You know the combat system is FUBAR when...

[Chamnix]

If it works this way, then it is indeed FUBAR. This means, as one example, that a modern paratrooper (6-11-1) attacking a spearman (1-2-1) on a mountain (def bounus 100%) only has a 6/6+4=60% chance of winning. The Bronze Age spearman has a 40% chance to win...

Another example: A berserk 6-2-1 attacking a modern armour (24-16-3) in the open, let us say on plains, has a 6/6+16=27% chance of winning. If it attacks a tank (16-8-0), the berserk has a 6/6+8=43% chance of victory ie more or less a 50-50 chance.

If this isn't FUBAR...

It does. Your examples are mostly correct – the only modification is that the chances you are giving are single round chances. When both units have 4 hps for example, then the stronger unit’s chances of winning the overall battle are greater than those figures, but not high enough to make you happy, I’m sure .

To me, it is simply about gameplay over realism. I don’t want the player with the research edge to win almost all the time. I think it makes for a more interesting game if the player with the backwards units can use tactics and terrain at least to make the more technologically advanced player struggle some. Think of it as the game punishing the strong and trying to challenge you .

I agree that there are plenty of things they could have done differently and possibly better, but I’m not sure that the way you thought it worked would have been an improvement.

Judging from the graphics/animations and when damage occurrs:

1. Defender free round -> 50% chance to damage x Defender's A/Attacker's D

2. Attacker's round -> Attacker's A + rng vs defender's D + bonuses + rng, the greater score wins the round

3. Defender's A +rng vs Attacker's D + bonuses + rng, the greater score wins the round

4. A

5. D

(etc til one party is "killed" or flees)

Ie. the combat system used in AD&D PC-games and early Military games.

Correct me if I’m misinterpreting this, but it seems to me that units’ having separate attack and defense values would be almost irrelevant under your system. The attacking unit is forced to use its defense value, and the defensive unit uses its attack value. As an example, if a 4/1/1 unit attacks a 4/1/1 unit, who should win? If there is to be any significance to separate attack/defense numbers, then the attacking unit should win that battle most of the time, but I don’t think your method would give that result.

 

[timerover51]

Chamnix, the problem I have with the combat system is two-fold. First, I design miniature and board war games, and the hardest thing in designing is the combat results system, trying to get it to "feel" right. Not necessarily to be absolutely historically accurate, although I normally keep that as an option for myself, but to be both playable and be fairly accurate historically. Playable and accurate tend to be antagonistic in design. The combat system in the game may be designed to make it more challenging going against the AI, but given the number of threads and complaints, for many it does not contribute to playability nor accuracy. I am enjoying playing Civ3 again only because I now have a Windows box for game editing, which allows me to work on getting somewhat more historical or accurate results. Prior to that, I had pretty much quite playing because of the AI headaches, or only played a solitaire game with launching the Space Ship as the victory condition.

The second problem is that there are better combat results systems out there. Yes, they are more predictable, but why is that bad? Given the difficulty range in Civ3, I cannot think that making the combat system a bit closer to what would really happen if an Spartan mora tried to tackle a battalion of mordern armor would affect the game that much. The other ones are still plenty challenging, and I do not have to play with the units near so much.

Note, I am a military historian, naval historian, and weapons effects consultant. I have spent a lot of time determining what happens when weapons go bang and bullets hit people.

 

[Chamnix]

… given the number of threads and complaints, for many it does not contribute to playability nor accuracy.

I may be suffering from my own “forum cognitive bias”, but it seems to me that more threads are about the “AI cheating” in combat than about the combat system itself. A given battle has exactly the same odds regardless of who owns which participating unit, and people’s claiming otherwise is what I have generally argued against.

As far as the combat system itself, there are many ways of designing it, and which you prefer is purely a matter of taste. Civ 4 changed the combat operation significantly – some people think it’s a major improvement, some think it’s a disaster .

The second problem is that there are better combat results systems out there. Yes, they are more predictable, but why is that bad?

It’s just a matter of personal preference. You could design a game where the stronger unit always won even if it were just a little bit stronger, or you could design a game where every battle is a coin flip, or anywhere in between. I think civ 3 does an adequate job striking a balance, but reasonable people may disagree. I’m guilty of drifting off topic a bit in my last post, but the most important thing about the combat system is that for better or for worse it is unbiased.

Given the difficulty range in Civ3, I cannot think that making the combat system a bit closer to what would really happen if an Spartan mora tried to tackle a battalion of mordern armor would affect the game that much.

Note, I am a military historian, naval historian, and weapons effects consultant. I have spent a lot of time determining what happens when weapons go bang and bullets hit people.

I am none of the above – I am a game player who just enjoys competing. I don’t really think in terms of a tank charging a spearman; I really just think of it as 16/8/2 unit attacking a 1/2/1 unit. I know what my odds of winning should be, and as long as the game is true to those odds, what the units are named doesn’t diminish my enjoyment of the game. I recognize people play civ for other reasons, and I don’t have a problem with that, but I guess the game can’t please all the people all the time.

(I know I split your quote apart pretty badly, and I apologize for that, but I don’t think I misrepresented anything you said – let me know if you disagree ).

 

[D0NIMATRIX]

Note, I am a military historian, naval historian, and weapons effects consultant. I have spent a lot of time determining what happens when weapons go bang and bullets hit people.

You're cool.

 

[timerover51]

Chamnix, when I work on designing a game, no unit is invulnerable, just some are more difficult to kill than others. It the Skirmish gaming rules that I publish, English riflemen are dead meat if the Zulus can close to hand to hand range, but it takes a lot of Zulus to get there. Same thing with naval rules. AD&D had the critical strike rule, which leveled things out a lot in there, as you do have a lot of odd things happen in combat. Harald Godwinson takes an arrow in the eye despite his helmet. Henry of France has a wood splinter go through his helmet eye slit during a joust, with fatal results. The Battle of Barnet during the English Wars of the Roses is one of the more bizarre ones on record. Friendly fire occurences galore. And do not get me on the subject of 15th and 16th century mercenaries.

When a spearman trashes my tank when I am attacking the town/city he is in, I have no problems with that. I spent part of the spring of 1991 working for the Marines on projects involving MOUT, short for Mechanized Operations in Urban Terrain, and my advice to the Marines was, do not do it. Mech vehicles in a city are expensive targets,
although the discarding sabot round of the Bradley/Grizzly would be useful for strafing buildings.

 

[timerover51]

You're cool.

Since you list yourself as Biohazard, I might as well let you know that I am pretty good on chemical, biological, and nuclear weapons and effects as well.

 

[Pyrrhos]

@ Chamnix.

I'd say that all those people complaining about the "AI cheating" is the result of three things:

i) A combat system that in certain, not uncommon, situations accords a neolithic warrior a 40-60% chance against the Screaming Eagles or All Americans of 2008 vintage. As most of us realise how utterly inaccurate that is, we perceive it as "cheating".

ii) The built-in checks against a run-away civ admitted to by the designers. Usually this civ is the player civ, so if many players feel that the AI is "cheating", they do have a point, even if the AI is just following its programming.

iii) Cognitive bias. It does play a part, but only a small one. I honestly cannot recall ever being the beneficiary of "spear" whereas in every game, there are multiple occurrencies of ridiculous combat results going against me. Since I know that, as an example, a Roman legionary or a knight would "wipe the floor" with a Bronze age spearman, I do not react when my knight or legionary kills the spearman on the mountain without losing any hitpoints. As the chance of that happening is [3/3+(2+2x100%)] to the power of hp of the spearman = 27/343 or 7.9% for a regular spearman, I have been lucky. But this is where cognitive bias enters the frame - as the result is what I expected and as I am not aware of the luck involved, I don't react to these instances of good, but not outrageous mind you, fortune.

PS. That's scary, timeover51. Myself, I am ex-army (with special training in NBC as a staff officer and yes, I've "played around" with mustard gas, making it burn by sprinkling it with chlorinated chalk etc). I too have studied history and have a degree. But I have only played games, never designed them.

 

[Chamnix]

ii) The built-in checks against a run-away civ admitted to by the designers. Usually this civ is the player civ, so if many players feel that the AI is "cheating", they do have a point, even if the AI is just following its programming.

If this is to say that the game designers made combat results depend on how each player was doing in the game and not based purely on attack/defense/terrain, etc., then I believe you are still just plain wrong. If that is not what you mean, then I am afraid I don't understand what you are getting at.

 

[D0NIMATRIX]

Since you list yourself as Biohazard, I might as well let you know that I am pretty good on chemical, biological, and nuclear weapons and effects as well.

I envy you.

 

[Marsden]

I have a question/comment that somewhat pertains to this regarding the longbowman unit. In game it's an ok attacker terrible defender. Didn't longbowmen in a castle or even in a field between their sticks (sorry I can't remember the proper name for those pointed sticks the bowmen stand between to break up charging enemies) make a powerful defender? Are they going by their lack of armor? It seems as if the defensive ability to hurt the enemy 100 yards away is completely ignored. As if every battle was a meelee and the longbowmen only shoot at point blank.


Let me draw on my limited historical knowlege to the 100 years war at Agnicourt. The English had ~5000 Longbows and ~1000 "mideaval infantry" vs. ~25,000 Knights. Nominally, I think the English were attackers, but they defended the field from the charges of the knights, IIRC. Again I ask, how would this play in a Civ 3 game. It seems like my 1 MI and 5 Longbows would be slaughtered by 25 AI knights. I realise even though they quit the field the knights were not all slaughtered, but they did loose a lot, I can't remember right now.

 

[CommandoBob]

Let me draw on my limited historical knowlege to the 100 years war at Agnicourt. The English had ~5000 Longbows and ~1000 "mideaval infantry" vs. ~25,000 Knights. Nominally, I think the English were attackers, but they defended the field from the charges of the knights, IIRC. Again I ask, how would this play in a Civ 3 game. It seems like my 1 MI and 5 Longbows would be slaughtered by 25 AI knights. I realise even though they quit the field the knights were not all slaughtered, but they did loose a lot, I can't remember right now.

The Face of Battle by John Keegan describes this battle quite well. It was an eye-opening read.

 

[D0NIMATRIX]

I have a question/comment that somewhat pertains to this regarding the longbowman unit. In game it's an ok attacker terrible defender. Didn't longbowmen in a castle or even in a field between their sticks (sorry I can't remember the proper name for those pointed sticks the bowmen stand between to break up charging enemies) make a powerful defender? Are they going by their lack of armor? It seems as if the defensive ability to hurt the enemy 100 yards away is completely ignored. As if every battle was a meelee and the longbowmen only shoot at point blank.


Let me draw on my limited historical knowlege to the 100 years war at Agnicourt. The English had ~5000 Longbows and ~1000 "mideaval infantry" vs. ~25,000 Knights. Nominally, I think the English were attackers, but they defended the field from the charges of the knights, IIRC. Again I ask, how would this play in a Civ 3 game. It seems like my 1 MI and 5 Longbows would be slaughtered by 25 AI knights. I realise even though they quit the field the knights were not all slaughtered, but they did loose a lot, I can't remember right now.

I think it was actually 7,000 longbows vs 14,000 knights, but my sources are rather ancient...

 

[Pyrrhos]

If this is to say that the game designers made combat results depend on how each player was doing in the game and not based purely on attack/defense/terrain, etc., then I believe you are still just plain wrong. If that is not what you mean, then I am afraid I don't understand what you are getting at.

Yes and no. What I refer to is the built-in checks against a run-away civ admitted to by the designers of which diddling with the combat system/results is but one possibilty.

 

[Pyrrhos]

None of the unit stats in relation to each other make a lot of sense historically. Except maybe warrior vs warrior.

Only units where the A and D values are the same such as the warrior make sense.

Reading this - "As an example, if a 4/1/1 unit attacks a 4/1/1 unit, who should win? If there is to be any significance to separate attack/defense numbers, then the attacking unit should win that battle most of the time, but I don’t think your method would give that result" - makes you wonder. The unit referred to is in all probability the longbowman, so let us look at a couple of longbowmen examples:

Longbowman vs longbowman. The attacking longbowman has a 4/4+1xdef bonus or 78.4% chance of winning each round on grassland, plains etc. The defending longbowman a 1/1+4xdefbonus or 18.5% chance. The chance of a defending longbowman to actually defeat an attacking veteran longbowman on plains etc is thus only 0.1%. FUBAR!

The result is the same when a longbowman is attacked by a veteran knight - the longbowman has only a 1/1000 chance of winning the combat. If attacking, the longbowman has a 4/4+3xdefbonus or 54.8%, i.e. 55-45 chance of winning. Again, FUBAR!

A Civ III "Battle of Agincourt" would look something like this:
25 French knights vs an English force of 2 MedInf + 5 longbowmen
[Knight attacking MedInf (as it has the better D) 4/4+2xdefbonus = 64.5% chance of knight winning]
If the French attack first, 2-3 battles will suffice to kill the MDIs and another 5 to kill off the remaining longbowmen. The French win 7-1 in lost units.

If the English attack first, with 55-45 odds, the seven englishmen will kill four knights and lose three longbowmen. When the French turn comes, the remining four already wounded English units will most likely be polished off
without loss. The French win 7 to 4 this time.

The Battle of Agincourt is indeed a good point Marsden!

 

[Chamnix]

Only units where the A and D values are the same such as the warrior make sense.

Reading this - "As an example, if a 4/1/1 unit attacks a 4/1/1 unit, who should win? If there is to be any significance to separate attack/defense numbers, then the attacking unit should win that battle most of the time, but I don’t think your method would give that result" - makes you wonder. The unit referred to is in all probability the longbowman, so let us look at a couple of longbowmen examples:

Longbowman vs longbowman. The attacking longbowman has a 4/4+1xdef bonus or 78.4% chance of winning each round on grassland, plains etc. The defending longbowman a 1/1+4xdefbonus or 18.5% chance. The chance of a defending longbowman to actually defeat an attacking veteran longbowman on plains etc is thus only 0.1%. FUBAR!

I was deliberately not referring to longbows – I was talking about a 4/1/1 unit. That is, a unit that is good when attacking and very poor when being attacked. In that case, it is obvious that the attacker should win a large majority of the time. You can say that there is absolutely no historical military unit like that, and I am not going to disagree with you. However, once you make the leap of faith that such a unit can exist, the combat results make perfect sense.

If you cannot accept separate attack and defense strengths, then I guess you have no real choice but to make warriors 2/2, spearmen and archers both 3/3, swords and longbows both 5/5, up to tanks at 24/24 and modern armor at 40/40. If you think that improves the game, then you are certainly entitled to that opinion, but I think separate attack and defense values add so much to gameplay that it is well worth the cost in historical accuracy.

Yes and no. What I refer to is the built-in checks against a run-away civ admitted to by the designers of which diddling with the combat system/results is but one possibilty.

OK, I will acknowledge that the game designers could have designed the game so that combat odds were partially dependent on the relative power of the combatants’ empires, and that would have been a built-in check against a runaway civ. However, the fact remains that they did not do this – the combat odds are the same regardless of how well the player is doing as has been shown over and over and over again through countless tests.

 

[Marsden]

I think what they did when they made Civ 3 it to oversimplify things to the point of nearly fantasy.
Spearmen, pikemen, muskets, infantry, etc are defensive units and can be generically thought of defender level 1,2,3, 5, and so on. Attackers also, so Attacker (armored) 1, or tank, vs. defender level 1 has nothing to do with technology difference. It's as if a warrior that can swing his ax 24 times better that plain warriors is attacking that spearman.

It's not historical, it's very simplistic. Just about every thing is.
Only temples and libraries add to a nation's culture, nothing else. Right. If my library has more culture than yours I get control of the coal mine even if you more infantry on to it? Sure.

My previous examples in this thread were trying to point out that things that wouldn't have looked good on paper have actually happened and I think they intentionally made the game as random as possible to try to allow these things to happen so you can have your own stories. "Remember when they attacked my city and all I had were 2 spears and they held back 4 cavalry" That kind of thing in the context of the civ you are playing remembering it's great battles, like we still talk about Thermopolae and Agincourt and others.

Also, you are a very small minority, they educated computer game player. They aren't concerned with satisfying your mental accumen when it's much more profitable to satisfy the throngs of less educated that just want to play a game. Please don't take it the wrong way, I like discussions, but your trying to make a silk purse out of a sow's ear. It's a trite little game. A game where longbows only shoot arrows about 6 feet away from them and guys with bronze tipped spears can shove them through 3 inches of steel plate. It's great fun, but there isn't any historical accuracy. Historical flavor is what they were going for. What if you could found a Roman Empire that never ends? That sort of thing. Not if a Roman Legion from the height of the empire were to face off with a regement of union riflemen c. 1862 who would win.

Does this make the combat system FUBAR? It's a matter of opinion. And like all opinions, it depends on the frame of refrence of the opinion holder. If historical accuracy is your frame of refrence, then yes would be the answer. If fun computer games where little guys poke each other with spears until one falls over, then maybe not.

 

[timerover51]

I would agree with the comment about the designers have a very poor knowledge of history, and especially military history. I also suspect that they never played a miniature war game using actual dice. A computer game combat system is essentially a computerized version of a miniature war game with the computer supplying the opponent, and the RNG throwing the dice. Given the enormous increase in computer power since Civ1 came out, it is very surprising that the combat system has not been improved.

Also, Marsden does have a point with his comment about "educated computer player." My problem is that I am a military and naval historian, and miniature and board war game designer, and also an expert on weapons, weapons effects, and the interplay between weapons and protection and terrain. Not exactly your normal gamer. So, I will complain about the RNG and combat system, but now having an editor, I will also be adjusting unit combat values, and tweaking wherever I can to get a game that I enjoy playing given my somewhat unusual background. To me, that is part of the challenge of the game, and also an indication that the designers did do a good job with much of the other aspects of the game so that I keep playing it.

 

[Pyrrhos]

I would agree with the comment about the designers have a very poor knowledge of history, and especially military history. I also suspect that they never played a miniature war game using actual dice. A computer game combat system is essentially a computerized version of a miniature war game with the computer supplying the opponent, and the RNG throwing the dice. Given the enormous increase in computer power since Civ1 came out, it is very surprising that the combat system has not been improved.

Also, Marsden does have a point with his comment about "educated computer player." My problem is that I am a military and naval historian, and miniature and board war game designer, and also an expert on weapons, weapons effects, and the interplay between weapons and protection and terrain. Not exactly your normal gamer. So, I will complain about the RNG and combat system, but now having an editor, I will also be adjusting unit combat values, and tweaking wherever I can to get a game that I enjoy playing given my somewhat unusual background. To me, that is part of the challenge of the game, and also an indication that the designers did do a good job with much of the other aspects of the game so that I keep playing it.

Amen to that!

PS. Good post, Marsden!

 

[anaxagoras]

I would agree with the comment about the designers have a very poor knowledge of history, and especially military history.

I completely disagree with this. Please remember that "knowledge of" and "adherence to" are not at all the same things. In the numerous interviews he has done about this game, Sid has always made very clear what his own design philosophy was: The game is informed by history, but where history and playability conflict, playbility wins every time. The decision to make the dreaded "spear" result possible is a direct, intentional consequence of that philosophy.

Just because it is not the design decision you would have made, that does equate to lack of knowledge on the part of the designer.

I also suspect that they never played a miniature war game using actual dice.

This is demonstrably incorrect. Sid was a dice gamer before he became a computer game designer. Jeff Briggs, who is considered by some to be the lead designer of CivIII, was a game editor and designer for West End Games. The designers wanted a simplistic and transparent combat system primarily because it is not solely a combat game. An updated, accurate combat system requires the user to be aware of the fundamentals of actual combat to figure out what kinds of attacks and defenses will work and what will not. Making a system that accurately portrays every possible combination of combatants from the bronze age to the present is a daunting task for the programmer, but using such a system is a daunting task for the player, as well. The designers did not want average players to have to work that hard.

My problem is that I am a military and naval historian, and miniature and board war game designer, and also an expert on weapons, weapons effects, and the interplay between weapons and protection and terrain. Not exactly your normal gamer.

Exactly, and I am sure there are pure combat games out there that will appeal more to your expertise. Try battlefront.com for a couple you might like, if you haven't already.

 

[anaxagoras]

The rest of the game is based upon a very poor understanding of historical events and processes.

Feel free to explain precisely what you mean.