Discussion Forum

Hi,

Is there a way to get COMSOL to report the time stamp for log entries during a time marching simulation? I know I get it on the Start/Finish of a step in the sequence but something more granular would be helpful. I am wondering if I'm missing a checkbox, though I did specify "detailed log" in the options. I'm running a longer simulation with a lot of timesteps. . The matrix is large enough that I don't want to wait around with a stopwatch, but I'd like to know the approximate solve time to get an ETA for it finishing, know if it's stalling, or make some wall clock estimates. It would be really nice to have a time stamp associated with each step.

Right now I see somethign like:

Time t = 0.123
Iter    SolEst    Damping    StepSize    #Res    #Jac    #Sol    LinErr    LinRes
1        1.2        1.00            0.061        1123    1124    1123    6e-9    4e-11
Solution error estimates for fields: etc, etc
2        1.1e-13    1.000      1.1e-9      1124    1125    1124    2e-8    3e-9
Solution error estimates ....

Step    Time    Stepsize    Res    JAc    Sol    Order    Tfail    NLfail LinErr LinRes
511    0.123    0.015       1125 1126  1125 2            4          0       2e-8 3e-9

I'd love to be able to get a timestamp either at the start or end of a major (or minor) timestep, something like:

Time t = 0.123 (2023-08-25 09:22:37 UTC)
Iter    SolEst    Damping    StepSize    #Res    #Jac    #Sol    LinErr    LinRes
1        1.2        1.00            0.061        1123    1124    1123    6e-9    4e-11
Solution error estimates for fields: etc, etc
2        1.1e-13    1.000      1.1e-9      1124    1125    1124    2e-8    3e-9  

Step    Time    Stepsize    Res    JAc    Sol    Order    Tfail    NLfail LinErr LinRes
511    0.123    0.015       1125 1126  1125 2            4          0       2e-8 3e-9

Or

Time t = 0.123 
Iter    SolEst    Damping    StepSize    #Res    #Jac    #Sol    LinErr    LinRes
1        1.2        1.00            0.061        1123    1124    1123    6e-9    4e-11
Solution error estimates for fields: etc, etc
2        1.1e-13    1.000      1.1e-9      1124    1125    1124    2e-8    3e-9  

2023-08-25 09:22:37 UTC
Step    Time    Stepsize    Res    JAc    Sol    Order    Tfail    NLfail LinErr LinRes
511    0.123    0.015       1125 1126  1125 2            4          0       2e-8 3e-9    

Is there something I'm missing? I'd rather not have to add a method call into the solver process to do it. I've toyed with creating a probe that calls out to Java's Timestamp(), but that also feels like a huge workaround.

Thanks, Peter