From owner-nwchem-users@emsl.pnl.gov Tue Nov 20 02:01:43 2007 Received: from odyssey.emsl.pnl.gov (localhost [127.0.0.1]) by odyssey.emsl.pnl.gov (8.14.1/8.14.1) with ESMTP id lAKA1gtv024731 for ; Tue, 20 Nov 2007 02:01:43 -0800 (PST) Received: (from majordom@localhost) by odyssey.emsl.pnl.gov (8.14.1/8.14.1/Submit) id lAKA1gql024730 for nwchem-users-outgoing-0915; Tue, 20 Nov 2007 02:01:42 -0800 (PST) X-Authentication-Warning: odyssey.emsl.pnl.gov: majordom set sender to owner-nwchem-users@emsl.pnl.gov using -f X-Ironport-SG: OK_Domains X-Ironport-SBRS: 5.7 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AgAAAKw/QkeW2QEdnWdsb2JhbACBW404AQEBAQsPCGg X-IronPort-AV: E=Sophos;i="4.21,440,1188802800"; d="scan'208";a="57074988" Date: Tue, 20 Nov 2007 11:00:47 +0100 From: Alessandro Bencini Subject: Re: [NWCHEM] print all movecs to a text file In-reply-to: To: nwchem-users@emsl.pnl.gov Message-id: <4742B04F.8080409@unifi.it> MIME-version: 1.0 Content-type: text/plain; charset=ISO-8859-1; format=flowed Content-transfer-encoding: 8BIT User-Agent: Thunderbird 2.0.0.6 (X11/20070728) X-IP-UNIFI: 150.217.152.21 Auth bsandro@unifi.it References: <20071117111634.BDD72532@mailstore.wm.edu> <473FF16C.7010403@unifi.it> Sender: owner-nwchem-users@emsl.pnl.gov Precedence: bulk Some time ago I converted eigenvectors (in cartesians) from nwchem to gamess by proper reordering the d orbitals. A test in sto-3g. Gamess never found the energy converged, but starts some cylces of scf. The final energy reaches is the same as nwchem, but the vector coefficients are different. May be you know why? Alessandro Wirawan Purwanto wrote: > Ah, thanks for pointing this utility out! > > Yes, the d and f (and higher) functions are ordered differently among > these programs: g98, nwchem, gamess. > > If we use Cartesian functions, as far as I know, the d functions are > ordered the same in g98 and gamess. I don't know for higher functions, > but seems like the wave functions produced by these 2 packages can > interoperate. > > If we use spherical functions (1s 3p 5d 7f ...), then none of them > interoperate. g98 will order a d function like this: > > m= example amplitude: > 0 0.3 > +1 0.2 > -1 0.1 > +2 0.4 > -2 0.5 > > but nwchem orders it > -2 0.5 > -1 0.1 > 0 0.3 > 1 -0.2 << see below > 2 0.4 > > As an example above shows, for the positive odd magnetic quantum > numbers (m = +1, +3, ...), apparently there is a additional negative > sign flipping between the wave functions produced by nwchem and g98: > you have to flip the sign of the amplitudes of these functions if you > want to export nwchem orbitals to g98 format, or vice versa. Gamess > unfortunately does not produce output in spherical basis directly; it > reverts back to the cartesian functions. > > Wirawan > > > On Sun, 18 Nov 2007, Alessandro Bencini wrote: > >> In the /contrib directory of the distribution source there is a small >> utility, mov2asc, that allows one to convert the .movecs file to >> ascii form in double precision. >> After that one can quite easily extract the eigenvetors. Please note >> however that d and f functions are ordered in a different way as >> respect to g98 and gamess, for example. I am also not sure if the >> eigenvectors have the same normalization. >> >> Alessandro >> >> >> Wirawan Purwanto wrote: >>> The vectors will be printed in the text output; you will need to >>> extract it manually or using tools like perl. I think at least 6 >>> digits are printed. I don't know the easy way of reading movecs >>> file. But it is nice if nwchem can print out these vectors in full >>> precision (double precision, 15 digits). Just a suggestion for >>> nwchem developer. >>> >>> Wirawan >>> >>> ---- Original message ---- >>> >>>> Date: Sat, 17 Nov 2007 11:14:05 -0500 (EST) >>>> From: Wirawan Purwanto Subject: Re: [NWCHEM] print >>>> all movecs to a text file To: nwchem-users@emsl.pnl.gov >>>> >>>> I think this is the way to do it: >>>> >>>> dft >>>> ... >>>> print "final vectors" >>>> ... >>>> end >>>> >>>> Good luck. >>>> >>> >>> >> >> >> -- >> ___________________________________________________ >> Alessandro Bencini >> Dipartimento di Chimica >> Polo Scientifico dell'Universit? di Firenze >> via della Lastruccia, 3 50019 Sesto Fiorentino (FI) >> tel: 00390554573271 >> fax: 00390554573372 -- ___________________________________________________ Alessandro Bencini Dipartimento di Chimica Polo Scientifico dell'Università di Firenze via della Lastruccia, 3 50019 Sesto Fiorentino (FI) tel: 00390554573271 fax: 00390554573372