Custom Search
--- > R1 = 0. R2 = 15.000 R3 = 15.000 R4 = 115.000 RK2 =5000.000 RK3 = 5000.000 --------------------------------------- possible FAILURE: check mdout.jar.000.dif /home/peker/amber10/test/jar_multi 87c87 < R1 = -85.000 R2 = 15.000 R3 = 15.000 R4 = 115.000 RK2 =5000.000 RK3 = 5000.000 > R1 = 0. R2 = 15.000 R3 = 15.000 R4 = 115.000 RK2 =5000.000 RK3 = 5000.000 ### Maximum absolute error in matching lines = 8.50e+01 at line 87 field 3 ### Maximum relative error in matching lines = 1.00e+00 at line 87 field 3 --------------------------------------- possible FAILURE: check mdout.jar.001.dif /home/peker/amber10/test/jar_multi 87c87 < R1 = -85.000 R2 = 15.000 R3 = 15.000 R4 = 115.000 RK2 =5000.000 RK3 = 5000.000 > R1 = 0. R2 = 15.000 R3 = 15.000 R4 = 115.000 RK2 =5000.000 RK3 = 5000.000 ### Maximum absolute error in matching lines = 8.50e+01 at line 87 field 3 ### Maximum relative error in matching lines = 1.00e+00 at line 87 field 3 --------------------------------------- None of those errors are located within the actual results of the calculation, just the functional form of the restraint potential. It's possible the test was changed, but the saved files were not updated. It's also possible that defaults were changed, but again not updated. In any case, R1 only signifies the point at which the potential becomes linear (rather than quadratic), so the two biassing potentials (the one used to generate the test and what's actually being run when you make test) differ only past 0 degrees for the torsion. The fact that none of the results differ probably means it's safe to use this functionality, you just have to be careful that the restraint function used by the program is what you expect it to be (i.e. check that I remember a while ago having this exact test failure, but I don't have a copy of amber10 lying around anywhere to refer back to. As to the tip5p error, the intel don't print out those last two lines, and I'm not sure why, but they are documented. I don't think there's anything actually wrong with that test. The only two tests that look worrisome are the LES 2temp.out (for some reason Etot is never getting computed for that frame), and the NEB test where a bunch of extra stuff is being dumped into the output file. Those two tests you'll want to be careful with. The only one I haven't mentioned (rdc) is just benign round-off, as has been mentioned in a previous post in this thread. All the best, Jason On Sat, May 22, 2010 at 8:28 PM, peker milas <pekermilas.gmail.com> wrote: > Thank you for those kind responses.and help. I will check them in the > way you explained. Have a nice weekend. > > peker > > On Sat, May 22, 2010 at 8:24 PM, case <case.biomaps.rutgers.edu> wrote: > > On Fri, May 21, 2010, peker milas wrote: > >> > >> Those test results belong to AmberTools1.2 with Amber10. I compiled > >> them with gcc(4.4.1)+gfortran on Ubuntu Karmic Koala. I applied all > >> bugfixes. Configuration and installation steps were just fine. I > >> couldn't find a complete document for explaining details of those > >> tests. I mean like if there is a test like jar_multi, what is its > >> actual function ? or if it fails which package will have a problem on > >> Amber10? So that was actually the main reason for my question. > > > > We don't have a document explaining all of the tests. You need to go to > the > > jar_multi subdirectory, and look at what is being run, and what the > nature > > of the failure is. These are non-equilibrium work calculations, so > failures > > may not mean you can't run other parts of Amber. > > > > ...dac > > > > > > _______________________________________________ > > AMBER mailing list > > AMBER.ambermd.org > > http://lists.ambermd.org/mailman/listinfo/amber > > > > _______________________________________________ > AMBER mailing list > AMBER.ambermd.org > http://lists.ambermd.org/mailman/listinfo/amber > -- Jason M. Swails Quantum Theory Project, University of Florida Ph.D. Graduate Student 352-392-4032 _______________________________________________ AMBER mailing list AMBER.ambermd.org http://lists.ambermd.org/mailman/listinfo/amberReceived on Sat May 22 2010 - 18:30:03 PDT