Source Code Project Mantis - FSSCP
View Issue Details
0003014FSSCPgraphicspublic2014-03-05 15:542016-12-11 20:37
ReporterKobrar 
Assigned ToDahBlount 
PrioritynormalSeverityminorReproducibilityalways
StatusresolvedResolutionfixed 
PlatformOSWindows 7 HP 64bitOS Version
Product Version3.7.1 
Target VersionFixed in Version 
Summary0003014: Environmental reflections are wrong
DescriptionSo it is a bug originally reported on the forum in thread:
http://www.hard-light.net/forums/index.php?topic=87019.msg1739006#msg1739006
Basically, some of the coordinates of the cubemap seem to be flipped.

Testcase here: https://www.mediafire.com/?s1rcbe6pcoi1o1q
Steps To Reproduce1.Let the engine produce his own cubemap[at least from cube skybox].
2.Watch the reflection on some reflective object.
TagsNo tags attached.
Attached Files

Notes
(0015803)
Goober5000   
2014-06-05 20:31   
Am I to understand that this is now fixed?
(0015804)
niffiwan   
2014-06-05 20:41   
(Last edited: 2014-06-05 20:50)
Did you mean to assign this one to me? I am an OGL neophyte...

Edit: ok - just some confusion due to me misnaming a commit on github - this one isn't fixed as far as I know.

(0016295)
Goober5000   
2014-09-24 02:42   
Removing the 3.7.2 tag since I only added that because I thought niffiwan had fixed it. :)

(0016759)
Echelon9   
2015-06-26 07:24   
I'll have a go looking at this one.
(0016791)
DahBlount   
2015-11-22 17:59   
Pull request to fix this is now available.
https://github.com/scp-fs2open/fs2open.github.com/pull/466
(0016796)
niffiwan   
2015-12-06 01:56   
This is fixed now?

Also - can haz DahBlount 'developer' status?
(0016797)
MageKing17   
2015-12-06 04:53   
The final PR that should have fixed this: https://github.com/scp-fs2open/fs2open.github.com/pull/468

Until DahBlount gets dev status, I've unassigned this from Echelon9 and I'm marking it fixed.
(0016798)
Kobrar   
2015-12-07 16:58   
As seen here: https://www.youtube.com/watch?v=OO4cxEh-kqk
There are still some issues
(0016800)
DahBlount   
2015-12-12 19:05   
Looking further into this, turns out the inverse matrix4 function wasn't even required. Removing the function call in gropengltnl.cpp gives proper results simply with the changes to the texture_mat setup in gropengltnl.cpp and freespace.cpp as well as an additional change in the main vertex shader to normalize the eye position.
(0016802)
MageKing17   
2015-12-15 09:23   
This should finally be fixed for real now: https://github.com/scp-fs2open/fs2open.github.com/pull/499

Issue History
2014-03-05 15:54KobrarNew Issue
2014-06-05 20:31Goober5000Note Added: 0015803
2014-06-05 20:31Goober5000Assigned To => niffiwan
2014-06-05 20:31Goober5000Statusnew => code review
2014-06-05 20:31Goober5000Target Version => 3.7.2
2014-06-05 20:41niffiwanNote Added: 0015804
2014-06-05 20:50niffiwanNote Edited: 0015804bug_revision_view_page.php?bugnote_id=15804#r805
2014-06-05 20:50niffiwanAssigned Toniffiwan =>
2014-06-05 20:50niffiwanStatuscode review => new
2014-09-23 23:16niffiwanDescription Updatedbug_revision_view_page.php?rev_id=918#r918
2014-09-24 02:42Goober5000Note Added: 0016295
2014-09-24 02:42Goober5000Target Version3.7.2 =>
2014-09-24 02:42Goober5000Note Edited: 0016295bug_revision_view_page.php?bugnote_id=16295#r920
2015-06-26 07:24Echelon9Note Added: 0016759
2015-06-26 07:24Echelon9Assigned To => Echelon9
2015-06-26 07:24Echelon9Statusnew => assigned
2015-11-22 17:59DahBlountNote Added: 0016791
2015-12-06 01:56niffiwanNote Added: 0016796
2015-12-06 04:53MageKing17Note Added: 0016797
2015-12-06 04:53MageKing17Assigned ToEchelon9 =>
2015-12-06 04:53MageKing17Statusassigned => resolved
2015-12-06 04:53MageKing17Resolutionopen => fixed
2015-12-07 16:58KobrarNote Added: 0016798
2015-12-07 16:58KobrarStatusresolved => feedback
2015-12-07 16:58KobrarResolutionfixed => reopened
2015-12-12 19:05DahBlountNote Added: 0016800
2015-12-15 09:23MageKing17Note Added: 0016802
2015-12-15 09:23MageKing17Statusfeedback => resolved
2015-12-15 09:23MageKing17Resolutionreopened => fixed
2016-12-11 20:37Goober5000Assigned To => DahBlount