Unstable Equilibrium Aversion topic

One of the constants in real-time strategy games is that if a player starts winning, they will gain advantages that make them likely to continue winning.

This leads to the fact that, if one player manages an early advantage, the other player is likely to experience a slow, grinding defeat, barring the first player either getting lazy or the second player getting lucky.

The first question is SHOULD we fight this. The second question is if so, how do we fight it?

I think that comeback mechanics would be a helpful thing, but the difficulty is in implementing them while maintaining certain things that Planetary Annihilation needs, like the encouragement of vast armies and great wars.

It's also important to avoid a stalemate-enforcing mechanic, where neither side can get a win.

If one player does gain an advantage that makes their likelihood of winning certain, then victory should not take a long time past that- It's not much fun to be forced to play for half an hour, knowing that you're going to lose.

I don`t think a "comeback" mechanic is of any good. If the guy can get a small advantage at the beginning it`s in the other side`s hands to turn the tables or pay for it`s mistake.

The game should not give an advantage for the one that did the mistake or take the advantage from the one that did something right.

If you want to avoid early advantages just make a non aggression timer agreed upon by all players.

There is no reason to reward mistakes.
Thats basically what any kind of come-back-gameplay-mechanic would do.

I wouldn't mind some high risk high reward and/or sneaky strategies (TML snipe anyone), but not any automatic system. Basically, if you're starting to lose, you should have options, but your enemy should know about those options too.

However, there should be a way for the player to recover from a mistake, if they use proper strategic thinking. Only the most extreme mistakes should be irrecoverable.

High-risk high-reward sounds like an interesting concept to pursue- The incorporation of strategies that will likely lead to loss if they fail, but can potentially turn the game around if you succeed.

What's the best way to incorporate this, though? Destruction of permanent resources? How about a system where you can send resource collecting engines into a destructive over-consumption mode- They will take damage, and can't be repaired, but are capable of producing more resources while it's in place? Put in place a brief 'shutdown' period to change, say, 5 seconds without production? A 'cannibalism' mode for factories that works similarly, increasing their build speed and reducing the resource cost of units, but also doing damage?

Balancing the shut-down time and the need to repair, and the increased vulnerability this gives your units, this would, ideally, have the following effects

Give a player who's fallen behind in mass and construction a chance to regroup, at the cost of making their position more fragile, by weakening their resource buildings.

Encourage risky behavior on the part of players who are already losing, which will either turn the game around, or speed their loss.

Discourage players with an advantage in production from using these techniques. How best to do this? We don't want players to be encouraged to use this mechanic because of its micro benefits. A greater 'turnover' period depending on how much mass/energy you've got? Perhaps by making this a global thing, so you can't strategically place certain factories and resource producers into overdrive, you must instead choose an all-or-nothing strategy, and having a greater lag time or greater damage or less benefit with more factories/resource producers?

Just tossing some stuff out.

Well, I'm not quite sure WHAT would work, but it would have to be something that makes a poor long term strategy. A "shut down time" wouldn't be enough. The permanent damage and destruction of resources ideas are more along the right train of thought here (not that it's a necessarily a good idea; others here are much better at thinking through these types of ideas than I am).

Sorry to disapoint you but thats the tactical advantage of war. If you gain an advantage you're going to be better off than you're opposition. And anyway, you dont need a reward system. What you could do is begin thinking diplomaticly, make an alliance to stand against someone who is fairly powerful. 1 Strong player vs 4 Weak players could become an effective method of turning the tide of war, especialy when you can harras their resources. Even being able to out manouver 1 player with 4 players would become an extremely useful tactic. This game should be as brutal as possible to people who dont think tacticaly. Its an RTS, you are ment to be at a disadvantage if you make a poor move, which can be corrected by doing a great move.

Scott, there are two things wrong with that.

First, it only works if you're in a free-for-all game, and not all games will be free-for-all.

Second, it just delays the problem till later. You just also make life inconvenient for anyone who takes an early lead.

Really, FA already had such things in place. Enemy's winning and starts sending units at you that you manage to repel? Great, because now you have a bunch of steaming wreckages you can reclaim. And if it keeps happening like that? Then build up your power and start spamming mass fabricators.

No matter what, any mechanic you use is going to work both ways.

Wreckage is pretty decent. A failed attack is more money for the opponent.

Energy is also effective. No matter how big an opponent is, without energy he's helpless.

Just reduce the slippery slope as much as possible. Don't actively help the disadvantaged player.

If the slope isn't very steep, then there's every possibility that advantaged will also make a mistake.

That is definately a good thing. Rushing and other kinds of early aggression favors high-APM players, and I don't want them to dominate the game. As to comeback mechanics, It is also tend to make games longer (which is also driving APMers away) and less dependent of random events. To beat someone this way, one needs to generally outsmart him, not just take advantage of opponent being distracted from PC a moment before his units had an encounter.

If you are distracted from the game, you should lose
Also it is not about driving APMer away. Why should it be? In fact anyone who puts a serious effort into improving will automatically speed his play up as much as he can.
A good player never tries to reach high APM, he just knows what he needs to do and therefore does it without long breaks to first think, this naturally results in "high" apm.

And no one wants to have the first 5 minutes about rebuilding the same infrastructure because there's no other way to play.

The game should start at the first interesting tactical decision. It could be as simple as building economy or a factory first.

In fact I dont know any RTS that does this. The first 5 minutes are quite important in games of the SupCom:FA type just as much as in SC2.

So I brought this up in another thread the other day

And yeah there are two ways you can deal with this.

First you can make sure that the losing side still has some hope that they can win (even if they probably won't) This is why I like the reclaimable wreckage mechanic because it ensures that rebuilding after an attack will always be easier for the defender than the attacker, because the defender has a bunch of free scraps outside.

The second way you can deal with this problem is to make a game mercifully short when it's clear that one side is going to win. This is the one good argument for a big expensive krogoth style experimental I have heard. Think of an experimental as the economic victory option out of a 4x strategy game. And it's far more satisfying to build a giant laser death spider and have it quickly tear through someone's base, than it is to wait for them to type "GG" in the chat window before disconnecting.

Battle Realms had some very interesting ideas in this regard. Your main source of units were peasants that were automatically spawned after a period of time, and would be trained into combat units as needed. If the player had fewer peasants, they would spawn faster, and if they had more, they would spawn slower. As a result, a player could still make a comeback after a significant loss, but only if they played intelligently.

I'm not saying this method should be used by PA in any way, but it is an interesting idea, and it worked quite nicely.

If the guy wants to give up he just GG and leave. There is people that don`t give up so they are willing to keep playing just for play`s sake. So let the game go on. Experimentals are a contradiction, since they take much time to be built if you are just starting them at the moment you guarantee the game. Rather make a swarm and finish the game off than to make an experimental.

This is how games were made a decade and a half ago. Nowadays this approach just drives players away because they get frustrated by being punished with defeat every time they miss a button. Game developers are very afraid that players will drop ther game in the flash of anger, and then lower its overall rating by negative reviews and downvotes on gaming resources, thus probably preventing others from even buying a game. Because of that, current generation game tend to be so forgiving that it literally playing itself. And it is not just a simplification, it is a different approach to the player: days back, player kept his interest in a game in spite of difficulty, and now game itself actively tries to entertain player and keep his interest alive.

Well, that's not true. APM is bound to fluid intelligence and reaction speed, and both of this characteristics are physical and mostly unimprovable after early childhood. People who have them high enough, definately, can react quickly to situation after learning the game mechanics. Ones who doesn't have them will never reach high APM, whatever they do, and ones from first category will never understand why, since it looks natural to them. In fact, favoring high APM in games should be considered as a discriminative design and therefore a bad practice. Unfortunately, modern wolrd, for a number of reasons, does not recognize an existence of an intelligence (as well as any CNS parameters-related) discrimination at all.