R-beta: Crashes

Jim Lindsey (jlindsey@luc.ac.be)
Mon, 16 Jun 1997 09:45:43 +0200 (MET DST)


From: Jim Lindsey <jlindsey@luc.ac.be>
Message-Id: <9706160745.AA31355@alpha.luc.ac.be>
Subject: R-beta: Crashes
To: r-testers@stat.math.ethz.ch (r-testers)
Date: Mon, 16 Jun 1997 09:45:43 +0200 (MET DST)

I am having problems with R crashing when I optimize complex
likelihoods with nlm. I have about 1500 observations and some very
nonlinear PKPD models with up to 28 parameters using various
distributions.
  With the standard memory size, I get the occasional core dump of
about 29mb. If I up memory to -n288000 and -v5, the dumps seem
somewhat rarer but are larger: 34mb. They occur both with 0.49 and
0.16.1 although I have primarily been using the former.
  They are two extra uncontrolled factors: I have a new portable and
have installed the latest Red Hat Linux 2.0.27. (The latter, for X
windows,
 is very much less
stable than my old Slackware a.out Linux 1.2.13!!!) The portable has a
Pentium 133 that gets very warm when doing such hard work.

 Any ideas? Jim
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=