[Wien] [Wien2k-announce] New Version of WIEN2k
Torsten Andersen
thor at physik.uni-kl.de
Thu Aug 26 08:01:18 CEST 2004
PS. After having slept on it, I added in the beginning of clmcopy.f:
logical kdelta
external kdelta
and now it compiles. I hope I did not do any damage by that...
Best regards,
Torsten Andersen.
Torsten Andersen wrote:
> Dear Peter,
>
> I downloaded this new version, since I found that the new features could
> be useful - especially the announced enhancements of mini and mixer. I
> get the following compilation errors when compiling on an IBM p630 with
> the same options as on the Wien2k faq page:
>
> ----<SRC_mini/compile.msg>----
> xlf90 -q64 -O5 -qarch=pwr4 -c drmngwien.f
> "drmngwien.f", line 162.29: 1513-041 (S) Arguments of the wrong type
> were specified for the INTRINSIC procedure "min".
> ** drmngwien === End of Compilation 1 ===
> 1501-511 Compilation failed for file drmngwien.f.
>
> ----> I changed line 162 to:
>
> f0=min(-0.5d0,fx+0.5d0)
>
> ----<SRC_mixer/compile.msg>----
> xlf90 -q64 -O5 -qarch=pwr4 -c mixer.f
> "mixer.f", line 295.23: 1513-041 (S) Arguments of the wrong type were
> specified for the INTRINSIC
> procedure "min".
> ** mixer === End of Compilation 1 ===
> 1501-511 Compilation failed for file mixer.f.
> make: 1254-004 The error code from the last command is 1.
>
> ----> I changed line 295 to:
>
> PRATT=min(0.1d0,qmx)
>
> ----<SRC_clmcopy/compile.msg>----
> xlf90 -q64 -O5 -qarch=pwr4 -c clmcopy.f
> "clmcopy.f", line 262.19: 1511-016 (S) Expression used in a logical IF
> or block IF must be of type logical. Respecify.
> "clmcopy.f", line 265.19: 1511-016 (S) Expression used in a logical IF
> or block IF must be of type logical. Respecify.
> "clmcopy.f", line 272.19: 1511-016 (S) Expression used in a logical IF
> or block IF must be of type logical. Respecify.
> ** clmcopy === End of Compilation 1 ===
> ** kdelta === End of Compilation 2 ===
> ** stern1 === End of Compilation 3 ===
> ** stern === End of Compilation 4 ===
> ** dmatcopy === End of Compilation 5 ===
> ** determinant === End of Compilation 6 ===
> ** vecprod === End of Compilation 7 ===
> ** dot === End of Compilation 8 ===
> ** euler === End of Compilation 9 ===
> ** find_rot_mat === End of Compilation 10 ===
> ** find_dlmat === End of Compilation 11 ===
> ** fact === End of Compilation 12 ===
> ** apply_inversion_ylm === End of Compilation 13 ===
> 1501-511 Compilation failed for file clmcopy.f.
> make: 1254-004 The error code from the last command is 1.
>
> ----> Line 262 reads: IF (KDELTA(Kvecvl(1,I),NST,krotst)) GOTO 50
>
> and I could not immediately get this to work. I noticed also that krotst
> has different dimensions in lines 262 and 272...
>
> What can be done?
>
> Best regards,
> Torsten Andersen.
>
>
>
>
> Peter Blaha wrote:
>
>> Dear WIEN users,
>>
>> A new version WIEN2k_04.8 is on the web.
>>
>> I'd like to thank in particular
>>
>> L.Marks for his improvements of mixer and mini (see
>> below) and
>>
>> S.Cottenier, T.Andersen, F.Nastos and K.Jorissen for their
>> continuous
>> effort in answering questions on the mailing list
>>
>> ... while I was on vacations.
>>
>>
>> The update contains significant changes in
>>
>> SRC: run*_lapw: (it will now check for the presence of broyden files and
>> delete them (after a warning and a delay of 60 seconds,
>> unless you specify -NI). By default it will now run
>> up to
>> 40 cycles.
>> min_lapw: will delete any *broyd* and *tmpM files and copy
>> .min_hess to
>> .minrestart (unless -nohess is not given) at startup
>> (unless
>> -NI is specified (eg. when continuing after a crash).
>>
>> When an scf file (with valence-forces!) is present, it
>> will
>> jump to mini, generate a new struct file and continue
>> with the
>> minimization. So you can now:
>> generate struct file
>> init_lapw
>> run_lapw -fc 1
>> check if forces are bigger than eg 5mRy/bohr
>> min_lapw
>>
>> New switches -p (adding parallel to the "default" job)
>> -sp (uses runsp instead of run in the
>> default job
>>
>> SRC_mixer: Modifications in the broyden scheme and optional additional
>> input
>> in case.inm (old input still works):
>> scale_pw, scale_clm: (reduce pw-scale, increase
>> clm-scale when
>> charge oszillations occur)
>> nbroyd (restarts the broyden after nbroyd (20) steps
>> (moving
>> the restart from run_lapw to mixer)
>> monitoring the broyden-performance (:DIRB), if "ANGLE" is
>> too big,
>> automatic restart
>> The new parameters are not yet tested for too many systems and
>> further
>> finetuning might be necessary. Nevertheless in some difficult cases
>> (magnetic surfaces,...) significant improvements were obtained.
>>
>> SRC_mini: PORT option further improved. New default inM file
>> (SRC_templates).
>> Please see new UG for details (min_lapw and mini sections).
>>
>> Regards
>>
>>
>> P.Blaha
>> --------------------------------------------------------------------------
>>
>> Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
>> Phone: +43-1-58801-15671 FAX: +43-1-58801-15698
>> Email: blaha at theochem.tuwien.ac.at WWW:
>> http://info.tuwien.ac.at/theochem/
>> --------------------------------------------------------------------------
>>
>>
>> _______________________________________________
>> Wien2k-announce mailing list
>> Wien2k-announce at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien2k-announce
>>
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>
>
--
Dr. Torsten Andersen TA-web: http://deep.at/myspace/
AG Hübner, Department of Physics, Kaiserslautern University
http://cmt.physik.uni-kl.de http://www.physik.uni-kl.de/
More information about the Wien
mailing list