From chemistry-request@server.ccl.net Sat Feb 10 07:19:10 2001 Received: from email03.aon.at ([195.3.96.112]) by server.ccl.net (8.11.0/8.11.0) with SMTP id f1ACJAw29976 for ; Sat, 10 Feb 2001 07:19:10 -0500 Received: (qmail 4309742 invoked from network); 10 Feb 2001 12:16:54 -0000 Received: from n773p005.adsl.highway.telekom.at (HELO pc1.eidolon.net) ([62.47.40.133]) (envelope-sender ) by qmail3.highway.telekom.at (qmail-ldap-1.03) with SMTP for ; 10 Feb 2001 12:16:54 -0000 Sender: hal@pc1.eidolon.net To: CCL Subject: comparability of different ab initio software packages From: Hannes Loeffler Date: 10 Feb 2001 13:11:56 +0100 Message-ID: <87lmreagab.fsf@pc1.eidolon.net> Lines: 20 User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (Cuyahoga Valley) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Hello, we use several quantum chemistry packages (Gaussian, TURBOMOLE, and currently I'm looking into GAMESS) and I wonder how to compare results. These programs use different convergence criteria (energy vs. density matrix) by default, different cutoffs, exploit of symmetry, possibly other approximations and assumptions, etc. This makes it somewhat difficult to compare these packages. Basically all these program should yield the same result for a fundamental quantity like the energy given a specific method and basis set. However, is it possible to achieve this goal especially considering the different ways how to manipulate the way these programs do the calculations? And what does it mean to get the "same" energy? Or to view it differently: what if I get different results? Where to start? Where to look? I know that my questions are very general but I still hope to get some advice. TIA, Hannes. From chemistry-request@server.ccl.net Sat Feb 10 07:24:52 2001 Received: from email04.aon.at (WARSL401PIP7.highway.telekom.at [195.3.96.115]) by server.ccl.net (8.11.0/8.11.0) with SMTP id f1ACOqw30000 for ; Sat, 10 Feb 2001 07:24:52 -0500 Received: (qmail 183042 invoked from network); 10 Feb 2001 12:22:51 -0000 Received: from n773p005.adsl.highway.telekom.at (HELO pc1.eidolon.net) ([62.47.40.133]) (envelope-sender ) by qmail4.highway.telekom.at (qmail-ldap-1.03) with SMTP for ; 10 Feb 2001 12:22:51 -0000 Sender: hal@pc1.eidolon.net To: CCL Subject: quality of gradients From: Hannes Loeffler Date: 10 Feb 2001 13:17:53 +0100 Message-ID: <87hf22ag0e.fsf@pc1.eidolon.net> Lines: 12 User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (Cuyahoga Valley) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Hello, in my other mail I asked about energy convergence criteria. Now I would like to know how they influence the quality of gradients. Decreasing convergence criteria naturally saves computation time but will this compromise overall quality? Are there any studies with more details? I should mention that I need the gradients for MD simulations. TIA, Hannes. From chemistry-request@server.ccl.net Sat Feb 10 11:09:57 2001 Received: from web9610.mail.yahoo.com (web9610.mail.yahoo.com [216.136.129.189]) by server.ccl.net (8.11.0/8.11.0) with SMTP id f1AG9vw30410 for ; Sat, 10 Feb 2001 11:09:57 -0500 Message-ID: <20010210160956.42579.qmail@web9610.mail.yahoo.com> Received: from [161.142.10.11] by web9610.mail.yahoo.com; Sat, 10 Feb 2001 08:09:56 PST Date: Sat, 10 Feb 2001 08:09:56 -0800 (PST) From: sohail qamar Subject: AMBER6*COMPILATION ON SUN* To: chemistry@ccl.net MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii I am getting errors with f77 compiler.I have installed the licenses properly(as far as I know) but still its giving me an output of missing licence file in flexlm/licences/licences.dat If I put the appropriate license files in path it still give me errors. It also give an error of setting domain name to 1 or to use etc/resolve.configfile but I have none in my sun. If any friend has compiled amber6 on sun.sparc please give me some idea. sohail. __________________________________________________ Do You Yahoo!? Get personalized email addresses from Yahoo! Mail - only $35 a year! http://personal.mail.yahoo.com/