I have some PGI results. I think some of these are more involved than what goes on the "Test Results" page.
I got these from Shan-Ho Tsai at UGA's central computing, where they have an elaborate setup and multiple compilers. So far, she is getting pretty typical results with gcc (not finished). But, I wanted to go ahead and let you know about her PGI results.
I have her contact info if you want to talk to her directly. She says she is happy to provide more info if you need it.
:-) Lachele
Dr. B. Lachele Foley
Complex Carbohydrate Research Center
The University of Georgia
Athens, GA USA
lfoley.uga.edu
http://glycam.ccrc.uga.edu
2. Using PGI 12.3 compilers: there were no errors in the compilation,
but several tests had errors, namely
For Amber:
(i) the following 15 tests had errors:
(cd crg_reloc/water && ./Run.water.min)
(cd crg_reloc/water && ./Run.water.md)
(cd crg_reloc/cyt && ./Run.cyt.min)
(cd crg_reloc/cyt && ./Run.cyt.md)
(cd crg_reloc/misc && ./Run.dist.type1)
(cd crg_reloc/misc && ./Run.dist.type3)
(cd crg_reloc/misc && ./Run.dist.type4)
(cd crg_reloc/misc && ./Run.dist.type5)
(cd crg_reloc/misc && ./Run.angle.type1)
(cd crg_reloc/misc && ./Run.angle.type2)
(cd crg_reloc/misc && ./Run.angle.type3)
(cd crg_reloc/misc && ./Run.angle.type4)
(cd crg_reloc/misc && ./Run.mix)
(cd crg_reloc/1-4elec && ./Run.opt0)
(cd crg_reloc/1-4elec && ./Run.opt1)
For all these, the error message is
0: COUNT: invalid DIM argument
(ii) the mtmdtest tests failed drastically
(generated lost of NaNs). This error also
occurred in Amber11 and the solution is to
compile multitmd.f with no optimization.
So in src/sander/depend I changed
multitmd.o: multitmd.F90
$(FC) $(FPPFLAGS) -c $(FOPTFLAGS) $(FFLAGS) $(AMBERFFLAGS)
$(RISMSANDER) -o $. multitmd.F90
to
multitmd.o: multitmd.F90
$(FC) $(FPPFLAGS) -c $(FFLAGS) $(FFLAGS) $(AMBERFFLAGS)
$(RISMSANDER) -o $. multitmd.F90
After this change, the mtmdtests passed (this solution
was suggested to me by a developer, back when I was
building Amber11).
(iii) the qmmm2/meoh-nh3_PM6-DH+ test produce a
possible failure. However, it appears that there
are some formatting problems in the output file
and there are no actual numerical errors, but it's
hard to tell for sure. Here is the beginning of
test/qmmm2/meoh-nh3_PM6-DH+/meoh-nh3_md.out.dif:
106c106
< QMMM: to hold min npairs of 0 and nquant per thread of
10
> QMMM: to hold min npairs of 0 and nquant per thread of
107c107
< QMMM:
> 10
108c108
< QMMM: SCF Convergence Information
> QMMM:
109c109
< QMMM: (*) = Pseudo Diagonalisation
> QMMM: SCF Convergence Information
110c110
< QMMM: Cycle Energy dE dP |FP-PF|
> QMMM: (*) = Pseudo Diagonalisation
111c111
< QMMM: 1 -39850.63 -39850.63 0.1797693+309 -1.000000
> QMMM: Cycle Energy dE dP |FP-PF|
112c112
< QMMM: 2 -43497.89 -3647.252 0.5047034 -1.000000
> QMMM: 1 -39850.63 -39850.63 0.1797693+309 -1.000000
113c113
< QMMM: 3 -43525.47 -27.58606 0.1358226 -1.000000
> QMMM: 2 -43497.89 -3647.252 0.5047034 -1.000000
For AmberTools:
The following 3 tests showed possible failures:
1. AmberTools/test/pbsa_rot/mdout.charge.min.dif
2. AmberTools/test/pbsa_memb_test/aquaporin_coil.mem.out.dif
3.
AmberTools/test/mmpbsa_py/07_Comprehensive/FINAL_RESULTS_MMPBSA_1.dat.dif
However, 1 and 2 appear to be just output file formatting
problems, similar to the meoh-nh3_md.out.dif file above.
But 3 had actual large differences. The content of
FINAL_RESULTS_MMPBSA_1.dat.dif is
17c17
< Vibrational 70.6419 10.6543
4.7647
> Vibrational 68.1106 4.6216
2.0668
18c18
< Total 95.7422 11.1286
4.9769
> Total 93.2564 5.0712
2.2679
38c38
< Vibrational 12.7828 10.6560
4.7655
> Vibrational 10.2511 4.6255
2.0686
39c39
< Total -8.7066 11.1313
4.9781
> Total -11.1926 5.0770
2.2705
### Maximum absolute error in matching lines = 6.06e+00 at line 18 field 3
### Maximum relative error in matching lines = 1.31e+00 at line 17 field 3
So to summarize, using the PGI compilers AmberTools appears to
have one relevant test fail, and Amber has 15 tests that had
errors.
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Thu Mar 22 2012 - 11:30:02 PDT
