would appreciate it!
>
>
> Cheers,
>
> Joseph.
>
>
>
> Joseph S. Brock | PhD
> Researcher
> Drew Lab
> Department of Biochemistry and Biophysics
> Stockholm University
> _______________________________________________
> PyMOL-users mailing list
> Archives: http://www.mail-archive.com/pym...@li...
> Unsubscribe:
> https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe



-- 
Jarrett Johnson
PyMOL Developer
Schrödinger, Inc.

Hi pymol users,


I've observed a weird behaviour with the open source pymol (v 1.7.2.1-2.2 running on ubuntu 16.04).


When I try to create some distance objects either via the command line of the actions menu, an object is created but no dashed lines appear. When I "zoom" this object it is very far in space from the atoms I selected to make the distances between, and there is still nothing I can do to make the dashed lines appear.


If anyone can let me know what I'm doing wrong I would appreciate it!


Cheers,

Joseph.


Joseph S. Brock | PhD
Researcher
Drew Lab
Department of Biochemistry and Biophysics
Stockholm University

Inviato da iPad

   FYI, the reason that the ubuntu pymol 2.1.0 package currently doesn't
use the PyQT interface in the presence of the libqt5opengl5 package is that
they are hard coding pymol to use pmg_tk with the patch...

$ cat 50_force_tk.patch
Description: Disable the Qt interface to ensure the Tk interface is used.
  Patch must be dropped when converting to the Qt interface.
Author: Stuart Prescott <st...@de...>
Bug: 902348
--- a/modules/pymol/__init__.py
+++ b/modules/pymol/__init__.py
@@ -436,16 +436,7 @@
         args = sys.argv
     invocation.parse_args(args)

-    if invocation.options.gui == 'pmg_qt':
-        if invocation.options.no_gui:
-            return _launch_no_gui()
-
-        try:
-            from pmg_qt import pymol_qt_gui
-            sys.exit(pymol_qt_gui.execapp())
-        except ImportError:
-            print('Qt not available, using GLUT/Tk interface')
-            invocation.options.gui = 'pmg_tk'
+    invocation.options.gui = 'pmg_tk'

     prime_pymol()
     _cmd.runpymol(_cmd._get_global_C_object(), block_input_hook)

Reverting that change in /usr/lib/python2.7/dist-packages/pymol/__init__.py
solves the problem.
            Jack

Please Unsubscribe Me
_______________________________________________
PyMOL-users mailing list (PyM...@li...)
Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users
Archives: http://www.mail-archive.com/pym...@li...
Unsubscribe:
https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe

If you don't want GLUT to be linked in as a fallback, compile with --no-glut
(See https://pymolwiki.org/index.php/Linux_Install#Compile_and_install )

Tcl/Tk could be loaded if there is an activated plugin which imports it. PyMOL does some trickery to provide the event loop for Tkinter from the PyQt thread to support old plugins.

Cheers,
  Thomas

> On Oct 16, 2018, at 6:59 PM, David Mathog <ma...@ca...> wrote:
> 
> On 16-Oct-2018 00:07, Thomas Holder wrote:
>>> On Oct 15, 2018, at 9:38 PM, David Mathog <ma...@ca...> wrote:
>>> There must be something about Qt that is a significant improvement over Tk/Tcl - where/what is that?
>> It's not only Tk/Tcl, it's also GLUT which got replaced.
> 
> GLUT still seems to be there though, and the link libraries are loaded to some extent even when Qt starts.
> 
> On Windows 7 ntldd shows _cmd.pyd linked to libfreeglut.dll.
> On linux (ubuntu 16.04 LTS) ldd shows _cmd.so linked to libglut.so.3.
> 
> Start the linux pymol, it finds and uses Qt, then:
> 
> lsof | grep glut
> python    3010            root  mem       REG        8,5       236564    1978217 /usr/lib/i386-linux-gnu/libglut.so.3.9.0
> QXcbEvent 3010 3012       root  mem       REG        8,5       236564    1978217 /usr/lib/i386-linux-gnu/libglut.so.3.9.0
> 
> lsof also shows
> 
> python    3010            root  mem       REG        8,5      1475144    1975104 /usr/lib/i386-linux-gnu/libtk8.6.so
> QXcbEvent 3010 3012       root  mem       REG        8,5      1475144    1975104 /usr/lib/i386-linux-gnu/libtk8.6.so
> 
> and it is clearly using Qt, so perhaps some of the libraries lsof/ldd show are linked but are not called if Qt is used.
> 
>> So on the
>> pure "bug-fix list", we have:
>> - HiDPI support (we've seen unusable graphics on Intel/Windows)
>> - viewport size issues with multiple screens on Windows
>> On top of that, we get:
>> - trackpad support (pinch zoom)
>> - file open with drag-n-drop
> 
> Which of those fixes, if any, require Qt?
> 
> Thanks,
> 
> David Mathog
> ma...@ca...
> Manager, Sequence Analysis Facility, Biology Division, Caltech

--
Thomas Holder
PyMOL Principal Developer
Schrödinger, Inc.

On 16-Oct-2018 00:07, Thomas Holder wrote:
>> On Oct 15, 2018, at 9:38 PM, David Mathog <ma...@ca...> wrote:
>> 
>> There must be something about Qt that is a significant improvement 
>> over Tk/Tcl - where/what is that?
> 
> It's not only Tk/Tcl, it's also GLUT which got replaced.

GLUT still seems to be there though, and the link libraries are loaded 
to some extent even when Qt starts.

On Windows 7 ntldd shows _cmd.pyd linked to libfreeglut.dll.
On linux (ubuntu 16.04 LTS) ldd shows _cmd.so linked to libglut.so.3.

Start the linux pymol, it finds and uses Qt, then:

lsof | grep glut
python    3010            root  mem       REG        8,5       236564    
1978217 /usr/lib/i386-linux-gnu/libglut.so.3.9.0
QXcbEvent 3010 3012       root  mem       REG        8,5       236564    
1978217 /usr/lib/i386-linux-gnu/libglut.so.3.9.0

lsof also shows

python    3010            root  mem       REG        8,5      1475144    
1975104 /usr/lib/i386-linux-gnu/libtk8.6.so
QXcbEvent 3010 3012       root  mem       REG        8,5      1475144    
1975104 /usr/lib/i386-linux-gnu/libtk8.6.so

and it is clearly using Qt, so perhaps some of the libraries lsof/ldd 
show are linked but are not called if Qt is used.

> So on the
> pure "bug-fix list", we have:
> 
> - HiDPI support (we've seen unusable graphics on Intel/Windows)
> - viewport size issues with multiple screens on Windows
> 
> On top of that, we get:
> 
> - trackpad support (pinch zoom)
> - file open with drag-n-drop

Which of those fixes, if any, require Qt?

Thanks,

David Mathog
ma...@ca...
Manager, Sequence Analysis Facility, Biology Division, Caltech

please unsubscribe me

Please unsubscribe me.

-- 

*Dr. Arturo Sánchez*

Full member of the American Society for Virology

Full Member of the World Society for Virology

Miembro de la Red Mexicana de Virología

Miembro del Consejo Consultivo Científico de la CIBIOGEM



*Centro de Investigaciones Biológicas del Noroeste S. C. (CIBNOR)*

Laboratorio de Referencia, Análisis y Diagnóstico en Sanidad Acuícola

Calle Hermosa 101. Col. Los Ángeles. CP 83106.

Hermosillo, Sonora. México.

Tel. +52 (662) 2-13-15-93. Ext 5004.

https://www.cibnor.mx/

E-mail: asa...@ci...

Skype: swine.flu3

<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html;
      charset=windows-1252">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <tt>See the footer. Safe to click. <br>
    </tt><br>
    <tt>Marko</tt><br>
    <tt></tt><br>
    <div class="moz-cite-prefix">On 16/10/2018 15:08, hy2565 wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:5bc...@mx...">
      <meta http-equiv="content-type" content="text/html;
        charset=windows-1252">
      <div>Please unsubscribe me.</div>
      <div> </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
PyMOL-users mailing list (<a class="moz-txt-link-abbreviated" href="mailto:PyM...@li...">PyM...@li...</a>)
Info Page: <a class="moz-txt-link-freetext" href="https://lists.sourceforge.net/lists/listinfo/pymol-users" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/pymol-users">https://lists.sourceforge.net/lists/listinfo/pymol-users</a>
Archives: <a class="moz-txt-link-freetext" href="http://www.mail-archive.com/pym...@li..." rel="nofollow">http://www.mail-archive.com/pym...@li...">http://www.mail-archive.com/pym...@li...</a>
Unsubscribe: <a class="moz-txt-link-freetext" href="https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe

Please unsubscribe me.

________________________________
Thomas Fouqueau, PhD.
Research Associate
RNAP lab
Department of Structural and Molecular Biology (ISMB)
University College London (UCL)
Gower Street
London, WC1E 6BT
Tel.: +44 20 7679 0477
https://sites.google.com/site/rnaplab/

________________________________
From: hy2565 <hy...@co...>
Sent: 16 October 2018 15:08:08
To: pymol-users
Subject: [PyMOL] Please unsubscribe me

Please unsubscribe me.

Please Unsubscribe Me