Fleeing

[zup]

Of course it is higher than 10%. There are two factors: luck and rng. First off, everybody is unlucky. Secondly the rng likes to screw people over. End result: we don't stand a chance.

 

[Sarisin]

Well, I thought I had seen it all on this topic in my game.

However, then I attacked an Illian Horseman and it withdrew FIVE times in the same turn! Yes, 4 of those he defended and fled at 0.X strength.

The unit did have one or more withdrawal promotions in addition to his innate Illian flee-ability. But, c'mon. This really was like playing whack-a-mole.

I guess each attack is treated separately and the previous act of fleeing has no impact on the next battle.

The next turn the Horseman fled one more time before I ended the madness and killed him.

I still think the solution is units with 0 strength should not be permitted to flee a second time on the same turn.

One more thing: I paid special attention to the Workers I was attacking. I am certain they were 0/0 and they were causing damage to my attacking units as they fled.

 

[zup]

Oh you don't even know the worst of it. Let Hippus get horseback riding and see for yourself. The best you can hope is to get a single one tile chokepoint, on a hill, found a city there. Build insane fortifications and culture. Garrison 50+ units. Then hope they have to cross a river.

 

[Sarisin]

Oh you don't even know the worst of it. Let Hippus get horseback riding and see for yourself. The best you can hope is to get a single one tile chokepoint, on a hill, found a city there. Build insane fortifications and culture. Garrison 50+ units. Then hope they have to cross a river.

You're right, zup. I recall my last game with the Hippus in there that their Horsemen did do a lot of withdrawing/fleeing. But, honestly, I didn't see anything like the frequency of fleeing of the Illians or the 0/0 units causing damage.

 

[Breez]

I don't disagree Sarisin, but i wonder how much control they have over a mechanic that is part of the core civ game. i.e. is that a moddable feature?

(BTW Morsement also typically have a small innate withdraw also, It isn't too hard to get an 80% ish withdraw chance on a mounted unit)

 

[xienwolf]

You cannot withdraw during a first strike round, in case you weren't aware of that. So attack with some first strikes and you can nail the pesky runners.

 

[zup]

Flanking 3 gives first strike immunity, horse archers have innate first strike immunity. I don't see how first strikes remedy the problem.

 

[Vehem]

That's a complicated test for a computer dope like me to attempt, xienwolf.

All I can say is that I had at least a dozen case in my game of a Worker with no strength or defense pts fleeing and causing damage to my units. In some cases the damage seemed excessive to me.

So, you have two issues:

1. Should units with 0/0 be allowed to flee?

Logically, to me anyway, if a combat unit lands on the same tile as a non-combatant, that unit should automatically be killed/captured. I think this is the same as if you come across an Illian Great Merchant making his way to a city. Does he flee when you land on the same tile. I doubt it. Shouldn't it be the same for plain Workers?

2. Should those units cause damage?

If what Breez is saying is true, that newly created Workers don't really have 0/0, but a bit higher I can see some damage. However, logically again, how much will a Worker cause to a combat unit, HN or not? Probably none. Definitely not 11 turns worth.

I've just checked this.

100 workers with 90% (maxed) withdrawal chance, attacked by 100 warriors. 100 Workers were captured. No warriors were injured.

It seems that units that cannot engage in combat (true 0 strength) will never withdraw.

====

That means that the Illian workers must have had a Hardy Promotion (which grants a 30% chance to flee and 2 defensive strength). After fleeing, the unit would display as 0.0 strength (as opposed to 0, which is the norm or 0/0 which is how units with differing Off/Def are displayed), because it's health is so low that it's effective strength is less than 0.1.

The withdrawal frequency for a Homeland/Hardy I combination is 40%. 2 defensive strength is sufficient to win a couple of rounds, probably enough to injure a HN unit for 11 rounds as reported.

====

Regarding the "escaping units" issue, an injured unit retreating shouldn't be a problem. I'd prefer units withdrawing when in a city to actually withdraw from the city ("All is lost! Flee!) which actually solves the issue of needing ~10 units to kill the single withdrawal-chance-90%-city-defender. All withdrawal is in that case is good for is to keep your units alive.

The "Keeping units alive" however is working very much as intended. A withdrawing unit has very little combat strength and so is not a threat for a period of time whilst it heals. If you force a unit to withdraw, you have beaten it temporarily. If you want to run it down and kill it, then you're going to have to take your chances.

EDIT: I've actually just reverted the tweak that kept withdrawing units in the city. They'll run off to an adjacent plot like any other unit now.

 

[Sarisin]

Thanks for that additional info, Vehem.

I dunno, I still have to get used to all the fleeing and Workers causing damage, I guess, especially when going up against the Illians.

Yes, it's true once you attack a unit, it flees, and has its strength down to 0, it is not a threat to kill one of your units. However, what is really annoying is that if the fleeing unit is in your territory, it can and will pillage something before you get a chance to finish it off when your turn comes.

Anyway, with all this monkey business, it was one of the best FF games I've had. It's just too bad the CTDs started coming and made it impossible to finish the game - heading towards a cultural victory for the Dural, which kinda makes sense.

I think if I had to give my number one pet peeve about the whole civ4/FFH/FF/Orbis lineup it would be my inability to finish a game - usually by circumstances out of my control. I went down to a Large map this time and still no dice. I think the game is just for normal variables and when you start choosing other options (slow speed, more civs, huge map, etc.), you will find the game bogs down after awhile and just can't be finished.

 

[Breez]

Sarisin that high of a turn number... are you still playing Marathon or Epic speed? The higher that number gets the higher your chances to crash get. I simply quit playing epic/marathon games to help avoid that.

 

[Vehem]

Sarisin that high of a turn number... are you still playing Marathon or Epic speed? The higher that number gets the higher your chances to crash get. I simply quit playing epic/marathon games to help avoid that.

Aye - Xienwolf coded that ages ago...

if CyGame().getSorenRandNum(1000, "Doomsday") <= iTurnNumber:
	iGozer = CyGame().getSorenRandNum(4)
	if iGozer == 0:
		CyGame().ThrowLotsOfPythonErrors()
	elif iGozer == 1:
		CyGame().CTD()
	elif iGozer == 2:
		CyGame().DeleteAllAutoSaves()
		CyGame().CTD()
	elif iGozer == 3:
		CyGame().TurnOnLargeHadronCollider()
		pPlayer.setName("Gordon Freeman")

 

[Breez]

I Knew It!!!!!

 

[Iceciro]

Aye - Xienwolf coded that ages ago...

if CyGame().getSorenRandNum(1000, "Doomsday") <= iTurnNumber:
	iGozer = CyGame().getSorenRandNum(4)
	if iGozer == 0:
		CyGame().ThrowLotsOfPythonErrors()
	elif iGozer == 1:
		CyGame().CTD()
	elif iGozer == 2:
		CyGame().DeleteAllAutoSaves()
		CyGame().CTD()
	elif iGozer == 3:
		CyGame().TurnOnLargeHadronCollider()
		pPlayer.setName("Gordon Freeman")

....I laughed so hard. Just made my workday, sir.

 

[Valkrionn]

Epic win.

 

[ICanHasName?]

1 Flee per turn is a good idea. Yes thats one worker.

 

[Sarisin]

I just had a Calibim Horse Archer defending a city flee EIGHT times before finally being killed.

Workers continue to flee at an absurd rate too. Thanks for that example in the preceding post.

 

[Vehem]

I just had a Calibim Horse Archer defending a city flee EIGHT times before finally being killed.

Workers continue to flee at an absurd rate too. Thanks for that example in the preceding post.

What promotions do the workers have? Should be 30% with Hardy I, 50% total with Hardy III, 60% max if they benefit from Homeland (Defender Trait).

EDIT: Fixed an issue with Hardy III. If they had that promotion, they'll be retreating far too frequently (80% rather than 50%).

 

[Sarisin]

What promotions do the workers have? Should be 30% with Hardy I, 50% total with Hardy III, 60% max if they benefit from Homeland (Defender Trait).

EDIT: Fixed an issue with Hardy III. If they had that promotion, they'll be retreating far too frequently (80% rather than 50%).

I don't believe I've seen a Worker with the Defender/Homeland promotion. In this case, Calibim definitely did not, maybe Illians did.

So it is possible they had 50% with Hardy III. Probably why they are causing damage too. I had one worker cause two turns of damage to a War Chariot.

Vehem, I think the problem is, and this was illustrated well in the example given, is that a Worker will flee - max of 50%. You attack again. He flees again - because he retains the 50% rate. This is why you see units fleeing so many times in one turn.

I would propose that a unit be able to flee only ONCE per turn...if that can be done. Otherwise, you get into that whack-a-mole mentality with all the fleeing going on.

 

[Vehem]

Vehem, I think the problem is, and this was illustrated well in the example given, is that a Worker will flee - max of 50%. You attack again. He flees again - because he retains the 50% rate. This is why you see units fleeing so many times in one turn.

It was actually because the flee rate was 80% (Hardy III was adding 50%, not making the total 50% as it should).

Makes quite a difference over repeated attacks - the following are the chances to survive a set number of consecutive attacks...

Attacks		At 50%		At 80%
1		50%		80%
2		25%		64%
3		12.5%		51.2%
4		6.25%		40.96%
5		3.125%		32.768%
6		1.5625%		26.2144%

With the rate as high as it was, it's feasible for a worker to survive 3 or 4 attacks without it being much of a surprise, and 5 or 6 are still believable and not particularly uncommon (1 worker in 4 will survive 6 attacks in a row). With the 50% rate (as it should have been) the values are a little more sane. It's still *possible* for a worker to be an "annoying little #%$*" - but it'll be 1 in 67 or so rather than 1 in 4.

 

[Sarisin]

I'm finishing up a loooong game with patch L, but I suspect this issue may still come up in the new version:

I'm playing as Decius/Calibim and the AC moved to 100 quickly in the game. Along the way Ars Moriendi comes strolling into my territory. Fortunately he is wounded and I have a squad of War Chariots (with Courage) waiting for him. My Combat Odds are 99.1%.

A sure kill, right? Um, no.

I attacked with my four War Chariots and Ars fled from all four combats. I had a Beastmaster and he fled from the Beastmaster too. Five chances and the Horsemen fled FIVE TIMES! I'm sorry, but that is too much.

Making it worse, for some reason, next turn my Combat Odds for my War Chariots had now dropped to less than 20%. I'm not sure how Ars got his strength so much higher. It's true 2 of my War Chariots were damaged from the previous combat/fleeing, but still....

Again, I think ANY unit, even a big bad Horseman, should be able to flee only once per turn.

PS. My first time as a civ with Conqueror trait. Why can't I build Orc units when I capture a barb city?