Closed valassi closed 5 days ago
I increased it to 6E-4. Enough for sse4, not enough for avx2...
https://github.com/madgraph5/madgraph4gpu/actions/runs/9807953772/job/27082866125#step:12:112
*** (2-avx2) Compare MADEVENT_CPP xQUICK xsec to MADEVENT_FORTRAN xsec ***
ERROR! xsec from fortran (2.6007004181875136E-006) and cpp (2.6034626920972927E-006) differ by more than 6E-4 (0.001062126914142647)
Ok fixed in #882 (for the master_june24 branch, to be merged there eventually). Closing.
PS I also immediately merged the patch into master in #890.
Another issue which appeared through #830 and being reviewed in #882.
In WIP PR #882 I added a workaround (NOT a fix) for crash #885 just to allow the CI tests to proceed further. Apart from other 'expected' failures, there is a xsec mismatch for ggttggg. This can be fixed by increasing tolerance https://github.com/madgraph5/madgraph4gpu/actions/runs/9807713341/job/27082079782