mutex source/colon convention

[song]

hi, 

    I have some fault sources that I have predefined some weight for their magnitude, slip rate etc and modelled it as mutex source with their weight in a source group for each fault. I didn't use the colon convention and just gave them different name in oq engine 317 version and I could calculate it, however when I change it to the colon convention, I got quite different answers, PGA increased from 0.44 to 0.77g,

  So I updated  the software to 320, it seems now we have to use the colon convention, for same assessment, I got a lower PGA value now at about 0.37 g.

   which is the correct one, shall I use the latest version results?

[kendra....@globalquakemodel.org]

Hello, 

Without seeing your source code, it's difficult to know if the changes in output values are due to corrections/bug fixes in the openquake engine, or due to incorrect implementation. Mutex sources have undergone a lot of improvements since 3.17.