Incomplete stick file on running xcross_dos.exe

[VishnuRadhakrishna]

Hi,

I have been trying to generate a stick file and out file for the CN trihybrid model on ExoMol using the .states and .trans file on ExoMol for CN (Trihybrid).
https://www.exomol.com/data/molecules/CN/12C-14N/Trihybrid/

I am running ExoCross on windows system and using the xcross_dos.exe in the terminal. However, I see that it is generating only 1,576,054 transitions. But should be 2285103 lines since that is the number of transitions in the trihybrid model.

Can you please let me know why I am getting an incomplete outfile?

[VishnuRadhakrishna]

my input file has the following lines:

Temperature 296
Range 0.0 6.0

absorption
stick
threshold 0

output 12C14N-stick-296K

States 12C-14N__Trihybrid.states
Transitions 12C-14N__Trihybrid.trans

[Trovemaster]

Dear VishnuRadhakrishna,

I suspect the missing transitions are either outside the frequency region or too weak. You will still get a slightly different numbers but these are with negative frequencies (due to the limitations of the trihybrid process)

Please try the following input:
Temperature 3000
Range 0 600000

absorption
stick
threshold -1

output 12C14N-stick-3000K

States 12C-14N__Trihybrid.states
Transitions 12C-14N__Trihybrid.trans

[VishnuRadhakrishna]

Thanks for the prompt response. I tried running the above-mentioned input file and I am still getting the same number of lines 1576054. 0 to 60000 is the frequency range in the trihybrid model. So I don't think the transitions are out of the frequency range.

[VishnuRadhakrishna]

I had obtained an output file from the authors of the Trihybrid model. Their out file had total intensity , T and memory report and timer data at the end of their out files. However, I do not see such things at the end of my out file and the last line is a just the stick data at some transition at 1576054 line. This made me suspect that something else might be going on here and may be the code is not running to completion on my end.

[Trovemaster]

Can you try a new xcross_dos.exe (just updated)? This is just in case there were recent fixes that affected this issue.

Cheers,
Sergey

[VishnuRadhakrishna]

I tried the new one. It is giving me 1575454 transitions. It is lower than the number I was previously getting. I am not sure if it still completing the simulation. The end of the .out file is still just a transition detail.

[Trovemaster]

I have just tested the T=3000 K input, it works fine on my Windows laptop. Can you try a different Windows machine with more memory? Have you tried reducing NRAM parameter:

NRAM 10
NPROCS 1

Cheers,
Sergey

[Trovemaster]

Dear VishnuRadhakrishna,

I have an idea. Can you please try the xcross_dos_debug.exe? It was compiled with larger stack size.

Cheers,
Sergey

[VishnuRadhakrishna]

Sure. I will try that. Also, Can you let me know where I must change the NRAM parameter? Should I just enter it in the .inp file? Sorry, I am relatively new to using ExoMol

[Trovemaster]

Thanks! You can put them anywhere in the input file:

Temperature 3000
Range 0 600000

NRAM 10
NPROCS 1

absorption
stick
threshold -1

output 12C14N-stick-3000K

States 12C-14N__Trihybrid.states
Transitions 12C-14N__Trihybrid.trans

[VishnuRadhakrishna]

Hi,

Using xcross_dos_debug.exe fixed the problem. it completed the simulation and I was able to obtain all the lines. Thank you so much. So should I just keep using the xcross_dos_debug.exe file to run?

[Trovemaster]

Excellent! yes, you can use xcross_dos_debug.exe (safer but potential slower option), but I will update xcross_dos.exe as well.

[VishnuRadhakrishna]

Okay! once again, thank you so much.