Single Player bugs and crashes v35 plus (SVN) - After the 18th of August 2014

[Arakhor]

The localisation for all the forced terraforming actions in the Weather Control is missing and the Chief's Pyre is missing a text string too. Further, the "Mesoamrican Temple" [sic] is misspelt.

On a different note, the Magic quote doesn't really fit what this technology is supposed to be. Setting aside why it's even in the main game and not in the Fantasy module, I'd suggest that a better quote might be Connor McLeod's "Hey, it's a kind of magic!" from Highlander.

 

[Thunderbrd]

No it can't because it wasn't used in C2C and i removed it some time ago.

Ah... some of it's old code may still be haunting us it sounds like anyhow. You've always had a better grasp on the discord there.

 

[alberts8]

Ah... some of it's old code may still be haunting us it sounds like anyhow. You've always had a better grasp on the discord there.

I think one reason is that the AI uses a different Odds calculation the other could be possible issues in some performance enhancements Koshling made years ago.

But if it would come from Koshlings changes to speed things up it should have been noticed right after he made those changes and i don't want to undo them because our turn times are already to long.

Edit:

Aren't bad units chosen in some situations as sacrifice just to damage enemy units? It would explain those things and is a feature not a bug.

 

[Sparth]

The Mining tech is still playing Leonard Nimoy's voice (which doesn't match the actual text displayed), instead of the quote that I recorded that does.

fixed

 

[Sparth]

On a different note, the Magic quote doesn't really fit what this technology is supposed to be. Setting aside why it's even in the main game and not in the Fantasy module, I'd suggest that a better quote might be Connor McLeod's "Hey, it's a kind of magic!" from Highlander.

Done
I took audio quote from the movie. It quite noisy, still better then synthetised voice.

 

[manivo]

I think one reason is that the AI uses a different Odds calculation the other could be possible issues in some performance enhancements Koshling made years ago.

But if it would come from Koshlings changes to speed things up it should have been noticed right after he made those changes and i don't want to undo them because our turn times are already to long.

Edit:

Aren't bad units chosen in some situations as sacrifice just to damage enemy units? It would explain those things and is a feature not a bug.

It's something that changed in the past couple of months - I've done an SVN update before starting a new game June 1st, after a couple months break from C2C - I haven't noticed this behavior last time I played.

Again, I don't think it's a calculation that involves sacrificed units to weaken the enemy, as one of the units in the selected group always has the exact odds displayed in the pre-attack calculation.

 

[alberts8]

It's something that changed in the past couple of months - I've done an SVN update before starting a new game June 1st, after a couple months break from C2C - I haven't noticed this behavior last time I played.

Again, I don't think it's a calculation that involves sacrificed units to weaken the enemy, as one of the units in the selected group always has the exact odds displayed in the pre-attack calculation.

So one unit had a combat outcome that was different from the odds displayed?
Or was another unit chosen to attack not the one that was chosen to display the odds?

We had reports of such problems last year and some players even said that they noticed it years ago after something was changed.

 

[Thunderbrd]

I think one reason is that the AI uses a different Odds calculation the other could be possible issues in some performance enhancements Koshling made years ago.

But if it would come from Koshlings changes to speed things up it should have been noticed right after he made those changes and i don't want to undo them because our turn times are already to long.

Edit:

Aren't bad units chosen in some situations as sacrifice just to damage enemy units? It would explain those things and is a feature not a bug.

I think the issue is that it IS a feature for the AI. But a selected stack for the player shouldn't behave this way because it makes the player feel like they've been lied to by the odds calculator when a weaker unit is chosen to go fight than the strongest unit that displayed its odds.

 

[alberts8]

I think the issue is that it IS a feature for the AI. But a selected stack for the player shouldn't behave this way because it makes the player feel like they've been lied to by the odds calculator when a weaker unit is chosen to go fight than the strongest unit that displayed its odds.

But even without it the odds displayed could be different from the odds used to select the best attacker or defender in the other stack. Because the best unit to attack or defend is chosen using another odds calculation.

 

[manivo]

Or was another unit chosen to attack not the one that was chosen to display the odds?

That one.

The displayed odds were of the unit with the highest odds in the selected group. Attacking used a different unit - I haven't thoroughly tested it, but the behavior seems to be that the attacking unit picked was just the first unit in the selected group.

 

[Harrier]

Something to remember.
When you hover over an enemy unit ( from a tile with lots of units) - it calculates the battle odds using all the units on that tile. If all those units are not stacked together, when you attack - the odds will be different. It will recalculate to include only the units that are combined in the stack that attacked.

As an example, you have a horse unit (not stacked) in a tile with lots of foot units, if you hover over an enamy unit 2 tiles away, you get good odds (because of the foot units) - but only the horse can move 2 tiles - so the attack fails. (because the odds are now differrent.)

I fall into this trap all the time, even though I know about it - memory loss.

Unless of course C2C has changed that part of Civ4.

 

[Thunderbrd]

But even without it the odds displayed could be different from the odds used to select the best attacker or defender in the other stack. Because the best unit to attack or defend is chosen using another odds calculation.

As we've discussed before, this is a bug. The display should be the same as what it would send in, though not necessarily the same as what the AI would send in to attack first as it evaluates more than just combat strength.

 

[alberts8]

As we've discussed before, this is a bug. The display should be the same as what it would send in, though not necessarily the same as what the AI would send in to attack first as it evaluates more than just combat strength.

The whose situation is just frustrating.
A set of changes a few years ago caused those issues and i don't even want to think about how messed up things where before those changes. Someone must have noticed the effects of those changes right after they where made and the issues we have now should have been fixed right away. But i quess all the requests to add new stuff where more important and i understand that because adding new things is alot more fun as trying to fix endless bugs..

 

[JosEPh_II]

@alberts2,
Depending upon the time frame of these errors, some bugs were introduced prior to having a competent "coder" on board. And the early stages as you said new things were being added in so fast that only the most glaring problems were attempted to be fixed. I remember a time when koshling early on warned that there were major bugs that needed serious attention an would the xml modders adding new content halt the additions. He was basically ignored. The reason given was that the modders adding content "needed" to get their stuff in while they had their "creative juices" still flowing. And If they had to stop they might not finish what they started. A certain modder was very very bad about this behaviour and attitude.

Later on when more coders got on board (AIAndy for example) some of these issues were brought up and by this time koshling was growing weary of having to correct xml as well as fixing the new bugs being introduced with every new addition. He warned that if you went too deep (in the coding" there were "dragons lying down there". Then AIAndy who made the Property system had a job change and his time was reduced and eventually he only pops in from time to time now, but not to code. Koshling was helping T-brd and ls612 get more in depth with the coding but was growing really weary of the never ending stream of bugs introduced with each new addition to the mod. And he burned out.

So in essence the coding has hidden many deep seated bugs that have been around for a long time. Constant New additions obscure the need to root them out, and only the newest introduced bugs (usually the graphical CTD variety) got any real attention.

So as a metaphor and an alarmist reaction, today's C2C's coding is like a Jenga Tower base that just needs one piece pulled away for the toppling of the whole tower. (Probably isn't That Bad, but then again???) You see this better than we can.

JosEPh

 

[Dancing Hoskuld]

The change to the Hunting Instruction building does not work well with the game option (and traits) that increase the number of exp you require to level.

 

[Toffer90]

The change to the Hunting Instruction building does not work well with the game option (and traits) that increase the number of exp you require to level.

The same could be said about the barracks, dog kennel, stable etc. as well

EDIT: I made a mistake and gave master hunter tradition 1 XP less than the two buildings it replace does; I will fix it the next time I make a commit.

EDIT2: Gamespeed should affect the amount of XP needed to level up and the amount of XP given by buildings. Alas, I don't know how to make this happen.

 

[Dancing Hoskuld]

The same could be said about the barracks, dog kennel, stable etc. as well

Not so, they did not give a free promotion.

 

[Toffer90]

Not so, they did not give a free promotion.

I knew I defied you when I made this change; sorry for that.
I would be okay with a free promotion if it gave an unique promotion that can't be taken at level up.

 

[Thunderbrd]

The whose situation is just frustrating.
A set of changes a few years ago caused those issues and i don't even want to think about how messed up things where before those changes. Someone must have noticed the effects of those changes right after they where made and the issues we have now should have been fixed right away. But i quess all the requests to add new stuff where more important and i understand that because adding new things is alot more fun as trying to fix endless bugs..

Well... that's the thing... I'm pretty sure the changes when made were intended to have this effect and came from the lead from behind methodology which may not have been well isolated when coded. It was intended to send in weaker units so the weaker units could soften the enemy and be sacrificed rather than the toughest ones and so they could otherwise begin to benefit from leveling so that not only the leaders in the stack would ever improve. But it's terrible for the player who feels lied to.

That said, you may well see more in it that I ever did - I found that particular portion a needlessly complex maze that was very hard to sort out and tried to avoid it except to inject what seemed to be necessary where it seemed to be. I suppose that could've caused some issues or compounded some as well and if so, my apologies.

 

[Toffer90]

Repeatable EoT crash on latest SVN. Happens when my scout wins over a wilderbeast in self defense.

View attachment 395368

The save game included in attachment is from a game with Josephs and my own modmods.
But I don't think the CTD is from those xml changes.