FfH2 0.25 OOS errors

Patch L.
Completed a whole game without a single OOS. No FFH option turned on except for Wildlands. I guess the issue is solved :D

Kioras, make sure you are both running v. 3.13 of BtS and patch L. It can't be the Operative System.
 
We are both using BTS 3.13, patch L for FFH2, nothing else added. No custom patchs for either.

We are also playing over gamespy. So unless it is something else, we are at our wits end on what is causing it
 
Well into a game. 400 turns plus. Been fighting wars all the time. No OOS errors at all for us. I would re-install and patch, and then start a new game if I was you.
 
Well we never had OOS when playing with direct connect, granted only like 60 turns or so.

Maybe it is gamespy
 
I just finished a direct IP game with a friend. I was Perpentarch and my friend was Keelyn. I completed the Mercurian Gate and chose to switch to Basium and the game went OOS.
 
just had a bunch more OOS with a direct connect game.

Going to try again with a new game post full civ reinstall
 
I just finished a direct IP game with a friend. I was Perpentarch and my friend was Keelyn. I completed the Mercurian Gate and chose to switch to Basium and the game went OOS.

for multiplayer, no one should ever choose to switch to hyborem or basium via the popup. instead, say No, then log out and log back in, and choose hyb or bas.
 
Maybe there should be a note in that message then? (I haven't actually tried either so I can't remember if there is one)
 
I just found an OOS cause from Governors Manors that will be fixed in 0.30. Some of the reason some people may see them and others dont may be because the Calabim are or aren't in the game.
 
Well that explains part of the problem, I think in most games we had calabin, and since we liked nice big games, there were roughly 14+ civs in total.

The OOS never were happening in combat, they always happened at the end/begining of turn, often much later in the game (120-150+ turns in)

yep, checked it the most recent OOS game had calabin in it
 
Started a new game, went OOS 2 times, after an advanced start, had barbs,2 players and 10 AI civs, none of them were cabalim.

One of them was grigori, and the only events known when it went OOS were either a hero born or a religion spread
 
I played a game with 8 AI e 2 human.
I played Calabim.
I had instead of OOS, in two different cases, the game freezed and then crashed with windows error, both host and other.
This has happened with counter at 40, when all tiles with factory and animals change, and with counter at 100, when was generated avatar and all his units!
I think, with huge map, that this two cases, are too heavy for the sistem!
 
Kael, out of sheer curiosity, could you quickly explain what caused the two OOS errors you isolated? (The fortify bug and the governor's manor bug) Were it changes not properly broadcast to other players? Random number related bugs? Something else?
 
Kael, out of sheer curiosity, could you quickly explain what caused the two OOS errors you isolated? (The fortify bug and the governor's manor bug) Were it changes not properly broadcast to other players? Random number related bugs? Something else?

Here is the best advice I've been given on OOS errors:

When you get an OOS, it means that one computer is doing something different than the rest of the computers in the game.

This usually happens in response to user input, i.e. you don't let the other computers know what the user selected. That's why we need to send messages for all user actions on the interface.

You can also get out-of-syncs when in the code you are doing things based on the active player. That results in different behavior on different computers because each computer has a different active player.

I hope this helps as a start. Let me know if you have any more specific questions.

Good luck!

Specific to those 2 errors, the attributes in question (the boolean that sets a unit as one that gets double the normal fortify bonus, or the boolen that sets a city as getting bonus production for unhappiness) weren't in the ::read or ::write functions for their respective objects.

As you would expect not being in those functions meant that reload of save games would cause problems. Which is why some players noticed that the governors manors stopped working in some games (because they had reloaded since building the manor). What I didn't know until tracking these issues down is somehow that process effects OOS's.

Either the ::read ::write streams are passed between servers to communicate changes, or the OOS checksum is calculated from the ::read and ::write streams so it wasnt being calculated correctly.

I suspect the first case since Sureshot noticed that they would have different winners in combats (so one computer knew the unit had a double fortify bonus and the other one didnt) and it wasnt just a "false" OOS.

Hopefully that helps.
 
I am just curious, on the people who are not getting OOS errors, do you play with huge maps and roughly 14+ civs?(12+ AI)?

I wonder if there is something happening with some of the racial combinations causing issue. Or if a particular race/building is causing issues.
 
I played max with 12 civ...10 AI + 2, huge pangea, with only one OOS, probably caused for one player went to dektop during a changing turn!
Calabim seems random to give some OOS, but in my game I haven't any.
With Huge map, I had only one problem.
It's very difficult for gamespy, for the game and for my pc, to alive in game during armageddon counter at 40 and at 100, caused by too heavy changing in game!
 
Tried a different game, it tends to OOS on turns, no where else. 8 AI civs, not using unrestricted leaders, huge big and small.

I wonder if it has anything to do with religon spread, as it doesn't start till then, or maybe modifiers on great leader production (spirit mana) As those are the only 2 things we have seen happening when it does go OOS.
 
I know this game isn't being developed any longer, but I still play it from time to time. Did anyone solve the OOS issue, or is it just too many triggers to fix them all? I get it all the time on both play by LAN and over gamespy.

Thanks!
 
I know this game isn't being developed any longer, but I still play it from time to time. Did anyone solve the OOS issue, or is it just too many triggers to fix them all? I get it all the time on both play by LAN and over gamespy.

Thanks!

I'm afraid that (as far as I know) I don't think there is a solution for the last version of the mod due to the amount of changes and possible triggers. If you can find an older version of the mod it worked a lot better, though you'd still encounter OOS very late in the game. The downside of that is obviously that you get less of the later functionality and content of FFH2, but the game was still enjoyable. The latest patch they happened quite quickly and regularly so it kind of ruled out multiplayer completely.
 
Top Bottom