#205 closed defect (fixed)
Resolve O'Mega speed issues by colorizing DAGs, not models
| Reported by: | Juergen Reuter | Owned by: | ohl |
|---|---|---|---|
| Priority: | P1 | Milestone: | |
| Component: | omega | Version: | 2.0.0rc2 |
| Severity: | critical | Keywords: | speed |
| Cc: |
Description
O'Mega is now quite slow in generating coloured matrix elements and jet matrix elements summed over flavours, compared to e.g. Madgraph. BTW, also the WHIZARD phase space generation is slow.
Change History (5)
comment:1 Changed 15 years ago by
| Milestone: | v2.0.1 → v2.0.0final |
|---|---|
| Status: | new → assigned |
comment:2 Changed 15 years ago by
| Summary: | O'Mega speed issues → Resolve O'Mega speed issues by colorizing DAGs, not models |
|---|
New name (subsuming ticket #255).
comment:3 Changed 15 years ago by
Status update: the mssm_ext-*.m4 checks go through with the new colorizer - except for mssm_ext-[ud]g.m4, but I understand the reason (not all couplings of three octets are gluons and I must handle their colorization separately).
Note: See
TracTickets for help on using
tickets.
The new algorithm for generating colored matrix elements in source:branches/ohl/omega-development/colorize_dags_not_models is orders of magnitude faster (it's also much more elegant and might be generalized to a faster implementation of the first step). As of r2114, enough color structures are implemetented to support the standard model. The rest will follow soon.
Needs testing, but should go into milestone:2.0.0final.