Savegame bug / pathfinder issue? (Likely related to Ticket #127) #177
Comments
|
mathew.eis@... uploaded file |
|
mathew.eis@... commented A little more looking thinks it's also related to #96 The difference with this one is you can run the game fine until you request the units specified to move long distances. |
|
mathew.eis@... commented Er, my bad, it's the units that are in the LOWER RIGHT part of the map, not lower left. |
|
Giel changed blocking which not transferred by tractive |
|
mathew.eis@... commented More speculation on my part here... As similar to #127, it appears to be an endian issue, however this time I've found an offending variable in sMove->bumpDir It looks like it is loaded properly (along with most of the rest of the droid's data) into psSaveDroid->sMove as it is read from the file, but the endian-ness is swapped at some point during or after it is copied into psDroid->sMove. I'm still trying to figure out exactly where this takes place... For example, the correct value (279 or 0x0117) is loaded into psSaveDroid, but then later in psDroid, it becomes 5889 or 0x1701. Everything runs fine because the direction is right, until the bug is then "activated" at some point later by requesting the droid to move a long distance, and it checks something against bumpDir, which is outside of the range. ... still learning the source code, so sorry if any of this is wrong. |
|
Giel changed status from |
|
Giel changed resolution from `` to |
|
Giel commented Closing, as this is a duplicate of #96. |
|
Giel changed resolution from |
|
Giel commented (In [6513]) Merged revision [6512] into the 2.1 branch via svnmerge from trunk ........
|
resolution_fixedtype_bug| by mathew.eis@...Hello,
Attached is a savegame from version 2.1 RC1 that can repeatably crash both RC1 and RC2 - the non debug versions. I'm running Mac OS X 10.4.11 on a Powerbook G4 1.25 Ghz
The easiest way to repeat it is to go to the lower left region of the map as soon as the game loads, select the large group of units there, and send them for recycling.
I believe the bug is related to the path finder (? - sorry, I'm not very familiar with the source code yet), because it can also be repeated by requesting that the units move to any area very far from where they are currently stationed, however, if you request that they only move small distances at a time, it does not crash.
In the Console, I get the following:
error : [dirDiff] Assert in Warzone: /Users/freddie/Documents/Programming/2.1_rc2/macosx/../src/move.c:252 (retval >=0 && retval <=180), last script event: 'N/A'
Assertion failed: (retval >=0 && retval <=180), function dirDiff, file /Users/freddie/Documents/Programming/2.1_rc2/macosx/../src/move.c, line 252.
I hope that helps...
Issue migrated from trac:177 at 2022-04-15 17:49:49 -0700
The text was updated successfully, but these errors were encountered: