Should I use different ids (@1,@2,...) in multiprocess proc_card or could I use the same ?

It is not necessary to make the process numbers different. The reason to do it is that the processes with different numbers are presented in different groups on the "results" web page, and in the event file (LHE file). So for processes that you prefer to have presented in one group, with the cross sections summed within that group, it is better to use the same number. The same applies for the event files - processes which you prefer to have grouped as one subprocess in the event file should have the same number, and vice versa. This feature can be particularly useful for matched processes, where it can be interesting to get the matching efficiency separately for the different multiplicities.

The only case when it is necessary to use different process numbers is if you have two different processes with identical final states (e.g. with different coupling orders) in the same proc_card. In this case, MadGraph will consider the processes as identical and only use the first process, unless they have different process numbers.

-- Main.JohanAlwall - 02 Mar 2009