From owner-chemistry@ccl.net Sun Feb 17 00:02:01 2008 From: "Chun-Yi Sung juneyi1^_^yahoo.com.tw" To: CCL Subject: CCL: Fixing MM atoms in ONIOM Message-Id: <-36316-080216234805-8773-V9W5HHUpjRc0RBUzd8pU/A|-|server.ccl.net> X-Original-From: "Chun-Yi Sung" Date: Sat, 16 Feb 2008 23:48:01 -0500 Sent to CCL by: "Chun-Yi Sung" [juneyi1:_:yahoo.com.tw] Hi, CCLers I tried to fix MM atoms in ONIOM like this: Si-Si3 -1 3.063865 -6.363456 10.491696 L and for the QM atoms I wanted to fix, I flagged them with 0 like this: Si-Si3 0 -3.978982 4.068174 2.511722 H but then I used "X n F" at the end along with "opt=modredundant" in the Route section. However, in the output file, I got a huge block of stuff like this: NRF= 0 NRA= 0 NVA= 0 HaveQM=F Convergence limit is 0.300E-04 MaxStp= 5000 StMxLn= 1.00D-04 StpMin= 1.00D-06. Convergence criteria 0.00004500 0.00003000 0.00018000 0.00012000 Step NS ND Rises OKQ Scale Max. Force RMS Force Max. Disp. RMS Disp. Energy Flag 1 0 0 F T 1.00D+00 0.00000000 NaNQ 0.00000000 NaNQ -0.0450025 +-+- NRF= 0 NRA= 0 NVA= 0 HaveQM=F Convergence limit is 0.300E-04 MaxStp= 5000 StMxLn= 1.00D-04 StpMin= 1.00D-06. Convergence criteria 0.00004500 0.00003000 0.00018000 0.00012000 Step NS ND Rises OKQ Scale Max. Force RMS Force Max. Disp. RMS Disp. Energy Flag 1 0 0 F T 1.00D+00 0.00000000 NaNQ 0.00000000 NaNQ -0.0450025 +-+- during each geom optimization step, and I always have "NaNQ" for the RMS for MM Force like this: Item Value Threshold Converged? Maximum Force 0.064776 0.000450 NO RMS Force 0.006255 0.000300 NO Maximum Displacement 0.164584 0.001800 NO RMS Displacement 0.013024 0.001200 NO Maximum MM Force 0.000000 0.000045 YES RMS MM Force NaNQ 0.000030 YES My questions are: 1. Did I fixed the MM atoms in the correct way? If not, what is the correct way? 2. If I did fix MM atoms correctly, is it normal to have the block shown above and NaNQ for RMS for MM Force ? and is there any command I can use so that it won't output the huge block shown above? thanks Chunyi From owner-chemistry@ccl.net Sun Feb 17 04:24:00 2008 From: "John Hearns john.hearns|streamline-computing.com" To: CCL Subject: CCL: Computing on graphical processors Message-Id: <-36317-080217034957-4665-D5Xbr2dKxWzVT9JZE8IaRg : server.ccl.net> X-Original-From: John Hearns Content-Transfer-Encoding: 7bit Content-Type: text/plain Date: Sun, 17 Feb 2008 08:16:34 +0000 Mime-Version: 1.0 Sent to CCL by: John Hearns [john.hearns(0)streamline-computing.com] On Sat, 2008-02-16 at 20:11 -0500, John McKelvey jmmckel,,gmail.com wrote: > Hello.. > > It seems that there is beginning to be significant efforts to do > processing using video card processors. Could someone comment on > this as to why this is being so successful [probably many factors, I'm interested in the Nvidia CUDA environment. I found that, certainly on SuSE Linux, you just need to download and install and you have a working environment - nvcc compiler, libraries etc. You can compile and experiment on a low end graphics card, but the same code can be transferred to one of their dedicated Tesla GPUs. the price of these has become more reasonable recently. We've had interest in clusters with GPUs from at least two chemistry departments, and there will probably be more. Its of course a matter of what goes around comes around - I remember as a baby graduate student being encouraged to make use of the brand new vector unit on the campus IBM 3090. John Hearns Senior HPC Engineer Streamline Computing From owner-chemistry@ccl.net Sun Feb 17 13:11:01 2008 From: "Anup Thomas anupiict\a/gmail.com" To: CCL Subject: CCL: Hardware availability to do electronic structure simulations Message-Id: <-36318-080216231317-30352-YpJe0mX3XftZffj4CekFZg|server.ccl.net> X-Original-From: "Anup Thomas" Date: Sat, 16 Feb 2008 23:13:13 -0500 Sent to CCL by: "Anup Thomas" [anupiict#gmail.com] Good evening Sir/Mam I want to do some CCSD , EOM-CCSD calculations on 90 and 78 electron species with a good basis set. Any one knows about any open computing facility availabe for doing computational chemistry work. If so please let me know. regards Anup Thomas JRF 7123 I and PC div I I C T India anupiict]~[gmail.com From owner-chemistry@ccl.net Sun Feb 17 13:46:01 2008 From: "Carlos Amador carlos.amador(a)servidor.unam.mx" To: CCL Subject: CCL: Computing on graphical processors Message-Id: <-36319-080217112304-6295-13p+nQGW1Lzsg+fSkLjSBQ ~ server.ccl.net> X-Original-From: "Carlos Amador" Date: Sun, 17 Feb 2008 11:23:01 -0500 Sent to CCL by: "Carlos Amador" [carlos.amador++servidor.unam.mx] Alejandro Pisanty asked me to comment on recent efforts to use the computing power of graphical processing units in quantum chemistry codes. I want to mention two very recent papers on the subject. One is by ourselves ---an effort lead by Aln Aspuru-Guzik at Harvard---: Accelerating Resolution-of-the-Identity Second Order Mller-Plesset Calculations with Graphical Processing Units L. Vogt, Roberto Olivares-Amaya, S. Kermes, Y. Shao, C. Amador-Bedolla, and A. Aspuru-Guzik Journal of Physical Chemistry A, 112 (in press, 2008) doi: 10,1021/jp0776762. (Published on Web 01/30/2008) and the other is by Todd Martinez from Urbana Quantum Chemistry on Graphical Processing Units. 1. Strategies for Two-Electron Integral Evaluation Ivan S. Ufimtsev and Todd J. Martnez* J. Chem. Theory Comput., 4 (2), 222 -231, 2008. 10.1021/ct700268q S1549-9618(70)00268-0 Web Release Date: January 25, 2008 These efforts show very promising results. I think this may be the means to another big step towards making computational chemistry real chemsitry... All best, Carlos Amador-Bedolla Facultad de Qumica, UNAM Mexico From owner-chemistry@ccl.net Sun Feb 17 14:25:00 2008 From: "Venable, Richard (NIH/NHLBI) E venabler()nhlbi.nih.gov" To: CCL Subject: CCL: Computing on graphical processors Message-Id: <-36320-080217133539-2862-c0A+02B88kAoo0HagHVThg(-)server.ccl.net> X-Original-From: "Venable, Richard (NIH/NHLBI) [E]" Content-class: urn:content-classes:message Content-Type: multipart/mixed; boundary="----_=_NextPart_001_01C87193.DE590CD4" Date: Sun, 17 Feb 2008 13:35:28 -0500 MIME-Version: 1.0 Sent to CCL by: "Venable, Richard (NIH/NHLBI) [E]" [venabler=-=nhlbi.nih.gov] This is a multi-part message in MIME format. ------_=_NextPart_001_01C87193.DE590CD4 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable As others have noted, wide availability, low cost (driven by the gaming = market), and existing toolkits have favored using GPUs, which typically = have several FPUs on board. Some tinkering is required to get the best = performance, based on a talk I heard recently. A current limitation is = that most of the FPUs are single precision (32-bit f.p.), which is = sufficient for graphics rendering. -- Rick Venable 5635FL/T906 Membrane Biophysics Section NIH/NHLBI Lab. of Computational Biology Bethesda, MD 20892-9314 U.S.A. (301) 496-1905 venabler AT nhlbi*nih*gov -----Original Message----- > From: John McKelvey jmmckel,,gmail.com [mailto:owner-chemistry()ccl.net] Sent: Sat 16-Feb-08 8:11 PM To: Venable, Richard (NIH/NHLBI) [E] Subject: CCL: Computing on graphical processors =20 Hello.. It seems that there is beginning to be significant efforts to do = processing using video card processors. Could someone comment on this as to why = this is being so successful [probably many factors, I suppose], what kinds of computations work well on them, and what kinds of work probably do not = work well on them.... and why? Thanks! John McKelvey ------_=_NextPart_001_01C87193.DE590CD4-- From owner-chemistry@ccl.net Sun Feb 17 22:19:00 2008 From: "Sina T reli sinatureli-,-gmail.com" To: CCL Subject: CCL: Self Consistent Reaction Field Message-Id: <-36321-080217221746-29007-blmwIXlZWwMdPzxhHfSJ2w\a/server.ccl.net> X-Original-From: "Sina T reli" Date: Sun, 17 Feb 2008 22:17:42 -0500 Sent to CCL by: "Sina T reli" [sinatureli\a/gmail.com] Greetings, I have a question about scrf. While calculationg UV-spectrum of molecules, if we take them to be inside protein enviroment ligated to some protein side-chain, is it a good idea to use scrf and if it is what should be the index of refraction and dielectric constant to be used? Since it is burried inside the protein I assume a dielectric constant of 4-6 would be okay but I am not sure about the refractive index. Thanks...