From owner-chemistry@ccl.net Wed Aug 12 07:11:00 2015 From: "MARK mark94025^_^comcast.net" To: CCL Subject: CCL:G: GPU enabled Computational Software Message-Id: <-51576-150811224829-14803-v3iabS/6xtj6//Kasl0B9Q^_^server.ccl.net> X-Original-From: MARK Content-Type: multipart/alternative; boundary="----=_Part_3124173_1449034476.1439347700928" Date: Wed, 12 Aug 2015 02:48:20 +0000 (UTC) MIME-Version: 1.0 Sent to CCL by: MARK [mark94025-$-comcast.net] ------=_Part_3124173_1449034476.1439347700928 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit I'll let Jimmy Stewart respond as MOPAC is his baby. My assumption would be that once enhancements are in the main source code tree that those stay there for future releases. But Dr Stewart should confirm. ----- Original Message ----- > From: "John Keller jwkeller .. alaska.edu" To: "Mark Berger " Sent: Tuesday, August 11, 2015 5:20:19 PM Subject: CCL:G: GPU enabled Computational Software Thanks Mark. What is MOPAC2013? I see only MOPAC2012 on the MOPAC website. John Keller On Tue, Aug 11, 2015 at 2:37 PM, MARK mark94025() comcast.net < owner-chemistry..ccl.net > wrote: I forgot a recent QC addition via OpenACC; LS-DALTON > From: "MARK mark94025/./ comcast.net " To: "Mark Berger " Sent: Tuesday, August 11, 2015 10:53:49 AM Subject: CCL:G: GPU enabled Computational Software After further mature reflection, I thought I'd send the actual list of GPU-accelerated MD, QC, docking and viz progrems. See below: GPU-accelerated quantum chemistry codes are: Abinit ACES III ADF BigDFT CP2K GAMESS GAMESS-UK Gaussian (in development via OpenACC) GPAW LATTE MOLCAS MOPAC2013 NWChem OCTOPUS Q-CHEM Quantum Espresso/PWscf QUICK TeraChem VASP GPU-accelerated MD programs are: ACEMD AMBER (PMEMD) CHARMM DESMOND ESPPesSo Folding%Home GPUgrid.net GROMACS HALMD HOOMD-Blue LAMMPS NAMD OpenMM Molecular Visualization and Docking programs are: AMIRA BINDSURF BUDE Core Hopping FastROCS Interactive Molecule Visualizer Molegro Virtual Docker PIPER Protein Docking PyMol VEGA ZZ VMD > From: "MARK" To: "CCL Subscribers" Sent: Wednesday, July 29, 2015 4:00:14 PM Subject: Re: CCL:G: GPU enabled Computational Software See this link on the NVIDIA website for GPU-accelerated applications: http://www.nvidia.com/object/gpu-applications.html Search by molecular dynamics, quantum chemistry, bioinformatics or visualization and docking Mark Berger Senior Alliance Manager, Life & Material Sciences NVIDIA Corporation > From: "John McKelvey jmmckel.%. gmail.com " To: "Mark Berger " Sent: Wednesday, July 29, 2015 1:59:04 PM Subject: CCL:G: GPU enabled Computational Software There is also MOPAC 2012, I believe.... John McKelvey On Wed, Jul 29, 2015 at 1:16 PM, Kristopher Keipert kwk% iastate.edu < owner-chemistry[*]ccl.net > wrote:
The most full featured is Terachem but it is not free. There is also LIBCCHEM for GAMESS-US. On Wed, Jul 29, 2015 at 10:22 AM, Jim Kress jimkress35-$- gmail.com < owner-chemistry-,-ccl.net > wrote:
In searching, I find a somewhat limited implementation of utilization of GPUs in Quantum Chemistry software. Does anyone know what QM/ QC package(s) (excluding Gaussian) have the broadest implementation of GPU acceleration? Thanks. Jim
-- John McKelvey 545 Legacy Pointe Dr O'Fallon, MO 63376 636-294-5302 jmmckel[*]gmail.com
------=_Part_3124173_1449034476.1439347700928 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable
I'll let Jimmy Stewart respond as MOPA= C is his baby.  

My assumption would be that once enhancements a= re in the main source code tree that those stay there for future releases. =  But Dr Stewart should confirm.

<= div style=3D"color:#000;font-weight:normal;font-style:normal;text-decoratio= n:none;font-family:Helvetica,Arial,sans-serif;font-size:12pt;" data-mce-sty= le=3D"color: #000; font-weight: normal; font-style: normal; text-decoration= : none; font-family: Helvetica,Arial,sans-serif; font-size: 12pt;">From:= "John Keller jwkeller .. alaska.edu" <owner-chemistry[#]ccl.net>To: "Mark Berger " <mark94025[#]comcast.net>
Sen= t: Tuesday, August 11, 2015 5:20:19 PM
Subject: CCL:G: GPU en= abled Computational Software

Thanks= Mark. What is MOPAC2013? I see only MOPAC2012 on the MOPAC website.
<= div>John Keller


On Tue, Aug 11, 2015 at 2:37 PM, MARK mark94025() = comcast.net <owner-chemistry..ccl.net> wrote:
I forgot a recent QC addition via OpenACC; LS-DALTON=

From: "MARK mark94025/./comcast.net"= <owner-chemistry*ccl.net>
To: "Mark Berger = " <mark94025*comcast.net>
Sent: Tuesday, Aug= ust 11, 2015 10:53:49 AM

Subject: CCL:G: G= PU enabled Computational Software

After further mature reflection, I= thought I'd send the actual list of GPU-accelerated MD, QC, docking and vi= z progrems.  See below:

GPU-accelerated quantum chemistry codes are:

 

=

Abinit

ACES III

ADF

BigDFT

CP2= K

GAMESS

=

GAMESS-UK

Gaussian (in development via OpenACC)=

GPAW

LATTE

MOLCAS

MOPAC2013

NWChem

OCT= OPUS

Q-CHEM<= /p>

Quantum Espresso/PWscf

QUICK

TeraChem

VASP

 

GPU-accelerated MD programs are:

 

ACEMD

AMBER (PMEMD)

C= HARMM

DESMOND

ESPPesSo

Folding%Home

GPUgrid.net

GROMACS

HALMD

HOOMD-Blue

LAMM= PS

NAMD

<= p style=3D"margin:0px" data-mce-style=3D"margin: 0px;">OpenMM

 

Molecular Visualization and Docking programs are:

 

AMIRA

BINDSURF

BUDE

Core Hopping

FastROCS

Inter= active Molecule Visualizer

Molegro Virtual Docker

PIPER Protein Docking

PyMol

VMD


From: "MARK" <mark94025%comcast.= net>
To: "CCL Subscribers" <chemistry%ccl.net>= ;
Sent: Wednesday, July 29, 2015 4:00:14 PM
Subject: Re= : CCL:G: GPU enabled Computational Software

See this link on the NVI= DIA website for GPU-accelerated applications: http://www.nvidia.com/ob= ject/gpu-applications.html 

Search by mol= ecular dynamics, quantum chemistry, bioinformatics or visualization and doc= king

Mark Berger
Senior Alliance Manager= , Life & Material Sciences
NVIDIA Corporation

<= /div>
From: "John McKelvey jmmckel.%.gmail.com" <owner-che= mistry%ccl.net>
To: "Mark Berger " <mark94025%comcast.net>
Sent: Wednesday, July 29, 2015 1:59:04 P= M
Subject: CCL:G: GPU enabled Computational Software

=
There is also MOPAC 2012, I believe....

John McKelvey

On Wed, Jul 29, 2015 at 1:16 PM, Kristopher Keipert = kwk%iastate.edu <owner-chemistry[*]ccl.net> wrote:
The most full fea= tured is Terachem but it is not free.
There is also LIBCCHEM for G= AMESS-US.

On Wed, Jul 29, 2015 at 10:22 AM, Jim Kress jimkress35-$-gmail.com <owner-chemistry-,-ccl.net> wrote:

In searching, I find a somewhat limited implementatio= n of utilization of GPUs in Quantum Chemistry software.  Does anyone k= now what QM/ QC package(s) (excluding Gaussian) have the broadest implement= ation of GPU acceleration?=

 

Thanks.

 

<= span style=3D"font-size:11.0pt;font-family:"Calibri",sans-serif;c= olor:#1f497d" data-mce-style=3D"font-size: 11.0pt; font-family: 'Calibri',s= ans-serif; color: #1f497d;">Jim


=



--
John McKe= lvey
545 Legacy Pointe Dr
=





------=_Part_3124173_1449034476.1439347700928-- From owner-chemistry@ccl.net Wed Aug 12 11:42:01 2015 From: "Jimmy Stewart MrMOPAC**att.net" To: CCL Subject: CCL: GPU enabled Computational Software Message-Id: <-51577-150812113848-10260-1k5q64YhZnYoHeg4nwovTA^-^server.ccl.net> X-Original-From: Jimmy Stewart Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=utf-8; format=flowed Date: Wed, 12 Aug 2015 09:38:47 -0600 MIME-Version: 1.0 Sent to CCL by: Jimmy Stewart [MrMOPAC(~)att.net] Replying to Mark Berger's invitation: The current MOPAC is MOPAC2012. Earlier MOPAC's were MOPAC2009 and MOPAC2007. Individual versions of MOPAC are also identified by a numeric code, e.g., 15.168, which can be read as the year (2015) minus 2000, and the Julian date within that year (day 168). This code can be seen on the right-hand-side of line 2 in the output, and on the RHS of line 5 in the archive file. Three types of MOPAC are supported: Windows, Mac (OSX), and Linux. These are identified by the letters W, M, and L, for example 15.168L for the Linux version of MOPAC. Bug fixes and minor improvements are being made continuously at the rate of about four a week. These are normally not documented; the only indication is that the Julian date has changed. Changes in the MOPAC program number, e.g., MOPAC2009 => MOPAC2012, occur when major changes, such as the addition of PM7 and the large expansion of protein-handling functions are made. Adding the ability to use the NVIDIA GPU increases the speed of MOPAC by a factor of just over 3 for bacteriorhodopsin. However, using a GPU would produce exactly the same results as not using it, therefore it did not justify any change in the year-name of MOPAC. Unless plans change, the next scheduled MOPAC after MOPAC2012 will be MOPAC2016. Jimmy Stewart From owner-chemistry@ccl.net Wed Aug 12 14:17:00 2015 From: "Mallika Khare mkhare=-=mtu.edu" To: CCL Subject: CCL:G: Geom=Connectivity Message-Id: <-51578-150812134752-30508-wTdlrhMhrcXmawWZ+Uzm6w],[server.ccl.net> X-Original-From: "Mallika Khare" Date: Wed, 12 Aug 2015 13:47:51 -0400 Sent to CCL by: "Mallika Khare" [mkhare*o*mtu.edu] 2108-A Upper Daniel Heights Woodmar drive Hello I wish to ask about the input command Geom=connectivity used in Gaussian. If I am using this command with "%OldChk= Checkpointfile_name.chk" command, I am able to view my input structure in GaussView. But, usually when we use checkpoint file we should use Geom= AllCheck. On using "Geom=AllCheck" I am not able to view my input file in GaussView. It shows me error. I would be very thankful if anyone of you can explain why is this happening. Thanks and Regards Mallika Khare mkhare]![mtu.edu From owner-chemistry@ccl.net Wed Aug 12 14:51:01 2015 From: "Eric G. Bowes egbowes/a\chem.ubc.ca" To: CCL Subject: CCL: Use of $CHOOSE with moderately large metal complexes Message-Id: <-51579-150812141832-17646-h2pJNHhM45N0B78AA3Oj6g*o*server.ccl.net> X-Original-From: "Eric G. Bowes" Date: Wed, 12 Aug 2015 14:18:10 -0400 Sent to CCL by: "Eric G. Bowes" [egbowes[*]chem.ubc.ca] Hello, I am trying to compare the degree of covalency in Ni-P bonds in a number of (dtbupe)NiX2 complexes by analyzing the FILENAME.47 file output from g09 with gennbo 6.0. Upon analysis of the NBO data, it seems that the lewis structure chosen does not have Ni-P bonds (or only has one). This seems to be the case especially when the P atom is located trans to a strong trans influence donor such as C. In order to obtain the information needed, I plan to manually choose a Lewis structure which corresponds to a complex with two Ni-P bonds. As such, I have two questions: (1) Am I able to automatically generate my $CHOOSE input somehow, or do I actually have to type out each bond and lone pair for a molecule this large? For something as simple as formaldehyde this is trivial, but when comparing a series of ~50 atom metal complexes this appears to be extremely time consuming. (2) Assuming that the non-Lewis density is larger in my chosen structure, is it valid to analyze the polarity of the Ni-P bonds given in this output? Thank you Eric G. Bowes PhD Candidate, Love Group University of British Columbia Vancouver BC, Canada