Armies: WOW!!!

Originally posted by zorven


Is the original attack value the average of the attack values of the units in the army?

I haven't tested how mixed-unit armies work with the new bonus. In PTW, mixed-unit armies displayed stats reflecting averages of the unit values, but actually employed real unit values when engaged in combat (i.e., a tank-cavarly-knight army would attack with a value of 16 unless the army lost 1/3 of its hit points in any one battle, and then the battle would continue with an attack value of 6 -- this even though the units stats in the info box showed the army with stats of 9.5.2 or 8.4.2 if my rounding is wrong). In C3C the above example army would have an attack bonus of 4, and based on the PTW implementation, I would speculate (repeat: speculate) that the above army would attack with 20 so unless it lost 1/3 its HPs in one battle and then would continue the battle with an attack value of 10.
 
Originally posted by Pyrkaige
Is this "artificial"? Perhaps, in the same way that a Golden Age is artificial, especially if started when some military unit formed 2000 years ago winning a conflict and suddenly sending its home civilization into years of extra growth. That isn't very "natural", last I checked.
That's right.
I ALSO have removed Golden Age from my mod, in fact :)
Fortunately for you, CivIII can be modified so that you don't have to deal with leaders or armies at all if you wish (more choices!) That might work best for you personally, instead of persistently whining about something that the vast majority of players seem to enjoy. :rolleyes:
It's what I do, and to be true I just can't play with the default rules now, I just have ten times less fun than when I play my own mod.

But the thing is, my main interrogation was mainly about : are people really liking armies and GL, or was they just using them because they're here, but would not care if they were removed ?
 
I like them a lot. I would miss them if they were removed.
 
Originally posted by warpstorm
I like them a lot. I would miss them if they were removed.
Well, then I suppose I can just shake my head in confusion and keep wondering ^^
 
Originally posted by Catt
The numerator of the 1/6 function is not the average of the constituent units' attack values, it is the sum of the values. A 3 sipahi army would be (8+8+8)/6 = 4. The original attack value of 8 is increased by 4, making an attack value of 12. The same army could attack up to 4 times per turn (since the army has an extra movement point and blitz ability).
Just where do you have this information from?

You do realize you are trying convince people that an army with 4 modern armors (base attack strength 24) will now have an attack of 48!!! - instead of 30 with the correct formula that I posted.
 
Originally posted by CyberChrist
Just where do you have this information from?

You do realize you are trying convince people that an army with 4 modern armors (base attack strength 24) will now have an attack of 48!!! - instead of 30 with the correct formula that I posted.

Do you have any basis to believe your formula is correct other than you think its outrageous that an MA army could have such a high attack? Note that a fortified mech infantry in a metropolis on a hill has a defense of 45 -- not even accounting for rivers, civil dense, radar towers, etc. Edit: The mech infantry's defense value is actually 49.5 -- I forgot to add the 25% forification bonus. End Edit.

I believe my formula is correct based on: (1) the explanation of the bonus in the SP strategy guide included on the Bonus CD; (2) the commentary in a beta-tester's reports published and then removed from the web; and most importantly (3) the actual testing I did to determine whether the bonus existed at all. From my earlier post (post #37 in this thread) please see the test parameters:

I established a small landmass where both Rome (human) and Egypt (AI) have one city. The land is RR'd. All unit experience levels have 10 HPs in order to eliminate the effects of promotions and in order to generate more separate combat results from each monotonous attack than 4 HPs provide. Longbows are modded to have an attack value of 22 and 2 moves (revenge of the longbows ); Infantry are modded to have a defense of 20. The Egyptian AI has a stack of ~40 infantry, unfortified on grassland -- with the grassland bonus, the infantry have an effective defense of 22. The Roman human has 30 longbows and 10 armies -- these are combined into 10 armies of 3 longbows each. Then it is just a matter of launching repeated attacks against the infantry stack with longbow armies, and keeping track of how many HPs are lost by each side.

If there were no army combat bonus, we should expect (with sufficient trials) a roughly 50 - 50 win - loss ratio between an army attacking at 22 offense against a defender with an effective 22 defense. If the 1/6 base bonus is present, we should expect the longbow armies to win about 60% of the time (3 longbows at 22 offense = 66 attack points; 66/6 = 11; 11 added to the attack of 22 means an effective atack of 33; 33 attack versus 22 defense = 33/55 = 60% win expected).

Out of 977 total individual combat die rolls (HPs), the armies lost 393 HPs while the infantries lost 584 HPs, meaning the infantries lost, and the longbows won 59.775% of the combats - pretty darn close to the expected 60 - 40 split.

Now, if your formula were correct, wouldn't we expect the moddified longbow army to have an attack strength of 25 or 26 (depending on rounding) and its expected win percentage to be closer to 53% or 54%?
 
Catt, I believe you are wrong. The Army shouldn't get 1/6th of each unit's attack and defense as bonus, it should get 1/6th of the Army's stats.

That means that a army of Cavalry would get a stat of 7 in attack and 3.5 in defense and if the movement is affected too, they would get 3.5 in movement too.

That is with a bonus of 1/6th of the army's stats. Not 1/6th for each unit as Catt implies, which would be 1 half for 3 units, and 2 thirds of the stats for 4.
 
Originally posted by Catt
Note that a fortified mech infantry in a metropolis on a hill has a defense of 45 -- not even accounting for rivers, civil dense, radar towers, etc.
I am sure most cities are not located on hills though, but apart from that it would have a defence of 49 actually (18+18+9+4) and an army with 4 Mech. Infantry would have a defence strength of 67 while fortified in a metropolis on a hill and using your formula.

This will increases the chance of the Modern Armors armies vs Mech. Inf armies (fortified in metro on hill) from a difference of 25(49-24) to 19(67-48), though your chances of softening the Mech. with artillery would be somewhat reduced. However armies of Modern Armors vs. non army Mech. Infantry would be almost 1:1 (48 vs 49), but really in favor of army due to increased HPs.

Having armies get as strong as this doesn't really seem to do much to improve on the possibilties of a peacefull game. Since you need armies of defenders to effectively defend against armies of attackers now (and even then at reduced strength), those with few or no armies will loose for sure. And those aiming for a peaceful game rarely have many (if any) armies.

Just for the sake of argument then if the formula you claim to be the right one is the right one, then that would HAVE to be the first thing they would need to fix/adjust in the next patch (imho) - unless they really want to make Civ3 even more focused on warmongering that is.
 
Originally posted by CyberChrist
... unless they really want to make Civ3 even more focused on warmongering that is.

I think that was the point of calling it "Conquests". My hotseat buddy agrees with me, and we both think that the few changes made add a sort of simplicity by neccesity.

Allow me to explain.

In Conquests, most of the scenarios don't have too many improvements or technologies for a civ to get bogged down trying to complete. Also if a player decides to focus on building up their empire, they lose the scenario because they get conquered- hence the name "Conquests".

This simplicity by neccesity truly improves an already in-depth and arse-kicking game by defining a role and focus for your empire on a grand scope while leaving the role and focus of cities and empire open in a much more narrow scope. Overall, you still have absolute control over your empire, but it is in your best interests to play "a certain way". How you choose to adapt to that role is open and fluid- the reason civ appeals to me. Tweaking the armies and adding Scientific Great Leaders have proven to be very cool additions in the games I've played so far.

I built a few light tank armies in the WWII in Pacific conquest, and all I can say is:
Tank Armies Rock!! :tank: :tank: :tank:
 
I'm unsure as to purchasing this expansion as alot of my homework +/- for the CIC is conducted on threads like this. So to help me, apart from the above threads on armies, can they be upgraded like other troops!!
 
Originally posted by Grey Fox
Catt, I believe you are wrong. The Army shouldn't get 1/6th of each unit's attack and defense as bonus, it should get 1/6th of the Army's stats.

That means that a army of Cavalry would get a stat of 7 in attack and 3.5 in defense and if the movement is affected too, they would get 3.5 in movement too.

That is with a bonus of 1/6th of the army's stats. Not 1/6th for each unit as Catt implies, which would be 1 half for 3 units, and 2 thirds of the stats for 4.

I don't really have a horse in this race - I'm just trying to figure out how the combat bonus works.

What is your theory based on (I stated what mine was based on a few posts above)? And how does it square with the empirical results I posted above?
 
I base my theory on the logic that the army unit gets the 1/6th bonus. Not each unit individually.

And an army based on 3 Siphai's has 8 in attack, not 24. (And a bonus of 1.33 of course)
 
I think the fact that MGLs no longer give you great wonders actually toned down warmongering. I say this because, in PTW and vanilla, you could get success at peaceful achievements - great wonders - by fighting a war. Now, the only thing war gives you is more success at war.

EDIT: It does make war more appealing, yes, but if they had taken away the Great Wonder option and left Armies unimproved, Military GLs would have almost no purpose.
 
Originally posted by Grey Fox
I base my theory on the logic that the army unit gets the 1/6th bonus. Not each unit individually.

And an army based on 3 Siphai's has 8 in attack, not 24. (And a bonus of 1.33 of course)

Grey Fox - no offense intended, but I don't see how that supports your position in any meaningful way. Yes, an army has certain unit stats, but . . . how in the world does that lead to the conclusion that the combat bonus formula must be based upon the army stats instead of the individual unit stats?

Similarly with respect to a distiction between the army unit and the constituent units, we know that an army still "acts" like individual units in certain ways. For example, an army of 3 units takes up 4 spaces in a transport. It acts as 3 MPs when in cities. And mixed unit armies attack and defend with the repspective units values -- not a blended army value. Put a tank and an archer in an army in PTW and its stats will show in the info box as 9.4.1 (or 9.5.1 if my rounding is wrong) -- but the army will attack with a value of 16 until such time as the tank is forced to yield to the archer and then it will attack with a value of 2.
 
Originally posted by Catt


Grey Fox - no offense intended, but I don't see how that supports your position in any meaningful way. Yes, an army has certain unit stats, but . . . how in the world does that lead to the conclusion that the combat bonus formula must be based upon the army stats instead of the individual unit stats?


Catt you seem to be loosing the argument here even though you have done the empirical tests to prove how it works. (there needs to be a head shaking smilie). This is obviously a perfect exampe of how hard it is to refute popular belief with science, even if the proof is conclusive.

To the others, it may seem more reasonable that the army bonus works based on the average army value, but that is not what Catt is talking about.

He is not saying how the army bonus should work, but how it actually works as determined by actual in game testing. Unless you can find a flaw in the tests Catt ran there is no question that the army bonus is calculated based on the total and not the average army strength. Hard numbers do not lie and that is what Catt has done.
 
:eek: :lol:

CyberChrist, Grey Fox, I don't follow your arguments at all. How can you dispute experimental evidence based on the fact that it seems unreasonable to you? Catt's test leaves no room for argument on how Armies work.

[Edit: cross-posted with etj4Eagle]
 
Sorry, but I've most have missed the fact that you've actually tested this in game.

I don't have the game myself so I can't make similar tests.

But if that is the fact, then I don't know if that is the way breakaway wanted it. At least it doesn't sound like it when they said it got a 1/6th bonus.
 
All this talk of armies is interesting. I never used armies before, probably because I really didn't understand how they would be a benefit in the times before Conquests.

I have a MGL just hanging out in a city right now. With Cossacks being a unit I have just been able to make in the last few turns, I think it's time to build an army to test this for myself.

Once I build an army, is it possible to add more units to it later?
 
Originally posted by CyberChrist
You do realize you are trying convince people that an army with 4 modern armors (base attack strength 24) will now have an attack of 48!!! - instead of 30 with the correct formula that I posted.

Not quite 48, more like 40.

24*4=96/6=16 bonus +24 base=40 attack

And considering that mech infantries in metropolises will often have a defense value of well over 40, it isn't that bad.

The most common armies that people will have is the 3-unit army, and most people put the same type of units into an army, so the simple way to remember the formula for those situations is +50% to the stats (knights go from 4 to 6, and Siphai go from 8 to 12).
 
I don't believe I was disputing anyting in my last post and saying there is no room for arguments is a rather narrow minded view.

If armies do indeed work as catt's experiment seem to indicate then the argument changes into whether this is really a good thing for Civ3.

Personally I say it is not a good thing and I will sincerely hope that they fix this - preferably to be using a formula similar to the one I posted earlier. All out war and the option of powerfull armies are all fine, but tripwiring the possibilities for peacefull games by making armies so powerfull that they become totally indespensible does the entire Civ3 community a great disservice - one I am sure would become painfully apparent to most in the long run.
 
Top Bottom