Re: [Bug-glpk] [Fwd: Inconsistent run times]

[Xypron]

Hello Marc,

for testing I created a lp file with:
glpk-4.44\w64\glpsol.exe -m glpk-4.44\examples\tsp.mod --wlp tsp.lp --check

and used the following coding (compiled to 64bit with VS 2010 Express) for 
testing:

#include "glpk.h"
#include <conio.h>
#include <time.h>
#include <stdio.h>

int main() {
  glp_prob *lp = NULL;
  glp_iocp iocp;
  int ret;
  int i;
  double diff;
  time_t tv1, tv2;

  for (i = 0; i < 5; i ++) {
    lp = glp_create_prob();
    ret = glp_read_lp(lp, NULL, "C:\\temp\\tsp.lp");
    glp_init_iocp(&iocp);
    iocp.presolve = GLP_ON;
    time(&tv1);
    glp_intopt(lp, &iocp);
    time(&tv2);
    diff = difftime(tv2,tv1);
    printf("Duration: %fs\n" ,diff);
    glp_delete_prob(lp);
  }
  while(!kbhit()) {
  }
  return 0;
}

I could not reproduce your observation.

Best regards

Xypron


On 30.10.2012 22:25, Marc Goetschalckx wrote:
>
> On 30-Oct-12 15:44, Xypron wrote:
>> Hello Marc,
>>
>> your mail had to be forwarded manually because you are not subcribed
>> to the GLPK bug list, see
>> https://lists.gnu.org/mailman/listinfo/bug-glpk
>>
>> Are solving the same problem in each round?
>
> Yes, exactly the same problem.  There is only one single lp file.
>> What time does glpsol need to solve your second problem?
>
> Exact same time since the glpsol program is restarted for every
> problem.  About 1.4 seconds.  This includes decoding the lp file,
> times for my program are only for the mipopt solution function.