RE: MAXIDS=10000

From: Thomas Ludden Date: December 01, 2006 technical Source: cognigen.com
Mark and Nick, I checked the reference copy of NONMEM V 1.1 and the Master CD used for the distribution of NONMEM V 1.1 dated 18 March 2005 and both have LIM6 0 in NSIZES. Perhaps the LIM6 value in NONMEM V 1.1 was changed from 200 to 600 for your respective installations. As already noted, the default value for this parameter was increased to 400 in both the NONMEM VI 1.0 Beta and the released version of NONMEM VI 1.0. Regarding MAXIDS: To sort the individual contributions of the objective function value prior to summation and to store these individual values in a common for later output by the user, it is necessary to provide a maximum size for the associated arrays. This is a limitation of FORTRAN 77. While it would have been possible to circumvent the limit when sorting was not selected, it would still have been necessary for storing these values for eventual output by the user. I would agree that a larger value should have been chosen. We have already had to increase the value from 1000 to 2000 for a current analysis. When NONMEM VI is first installed, users may want to adjust this value, the value for LIM6 and any other parameter values in SIZES to meet their specific needs or to match the values they have been using in NONMEM V. Tom Ludden Icon Development Solutions
Quoted reply history
-----Original Message----- From: owner-nmusers On Behalf Of Mark Sale - Next Level Solutions Sent: Thursday, November 30, 2006 8:40 PM To: Nick Holford Cc: nmusers Subject: RE: [NMusers] MAXIDS000 Nick, There is a buffer 6 size parameter (LIM6) in NSIZES in v5 and in SIZES in v6. I don't recall what array(s) are dimensioned with this. But, looks like the default size in v5 is 600 and default is 400 in v6. Another upward compatiability problem, perhaps. Do you have an individual with between 400 and 600 observations? Mark Sale MD Next Level Solutions, LLC www.NextLevelSolns.com > -------- Original Message -------- > Subject: [NMusers] MAXIDS000 > From: Nick Holford <n.holford > Date: Thu, November 30, 2006 7:38 pm > To: nmusers <nmusers > > Mark, > > I asssume, unlike the typical male, read all the instructions. So perhaps you can help... > > I have now increased MAXIDS to 10000 and get this error message: > 0TOT NO. OF PRED-DEFINED RECS IN BUFFER 6 IS LESS THAN NO. OF DATA RECS WITH SOME INDIVIDUAL > SEE INSTALLATION GUIDE > [This problem/data set runs fine with NMV] > > Where exactly is the installation guide that gives me instructions to deal with this problem? There is no guide by that name provided in the Guides folder. > > Nick > > -- > Nick Holford, Dept Pharmacology & Clinical Pharmacology > University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New Zealand > email:n.holford > http://www.health.auckland.ac.nz/pharmacology/staff/nholford/ _______________________________________________________________________________________________________________________________________
Nov 30, 2006 Nick Holford MAXIDS=10000
Nov 30, 2006 Mark Sale RE: MAXIDS=10000
Dec 01, 2006 Nick Holford Re: MAXIDS=10000
Dec 01, 2006 Mark Sale RE: MAXIDS=10000
Dec 01, 2006 Bob Leary RE: MAXIDS=10000
Dec 01, 2006 Thomas Ludden RE: MAXIDS=10000
Dec 01, 2006 Mark Sale RE: MAXIDS=10000
Dec 01, 2006 Thomas Ludden RE: MAXIDS=10000
Dec 04, 2006 Bob Leary MAXIDS=10000