The Way We Play Multiplayer

[TheJopa]

I'm calling . .. .. .. .. .. .. .. . on that. I'll tell you why when I debunk your second paragraph.



Exactly. So, listen: no one wants to remove all randomness. I never said that, my mod goes nowhere near even trying that, and as you point out yourself, the very notion doesn't even make sense.

My mod removes unpredictability from combat results. Everything else is as in the normal game (as far as luck is concerned). Civ4 (and FFH even more so) is such an impossibly complex game that there's always a whole lot of randomness involved. Maps, as you pointed out, are random each time. But even if you played on the exact same map with the exact same starting positions, no two games would be alike. Every little decision you take differently opens up a whole new decision tree, meaning that no two games will even be remotely similar.

And if you take a moment to consider that, you'll see how beautifully it devalues your first argument. Combat randomness adds nearly zero replayability because in the great picture, the little variations that are added by losing a combat you should have won etc are negligible (as far as number of possible game situations goes). That is not to say they do not matter for the game you're in--they matter greatly. But what I'm saying is that they do NOT restrict the number of game situations you can experience and as such add no replayability.

I was not talking about your mod but general theory

In fact I like your combat idea VERY MUCH. I myself get puzzled after ridiculous combat results. My friends and I often talked about how it sucks and that best solution would be that battles over 90% odds should never be lost, even if lost attacker should win but have 1 hp left. So battles over 90% would be what they are basically in my mind - sure victories with only one question: How much damage will attacker suffer?

About your mod : I'll try it when I'll have more time (And hopefully you will fix first strikes somehow)
What I don't like is messing with mobility. Horses should start with more movement than infantry. Plus without railroads stuff will just move too slowly.

Could you implement (since I see you understand programming) battle system from Afterworld, where it takes few combats to kill unit (At least for heroes and T$ units?)

 

[xienwolf]

One question I have though: Does your combat system tend to result in a flat out win for the higher strength unit, or appropriate scale by strength.

ie - Will a Strength 3 unit beat a strength 4 unit 25% of the time? (assuming those strengths are post-modifiers of course). If someone ever works that one out I'll be in love since then your early scouts can still survive the occasional bear or spider, but on the whole you can predict what a battle should result in.

 

[TheJopa]

ie - Will a Strength 3 unit beat a strength 4 unit 25% of the time? (assuming those strengths are post-modifiers of course). If someone ever works that one out I'll be in love since then your early scouts can still survive the occasional bear or spider, but on the whole you can predict what a battle should result in.

As far as I understand it is not the case, it is built just to avoid that 25%.

 

[xienwolf]

That is what I understood as well and what would keep me from using such a change. David does beat Goliath from time to time in the real world.

 

[TheJopa]

Yesterday I lost Rantine in 93% odds. Go figure. Just as I said, more than 90% should be a win, just how damaged unit will be is question. This is annoying in vanilla civ4, but here, where heroes are irreplaceable its even worse.

 

[Verdian]

One of my favorite videogame designers, David Kennerly, wrote a very interesting article on the subject of probability in combat in videogames. It specifically relates to RPGs, but the principle is the same. Forgive my plugging a link, but this fits here too perfectly.

Randomness without Replacement

From the article:

"Game mechanics" is one of those terms that every designer talks about. Everybody agrees that game mechanics is an important subject within game design. Yet almost nobody discusses, in detail, how the design of a mechanism may satisfy a player. In this article, I'm going to dissect a mechanism germane to role-playing games, the randomization function for determining whether a player's attack hits or misses its target. Roll up your sleeves for a modest amount of mathematics—just enough to demonstrate what a game mechanism is.

 

[Darkheart]

Interesting article, and the math is not too heavy. One thing it doesn't mention about non-replacement theory is that once you've had your 10 hits out of the pack, even if it's a measly goblin standing in front of you, it's probably a good idea to run, coz for the next 10 hits that goblin's gonna take out all of his frustrations on you!

The other thing I picked up about it, and maybe it's just my perverse cyniscism about MMORPG, it seemed to have a commercial bias to it, based around the concept of;- no matter how rash or foolish the players attack maybe, we must not allow him to feel frustrated, as he'll probably go and play a different game. Some of the MMORPG's that I've played have started out with really great ideas and values, only to degenerate into silly kit-fests, powergaming and "You too can buy your level 80 custom built character on E-bay!" Simply because what appeals to the "masses" also appeals to the stock-holders, in that equation who cares about quality?

My point about randomess, and one of the reasons why, I believe, it shouldn't be reduced is this:-

You are the Commander in chief, you have complete control over not just the order of battle assigned to your theatre, but also control over production and what's more even the politics. YOU choose when, how and who you go to war with. You already have far more autonomy then any general could ever hope for.

If, with this criteria, you go to war hoping to win, then you've already lost. If your battle plan is so dependant on 1 large powerful unit, then you're doing something wrong. (OOOOOOOOh gonna get flamed for that one!!!! )

It's the mark of a good general, commander or strategy gamer, to create circumstance. To create a position where he can exploit good fortune whilst minimising the effect of bad fortune, without affect to the end objective, merely a delay.

Randomness isn't an imposition.... it's a challenge.

 

[xienwolf]

Well, I do not agree that a 90% odds should mean that you always win, but wind up damaged. I think that 10% of the time you should die. But I do agree that the other 90% of the time should NOT result in you walking away unscathed.


What would be kinda nice is if instead of doing the combat rounds till you run out of Hp, it was a single roll, and the results dictated not only the winner, but the winner by what degree. So if Mr. 90% needed only roll under a 90 on a d100, and rolls an 89, then he loses 89/90th's of his overall health. Now, if he were to roll a 0, then he would lose no health at all. And if he were to have rolled a 99, then his opponent would lose 1/10th of his life, and win.

This way, the general result would be that you roll somewhere around a 50. So in a 50% battle, you should win half the time, but come away almost dead. If you do lose, your opponent comes away almost dead. But in those 90% battles, you will typically come out of it with only half your health remaining.


Hrm... now that I have written that I see some problems with it. Namely that with a couple of STR 2 units I could cut Acheron's health in half, then in half again, then take him fairly easily with a real unit...

BAH! Hard to find a system that will satisfy everyone, neh?

 

[Darkheart]

I think the reason for Hps and multiple rounds, and this goes back to Civ2 was to try to even out the randomness factor, rather than the old compliant of chariots defeating tanks.

Though saying that there's a really funny screenshot from Vorshlumpf showing a blow by blow account of a barb orc spearman knocking (small) pieces out of an obviously well promoted elven warrior. He quoted the odds at 0.11%!

LOL.... S**T Happens!

 

[sylvanllewelyn]

The problem is not randomness in combat, but the idea that heroes are used for direct combat. Since heroes are not replacable, and that some civs are defined by their single hero, I believe that heroes should be spellcasters, summoners or units with unique abilities, but never someone whose primary value comes from direct combat (although they should be able to, since if you don't protect them, they should die at the hands of his/her/its enemies).

 

[Tarquelne]

Compared to a "real wargame" - one with little cardboard pieces and NATO symbols and a rules booklet of at least a couple of dozen pages - CIV4's combat system just sucks. It's terrible. Measured against the lack of logistics, morale, combined arms, etc. any problems with the randomness of the game are insignificant. Except for "feel" - how it appeals to different tastes.

Of course Civ4's combat system isn't supposed to be a "real wargame". For a 4x game it's fine. Though I think it has lagged behind the other aspects of the game. (OTOH DCM seems to put combined arms in, which is a huge step forward.)

I think the single biggest step toward making Civ4's system like a "real" system might be to have combats commonly end with both units hurt, not one unit hurt and the other destroyed.

Getting back to this thread, such a system might act something like the "fair combat" mod for people who don't want to lose the randomness of combat: You aren't guaranteed a win or a loss, but an unwounded unit is likely to be alive at the end of a combat. Unlucky units should be able to retreat behind lines - they won't just be destroyed.

Such a system might even be fairly easy to implement in FFH: Just give every unit a withdraw chance, both offensively and defensively. (Maybe fairly large def. chance and lower off. chance?)

Would battle-lines become important? You'd want to protect the wounded units that withdrew from combat. We might see armies advancing and retreating across the map rather than stacks advancing and destroying or being destroyed.

Hmm... I think I'll try it and see what happens... though probably not till next week.

 

[xienwolf]

I think a better approach to having units remain injured instead of dead is to limit the rounds in combat.

Now, it checks Fill some one dies, but then checks if they retreat instead of die.

Your method with everyone able to withdraw means you windup at 0.1 health/Power.

Instead, allow only 10 rounds of combat. Then a full health unit with first Strikes cannot die in as ingle attack unless the difference in strength is enormous (allowing more than 12 damage per round). But, with no first strike (or Stoneskin) you most lose every round of the 10 to die in a single attack (if starting at full life).

 

[Tarquelne]

Your method with everyone able to withdraw means you windup at 0.1 health/Power.

I think it'd also mean the AI is screwed. It's not going to know how to protect all those wounded units. Cavalry will be able to pick them off, and then escape because they'll just get wounded in the counterattack.

Instead, allow only 10 rounds of combat.

How do I do that? (The _biggest_ advantage of the method I outlined above is that I already know how to do it. )

Alternatively, do you know how or where I'd change withdraw behavior?

 

[xienwolf]

Not sure where you could do it at. I believe that it would be a Python change at the least, but it is most likely going to require changes in the DLL.

You could possibly also alter it to allow a check for withdrawal on each combat for the unit who has the lower relative (health corrected) strength. Then your high withdrawal powerful units won't try to run away on the first round of an easy win, but each unit would try to break off from any combat in which they do not have the upper hand. Maybe also make the retreat check only happen if they LOSE that round as well (then you can still have a VERY lucky streak keep your weakened unit in the battle).

 

[Tarquelne]

but each unit would try to break off from any combat in which they do not have the upper hand.

That'd be very good. All sorts of interesting possible modifiers.

But I'd rather use the present system than mess around in the SDK. Well... Dale's Combat Mod probably either has the code I want or something close enough. At some point I may try merging it into FFH rather than playing a game, though that doesn't seem too likely. I think the present system is good enough... at least as long as it takes more than easy XML or python tweaks to change it.

But a "single combat" possibility between heroes as part of DCM's CASA resolution could be a nice FFH feature...

 

[rief_s]

The problem is not randomness in combat, but the idea that heroes are used for direct combat. Since heroes are not replacable, and that some civs are defined by their single hero, I believe that heroes should be spellcasters, summoners or units with unique abilities, but never someone whose primary value comes from direct combat (although they should be able to, since if you don't protect them, they should die at the hands of his/her/its enemies).

I think this the main reason that raise the problem. But not all hero is spellcaster/summoner. Some are direct combat. So, how about heroes become immobilize unit like item. After they created, they can join unit. If the unit dies, the hero become immobilize unit again. Make the hero invisible so they won't defend a tile.

 

[[NWO]_Valis]

I think this the main reason that raise the problem. But not all hero is spellcaster/summoner. Some are direct combat. So, how about heroes become immobilize unit like item. After they created, they can join unit. If the unit dies, the hero become immobilize unit again. Make the hero invisible so they won't defend a tile.

Nei, it could be itemized and attached to a unit and if that unit dies then it should have a chance to escape to a different unit in the stack or to the capital. The chances could be different for different heroes. There would also be a chance that the hero was not able to escape and was taken captive. As a player you could have then the option to execute it [in a town for happiness or other boni is for example you are Order and captured Rosier], let him go, keep as prisoner [good civis could not execute good and neutral heroes]. Such hero could be then rescued. Lots of possibilities

I want to try out a 10 turn combat mod. First strikes would gain on value, a biochemical war could be fought using diesied corpses, WW1 battle fronts could be a reality.

I would only like to point out that FFH2 is 100% better over vanila cause it has the attack/defence split