I think the bigger problem is with hull sizes myself. As far as I can tell the combat system is very well balanced in the "everyone uses medium fighters" level. That is one reason I want slower and slower research level options, I want to force everyone to this level of warfare. A medium fighter can't be made invincible from defense, but they also don't pack enough firepower vs. hp ratio so that defense becomes meaningless. But Iztok I know you and I have discussed whether increasing the logistics needed for a huge to 12 might help that problem.

The problem is the additive nature of the combat system. 100 defense is not twice as good as 50 defense in many cases, it can be almost infinitely better. When attack values are in the 70-200 range (so that 1/2 attack power is still not greater than hp on one huge ship) even I think the combat system works well. But, you can easily pack 6x the hp of a ship in offense. That means individual suicide ships. You will die, but you can take down many times your number. Then the question becomes cost. Stardock responded by lowering the cost of defense... this was a good decision.

Logistics does still play a role in the whole rock/paper/scissors of combat. I'm sorry if some of you have not seen that. Here I posted a quick description of a game I played a while back. Note, I was able to field huge hulls with 800+ defense myself. These kinds of ships pretty much stop the individual suicide strategy. I rarely see the AI use this tactic however. To beat the individual all defense ships take a fleet (well, two) all offense huge hulls. In a long protracted war against an AI where you both have the tech tree matched there is a lot of fleet micromanagement needed to get the most from your ships. In some situations you need suicide all attack ships. In some situations you want fleets of 4-5 defense heavy ships.

So against fleets of defensive ships, you need to use fleets of offensive ships. Against fleets of offensive ships, you need to use individual suicide ships. Against individual suicide ships you need to use fleets of defensive ships. Rock. Paper. Scissors.

It rarely gets that complicated because the AI likes really only one of the above three play styles - fleets of offensive ships. So, endgame the counter is suicide all attack ships. This is (and I hate to say it) really less a weakness of combat balance and more a weakness of AI strategy.

Also, I find it interesting that people support mutual destruction but that mutual survival never gets more attention. The few times the AI does field the all huge defense ships you can take them out not be switching to all offense, but simply by building all defense ships that carry 1 more attack point than the AI fleet. You win without taking damage... or DOING damage.


But what exactly would you have fixed? Stardock has already done plenty of tweaking of the combat system since DA was released. I admit I haven't played under the latest patch, but it seems to me they have struck about the best balance they can at this point. Exploits at this point revolve more around how the AI fields fleets (or fails to use fleets in planet defense). But the AI can make vastly superior ships to yours at the end of the tech tree on the hardest levels. Without these exploits, how would you beat them?

Just my opinion.
~ Wyndstar

I could absolutely go for that -- in fact, it makes an awful lot of sense (both in earth-based naval battles and theoretical sci-fi battles... the whole "loser gets utterly wiped out" thing just doesn't happen that often).

The problem would be implementation --- does it make battles 'tactical' rather than just watch and enjoy? I.e., would I watch the first salvo knock out 1/3 of my fleet and decide to flee on the spot? Would it be random? Or would it be pre-programmed that a ship/fleet flees when a battle is obviously hopeless?

Well, you could just use the Civ-like system where attack is matched against defense, and if attack wins, it deals damage. If defense wins, no damage is dealt. I am really not sure at the moment how the combat system works in DA, but it just seems overly complicated. Any link explaining the exact system?

I don't particularly like the Civ system, I prefer having my combat be a little more dynamic. But I can support the whole mutual survival mechanic: perhaps each combat consists of only a set number of volleys (I suggest 5), and then instead of Victory or Defeat, we get a Draw result showing how badly each side was mauled? Only problem: I can see planetary defense ships being built with all defenses and no weapons to force the enemy to take weeks and weeks shooting them down before they can land transports: obvious solution is to have planetary defense fleets only work against transports if they actually have some kind of weapon on them. . . or even allowing a transport that survives that five-round combat drop its troops. i can see players building huge-hulled troop transports with massive armor and barrelling through enemy defense fleets now. . . hmmmm, more contemplation is needed.

Hi!



Indeed. Since in late game defenses are usefull only in very specific situations, and I have yet to see individual suicidal ships ot he AI, I'm affraid RPS is only for human player, and there's only Rock for AIs.

There aro only two solutions to this problem: improve AI to recognize what it has to use in any given situation, or change the combat model to be more like the old one, for which the AI was originally designed, and that worked significantly more intuitively, without such peculiarities like bigger fleets being worse, ultimate defenses not usefull, single ship beating whole fleet and surviving...

At the end it's Stardock's decision, but if we provide them good points from where to work on, we all could benefit.

BR, Iztok

Agreed. The combat model needs to be more intuitive and make sense in general.

As I recall, the complaint with the original system was that fleets of tiny fighters were annihilating the big dreadnaughts without a hitch, which had people wondering why build the giant ships when they're just gonna die anyway, and it's a tossup between losing a couple of 150 bc fighters or your uber-expensive Huge?

I'm not sure what tack to take, but I know what I'd like the end result to be: an endgame where a good spread of ship types is vital to the war effort. I don't just want to walk all over the enemy with a couple of big ships, or a couple of fleets of tiny fighters, or send in huges one at a time against fleets because it's more efficient that way. I want to have to construct a fleet of a Huge with several Large and Medium Escorts to guard against the packs of Small and Tiny fighters that could otherwise wipe the big cap ships out. I want to create roving packs of escorts hunting down the fighter wings that threaten the big ships, rather like the destroyers of WWII hunting down submarines, or guarding carriers against battleships and cruisers. In short, I want an RPS relationship like:

Huge kills Medium

Medium kills Tiny

Tiny kills Huge

Small and Large are intermediate forms: Large gets hurt but can still take on Huges, and tends to win against Medium and Small, but still gets smacked up by Tiny. Small is almost as good against Large and Huge, can't really get up the numbers to take on Medium (but can try) and wins out in the long run against packs of Tiny.

But that's just me, your mileage may vary.

Hi!

In my games that almost never happened, because I used well-defended ships with proper defenses for my current enemy. My gripe with DL was the loss of a superiorily-defended super-expensive huge ship losing to a bunch of all-attack mediums, because my ship rolled some bad defensive rolls. At the end I adapted to using big fleets of properly-defended ships, in order to destroy opponent's ships ASAP to lower the damage to my ships.


We have 3 weapon and 3 defense types instead. But I too would like to see some more important role for smaller hulls, even after bigger hulls are common.

BR, Iztok

I'd like to see weaponry designed to take out specific hull sizes. Weapon effectiveness would change depending on the size of the target and the size-rating of the weapon.

10% for Size +/- 3
25% for Size +/- 2
50% for Size +/- 1
100% for Same Size

So if you put a Size Large Fruit Salad Shooter on a ship it would hit Large hulled ships for 100% damage but Medium and Huge for only 50%, small for 25% and tiny for 10%. This would prompt you to arm ships specifically for certain sized targets. Huge hulled ships would then do well to put an array of weapon sizes on board to deal with various target sizes. Smaller ships wouldn't have the space available so they'd want to specialize more.

Oh, I absolutely agree. I almost made this observation in my first post. I think its because the AI uses way points, and then auto fleets and auto attacks. Not a lot of strategy level code needs to be written to do this... but it is a brute force blunt approach with serious weaknesses. Fixing it... well, fixing it is almost certainly a problem for the next game.


Here here. I'm on board with you here as well. The reason I spoke up is because at this point the combat IS balanced as a stand alone system. Making changes to it to help the AI will have the inevitable effect of unbalancing the system. Which, in the long run will not help the AI, it will just give the player a new set of "optimal" ways to approach combat.


And this still works very much in DA, on the medium fighter level. See, at the Huge level there are only 5 ships you are talking about, every combination is easy to figure out. Combat is very all or nothing. At the medium fighter level, attrition is a fact of life (unlike with huge hulls) BUT you can significantly reduce attrition with matching defense. RPS of weapon types plays out perfectly. More logistics = more fighters = higher attack power = even less attrition. If there was only one size of ship (medium) this combat system is... spectacular. There are trade offs in different weapon and defense types, only so much space on each ship, and research strategy actually effects overall wars. Sadly...

The medium fighter level is almost never hit. The AI sticks with small and tiny hulls for too long, and when it DOES decide to go up in level it goes all the way to the top. Huge hulls are incredible against fleets of smaller craft, so the player is also motivated to move past the med fighter level and go straight to the capital ships.

Personally, and I know this will sound like crazy talk, but the solution I would like is only 1 capital ship per fleet - an additional restriction apart from their logistics cost. In this system, individual suicide capital ships lose their lustor, they are not taking out 3 enemy cap ships (and giving the player an attrition advantage). All defense huge hull ships are not going to kick out much damage, and will still lose to (large) fleets of enemy fighters. All offense huge hulls are going to knock each other out and be targeted first.

I know, I know, adding restrictions to the game typically just limits play styles. But I have thought about this and I think that would solve most of the current endgame problems. Tiny ships are still situational at best. And I'm not exactly sure how to balance large hulls (you don't want to turn the game into a bunch of fleets with 1 huge and all larges, that will lead to the same problems and further erosion of effective defense). But if you limit the player too much all fleets will look the same.

On the other hand, already all fleets look the same. 5 huge hulls. The thing I like the MOST about this is it is not only the best combat option, it comes with the least micromanagement. But obviously it has its weaknesses, chief amoung them being that the AI does not optimize its play for the current system.

As to your suggested solutions:

The better solution, but highly unlikely given the amount this would entail for Stardock. And lets be honest, 1.7 sounds great, but they DO have a new game coming out, and new projects they want to be spending their time on. I've been a programmer, I know how it is. At some point, you need to let old projects lie, warts and all.


Sadly, I don't really agree with you here. I liked the old model. But I like the dependability of the new model. The new system really enhances certain tactical and ship designing decisions in a way the old one did not. I like my capital ship knocking out several fighters in a round. I like that large fleets with weak weapons can wear away and do damage to moderately armored enemies (like you see at the med fighter level). The problem with the new system is also one of its coolest features. Huge hulls. And powerful capital ships are sweet. Problem is, they are all that exists.

Again, just my opinion.
~ Wyndstar

I like your idea of causing a penalty to damage the larger the attacker is than the defender: it gives Medium ships a role: big enough to take on fighters, but small enough to actually hit them. Also, how about adapting the "Weapon Types" idea from the Tactical Combat thread?

Sniper: Accuracy penalty reduced by two steps, takes up 3x space for the targeting system and turrets.

Accurate: Accuracy penalty reduced by one step, takes up 2x space for targeting system and turrets

Standard: Normal weapons, damage penalties as normal, least cost.

Heavy: Causes more damage, but incurs the same size penalties as a weapon one step larger than the ship, takes up 1.5x space

Superheavy: Causes even more damage, treated as a weapon two steps larger, takes up 2x space.

Medium ships would be the only ones who could mount all five weapon types. Tiny could mount standard, heavy, superheavy. Small could mount accurate, standard, heavy, superheavy. Large could mount sniper, accurate, standard, heavy. Huge could mount accurate, sniper, standard.

I don't expect to see any of this in GC 2, by the way, but it's fun to speculate.

Don't talk about the civ combat system! hehehehe it is so funny to watch a knight manage to defeat a tank with it's little sword waving around.

Obviously it is not possible for a person with a sword to defeat a tank so yea, the combat system in civ3/4 is basically just a funny joke as far as i am concearned.

hahaha

That is about as funny as watching news reports of terrorists training soldiers. There they are holding a gun and crawling under barbed wire etc etc, when what they should be teaching them is how to use radar and then run realy realy fast when they detect planes coming!

I have always wondered about objections like this to the visual conventions used in otherwise relatively abstract, strategic or grand-strategic games. After all, these visual conventions were forced upon game designers because fancy specific 3-D iconographic representations seemed to be necessary to attract the attention of buyers who couldn't quite grasp more flexible symbology.

Recall that low-tech forces have always been able to defeat hi-tech forces, in the right circumstances or in great numbers or sometimes just because they were very, very good compared to the troop quality they were fighting (Apaches come to mind.)

Would the objectors be happier if disparate technology battles in their strategy games were represented by an X in a box (with Tech II / Qual A printed next to it) defeating an Oval in a box (with Tech V / Qual F printed next to it)?

I actually wouldn't mind much; I think I can deal with the differing abstractions required in either case, but maybe that is the grognard gamer in me.
(For those of you who don't understand "grognard" in this context, give it up - you aren't going to understand the whole premise of the post anyway.)

drrider

It has nothing to do with Graphics, but just the simple realities of combat. Obviously if you want to defeat a tank in combat, you need some kind of weapon to penetrate its armour, and a sword is definately not one of them! Such units shouuld be unable to inflict damage on tanks in civ games, just so it is a bit more realistic.

I don't think you are following me. The unit represented by the little figures with swords is just that, a representation of a unit that was originally formed with basically older technology...in a very abstract and strategic game. That picture does not even represent the equipment of every one of the thousands of soldiers in the unit, just a general notion of what the preponderance of the troops in a similar weight unit might have carried at a comparable point in real-Earth military evolution.

Is it inconcievable that as armored vehicles began to be employed by rival nations, our heroes developed a light infantry doctrine (over the months/years represented by a turn) to deal with that to some degree? Maybe some AT mines in each sub-element? Some IED training, ala the Iraqi insurgent forces? They possibly have some anti-tank guns that are integrated into the light infantry? Perhaps some units got very good at it? All of that fits easily into the scope of a GRAND STRATEGIC scale game.

They are NOT necessarily pounding on the turret of a tank with swords just because the little picture of a guy with a sword was chosen to indicate their original technology level. They represent a military unit; they are combating their opposition with whatever the best means is they can come up with, even if they do not have a preponderance of mechanization or long range weapons themselves.

Don't get hung up on the icons; that's not what the game is about.

drrider

Hi!

Ermmm, the icon on the screen is the only thing that diferentiates that unit from some other one, without checking its stats. Following your argumentation would mean all chess pieces could look equal. Ever played Stratego?

If the game is an open-information game, the icon on the screen simply must represent unit's correct value. When unit is upgraded, its icon also needs to change, to represent its new value, at least after I had a new scan/look on it. So if swordsmen now wield bazookas, I want their icon change to anti-tank infantry.

BR, iztok

Which is exactly the way it works in Civ3+4 .
Sorry to correct you there drrider, as the rest is clearly very well thought out and written.
In Civ generally speaking the icon updates with new skills.
So you build a Spearman in Ancient times. As soon as this unit gets access to the longer and pointer Pikes you can upgrade it to Pikemen, thus also changing the stats in the background.
Of course, its only an abstraction, but as soon as a Spearman gets access to equipment meant to combat tanks like AA mines, its no longer called a Spearman by definition. It would be called a Guerilla or, in game, a Rifleman, Musketman or Infantry.

Basically, if an ancient unit gets new toys or learns new skills like modern warfare, its "stats" change, as it gets better and more knowledged in the way of killing.
You wouldnt call an Amazonian Indio a Spearman when he wields mines and guns either.


While I agree on the fact that an Ancient unit CAN defeat a modern one with alot of luck and tactics ( digging holes and camouflaging them comes to mind, e.g.) in Civ it happens all the time. You wouldnt want to know how many tanks of totally superior firepower I lost to Spearmen defending some lowlife city.

Its been one of the major points of frustration in the civ line of games since the game started and is getting even more pronounced with the new system of XP awards in Civ3 and 4.
So I wholeheartedly agree with Mystikmind there.

An even worse parallel...I have lost undamaged Stealth Bombers and Battleships to Phlanax units. Given that they haven't been upgraded to a higher technology with some form of ranged weapon like musket men (who I still will not buy for stealth bombers...I want to see a flying unit of something that might remotely have AA) and even assuming an "elite" version of a Phlanax this is a completely ridiculous situation.

Most game combat systems baffle me, especially civ's, in this regard. Yes, some of a tank's/starship's/etc defense is luck-based (e.g. weak points, random system fluctuations) but a lot of it is guaranteed. Spearman defeating a tank is nonsensical because the guaranteed part of a tank's defense nullifies any possible spearman attack. This is why I prefer combat systems that include some form of flat subtraction due to defense. Spearman rolls 0-5 attack, tank rolls 20-24 defense, the end. I really liked the MoO combat system in this regard.

I understand that there has to be a quality/quantity tradeoff, but it really should be the attacker that introduces this variability. If it's important for units to not go obsolete, just increase experience bonuses to offset technology differences. Why not allow the attacking unit to have a small chance, given enough experience, to perform a "brilliant maneuver" that greatly increases its damage for the duration of the combat? That makes a lot more sense to me than having a defending unit's ability to defend change drastically from round to round.

In a strategic melieu, with turns representing years or months, no TACTICAL technique/outcome is guaranteed; indeed discussion of whether the accepted battle field tactics of the lower tech unit will stand up to the higher tech unit on a conventional battlefield are meaningless. At the micro level, way below what the game is simulating, it is only going to happen maybe one or two times; then the lower tech force is going to find other means to resist/attack. Witness current Iraq; witness Finland. No tank is immune to a spearman if the night before the tankers plan to fight, 50 spearman martyrs sneak into their camp, smash the crews heads with rocks, and set the fuel on fire. But the game isn't simulating that, either, it is simply simulating that, over the course of a very long turn, there are options for lighter forces vs heavier forces.

drrider

Your point is a good one. However, this notion of alternative battle tactics must be kept within its scale. I could accept a surprise attack on an unprepared foe causing an atypical outcome, but this should happen only very rarely within the conflict being modelled. A dent is made in the tank batallion, causing a loss of 1 HP out of 100, but total destruction? Spearman special ops destroying thousands upon thousands of tanks, all of the watchmen in the superior army asleep at their posts? This would indicate a systematic weakness that should certainly be modelled at the strategic level.

I agree that the archetypical battle would only play out a few times. However, I think it's much more likely that the underdog would completely surrender/disband rather than develop some brilliant resistance strategy. Guerilla warfare is all well and good when you're an insurgency defending against foreign control; it doesn't work out so well when you're in the field against a force willing to accept your complete destruction. I feel that modern real-world examples are more about politics and diplomacy than about the ability for the inferior to defeat the superior.

I suggest looking at how well Polish horse cavalry did against German tanks before World War II before we start talking about asymmetrical warfare.

Alot of good posts i see, seems i am not the only one to find the civ3/4 combat system to be a joke.

In my current game of civ3 i am finding that i will need twice as many tanks to take a city as what the AI needs to retake a city? There is a consistant and obvious inbalance in the results the AI gets when attacking compared to the results i get when attacking.

Just as an example, say i attack an AI city with 5 modern armour inside.... i will typically loose 2 or 3 modern armour for each one of theirs. However, if they attack my city with 5 modern armour defending, then the AI will most definately only need 6 modern armour to take the city.

Am i just consistently unlucky or has anyone else noticed this combat imballance?