Combat systems / interactivity

[Nox]

I hope 'm not too spammy lately
've been thinking a LONG time about combat systems in terms of interactivity
finally coming with this list

Instant / non-interactive

+ easiest implementation
+ instant result for player
+ instant feedback for player
+ lowest performance cost

- very common
- lesser opportunity for tactics
- shallow?


Semi-interactive (one-sided)

+ instant feedback
+ tactics
+ not as common?
+ low performance cost

- bit more difficult implementation
- player/unit locking
(defender can't perform game actions with units engaged in combat...possibly all -> defender can't play)
- advantage for attacker, defender only pre-defined actions
- impossible for 3+ players in combat



Fully interactive (both sides)

+ tactic opportunity for both sides
+ not common/more interesting ?

- forced immediate participation in game OR no attacking offline players
- more difficult implementation
- more difficult design, impossible? to set ideal turn limit
- possibly very long combats
- unpredictably long gaps
- higher performance cost

Fully-interactive have far too many very grave ssues and I've come to conclusion is just impossible to create gamedesign-wise

So far the semi-interactive seemed tempting, but those issues: locking, no 3+ player combat, (possibly) disadvantaged defender
are quite heavy

Are the underlined issues possible to remove/suppress? I fear it's impossibe...

(3rd issue probably depends on the specific combat system, so we probably can't discuss that ...possibly special actions/passive bonus for defender/increased recovery)

[Doidel]

Hey Nox

Your post involves too much thinking. Nonetheless here are my two cents

You managed to somehow categorize battle systems. For whatever reason... Since your post lacks examples please correct me if I get it wrong:
1)
I guess the first one could be something like the "Gladiators", "Monsters" or "Legend" battle system. You have your hero on the left side, opponent is on the right. If advanced you see some text and a little animation added, and all systems end with a "You won" or "You lost" in text form.
2)
This would be somehow an automatic "Ogame" battle system? I.e. your ships on tour can't defend. And if you're attacked, a KI does basic stuff for your protection..
3)
This advanced battle system has either a more advanced KI or needs both players to be online. Right?
I don't know what you mean by "impossible? to set ideal turn limit" - can't you just define a maximum time span and if the player won't move he loses?
possibly very long combats.. well, depends heavily on the system I'd say.
Why would there be long gaps? You just wait for the enemy move or until the max move time is exceeded.
Higher performance cost again depends on the system. Usually one request per attack.. I guess that's not too much of a load..

In order to judge your statement that fully interactive systems would be impossible to create I'd need an example.. it's quite hard to imagine yet what you see as impossible.

I don't know what you mean by "locking", however I can imagine 3+ players quite well. Either they ally and then it's 1v1 again or they all clash on the same battlefield, like just randomly shooting enemies. All vs all

Conclusion: Need input!

Sorry for my not that well styled and not that greatly elaborate answer

All the best
Doidel

[Nox]

Hi, I think it's well styled, but...ok, explaining

Reason is - when we want to choose or think about a battle system/solve some problem, it's good to have precise idea what actually we're dealing with

1) I'm not sure what you mean, I mean this:

Defender has some predefined rules, attacker set attack options, click once and the *whole* combat is *immediately* calculated and result displayed.
Ogame has this too, you don't assign actions etc. in mid-combat.

Actually probably most of BBGs has this system.

2) Attacking player chooses actions during combat, defender's actions are chosen by script (for example based on predefined rules and strategy)

Problems
a) Attacker as (probably) intelligent being should be able to choose actions better than script, so this might be advantage
b) Locked defender

Problem:
what if I attack some player and fight him and in the meantime he attacks someone, ending with 3/15 units and 30%HP,
while I am actually in the middle of fighting him at that time and we both still have 14/15 units and 80%HP, then combat ends and what?
He actually should have 3/15+30%HP and 14/15+80% HP in the same time...

Or what if I kill him before he managed to finish his fight?

One obvious (yet problematic) solution: locking = player can't do any action during the time he's attack that might affect something related - move from area, release troops, attack someone...etc.

This is similar to Database/resource locking, atomicity etc. in programming

c) It's actually not 3+ players, but 2+ attackers, my bad. The system then would have to be the same as (3) or there will be only 1 out of X interactive players, which would be weird

3)
KI? you mean AI? yea, right

I can... but the time will never be optimal, it will either be too short so player would have to sit there all the time (~3 minutes are too short to do anything else in the meantime) or too long and every fight would take forever

There might an optimisation when you make your turn, then the turn is automatically given to opponent, but the longer limit, the more unpredictable therefore unusable this is

Higher perfomance cost is rather related to the sentence before, that you need to monitor for change
but you're right, it's not that bad


What turned out on IRC:

Harkins mentioned that I don't have to care about the problem with concurrency, meaning I would have to discard all the consequences of combat on defender
...which is rather the opposite what I was intending, so I kinda have to think about this

So what I am considering now is (2) with some modification of what Harkins mentioned, devise some change to locking approach or just locking with giving player something to do in the meantime

[Doidel]

Hehey

Now I understand
Hmm it's a matter of concept, isn't it? Of course you can't force a player to be attentive to attacks anytime and he has to respond within 20 seconds. I'd say the maximum time span for movement should only be implemented if the defender as well as the attacker have expressed their will to be attentive (like in a duel. One challenges, the other accepts).

An approach for (2) could be an attack and a defense squad. Every player has both of them. Or a duelling system. Or with locking, as you said.
Why does the attacker have to be better than the script? It heavily depends what information you can get and how much effort you put in it. In earlier game's "hard mode" opponents often had tons of information the player didn't have and could therefore calculate the most efficient way -> they were overpowered.

3) That's true, (3) can therefore only work with either both players being attentive or the seperate squads, as mentioned.

I dislike the idea of locks.. had to give my personal opinion in the end. How do you say? *scnr*, think that was it...

Best regards
Doidel

[Nox]

That duel agreement is good, I forgot that, thanks

Attack and defend squad.... originally there wes only hero and units were for spicing up, the hero/player is important so 'm not sure how that would work
(I should note that 'm building something a bit different then before, in previous ones there were battlegroups, but I lost that now)

I wrote "should", doesn't have to be, hopefully the result will be balanced AI

Well, I don't like locking players either, but... it's probably gonna have to be some compromise

Thanks for input

[Chris]

- I'm fan of instant combat for numerous reasons
- Fully interactive causes so much problems it is at the edge of being impossible to implement in any reasonable way (I won't even try to address the problems of this system, it is just too many of them).
- You can make OrderOfBattle (top level orders given in case of battle, combat executecd by AI), this way both sides have a feeling of interactivity without most problems
- You could make 3+ sides in semi interactive battle, you would need to check boardgames, these deal with such problems. The most common, I think, is to make separate battles between sides until only one is standing (I would also give "StrategicOrder" attribute and the player with lowest one is the last one to enter battle)
- Do you really need 3+ sides combat in the first place?
- Interactivity can be done by other elements than combat, so non interactive combat does not need to be as bad as it sounds
- Timing of attack in instant combat might give a lot of interactivity (hard to explain, my players call it "sniping" and some seem to like it a lot)

what if I attack some player and fight him and in the meantime he attacks someone, ending with 3/15 units and 30%HP,
while I am actually in the middle of fighting him at that time and we both still have 14/15 units and 80%HP, then combat ends and what?
He actually should have 3/15+30%HP and 14/15+80% HP in the same time...

Give him the lowest of the two.
When combat starts make a separate table entry with both sides forces, these are not affected by any external interferences. At the end of combat apply loses from the most pessimistic battle to the permanent table.

[Nox]

- I'm fan of instant combat for numerous reasons

Sure, there are advantages...on the other hand I'd try to do it different way than usual, but... I'll consider it

- Fully interactive causes so much problems it is at the edge of being impossible to implement in any reasonable way (I won't even try to address the problems of this system, it is just too many of them).

Agree, that's what I said

- You could make 3+ sides in semi interactive battle, you would need to check boardgames, these deal with such problems. The most common, I think, is to make separate battles between sides until only one is standing (I would also give "StrategicOrder" attribute and the player with lowest one is the last one to enter battle)

that's interesting! don't think I'd ever get such an idea (btw I agree that board and bb games are quite close)

- Do you really need 3+ sides combat in the first place?

That was rather theoretical, no, not that much

Give him the lowest of the two.
When combat starts make a separate table entry with both sides forces, these are not affected by any external interferences. At the end of combat apply loses from the most pessimistic battle to the permanent table.

Hmmm, that actually seems to be a nice solution, thanks