Single Player bugs and crashes v37 plus (SVN) - After the 24th of December 2016
- Thread starter Dancing Hoskuld
- Start date
...? no... is there that option?
Thunderbrd
C2C War Dog
Yes, in the game setup. I'll have to take a quick look and evaluate but you explained the game to be experiencing the exact effect that option is supposed to provide. I'll have to make sure it's not affecting games it shouldn't for some reason.
Thunderbrd
C2C War Dog
A couple of asserts I don't have time to look into right away:
Fixed
Code:
Assert Failed
File: CvGlobals.cpp
Line: 5463
SVN-Rev: 9535
Expression: stricmp(szType, "NONE")==0 || strcmp(szType, "")==0
Message: info type 'INTERSTELLAR_COLONIZATION' not found, Current XML file is: xml\Technologies\CIV4TechInfos.xml
----------------------------------------------------------
Code:
Assert Failed
File: CvGlobals.cpp
Line: 5463
SVN-Rev: 9535
Expression: stricmp(szType, "NONE")==0 || strcmp(szType, "")==0
Message: info type 'DATA_ARCHIVAL' not found, Current XML file is: xml\Technologies\CIV4TechInfos.xml
----------------------------------------------------------
Last edited:
Thunderbrd
C2C War Dog
Would've been due to the same problem I reported in the assert above. I've fixed it now anyhow. Had a moment I didn't expect to get here.
SVN 9580 and previous 40+ or so. I tried a few to see if I could find the problem. All graphics at lowest level.
Current SVN is unplayable - for first time in years I can only play 1 to 3 turns before CTD, even in Prehistoric, (before I could get to Medieval before problems occurred - and I could get round them). Game now crashes after about 5 minutes.
Can not supply a save game as a reload continues a bit further before the next crash.
Are the mini dumps on their own useful?
Current SVN is unplayable - for first time in years I can only play 1 to 3 turns before CTD, even in Prehistoric, (before I could get to Medieval before problems occurred - and I could get round them). Game now crashes after about 5 minutes.
Can not supply a save game as a reload continues a bit further before the next crash.
Are the mini dumps on their own useful?
Last edited:
Thunderbrd
C2C War Dog
They can be. But this sounds like MAFs impacting your machine.
It's sounding like we may need to find some ways to run more efficiently on memory. Still, it's a little odd that the few recent additions have had this drastic an effect on your system. Of course it's possible you've got our canary in the coal mine there in your system. I mean, a LITTLE memory overhead was added for the Outbreaks and Afflictions. Just not that much given that most of it was already there and there was just a little adjustments to improve things, some of which would also have reduced mem costs a bit.
However, I suppose the inclusion of the huge new amount of techs might have had an overhead impact, but again, I wouldn't think it would've been so much as to have that great an affect.
You might have to go back and figure out, by playing a bit then updating to the next revision, which one really created the problem for you.
If we're lucky, my hunch is wrong entirely and you're just frequently finding a crash point for some reason that needs to be resolved. If this is the case, yes, those minis could really help.
Well the version used for the new PBEM's works OK. I have just started a new game using it - with no problems. (different folder name and used the same UserSettings - so won't effect the PBEM's). I can start from there and see how far I can progress on SVN updates.
Thankfully my SVN download speed does not take to long, it is everything else that takes time.
So I will see how we (I) get on.
Thankfully my SVN download speed does not take to long, it is everything else that takes time.
So I will see how we (I) get on.
Last edited:
cammcken
Prince
I'm getting a CTD somewhere during this turn. It must be one of my automated units because I'm not clicking anything. A slight suspicion it's an outrigger that got ambushed, based on the event message I got right before the crash. But I only noticed once on my last play, when I was looking for it.
It began on an earlier SVN. I found the crash, updated SVN, cleared the cache, loaded the save, recalculated modifiers, and still got stuck.
I've attached a save from the same turn, a save from the turn before, and the most recent minidump.
Edit: Yes, it seems to be triggered by something the outriggers are doing. I loaded it up again and went through all the outriggers, giving them new orders manually, and I made it through the turn.
It began on an earlier SVN. I found the crash, updated SVN, cleared the cache, loaded the save, recalculated modifiers, and still got stuck.
I've attached a save from the same turn, a save from the turn before, and the most recent minidump.
Edit: Yes, it seems to be triggered by something the outriggers are doing. I loaded it up again and went through all the outriggers, giving them new orders manually, and I made it through the turn.
Last edited:
Thunderbrd
C2C War Dog
I'm glad you can. I'm in the middle of a project here. I'll try to remember to address this soon but I know I won't be able to tonight either.
Well that's not altogether tooooo surprising. This is where the memory balance tips apparently. It's interesting that its from bonuses in particular. If I was any good at finding causes for memory leaks I might be able to investigate if bonuses are somehow more problematic than other objects in particular. Unfortunately this is outside of my knowledge base. But I'll keep it in mind and be on the lookout for something along those lines. At some point soon I'll try to look through it all and see if anything jumps out at me.
It might be that we'll need to take the last couple of eras and put them aside in a second download for now. I'm not sure. I don't like that you're stuck here but I know I've got a lot coming that's probably only going to put us right back past this limit again at some point not too far off.
Dancing Hoskuld
Deity
I ran XML Validator over the XML folder and there are lots of problems (>18000) the first looks like a design bug. In buildings the group tag SpecialistCommerceChange actually changes Yields not Commerces. The Atonium has this problem in the Great Wonders file.
Most however are because the schema being pointed to is not correct in some way.
edit I am fixing the XML\TEXT folder as most of the errors here are with the files pointing to the wrong folder for the location of the schema. It has most of the 18000 errors.
edit2 yes that gets it down to just over 1000 errors!
edit3 It appears that in XML\EVENTS file Event Trigger Infos the tag ERA in the TRIGGERTEXT area is not optional and you need to put <Era>NONE</Era> in there if the text is valid for all eras.
The attached file contains the errors that remain after I fixed all the schema file locations. Note this is just those in the Assets/XML folder I have not looked in the modules folder. There are a lot of errors in the Terrain files
Last edited:
strategyonly
C2C Supreme Commander
So thats another area where "none" has to be listed, thats why everyone has to be careful, not all NONEs are useless, some might have to come back????
Dancing Hoskuld
Deity
There appears to be an error in the Unit Schema file. The tag iDistance is defined twice, once at line 55 and again at line 1127. This is causing the XMLValidator to report an error for every unit in the files.
Thunderbrd
C2C War Dog
Great catches DH!
Dancing Hoskuld
Deity
I am not going to touch the actual schema file. You probably need to do it just in case what the error is is that you meant to define something else there.
Thunderbrd
C2C War Dog
Its the core's unit schema right? I'll look as soon as i can which will be late tonight.
Dancing Hoskuld
Deity
Yes the core one. I updated one of the modules with the core one to see if that got rid of the errors and it did except for that one. It is the required copying to all the folders and renaming after you fix it that I dread.
Similar threads
- Sticky
- Sticky
