Opened 8 years ago

Last modified 8 years ago

#135 new defect - failed assert

assert inconsistent radiative quantities transition.cpp:74

Reported by: Gary J. Ferland Owned by: Gary J. Ferland
Priority: major Milestone: C13 branch
Component: radiative transfer Version: trunk
Keywords: Cc:

Description

due to lag between level population update and radiative quantities update

this happen infrequently - sensitive to platform and sim.

we need to identify which transition is causing this.

Attachments (2)

simple.out (535 bytes) - added by Gary J. Ferland 8 years ago.
rt.diff (5.2 KB) - added by Gary J. Ferland 8 years ago.
redo RT update within level2 and level3 solvers in attempt to make consistent

Download all attachments as: .zip

Change History (5)

comment:1 Changed 8 years ago by Gary J. Ferland

in linda.in sim (reported by linda strubble) assert thrown at rt_diffuse.cpp:562 line index i=72. wl=833.8, nelem=8, ion=2

ipT834

originally set at cool_oxyg.cpp:433

Changed 8 years ago by Gary J. Ferland

Attachment: simple.out added

comment:2 Changed 8 years ago by Gary J. Ferland

simple.out throws same assert but very quickly

i = 96

wl=360140 nelem=10 ion=3

comment:3 Changed 8 years ago by Gary J. Ferland

with r3836 increased tolerance by factor of three - with this the trunk runs all auto without tripped assert in transitions.cpp

the trips are FUV lines at low temperatures - these transitions are very weak and predominantly photoexcited. Apparently the photo rates are changing and introducing the noise that is detected.

the lines that trip are typically cool_carb.cpp:452 - 468 and cool_oxyg.cpp:598

Changed 8 years ago by Gary J. Ferland

Attachment: rt.diff added

redo RT update within level2 and level3 solvers in attempt to make consistent

Note: See TracTickets for help on using tickets.