asserted H^0 temperature depends on platform (trac #60)

[cloudy-bot]

reported by: @CloudyLex

comments for change r1865 in repository

these change asserted values of mean H^0^ weighted kinetic temperature.

Changes introduced over past few reves have improved physics of line overlap, fixed bugs in transfer of H-like overlapping lines. One most platforms the H^0^ weighted temperature changed to the value Ryan asserted.

orion_hii_pdr.in - VS08 value did not change. icl on same platform gets new answer

orion_hii_pdr_pp.in - gcc / cygwin did not change. VS08 and icl on same platform get new answer.

asserted value is in middle of range and error is increased. this is being posted to the trac site as an issue.

fix may be to move to all doubles - did not test whether this fixes the problem

Migrated from https://www.nublado.org/ticket/60

{
    "status": "closed",
    "changetime": "2009-02-08T16:39:30Z",
    "_ts": "1234111170000000",
    "description": "comments for change r1865 in repository\n\nthese change asserted values of mean H^0^ weighted kinetic temperature.\n\nChanges introduced over past few reves have improved physics of line overlap, fixed bugs in transfer of H-like overlapping lines.  One most platforms the H^0^ weighted temperature changed to the value Ryan asserted.  \n\norion_hii_pdr.in - VS08 value did not change.  icl on same platform gets new answer\n\norion_hii_pdr_pp.in - gcc / cygwin did not change.  VS08 and icl on same platform get new answer.\n\nasserted value is in middle of range and error is increased.  this is being posted to the trac site as an issue.  \n\nfix may be to move to all doubles - did not test whether this fixes the problem",
    "reporter": "gary",
    "cc": "",
    "resolution": "fixed",
    "time": "2008-03-19T08:23:14Z",
    "component": "radiative transfer",
    "summary": "asserted H^0 temperature depends on platform",
    "priority": "minor",
    "keywords": "bistable",
    "version": "",
    "milestone": "",
    "owner": "nobody",
    "type": "defect - convergence"
}

[cloudy-bot]

• @peter-van-hoof-noaccount commented:

This problem has been resolved by merging the newsolvers branch in r2539. These are the results from the trunk at r2618 on various platforms:

orion_hii_pdr.in:

run1/orion_hii_pdr.out: ChkAssert        HYDR      1   695.8130 =   694.4000  -0.002   0.100
run2/orion_hii_pdr.out: ChkAssert        HYDR      1   692.7369 =   694.4000   0.002   0.100
run3/orion_hii_pdr.out: ChkAssert        HYDR      1   693.9953 =   694.4000   0.001   0.100
run4/orion_hii_pdr.out: ChkAssert        HYDR      1   697.3259 =   694.4000  -0.004   0.100
run5/orion_hii_pdr.out: ChkAssert        HYDR      1   695.7970 =   694.4000  -0.002   0.100
run6/orion_hii_pdr.out: ChkAssert        HYDR      1   691.9440 =   694.4000   0.004   0.100
run7/orion_hii_pdr.out: ChkAssert        HYDR      1   694.6364 =   694.4000  -0.000   0.100
run8/orion_hii_pdr.out: ChkAssert        HYDR      1   693.8699 =   694.4000   0.001   0.100
run9/orion_hii_pdr.out: ChkAssert        HYDR      1   695.0556 =   694.4000  -0.001   0.100
runa/orion_hii_pdr.out: ChkAssert        HYDR      1   695.2213 =   694.4000  -0.001   0.100

orion_hii_pdr_pp.in:

run1/orion_hii_pdr_pp.out: ChkAssert        HYDR      1  1126.1146 =  1113.0000  -0.012   0.100
run2/orion_hii_pdr_pp.out: ChkAssert        HYDR      1  1120.9644 =  1113.0000  -0.007   0.100
run3/orion_hii_pdr_pp.out: ChkAssert        HYDR      1  1116.4166 =  1113.0000  -0.003   0.100
run5/orion_hii_pdr_pp.out: ChkAssert        HYDR      1  1123.9246 =  1113.0000  -0.010   0.100
run6/orion_hii_pdr_pp.out: ChkAssert        HYDR      1  1118.5914 =  1113.0000  -0.005   0.100
run7/orion_hii_pdr_pp.out: ChkAssert        HYDR      1  1116.2659 =  1113.0000  -0.003   0.100
run8/orion_hii_pdr_pp.out: ChkAssert        HYDR      1  1121.2456 =  1113.0000  -0.007   0.100
run9/orion_hii_pdr_pp.out: ChkAssert        HYDR      1  1114.0925 =  1113.0000  -0.001   0.100
runa/orion_hii_pdr_pp.out: ChkAssert        HYDR      1  1121.9968 =  1113.0000  -0.008   0.100

The error has been reset to the standard value in r2716.

This fix is to invasive for c08_branch, so closing this PR as fixed on the trunk and a wontfix on c08_branch.

• @peter-van-hoof-noaccount changed description from:

comments for change rev1865 in repository

these change asserted values of mean H^0 weighted kinetic temperature.

Changes introduced over past few reves have improved physics of line overlap, fixed bugs in transfer of H-like overlapping lines. One most platforms the H^0 weighted temperature changed to the value Ryan asserted.

orion_hii_pdr.in - VS08 value did not change. icl on same platform gets new answer

orion_hii_pdr_pp.in - gcc / cygwin did not change. VS08 and icl on same platform get new answer.

asserted value is in middle of range and error is increased. this is being posted to the trac site as an issue.

fix may be to move to all doubles - did not test whether this fixes the problem

to:

comments for change r1865 in repository

these change asserted values of mean H^0^ weighted kinetic temperature.

Changes introduced over past few reves have improved physics of line overlap, fixed bugs in transfer of H-like overlapping lines. One most platforms the H^0^ weighted temperature changed to the value Ryan asserted.

orion_hii_pdr.in - VS08 value did not change. icl on same platform gets new answer

orion_hii_pdr_pp.in - gcc / cygwin did not change. VS08 and icl on same platform get new answer.

asserted value is in middle of range and error is increased. this is being posted to the trac site as an issue.

fix may be to move to all doubles - did not test whether this fixes the problem

• @peter-van-hoof-noaccount changed resolution from "" to "fixed"
• @peter-van-hoof-noaccount changed status from "new" to "closed"