[phenixbb] --unused_ok, .def, documentation,...

Pavel Afonine PAfonine at lbl.gov
Tue May 12 08:12:34 PDT 2009


- As things are going more and more settled, the parameters are not 
changing that often anymore as it was at the beginning. So obviously if 
you download the new version a few times a week it is unlikely that you 
need to change your parameter files the same amount of times, or even once;

- Sometime new options/features come along with the new parameters for 
them. I'm wondering how people who download the newer version to use 
these options, will actually use them without updating their parameter 
files. In this case they just pretend using them -:)

- If we really want to keep "--unused-ok", then we should probably make 
it less obvious to use for newbies. If an expert knows it exists (which 
will be mentioned in the docs), then great - use it at your own risk. 
But for new users it is often the first thing they see when they 
mis-format they parameter files.

- One day running phenix.refine will be matter of running one command or 
hitting one button with no need to edit any parameters. That will be the 
day when the *true* automation of refinement won.


On 5/12/09 5:36 AM, Schubert, Carsten [PRDUS] wrote:
> Couple of comments:
>> I keep seeing people having trouble refining their structures in
> phenix.refine just because they were using "-->unused-ok" option. 
> Well, with every powerful tool come caveats. The unused_ok option is a
> expert option and if people who have difficulties understanding what
> they are doing use it and encounter problems then it begs the question
> why expert users need to suffer the consequences of their inabilities. I
> use this option often and it is warranted in the current environment in
> which phenix goes through a lot of revisions, which break compatibility
> with previous versions, but provide new/better functionality I'd like to
> use. And yes, if you are doing a couple of structures a day, reediting
> the parameter file(s) is a serious drag.
>> Finally, I don't understand what the trouble is to run this command:
>> phenix.refine --diff-params your_old_parameters.def >
> your_new_parameters.def
> It is NOT usable in a automation environment. Since phenix goes through
> so many releases, which change the parameter file and bails by default
> if the parameter file does not match the current version, automation is
> difficult to achieve. Correct me if I am wrong, but I believe it would
> almost be impossible to pick up a previous refinement done with a old
> phenix version again via an automated approach if the parameter file has
> changed and phenix bails. Granted I could freeze the phenix version used
> for automation and could maintain a couple of different versions for
> different purposes, but that quickly becomes impractical. 
> So in essence, for the "high-throughput-bleeding-edge-junky-users" the
> --unsed_ok option is essential and I hope that it remains.
> Cheers.
> 	Carsten
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://www.phenix-online.org/mailman/listinfo/phenixbb
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20090512/c7984f20/attachment-0003.htm>

More information about the phenixbb mailing list