Impossible to set ignored_portion less than 1e-5 in touschek_scatter function

Moderators: michael_borland, soliday

Post Reply
Skamarokha
Posts: 23
Joined: 14 Apr 2021, 09:42

Impossible to set ignored_portion less than 1e-5 in touschek_scatter function

Post by Skamarokha » 03 Jan 2023, 02:33

Found out that if one will set ignored_portion in touschek_scatter less than 1e-5 (1e-6, for example), all simulated particles will be put to tracking, as if ignored_portion were 0.

UPD: elegant 2022.2.0, Nov 15 2022 on Windows was tested.
Respectfully,
Mikhail A. Skamarokha.

michael_borland
Posts: 1927
Joined: 19 May 2008, 09:33
Location: Argonne National Laboratory
Contact:

Re: Impossible to set ignored_portion less than 1e-5 in touschek_scatter function

Post by michael_borland » 06 Feb 2023, 13:48

Mikhail,

I see in the code that values less than or equal to 1e-6 are treated as zero. I think the assumption was that for ~1 million particles, the algorithm that ignores particles probably doesn't give very precise results when ignored_portion is so small. However, it shouldn't hurt anything, so I've reduced the cut-over value to 1e-9.

Thanks for reporting the issue.

--Michael

Post Reply