<html><head></head><body><div style="color:#000; background-color:#fff; font-family:HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif;font-size:10px"><div id="yui_3_16_0_ym19_1_1484687367745_3907">Dear Werner</div><div id="yui_3_16_0_ym19_1_1484687367745_3910"><br></div><div id="yui_3_16_0_ym19_1_1484687367745_3857" dir="ltr">Thank you for the suggestions! I guess it is related to the choice of the basis. Since I used a aug-cc-pcwVQZ basis, there is no corresponding JK fit ones. I used an aug-cc-pV5Z JKFIT (see attached input). Do you have other suggestions for the choice of ri, df, and jk basis for the aug-cc-pcwVQZ orbital basis? BTW, the molpro version I used is 2012.1. Thanks again!</div><div id="yui_3_16_0_ym19_1_1484687367745_3857" dir="ltr"><br></div><div id="yui_3_16_0_ym19_1_1484687367745_3857" dir="ltr">Best wishes</div><div id="yui_3_16_0_ym19_1_1484687367745_3857" dir="ltr"><br></div><div id="yui_3_16_0_ym19_1_1484687367745_3857" dir="ltr">John</div> <div class="qtdSeparateBR"><br><br></div><div class="yahoo_quoted" style="display: block;"> <div style="font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size: 10px;"> <div style="font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size: 16px;"> <div dir="ltr"><font size="2" face="Arial"> On Tuesday, January 17, 2017 8:13 AM, Werner Győrffy <gyorffy@theochem.uni-stuttgart.de> wrote:<br></font></div> <br><br> <div class="y_msg_container">Dear John,<br clear="none"><br clear="none">There might be some problem with the choice of the basis sets you use. <br clear="none">It is recommended to use the optri basis sets, for example, for <br clear="none">basis=avtz use ccsd(t)-f12b,ri_basis=avtz/optri. Could you please <br clear="none">provide an input for your calculation failed?<br clear="none"><br clear="none">Regards,<br clear="none"><br clear="none">Werner.<br clear="none"><div class="yqt7423207528" id="yqtfd48410"><br clear="none">On 01/16/2017 05:41 PM, John Travers wrote:<br clear="none">> Hi<br clear="none">><br clear="none">> I was trying to do a numerical frequency calculation using f12-ccsd(t)<br clear="none">> in c1 symmetry for a small molecule. The program always produces some<br clear="none">> warning messages like : "Poor overlap with frozen basis set. May have<br clear="none">> deleted the wrong vectors!<br clear="none">> ? The problem occurs in SvdConstructOrthoBasis"<br clear="none">><br clear="none">> When checked the detailed output, I also found some possible numerical<br clear="none">> errors, something looks like :<br clear="none">><br clear="none">> Construction of CABS:<br clear="none">> Pseudo-inverse stability NaN<br clear="none">> Smallest eigenvalue of S -3.71E-13 (threshold= 1.00E-08)<br clear="none">> Ratio eigmin/eigmax -3.71E-13 (threshold= 1.00E-09)<br clear="none">> Smallest eigenvalue of S kept -3.71E-13 (threshold=-3.71E-13, 0<br clear="none">> functions deleted, 658 kept)<br clear="none">> Maximum overlap with frozen deleted vectors: (nDel= 44)<br clear="none">> Vector: 1 2 3 4 5 6 7<br clear="none">> 8 9 10 11 12 13 14 15 16<br clear="none">> 17 18 51 25 67 44 36 35 43<br clear="none">> 32 40 38 33 39 55 52 34 22<br clear="none">> 37 21 42 50 20 41 74 19 53<br clear="none">> 54 26 23 231 28 24 95 93 27<br clear="none">> 158 75 61 123 68 108 31 63 163<br clear="none">> 121 203 152 62 29 157 45 94 72<br clear="none">> 97 77 65 204 103 30 115 126 66<br clear="none">> 250 73 131 101 162 116 91 200 122<br clear="none">> Overlap: NaN NaN NaN NaN NaN NaN NaN<br clear="none">> NaN NaN NaN NaN NaN NaN NaN NaN NaN<br clear="none">> NaN NaN 1.436 1.248 1.102 1.000 0.999 0.997 0.996<br clear="none">> 0.994 0.985 0.983 0.973 0.964 0.962 0.912 0.896 0.871<br clear="none">> 0.858 0.853 0.827 0.825 0.742 0.596 0.594 0.537 0.523<br clear="none">> 0.485 0.465 0.386 0.375 0.373 0.346 0.327 0.321 0.310<br clear="none">> 0.307 0.268 0.259 0.257 0.249 0.247 0.216 0.204 0.194<br clear="none">> 0.191 0.183 0.166 0.164 0.159 0.135 0.134 0.130 0.126<br clear="none">> 0.118 0.114 0.110 0.100 0.095 0.093 0.092 0.092 0.091<br clear="none">> 0.089 0.080 0.079 0.079 0.078 0.076 0.075 0.072 0.070<br clear="none">><br clear="none">><br clear="none">> My questions are related to these "NaN" and warnings. Do these matter<br clear="none">> and how to avoid this? Thanks!<br clear="none">><br clear="none">> Best wishes<br clear="none">><br clear="none">><br clear="none">> John</div><br clear="none">><br clear="none">><br clear="none">><br clear="none">><br clear="none">><br clear="none">> _______________________________________________<br clear="none">> Molpro-user mailing list<br clear="none">> <a shape="rect" ymailto="mailto:Molpro-user@molpro.net" href="mailto:Molpro-user@molpro.net">Molpro-user@molpro.net</a><br clear="none">> <a shape="rect" href="http://www.molpro.net/mailman/listinfo/molpro-user" target="_blank">http://www.molpro.net/mailman/listinfo/molpro-user</a><br clear="none">><br clear="none">_______________________________________________<br clear="none">Molpro-user mailing list<br clear="none"><a shape="rect" ymailto="mailto:Molpro-user@molpro.net" href="mailto:Molpro-user@molpro.net">Molpro-user@molpro.net</a><br clear="none"><a shape="rect" href="http://www.molpro.net/mailman/listinfo/molpro-user" target="_blank">http://www.molpro.net/mailman/listinfo/molpro-user</a><div class="yqt7423207528" id="yqtfd90110"><br clear="none"></div><br><br></div> </div> </div> </div></div></body></html>