Bug Reports

Oh sorry, I didn't see your reply (and that could well be it). Thanks, I'll be waiting for the 2.13 version

Ok, so I'm getting a crash at this particular turn in the BTS version. It's happened 3 times at the same turn, so I'd say it's reproduceable, I'm attaching the save to see if you can get anything out of it. If it somehow doesn't crash after pressing "End Turn," could you please upload the save so I can continue?

Here's the save:

Try installing the new dll with Solver's unofficial patch from the download page. He claims to have fixed several crash bugs, so it might help you out.

I'll take a look at the save and report how it goes. [EDIT: No luck, crashes for me too. Doesn't immeadiately appear to be Revolution related ... also, doesn't seem to work with the Solver's patch Revolution DLL for unknown reasons.]

This mod is incredible - very clever stuff.

One problem I have, however, is every since I put Solver's Patch in the Rev Mod folder the AI literally kills itself every time AI automation is turned on. All units and cities just disappear and one loses the game. Quite odd.

@BobTheTerrible

Is the crash happening at about 1300AD? if so, I'm having the same problem it keeps crashing on me at the very same turn every time i load and i had a darn good game going to

Well I just tried the save with solver's dll, but it still crashes on that turn.

Having the same CTD problems in two separate games but around the same time frame - one in the 1620s as the Germans (first with Solver's unofficial patch) and then another in the 1480s as the Celts using the stock Revolutions mod for BTS 1.01.

Reinstalled Vanilla CIV, Warlords & BTS and patched up to 3.02 for BTS and this time I didn't even get past 1020 AD with the Revolutions 1.01 with Solver's patch as CTD again just to let you know. Looking forward to the update though to fix this.

In 1.61 you could play the 1000ad scenario with Revolutions, yet in 1.01bts the game crashes half-way though the scenario load if you attempt to load either the Civ 4 or BtS 1000AD or indeed any of the map scenarios.
just thought id report it.

orinsul said:

In 1.61 you could play the 1000ad scenario with Revolutions, yet in 1.01bts the game crashes half-way though the scenario load if you attempt to load either the Civ 4 or BtS 1000AD or indeed any of the map scenarios.
just thought id report it.

Click to expand...

This probably has to do with the number of civs defined for the map versus for the DLL. My guess would be you could open the map file and add a bunch of blank player and team definitions to fill the total out to 34 and then it would work. People have had this issue in the past with maps and number of civs defined.


As for the CTD syndrome, I haven't found any clues yet as to why it's happening. It doesn't seem that crashes occur during the execution of the mod code, so the options are that something isn't set up write or that the mod makes existing bugs in Firaxis's code more likely to bite. Solver's patch claims to fix several bugs that can cause crashes, we know there are issues out there ... I wish we knew when a real bug fix patch was expected for BTS, anyone heard anything?

The word on the street is that "BTS has a stabilty problems if number of civs is > 18", which perfectly explains why crashes would be both more likely with Revolutions and also not directly caused by the mod. If your save game already has > 18 civs, you're probably screwed until BTS gets patched.

However, I'll post a dll with a cap of 18 (the default even in BTS) soon which should allow you to start and complete a new game ...

Hello. I seem to have the same problem as mentioned a few posts ago. I'm in year 1375 AD and the game always crashes after clicking "end turn". I use the BTS version. Crashes both with or without Solver's pack. Oh, and it's a small map with only 5 civilizations, so it can't be because of this.

I can post the save game file, if it helps.

Having the same problem as everyone else......game crashes in 1130 every time, i think this could have something to do with the revoloution mod because i played an entire game without revoloution using 31 civs with no problems on the same map that my Revoloution game crashes on.

I've uploaded my save game and the map i was playing on just in case you can help.....the leader names for the map are all in Italian. I changed them all to english on my PC as well as changing the Mogol leader from Genghis to Kublai Khan but i couldnt upload it......i dont know if not having my edited version of the map would stop you from loading my save game, but if it does and u find a way to fix the saves, i guess i could e-mail it to you

http://forums.civfanatics.com/uploads/111893/Kublai_Khan_AD-1130.CivBeyondSwordSave

http://forums.civfanatics.com/upload...th_31_Civs.rar

I think I found a workaround. It is a nasty one, but it works.

First, I loaded the game one turn before it crashes. Then switched into worldbuilder (Ctrl-W). Deleted one of the opponent civilizations, then switched back and tried to end the turn. Most of the time it would crash. I reastarted the game, deleted a different civilization, and tried again. And so on...

This way I found the crash was triggered by the nasty Babylonian civilization. When I delete this civilization in the world-builder, the game doesn't crash. That's why I call this a nasty workaround - I can play on, but only if I delete this civilization.

Strangely enough, when I switch to play Babylonians (Ctrl-Shift-L), I can play as them and the game doesn't crash. But as soon as I switch back to my original civ and end a turn, it crashes. I played only a few turns as Babylonians, though. It might work if you play longer.

At this time, I'm trying to locate the bug more precisely, since I don't want to delete a whole civilization. When (and if) I find something, I'll let you know. But I guess you can try to help me - using worldbuilder and a little bit of patience, I think you can detect your "nasty civ".

Good luck and let us know if someone of you finds something.

EDIT: A quick quide how to find the nasty civ. Try switching civilizations (Ctrl-Shift-L) one after another, from tom to bottom. The last one that works should be the buggy one. Try to delete it in worldbuilder and see if it works.

I found it. At least almost.

This bug seems to be caused by units. But I don't know which units specifically. When I knew it was caused by Babylonians, I opened the worldbuilder and deleted all Babylonian units outside cities. There were five of them - an axeman, two swordsmen, a longbowman and a chariot. After deleting these the game doesn't crash. Weird, huh?

Please note that the crash is probably caused only by one of the above mentioned units. I simply ran out of patience (after like 30 crashes) and decided not to test in any more. I don't know what was the cause... maybe the AI tried to upgrade a unit? But where's the conflict with Revolution? Well... I don't know. I can play now.

Thanks for the sleuthing Markus, there are some unit related changes that are unused so I'll try removing them and see if the crashes relent.

Confirm the crash anywhere around 1000AD. My systems is BTS plus Revolutions 1.01 and Solvers 48civ DLL.


I have tested it say five times in two different combinations using Jdog's auto turns feature.

I have turned off ALL of the Revolutions code except change nation, auto turn.

The combination that crashes is:

Barbarian Civ on: Crash around 1000AD every time
Barbarian Civ off: No crash.

I fiddled with one of the Barbarian civ configurations ie) whether civs form only after a unit is landed on a new continent. That made no difference.

I'm thinking that it has something to do with the code that turns a Barbarian civ into a fully fledged civ when they are on their own isolated continent.

I haven't tested it thouroughly, but a few times on each combination is pretty good odds. I hope not to throw you off on a tangent lead.

Cheers.

Apparently, the next patch(a real patch, mind you) is currently in QA. That's second-hand, but it came from this board, so it's probably true. So it ought to be out reasonably soon. Sucks about the crash...I had a Lizzie crash on me, just as I was starting to build a massive army of Redcoats and Cavalry to colonize Gandhi. It's always interesting playing this mod. That particular one was fairly easy because my neighbors had major stability problems, probably because the AI doesn't balance its economy well.

Update: http://forums.civfanatics.com/showthread.php?t=243264
I'm guessing the next patch will be out...next week. Changelog is....extensive, to say the least.

Good news! Many thanks to Markus for his excellent hunch on units, a new DLL that appears largely crash free will be posted soon in the Downloads thread. It has been tested several times and every time gotten into the 21st century I'm still not sure why the crash was happening, but it was in currently unused code on units and by commenting those pieces out everything seems to have returned to normal.

Also good news that a real BTS patch is coming soon, I'll post a new full version with that shortly after it comes out (although there won't be a large number of changes, as I've been travelling a lot).

(EDIT: Please post if you have success or not on a savegame!)

jdog5000 said:

Good news! Many thanks to Markus for his excellent hunch on units

Click to expand...

Thanks so much for posting an interim DLL. I've auto-turned a few games with no crash issues. I'm actually a combination of shocked that the developers could have allowed unused code to execute for no reason and awed that you found that faulty code. I'm guessing that the unit code must be immense for that "bug" to have escaped QA.

I am surprised that the AI is still building excessive caravel's in light of Solver's updates.

Your own mod is a poison chalis Jdog. I can no longer play vanilla civ with nearly the same pleasure without at least BarbCiv going. I continue to assume that turning on the Revolutions code block is not worth it at 1.01beta because the code is entirely a Warlords port.

Good one

glider1 said:

Thanks so much for posting an interim DLL. I've auto-turned a few games with no crash issues. I'm actually a combination of shocked that the developers could have allowed unused code to execute for no reason and awed that you found that faulty code. I'm guessing that the unit code must be immense for that "bug" to have escaped QA.

Click to expand...

It was actually unused pieces of code for the mod, pieces that were planned for the future ...

Anyway, as soon as the patch comes out and I have a chance to merge changes, I'll post a new version with several BTS adaptations and other enhancements that will be the first non-beta BTS release.