RB1 - Cuban Isolationists

[Arathorn]

In ALL FOUR rifleman battles this turn, we were behind in the odds, yet we won all of them. That's incredibly rare in Civ4, if you're not familiar with the combat system.

I find it ironic how misinformed that statement is.

An attacking rifle (str:14) against an infantry with 50 hps in a town with zero defense bonus on a hill, and Combat I and City Defense I promotions is adjusted strength 14 against 15.5. But the attacker only needs 3 hits and the defender 5, because of hps, so the attacker actually has a 72.96% chance of winning. Against the odds? Nope! Not if you understand the combat system.

The second battle, unless there's some other bonus not shown, the riflemen even displayed better odds. 14 vs. 14 (but really, 13.95). With the hp situation, the attacking rifle had an 85.66% chance of success.

At Anda, there was no hill defense. I don't know what the terrain bonus for ice/tundra is, but unless it's very high (well over 50%), the attacking rifles were more likely to win than not.

Good story-telling, yes. Factwise, a bit lacking. And, then, preaching about how you need to know the combat system when displaying a lack of understanding? The height of irony.

Arathorn

 

[ToddMarshall]

As far as I understand it there is some underlying (invisible) Health (Hitpoints). So a fully healed str 10 Unit (ex. Knight) has 100 Health but a wounded str 20 Unit which has now only str 10 has only 50 Health. With each round a certain amount of damage is done is the health down to 0 the unit dies. Since he has less health the reduced str 20(now str 10) Unit is weaker than the fully healed str 10 Unit. The strength just being a sign how good the chances of the Unit are to win one round. If each round takes away 10 health from the loser the healthy unit only needs to win 5 times to kill the wounded one while the wounded needs to win 10 times to win the battle.

Caveat: Thats how i understand it so this can well be complete BS.

I had read exactly the opposite somewhere from an *I thought* reliable source. I was under the impression that a fully healed str 10 unit with no modifiers was the exact same as a 10 of 20 unit with no modifiers.

I for my part would not give up a City for peace not in SP and for sure not in MP.

I wouldn't make a habit of giving up a city for peace, but I HAVE done so in Civ 3 (twice) when I thought it was preferable to losing a better city I couldn't have held on to baring extreemly lucky rolls. (won both games).

Considering what Arathorn says about AI who, while being systematically taken overf by him, demand core cities for peace..... thats clearly a NEGATIVE feature to any SP games.

What is the point of this feature then??? The ONLY point I can see in it is to gift cities to friends in team/mp games. I too doubit I'd give up cities in MP for peace, (caveat: I've never played MP in Civ) though maybe.....

- Maniac

 

[Arathorn]

I had read exactly the opposite somewhere from an *I thought* reliable source.

The original DeepO/Solver article on this made the statement. So did my original combat article. Further testing revealed that they were wrong. My combat article is mostly corrected for this now (but a bit of the old stuff still remains -- it's not fully clean yet). Sulla wasn't the only beta-tester to not understand the combat system. I just find the irony of preaching incorrect information to be particularly delicious.

Anyway, to be completely clear, a 10.0/20 strength is a LOT weaker than a 10/10 unit. The latter will win about 77% of the time. A strength 8 unit will beat a 10.0/20 unit about 2/3 of the time but will only beat a 10/10 unit about 1/4 of the time. They're VERY different beasts. Current hps plays a huge role.

Arathorn

 

[ChrTh]

An attacking rifle (str:14) against an infantry with 50 hps in a town with zero defense bonus on a hill, and Combat I and City Defense I promotions is adjusted strength 14 against 15.5.

I'm confused, I thought infantry's base strength is 20? And then versus rifles, it gets +25%, making it a base 25. Where are you getting the 15.5 from? Am I missing something about defense strength (admittedly I haven't read all the combat articles yet)?

EDIT: 25 + 10% Combat 1 = 27.5 + 20% City Defense = 33 ... you mentioned 50 hps, is defense decreased by a percentage equal to the HPs lost?

 

[Kylearan]

Hi,

Sulla wasn't the only beta-tester to not understand the combat system.

I guess the beta testers hadn't access to the algorithms how the game handles combat internally, so it's not really their fault. Most of the time you fight with healed units anyway, so this is not something you can easily learn by simply playing a lot.

It's a shame that the displayed combat odds are not the real odds of winning a particular battle. Too bad things like the dreaded LCS (Lying Character Screen) from Diablo II make their return in other games...

I just find the irony of preaching incorrect information to be particularly delicious.

Amusing and ironic, yes, but then not everybody has an above average interest in combat combined with good skills in mathematical analysis and a lot of patience like you do...

-Kylearan

 

[ChrTh]

I'm confused, I thought infantry's base strength is 20? And then versus rifles, it gets +25%, making it a base 25. Where are you getting the 15.5 from? Am I missing something about defense strength (admittedly I haven't read all the combat articles yet)?

EDIT: 25 + 10% Combat 1 = 27.5 + 20% City Defense = 33 ... you mentioned 50 hps, is defense decreased by a percentage equal to the HPs lost?

Ok, I see it now on the screenshot: 10/20. So 10+25%=12.5+10%=13.75+20%=16.5 ...

you're off by 1 ... unless I'm missing something else. What calc are you do for overall odds of success?

 

[ChrTh]

Ok, I see it now on the screenshot: 10/20. So 10+25%=12.5+10%=13.75+20%=16.5 ...

you're off by 1 ... unless I'm missing something else. What calc are you do for overall odds of success?

Unless, of course, % are applied concurrently not cumulatively, in which case it's 10+55% = 15.5... never mind, the lesson, as always, is I'm an idiot.

 

[Rowain deWolf]

I wouldn't make a habit of giving up a city for peace, but I HAVE done so in Civ 3 (twice) when I thought it was preferable to losing a better city I couldn't have held on to baring extreemly lucky rolls. (won both games).

Ah yes Civ3 but there you had usally more than just 5, 6 Cities and thanks to corrruption several junk ones too.
Here in Civ4 losing 1 City is a huge loss most of the time. Very, very seldom you get as much space to expand peacefully as Sirian and Sulla got in this game. Thats what really crazy about this game. Room for ~20 Cities on a standard map with standard numbers of Enemies .

 

[ChrTh]

One last question. How do we determine the "number of hits" required to kill a unit? Arathorn, you said the attackers only need 3 while the defenders need 5. What causes those particular numbers?

Thanks

 

[Kylearan]

Hi,

How do we determine the "number of hits" required to kill a unit?

According to an article someone wrote , the damage per hit is fixed for the duration of combat and is given by floor(20*(3*A+D)/(3*D+A)) for the attacker, and floor(20*(3*D+A)/(3*A+D)) for the defender, with A and D the modified attacker's/defender's strength.
In the 3 hits vs. 5 hits case (rifle vs. 50hp inf...), A is 14 and D is 15.5, so attacker's damage is 19 and defender's damage is 21. The wounded infantry starts with 50 hitpoints, so 3 hits of 19 damage each will kill it, while the attacker is healed and starts with 100 hitpoints, so 5 hits of 21 damage each are required to kill it.

-Kylearan

 

[Arathorn]

ChrTh, check out http://forums.civfanatics.com/showthread.php?t=137615 for some reasonably detailed description of how combat is handled.

I do think I missed the +25% for infantry vs. gunpowder, though. There's +25% for being on a hill. That makes the infantry an 18 strength defender (10 + 25% for hill + 25% for gunpowder + 10% for Combat I + 20% for City Garrison I). Attacking rifle still has a 65.88% chance of victory. Even at 14 vs. 18 strength.

It's a shame that the displayed combat odds are not the real odds of winning a particular battle.

Agreed 100%. In the article above, some other instances where the inability to see all the details of a unit (including its hps) can make a reasonably significant difference in strategy. I'd much prefer a WYSIWYG interface, but, alas, it's not to be (in the original version, anyway).

Arathorn

 

[ChrTh]

Hi,



According to an article someone wrote , the damage per hit is fixed for the duration of combat and is given by floor(20*(3*A+D)/(3*D+A)) for the attacker, and floor(20*(3*D+A)/(3*A+D)) for the defender, with A and D the modified attacker's/defender's strength.
In the 3 hits vs. 5 hits case (rifle vs. 50hp inf...), A is 14 and D is 15.5, so attacker's damage is 19 and defender's damage is 21. The wounded infantry starts with 50 hitpoints, so 3 hits of 19 damage each will kill it, while the attacker is healed and starts with 100 hitpoints, so 5 hits of 21 damage each are required to kill it.

-Kylearan

Got it. Thanks (and maybe one day I'll get around to reading the combat articles...that's the problem with being a lover, not a fighter... )


EDIT: Thanks too Arathorn

 

[weakciv]

i guess i missed it somewhere. how do you know the current HP of a unit? I know max is 100 for each unit. Is there a calc based on X strength out of Y Full strength?

 

[ChrTh]

i guess i missed it somewhere. how do you know the current HP of a unit? I know max is 100 for each unit. Is there a calc based on X strength out of Y Full strength?

If you look at this pic:

...you'll see that the defending Infantry are at 10.0/20 strength. If I understand what everyone is saying above correctly, the base strength ratio (i.e. before modifiers) is equal to your HP ratio. So in this case, since they're 10.0/20=50% strength, they'd be 50% HP, i.e. 50.

 

[Arathorn]

how do you know the current HP of a unit?

Excellent question. Sometimes you don't. In this case, the strength of 10.0/20 tells you it must be at exactly 50 hps. 51 hps would be strength 10.2/20 and 49 hps would be 9.8/20. In general, current_strength/full_strength gives a good guide to hps.

In this case, I also know that the limit of collateral (and other types of damage) for a stealth bomber is 50. That is, it can't do more than 50 hps of damage. Since I know Sulla would bomb until he was blue in the face, I could predict the 50 hps, too. That's essentially "red-lined", although different units can redline to different values. It's an open question whether that redlined is more or less powerful than a Civ3 redlining. Whoever wrote that combat article should investigate that....

Arathorn

 

[ThERat]

Agreed 100%. In the article above, some other instances where the inability to see all the details of a unit (including its hps) can make a reasonably significant difference in strategy. I'd much prefer a WYSIWYG interface, but, alas, it's not to be (in the original version, anyway).

Now I understand why in the AW game I play I often lose despite the 'displayed' odds. This is extremely bad since you don't expect anyone in a AW game with so many fights to take out a calculator and check all those conditions.
But a wrong 'odd' display is actually the worst case scenarion in my mind

Now IIRC someone said the advantage of the single strength value is a simpler combat model. Well, well

 

[weakciv]

Excellent question. Sometimes you don't. In this case, the strength of 10.0/20 tells you it must be at exactly 50 hps. 51 hps would be strength 10.2/20 and 49 hps would be 9.8/20. In general, current_strength/full_strength gives a good guide to hps.

In this case, I also know that the limit of collateral (and other types of damage) for a stealth bomber is 50. That is, it can't do more than 50 hps of damage. Since I know Sulla would bomb until he was blue in the face, I could predict the 50 hps, too. That's essentially "red-lined", although different units can redline to different values. It's an open question whether that redlined is more or less powerful than a Civ3 redlining. Whoever wrote that combat article should investigate that....

Arathorn

ok thanks Arathorn and ChrTh.

So we will usually not know for sure what the exact HP is but we can make a very educated guess by the ratio of curr-str/full-str.

 

[weakciv]

(double post)

 

[Sullla]

Thank you for explaining to us a bit more about how the combat system works, Arathorn. The combat odds were against us in every battle fought by our heroic riflemen, so even if their opponents had fewer hit points, I think they still qualify as noteworthy.

PS There are also many ways in which to make a valid point. Posting messages to this effect:

Good story-telling, yes. Factwise, a bit lacking. And, then, preaching about how you need to know the combat system when displaying a lack of understanding? The height of irony.

Sulla wasn't the only beta-tester to not understand the combat system. I just find the irony of preaching incorrect information to be particularly delicious.

even if technically true, will not win you many friends. I make mistakes all the time, but this could have been done a bit more tactfully, I think.

 

[GreatMPL]

In ALL FOUR rifleman battles this turn, we were behind in the odds, yet we won all of them. That's incredibly rare in Civ4, if you're not familiar with the combat system.

I find it ironic how misinformed that statement is.

The statement, read carefully, is not misinformed:

In ALL FOUR rifleman battles this turn, we were behind in the odds, yet we won all of them.

Sullla probably means that the riflemen were behind in the displayed odds, which as we know are not the same as the real odds of winning.

That's incredibly rare in Civ4, if you're not familiar with the combat system.

That is: someone who doesn't know how the real odds are calculated is not likely to win 4 straight battles. Sullla has had enough experience with the combat system to be able to win anyway.