Follow-up from process group separation, for merged samples
The following discussions from !115 (merged) should be addressed:
-
@fsiegert started a discussion: I think in the current form this is nice for fixed-order calculations, but not very useful yet for multi-jet merged setups, because they would need something cumbersome along the lines of:
Process 93 93 -> 11 -11 93{1} Order (*,2); CKKW 30 Special Group(0-1) {3} End process; Process 93 93 -> 11 -11 93 Order (*,2); CKKW 30 Special Group(2-7) {3} Cut_Core 1 End process;
(+ combinations for higher multiplicities)
Would it be straightforward to make this work with a syntax like one of these?
a) Process 93 93 -> 11 -11 93{1} Order (*,2); CKKW 30 Special Group(0-1,2-7) {3} End process; b) Process 93 93 -> 11 -11 93{1} Order (*,2); CKKW 30 Special Group(0-1) {3} Special Group(2-7) {3} End process;
This would be useful also for cases like 93 93 -> 90 90 91 91 93{n} which one probably wants to split up into eeveve+mmvmvm+ttvtvt vs. eevmvm+eevtvt+mmveve+mmvtvt+ttveve+ttvmvm for integration.
If it's not straightforward, then I would approve this MR for the time being, and file a separate issue for this to be improved later.
-
@fsiegert started a discussion: And one more discussion item on this...
This feature has some overlap with the parentheses notation for splitting up process groups according to containers, e.g.
... PARTICLE_CONTAINER 994 vq 1 2; PARTICLE_CONTAINER 995 sq 3 4 5 -1 -2 -3 -4 -5; }(run) (processes){ Process (21,994,995) (21,994,995) -> 11 -11 93{2} ...
Do I understand correctly, that everything that can be achieved with the container notation can also be achieved with the
Special Group
notation, but in addition the latter is more flexible? Or will we ever want to combine both ways of splitting up process groups for integration?