DataIO operation System Error: CTZN

[AbU5e]

Reloading a save game from a Conquests game(just a regular random map), i've got this error :-/

All the recent savegames have the same problem, I have to go back almost 300 years to get a working savegame... just a little annoying.

The most annoying thing is i've had this problem several times before, same error, but this was way way back in vanilla Civ3 (though i'm fairly sure its happened in PTW as well)

Any ideas? :-/

I've googled around and found stuff about other similar savegame corruption errors, but none specific to this (Citizen related) error.

I can make the savegame available if some1 wants to have fun poking around to find the culpret invalid value...
(i'm guessing its got something to do with all the scandinavian cities i just finished capturing )

[Gramphos]

Have you beed using any savegame editors or trainers with the said game?

[AbU5e]

nope, just a regular game, on a regular map, no trainers, mods, scenarios or anything.

[Gramphos]

What patch version are youusing?

[AbU5e]

Conquests unpatched, tbh I wasn't aware of a patch for conquests until I read about it on this forum ^_^

The patch may prevent the error occuring in the future (though I doubt it, it isn't mentioned in the list of fixes, and none of the patches so far have addressed the issue :/)
What I realy want is a way to repair the savegame :-/
I spose I could get a savegame editor..... but I kinda hoped some1 else would have encountered this problem, and already found a solution.

[Gramphos]

If you submit the save I might be able to take a look at it. DATAIO errors are mostly strange forward, and are probably caused by a datacorruption in a counter on the number of citizens in a city. But there may be other corruptions as well. A single bit off in the memory on the wrong place could cause the error, but the game would still be playable until you try to reload. Than you will find that all saves since the error occured will be corrupt.

I can give your game ten minutes of my time and see if I can repair any incorrect counters. If there are more major corruptins, as missing datablock I'll not try to correct it.

[AbU5e]

thanx very much, Incidentally, why are autosaves 10 times bigger than normal saves?
is it compression that would take too long if performed on the autosaves?

Here are the 2 saves(1 normal, 1 auto) from the turn in question,

(normal save)

http://www.pkl.net/~rsc/downloads/Tokugawa%20of%20the%20Japanese,%201080%20AD.SAV

(auto save)

http://www.pkl.net/~rsc/downloads/Conquests%20Autosave%201080%20AD.SAV

[Gramphos]

You are right on the compression. autosaves are uncompressed (and therefore slightly easier to edit)

[Gramphos]

...

[Gramphos]

Sorry for the short message. had to run.

There were two erros in the file telling that two cities should have more citizens than there were citizens in the save

[AbU5e]

wow!, thanx alot!

Any idea what could have caused it? :-/

Can you remember which nation the cities belong to?

:edit:

um, did you resave it with a newer version than 1.00 Conquests?
What version do I need to patch upto, to be able to load it?

btw, thanx again.

[Gramphos]

I did not resave with a vewer vesion. are there any error?

And no, I didn't even check what cities that had the error.

If it isn't loadable, please post the error message and I might be able to find a solution. (The fix were made with my tool development version, and there mightbe problems with the code that can cause other corruptions)

[AbU5e]

ahh, nvm I was being stupid ^_^

Works great, thanx alot.

Still wish I knew what caused it.... but anyhow

[AbU5e]

urg, those 2 cities you 'fixed', could they perhaps be Satsuma and Kagoshima :-/

They now appear in civ3 as a city population 12 (which is correct), but have only 4 assignable citizens within (12 squares are being worked on, but only 4 of them can be deallocated :S)


oooh well, nvm, 1 more reason to not bother playing civ3 nemore :-/

thanx for trying ^_^

[Gramphos]

It is possible. Than it was a loss of the CTZN data for theose cities that caused it. the fix I did was to change the reference that tells how many datablocks there is, to the nomber that actually were in the file.