#91 closed defect (fixed)
Revise Return_Value::Retry_Event (cf. ticket #90)
| Reported by: | Stefan Hoeche | Owned by: | support@sherpa-mc.de |
|---|---|---|---|
| Priority: | critical | Milestone: | |
| Component: | Unknown | Version: | 1.2.2 |
| Keywords: | Cc: |
Description
The task of Return_Value::Retry_Event is to restart event generation with the hard process already in place. This does not work for weighted events starting from v1.2.0 as the Sudakov weight in the METS procedure is either zero or one and this leads to either a valid event or a rejection. In fact, the same problem should have occurred in earlier versions for unweighted events, as the code decided to either keep or reject the event based on an analytic Sudakov weight. (In both cases the problem is a bias towards event rejection if Retry_Event is called.)
As far as I understand, only event phases after the Jet_Evolution can/should request a Retry_Event. I therefore propose to revise this functionality an generalise the solution implemented by Frank for ticket #90. If Retry_Event is returned, the complete perturbative event must be restored, which includes the hard matrix element _and_ the shower evolution.
Attachments (0)
Change History (3)
comment:1 Changed 14 years ago by
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
comment:2 Changed 13 years ago by
| Milestone: | → rel-1.2.3 |
|---|
This issue is solved on trunk. If necessary, please reopen the ticket.