Error while opening mrkxxxxz.tmp

announcements (new versions, changes, bugs, installation problems...) related to program MARK

Error while opening mrkxxxxz.tmp

Postby RSAWhite » Sun May 26, 2013 5:15 pm

Hello,

I am trying to fit a CJS model constrained to an individual covariate (fish length) to a data set consisting of 871 individuals and 10 encounters. I had no problem fitting the phi(t)p(t) model before I added the individual covariate to the inp file, however, since I added this covariate to the file I get the "error while opening mrkxxxxz.tmp" issue after creating any model.

Here are my symptoms.
If I run any model with all 871 individuals without the covariate in the .inp file, all is good.
If I run any model after adding the covariate to the .inp file, I get this error.
However, if I reduce the number of individuals in this +covariate .inp file to 767, all is good.
Any more individuals than 767 when covariate is included in the inp file, will cause this error.
Finally, if I move the full 871 indivdual+covariate inp to a flash drive plugged into my laptop, all is good.

I am having these issues on a laptop with Windows 7, 3.86 GB RAM, 2.0 GHZ CPU, MARK 7.1, run as administrator.

If I analyse the inp file with all 871 individuals +covariate in another PC running Windows XP, 3.86GB RAM, 2.8 GHZ CPU, MARK 7.1, all is good.

I have followed inp formatting instructions to the letter. Here is a snippet of my file.

/* 0001 */ 1000000000 1 106.0 6.1;
/* 0002 */ 1101000101 1 121.2 9.0;
/* 0003 */ 1001100101 1 152.6 21.4;
/* 0004 */ 1010100111 1 99.6 5.6;

I am confident it is not my formatting because I run this .inp in MARK on another computer or on a flash drive plugged into any PC without any issues.

It seems to me to be something wrong with using windows 7, however, I have followed all suggestions to correct problems with this OS. Moreover, the file runs successfully in this environment if I remove the covariate from the INP file, or if run <768 individuals with covariate included, or if I move the .inp file to an external drive. Thus it will only run on the windows 7 installation drive if I remove the covariate, or reduce the number of fish in the +covariate file to <768.

Consequently, I cant see how this issue can be caused by computing power (because it will run on any PC if .inp is in an external HD). It therefore cant be .inp formatting since it can run well in some PC setups. Thus it seems to be a consequence of running on the drive that windows 7 is installed on, but then why can I run the .inp file sucessfully when number of indviduals is <768? It cannot be adminstrative priveledges as I am the admin.

Thus my symptoms seem to contradict any one cause I can think of. If any one has any ideas on the cause of my problem, I would be greatful to hear them.

Thank you,
Richard.
RSAWhite
 
Posts: 10
Joined: Fri May 24, 2013 10:47 pm

Return to software problems/news

Who is online

Users browsing this forum: No registered users and 11 guests