From owner-nwchem-users@emsl.pnl.gov Fri Feb 8 04:25:02 2008 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 m18CP1Ts028121 for ; Fri, 8 Feb 2008 04:25:02 -0800 (PST) Received: (from majordom@localhost) by odyssey.emsl.pnl.gov (8.14.1/8.14.1/Submit) id m18CP1Qs028120 for nwchem-users-outgoing-0915; Fri, 8 Feb 2008 04:25:01 -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: 5.9 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AgAAAK/Yq0eK+z2un2dsb2JhbACBWY5cAQEBAQEGBAYJCBibNg X-IronPort-AV: E=Sophos;i="4.25,321,1199692800"; d="scan'208";a="65987001" Message-ID: <47AC4A01.5090406@st-andrews.ac.uk> Date: Fri, 08 Feb 2008 12:24:33 +0000 From: Herbert Fruchtl User-Agent: Thunderbird 1.5.0.12 (X11/20070530) MIME-Version: 1.0 To: Erik McNellis CC: NWChem Users Subject: Re: [NWCHEM] The RI-SCF code in NWChem References: <47AC3D2C.6020209@fhi-berlin.mpg.de> In-Reply-To: <47AC3D2C.6020209@fhi-berlin.mpg.de> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-StAndrews-MailScanner-Information: Please contact the ISP for more information X-StAndrews-MailScanner: No virus detected X-StAndrews-MailScanner-SpamCheck: not spam, SpamAssassin (score=0, required 5) X-StAndrews-MailScanner-From: herbert.fruchtl@st-andrews.ac.uk Sender: owner-nwchem-users@emsl.pnl.gov Precedence: bulk Hi Erik, I'm surprised it's still in there. It is based on dense matrix multiplications, so the scaling isn't actually that good if you go to more sparse systems. I should know; I wrote it... With a bit of googling I found an old 4.0 manual on the web: laplace.physics.ubc.ca/VN/nwchem/doc/user/user.001.ps No guarantees that it works, since the code is no longer supported. Cheers, Herbert Erik McNellis wrote: > Dear All, > > I am doing a series of large HF based calculations, and would like to > minimize my basis set errors, which gets prohibitively expensive on the > hardware available to me. > > In an attempt to improve the scaling of my calculations, I read through > the Resolution of Identity SCF (RI-SCF) code distributed with NWChem. I > managed to figure out options and settings, and have now run a series of > tests, so far with good results. > > However, I cannot but interpret the complete lack of mention of the > RI-SCF code in the NWChem documentation and support resources, as this > functionality being unsupported and its use discouraged by the NWChem > developers. > > Is there a specific reason I should not use the RI-SCF code in NWChem? > > Thank you so much, > > Erik McNellis > -- Herbert Fruchtl EaStCHEM Fellow School of Chemistry University of St Andrews