New Combat System?

[weltraeumer]

I don't understand the "one unit type per hex" - given the scale of civilization. A traditional war game would have one unit per hex but its scale would be a much smaller area. Also, given that scale, historical armies (of all eras) did include mixed arms units. Even today, you have attack helos moving with ground units - in the same space.

look at it like both having the combat map and the tactical map in one thing... an ancient army of course had various troops mixed, but deployed into a formation... the spearmen next to another spearman, behind them the archers, the horses are flanking... nothing else in civ... (beside different terrain bonuses and yes you got to shut down your geographical sense for distance) It´s awesome.

 

[Ahriman]

IIRC they've already implied that a single battle will not completely destroy a full health unit.

I'm just struggling to relate that to the quoted paragraph.....

 

[KrikkitTwo]

My guess is that his reasoning is that Firaxis is likely to implement Panzer General's "more than one battle to totally kill units" rule, which does seem likely. Trouble is, if they don't use that rule, then using Civ IV's combat system in a 1 UPT Civ game would mean that the AI's occasional moon logic results (3+ losses at >95%) could mean your entire civilization could collapse. Well, maybe not, but a few heinously bad rolls -which happen all the time in this game- could cause you to lose a war that you had no right to lose, solely based on some longbowman taking out three divisions of musketeers or whatever.

Who says combat is random.... If a 5 Strength unit attacks a 20 strength unit.... perhaps they just both go down by 3 Strength to make a Str 2 and Str 17. No random involved... make sure you attack when you have the advantae...and that you don't put yourself in a bad position.

At best its probably semi random
ie
Str 5 v. Str 20.... the Str 5 is only going to do 20 average hp worth of damage to the Str 20 5/(20+5)
but it can do + or -10 hp randomly

the Str 20 will do 80 + or -10 hp to the Str 5... 20/(20+5)

So the chance of either one killing the other in one attack is exactly 0... ie randomness is there but limited
The chance of the Str 5 killing the Str 20 over multiple attacks (no healing) is exactly 0... at best it does 30 hp, and the Str 20 does 70 hp at worst.

So the str 20 will win... the issue is if it does this with 80 hp remaining (it only loses 10 hp per battle) or 40 (it loses 30 hp per battle)

 

[nmcul]

Who says combat is random.... If a 5 Strength unit attacks a 20 strength unit.... perhaps they just both go down by 3 Strength to make a Str 2 and Str 17. No random involved... make sure you attack when you have the advantae...and that you don't put yourself in a bad position.

At best its probably semi random
ie
Str 5 v. Str 20.... the Str 5 is only going to do 20 average hp worth of damage to the Str 20 5/(20+5)
but it can do + or -10 hp randomly

the Str 20 will do 80 + or -10 hp to the Str 5... 20/(20+5)

So the chance of either one killing the other in one attack is exactly 0... ie randomness is there but limited
The chance of the Str 5 killing the Str 20 over multiple attacks (no healing) is exactly 0... at best it does 30 hp, and the Str 20 does 70 hp at worst.

So the str 20 will win... the issue is if it does this with 80 hp remaining (it only loses 10 hp per battle) or 40 (it loses 30 hp per battle)

Well, yes, they could make a system like that. But it would be wrong! Warfare, particularly at the level of abstraction used in Civ, is not a game of mathematical (or any other sort of) certainties. A battle cannot be said to uniformly affect attackers and defenders in the same way. The only reason why games such as Panzer General or Advance Wars can get away with such models is because they have a tactical breakdown operating over very short increments of time and with outside factors taken into consideration. If your model breaks time down into so much as a whole day between engagements, much less the kind of time gaps used in Civ games, there are too many variables to attempt to justify lack of random results in combat.

I'll put it to you another way. The Spanish Armada was defeated due to a communications error coupled with some bad weather. It rained the night before Waterloo. If it hadn't, it's quite possible that the French would have carried the day. A Confederate officer lost a cigar box with Robert E. Lee's plans for the Antietam campaign inside, plans that were then found by Union soldiers. Stormy conditions made German advances in the Battle of the Bulge operational-and the weather lifting eventually sealed the fate of the operation. And, while these emphasize situations in which whole armies were affected, it's not hard to understand how smaller, more limited factors could swing a single engagement one way or another.

In war, there are no "sure things," and that's certainly true of a battle of the scale that unit-vs-unit combat will apparently represent. We need randomness.

 

[KrikkitTwo]

Well, yes, they could make a system like that. But it would be wrong! Warfare, particularly at the level of abstraction used in Civ, is not a game of mathematical (or any other sort of) certainties. A battle cannot be said to uniformly affect attackers and defenders in the same way. The only reason why games such as Panzer General or Advance Wars can get away with such models is because they have a tactical breakdown operating over very short increments of time and with outside factors taken into consideration. If your model breaks time down into so much as a whole day between engagements, much less the kind of time gaps used in Civ games, there are too many variables to attempt to justify lack of random results in combat.

I'll put it to you another way. The Spanish Armada was defeated due to a communications error coupled with some bad weather. It rained the night before Waterloo. If it hadn't, it's quite possible that the French would have carried the day. A Confederate officer lost a cigar box with Robert E. Lee's plans for the Antietam campaign inside, plans that were then found by Union soldiers. Stormy conditions made German advances in the Battle of the Bulge operational-and the weather lifting eventually sealed the fate of the operation. And, while these emphasize situations in which whole armies were affected, it's not hard to understand how smaller, more limited factors could swing a single engagement one way or another.

In war, there are no "sure things," and that's certainly true of a battle of the scale that unit-vs-unit combat will apparently represent. We need randomness.

If there is PURE randomness, then Spearmen WILL defeat tanks... for that matter a damaged spearmen will defeat a Full strength tank. (one time in a trillion)
They indicated that will not happen, indicating semirandomness

I'm saying there should be and speculating that there is semi-randomness.

ie Str 6 v. Str 4

The Str 6 unit does 60 hp of damage... lets make it +- 20 (so either 40 or 80)
The Str 4 unit does 40 hp of damage... same +-20 (so either 20 or 60)

so in 2 rounds, the Str 6 unit could be
dead, (25%)
at 20 hp (50%)
at 60 hp (25%)

the Str 4 could be
dead (75%)
ay 20 hp (25%)

result is still random, even to the degree of winner and loser (or even if there is no loser..ie no unit dies) That is because those results are close together.

But this would not happen with a Str 4 and a Str 16 unit (the Str 16 would to 80 hp..ie 60-100 so it would kill the Str 4 in 2 attacks guaranteed, and the str 4 would do 0-40 hp /attack.. so even if the worst happened, the Str 16 would survive... just badly damaged at worst 20 hp..but possibly none.)

War would still be random, but to a limited degree.. a unit would be At least so good, and at most so good.. randomness would be within a range.

In the previous civ engines, it was theoretically possible for any unit to defeat any other one (assuming they could attack each other).

Now you can change the degree of randomness, and they might do that, but I don't think there will be many 1% chance of winning type situations (unless you are talking about 1 army v. another)

 

[Chris Stalis]

One of the principal purposes of one-unit-per-tile is that when you attack a tile, there is only one unit to defend that tile. As soon as you go to more than one-unit-per-tile you introduce an ambiguity: which unit gets to defend?

Traditionally, the game has chosen the best possible defender in the stack to defend. Why is this bad? Because it eliminates all tactical planning. No matter what you do, the game will choose the best defender to counter it.

I provided a more detailed explanation of the value of single unit combat in this specific post of this thread. I'd appreciate additional comments on the topic there

Who says combat is random.... If a 5 Strength unit attacks a 20 strength unit.... perhaps they just both go down by 3 Strength to make a Str 2 and Str 17. No random involved... make sure you attack when you have the advantae...and that you don't put yourself in a bad position.

Sorry to be cheeky, but I want to play "Sid Meier's Civilization 5", not "Sid Meier Chess". I strongly suspect the developers wouldn't go in that direction.

 

[Colonel Kraken]

If the developers move in the direction of Panzer General style combat mechanics, it would certainly make the combat system much more interesting than its current implementation. I am particularly intrigued by (which has been hinted at) the prospect of "developing" your units in the same manner one did in the Panzer General series (of which I am a great fan).

In regards to archers shooting over the heads of spearman in a hex representing some 50 miles (or whatever), I think implementing the map as if it were both strategic and tactical at the same time is quite ingenious. Why not have the map represent what happens on a grand scale (cities, economics, resources, etc.) and represent the tactical/grand tactical/(and strategic choke points) considerations of prosecuting a battle/war?

On the other hand, the strategic map versus tactical map system of the Total War series is, in the words of our great friend Tigger, “splendiferous”.

Hmm . . . I always have thought that mixing Civ with Total War would be very cool . . .

 

[KrikkitTwo]

Sorry to be cheeky, but I want to play "Sid Meier's Civilization 5", not "Sid Meier Chess". I strongly suspect the developers wouldn't go in that direction.

Wll they already did with 1 UPT and more tactical combat. I don't think they will totally eliminate randomness, but I wouldn't be surprised if they limited it.

 

[Buccaneer]

I think that is what I was getting at: a strategic and a tactical map. Total war of course did that brilliantly (at least the early versions) but with a timeline of civ, that could really slow down the game. Alternatively, in my opinion, the best turn-based implementation of combat was Conquest of the New World. I wouldn't mind that for Civ5 since each battle does not take that much time.

In any event, this is certainly going to be a culture shock after 20 years of Civ1-Civ4.

 

[Willem]

Also, please don't give any speculation/rumors. I just want the facts.

There aren't that many facts to go by right now, just some screen shots and a couple of reviews. And with so much time to go before release, alot of things can change. So you'll have to make do with speculation for the most part.

 

[Willem]

On the other hand, the strategic map versus tactical map system of the Total War series is, in the words of our great friend Tigger, “splendiferous”.

Yeah, I was kind of hoping they'd go in this direction more. While I like the idea of the one unit per tile and ranged attacks on the whole, I'm not crazy about having it take place on the world map. It's just too much of a stretch of the imagination to presume that an Archer can shoot over a tile that might represent 100 square miles. Or hex miles in this case. Also, with a strategic map you could have various terrain elements that need to taken in consideration, like cliffs, gullies or maybe stream or pond etc.

 

[Desmond Hawkins]

It's called GAMEPLAY!

I was being snotty about the range thing, but making the game tactical doesn't necessarily improve gameplay. The way Civ abstracted things before was actually quite elegant... I don't really trust them to beat it.

I do respect you for your work in Colonization, so if you think the gameplay might work... well that eases my mind a little.

Then don't buy the game.

What kind of a comment is that? Isn't one of the advantages of these sorts of sites to give feedback as to what fans like and don't like? Besides, I have consistently pumped my money into the game throughout the franchise, and will give my opinion if I want.

Its here to stay for Civ 5 if you don't like it then maybe it will be different for Civ 6 but don't count on it. They wanted to get rid of the damn SoDs forever.

Be careful what you wish for.