<div>I think it's not the bug you meet before. My Molpro is 2002.6 RPM package with </div>
<div>patch level 91. This problem occur when the "NUMBER OF PRIMITIVE AOS" larger </div>
<div>than a number, about 250.<br> </div>
<div>If you feel interest, you can try a SOC calculation with large basis set.<br> </div>
<div>Thank you.</div>
<div> </div>
<div>Xinzheng</div>
<div> </div>
<div> </div>
<div><span class="gmail_quote">On 9/14/05, <b class="gmail_sendername">Matt Thompson</b> <<a href="mailto:thompsma@jilau1.colorado.edu">thompsma@jilau1.colorado.edu</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">On Wed, 2005-09-14 at 14:53 -0500, Xinzheng Yang wrote:<br>> Dear Molpro users,<br>><br>> When I perform a spin-orbit coupling calculations with a large basis
<br>> set, I always meet this problem:<br>><br>> ?WARNING: TOO HIGH ADDRESS IN CORLSI: LT= 126481559 IBASE=<br>> 126261521 LTOP= 126372191 MEMSTACK= 326261521<br>> SYMMETRIES OF SO OPERATORS: 3 2 4
<br>><br>> But this calculation can be done with a smaller basis set.<br>><br>> My Molpro is a 32-bit version. Does this problem can be solved by<br>> using a 64-bit version?<br><br>This sounds suspiciously like a bug I filed a couple years ago (couldn't
<br>do SO with a Sadlej basis) that got fixed Oct. '04.<br><br>Assuming you run 2002.6, you should check to see if you have applied the<br>lsint patch, which is around patchlevel 70 or so. If you use the RPMs<br>Molpro provides, they have all been patched above this, so they are
<br>good. If not, use patcher and a make should fix it.<br><br>HTH,<br>Matt<br>--<br>The mayfly lives only one day, and sometimes it rains. - Geo. Carlin<br> Matt Thompson -- <a href="http://ucsub.colorado.edu/~thompsma/">
http://ucsub.colorado.edu/~thompsma/</a><br> 440 UCB, Boulder, CO 80309-0440<br> JILA A510, 303-492-4662<br><br></blockquote></div><br><br clear="all"><br>-- <br>Xinzheng Yang<br>1071 Clayton Ln, Apt. 1406<br>Austin, TX 78723