|
From: Martin B. <bal...@do...> - 2012-04-24 09:43:58
|
Dear Edward and everybody else, sorry for responding so late. Opening the app works flawlessly for me on Lion, there are no apparent errors when clicking around but I didn't have time to test anything else. I will test the command-line portion with the test suite data, but I can't say when I will have time. Cheers Martin On 20.04.2012, at 16:09, Edward d'Auvergne wrote: > Dear all, > > If you are a Mac OS X user, I would like to ask if you could help with > relax's support for Mac OS X? I have created a special relax > application which targets 10.5, 10.6, and 10.7 systems and which is a > 3-way universal binary build for ppc, i386, and x86_64 systems. It is > using a full Python 2.7 framework and the wxWidgets used to create the > graphical user interface is using Cocoa. Unfortunately I do not have > access to enough test systems covering all these OS versions and CPU > architectures to be able to comprehensively test the Mac support. I > have now successfully tested on real hardware (Intel x86_64) with both > Lion and Snow Leopard, but more testing would be useful, especially if > you have an old PPC system lying around. > > I have created a special relax version 1.3.15_rc1 release just for Mac > users. This can be found on the download page at > http://download.gna.org/relax/. This is a DMG file containing the > relax app. If you could download this 130 Mb file and see if relax > will simply open, taht would be very useful information. I have more > instructions below for performing full analyses using the data in the > relax test-suite, if are interested in helping to improve Mac OS X > support for relax. Any help would be appreciated! > > Cheers, > > Edward > > > > > P. S. Here are some instructions for extensively stress-testing > relax, which is also useful for learning how to use relax The data > from the test-suite is all truncated, and sometime randomised, but can > be used for a full analysis. > > > 1) Testing relax via analyses. If some value is not give below, then > it can be invented without consequence. All the data is in the > relax.app/Contents/Resources/test_suite folder. > > 1a) NOE analysis. > > There are two Sparky peak lists in > ./test_suite/shared_data/peak_lists/ called 'ref_ave.list' and > 'sat_ave.list'. To load the spin systems for these lists into relax, > the PDB file test_suite/shared_data/structures/Ap4Aase_res1-12.pdb can > be used. This is sufficient for a full analysis. > > > 1b) R2 analysis. > > A full set of Sparky peak lists for R2 data can be found in > ./test_suite/shared_data/curve_fitting/. The relevant sequence data > can be extracted from the test_suite/shared_data/Ap4Aase.seq file. > You can invent the delay times of these data sets for fun. > > > 1c) Model-free analysis. > > One of the many examples from the model-free shared data directories > is for a synthetic spherical diffusion model. The data can be found > at ./test_suite/shared_data/model_free/sphere. The PDB file and > relaxation files are all there, no other files are required. > > > 2) Stress testing. > > For this type of testing, randomly clicking everywhere and anywhere in > any order would be useful. Or throwing in any random data or files. > If relax responds with an error that is not a special 'RelaxError', > then this is a bug. The nastier you can break relax, the better! > > > 3) relax information > > Before reporting any problems, a copy of the relax information print > out would be incredibly useful (assuming you can open relax to start > with). You will find this under the Tools->System Information menu. > The output will go to the relax controller, and this text can be > copied and pasted. > > > 4) Reporting problems. > > If you notice any problems, could you please report this as a bug > report (https://gna.org/bugs/?func=additem&group=relax). Please try > to report all steps required to reproduce the problem, Include the > information print out, and copy and paste all error messages > (including most of the information before the error occurred). Note > that any strange behaviour outside of what you would expect is to be > considered a bug. Any suggestions would also be appreciated (though > if it involves months worth of work, it may not happen straight away > ;). Thank you very much! > > _______________________________________________ > relax (http://www.nmr-relax.com) > > This is the relax-users mailing list > rel...@do... > > To unsubscribe from this list, get a password > reminder, or change your subscription options, > visit the list information page at > https://mail.gna.org/listinfo/relax-users -- Martin Ballaschk AG Schmieder Leibniz-Institut für Molekulare Pharmakologie Robert-Rössle-Str. 10 13125 Berlin bal...@do... Tel.: +49-30-94793-234/315 Büro: A 1.26 Labor: C 1.10 |
