View Full Version : It crashes almost every turn!


Incoming9000
Dec 01, 2003, 03:08 PM
I've been playing a random map game for several days and lately it has been crashing almost every turn. It either hangs mid turn, forcing me to make a cold reboot, or goes back to desktop.

I tried reinstalling the game, Civ3+Conquests, disabled background apps and checked drivers but nothing worked.

My rig is a Athlon XP 2100+ with 756Mb ram, SB live! Value and a GF5600 on Det 5303.

Any help would be appreciated.

Thanks!

PS: Save game file uploaded here! (http://www.civfanatics.net/uploads6/Henry_of_the_Portuguese,_1966_AD.SAV)

PSPS: Why do the autosave files are much bigger than the standart save files? Shouldn't they have the same size?:confused:

Bamspeedy
Dec 01, 2003, 03:25 PM
Did you install the 1.29 patch to Civ3, before installing Conquests? You'd think that you shouldn't have to, but you do need to.

Incoming9000
Dec 01, 2003, 03:26 PM
Really? Damn!

I'll check that!

Incoming9000
Dec 01, 2003, 04:30 PM
No good.

It lasted a while longer but it still hanged after 2 turns.

Still sumped! :(

AngryGerbil
Jan 03, 2004, 08:25 AM
I'm also having the same problem. It doesn't seem to happen in early games. It locks up sometimes, but usually it's just an insta-boot to desktop and it happens more often in later games with lots of war. Alot of times it happens when I capture a city and order it razed. This makes me think it is RAM but absolutly no other game has this problem, and Civ is probably the least demanding game for my PC that I play, or at least, I thought it was. Didn't ever used to have this happen back in PTW either, it's new for C3C.

Incoming9000
Jan 08, 2004, 12:41 PM
It took me a while but I did find a solution. I decided to disable the background music and guess what? No more crashes!

BTW you wouldn't happen to have a Via mobo with the K133 chipset? I discovered that there's a bug with that particular chipset that causes data corruption when large ammounts of data are being processed. This seems to be more notisable when a SBlive! is present.