From chemistry-request#* at *#ccl.net Mon Nov 29 18:35:27 2004 Received: from mtiwmhc12.worldnet.att.net (mtiwmhc12.worldnet.att.net [204.127.131.116]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id iATNZNKV029913 for ; Mon, 29 Nov 2004 18:35:24 -0500 Received: from laptop.us.fujitsu.com (128.denver-05rh15-16rt.co.dial-access.att.net[12.73.180.128]) by worldnet.att.net (mtiwmhc12) with SMTP id <2004112923500211200lu35te>; Mon, 29 Nov 2004 23:50:03 +0000 Message-Id: <6.1.2.0.2.20041129164437.028c7178=at=postoffice.worldnet.att.net> X-Sender: mrmopac=at=postoffice.worldnet.att.net X-Mailer: QUALCOMM Windows Eudora Version 6.1.2.0 Date: Mon, 29 Nov 2004 16:48:06 -0700 To: "Axel Mathieu" , "CCL" From: "James J. P. Stewart" Subject: Re: CCL:Line minimization failed twice In-Reply-To: References: Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="=====================_180607029==.ALT" X-Spam-Status: No, hits=3.6 required=7.5 tests=HTML_30_40,HTML_MESSAGE, RCVD_IN_DYNABLOCK,RCVD_IN_SORBS autolearn=no version=2.61 X-Spam-Checker-Version: SpamAssassin 2.61 (1.212.2.1-2003-12-09-exp) on servernd.ccl.net --=====================_180607029==.ALT Content-Type: text/plain; charset="us-ascii"; format=flowed Geometry optimization in MOPAC proceeds by a series of line minimizations. If the line search can't find a point that is lower than the starting point of the line search, then an error must have occurred (in the algorithm), and the warning message is printed. This occurs only in the BFGS optimizer - more recent versions of MOPAC use Baker's EF by default - it's more robust. I hope this helps. Jimmy At 02:43 PM 11/29/2004, Axel Mathieu wrote: >Greetings, > >I've been trying out some PM3 and AM1 calculations on some of my molecules >and have obtained the following "error" in some of the log files: > >THE LINE MINIMIZATION FAILED TWICE IN A ROW. TAKE CARE! > SCF FIELD WAS ACHIEVED > >I understand that this has already been asked before on CCL (even up to >1995) and have found solutions to it in the archives (reference: >http://www.ccl.net/cgi-bin/ccl/message.cgi?1995+08+09+008 >). However, it does not explain the source of this error. I'm relatively >new to MOPAC and would appreciate your help in aiding me to understand the >meaning of this. How does "minimization" fail twice in a row? > >Many thanks, > >APM > ( |-at-| |-at-| ) .-----------------oOOo----(_)----oOOo-------------------------------------. | James J. P. Stewart | | | Stewart Computational Chemistry LLC | E-mail: jstewart=at=us.fujitsu.com | | 15210 Paddington Circle | 39/03/15 N, 104/49/29 W | | Colorado Springs CO 80921-2512 | | | USA .ooo0 | Phone: USA +(719) 488-9416 | | ( ) Oooo. | | .------------------------\ (----( )-------------------------------------. \_) ) / (_/ --=====================_180607029==.ALT Content-Type: text/html; charset="us-ascii" Geometry optimization in MOPAC proceeds by a series of line minimizations.  If the line search can't find a point that is lower than the starting point of the line search, then an error must have occurred (in the algorithm), and the warning message is printed.

This occurs only in the BFGS optimizer - more recent versions of MOPAC use Baker's EF by default - it's more robust.

I hope this helps.

Jimmy

At 02:43 PM 11/29/2004, Axel Mathieu wrote:
Greetings,
 
I ve been trying out some PM3 and AM1 calculations on some of my molecules and have obtained the following error in some of the log files:
 
THE LINE MINIMIZATION FAILED TWICE IN A ROW.   TAKE CARE!
     SCF FIELD WAS ACHIEVED                                  
 
I understand that this has already been asked before on CCL (even up to 1995) and have found solutions to it in the archives (reference: http://www.ccl.net/cgi-bin/ccl/message.cgi?1995+08+09+008 ). However, it does not explain the source of this error. I m relatively new to MOPAC and would appreciate your help in aiding me to understand the meaning of this. How does minimization fail twice in a row?
 
Many thanks,
 
APM
 


                         ( -!at!- -!at!- )
 .-----------------oOOo----(_)----oOOo-------------------------------------.
 | James J. P. Stewart                 |                                   |
 | Stewart Computational Chemistry LLC | E-mail:  jstewart=at=us.fujitsu.com  |
 | 15210 Paddington Circle             |  39/03/15 N, 104/49/29 W          |
 | Colorado Springs CO 80921-2512      |                                   |
 | USA                   .ooo0         | Phone: USA +(719) 488-9416        |
 |                       (   )   Oooo. |                                   |
 .------------------------\ (----(   )-------------------------------------.
                           \_)    ) /
                                 (_/ --=====================_180607029==.ALT--