Game Inertia

Discussion in 'Game Play' started by spellman23, Nov 10, 2009.

  1. spellman23

    spellman23 Member

    Messages:
    861
    Likes Received:
    0
    Trophy Points:
    0
    Something we like to argue about around here is the "Slippery Slope". For those of you uninitiated, behold perhaps the most over cited article evar on the subject:
    http://www.sirlin.net/articles/slippery-slope-and-perpetual-comeback.html


    Now, today I'd like to address another point of concern. The system of Game Inertia.

    As many of you know, in physics inertia is a measurement of how much energy is required to create change in a moving object. Higher inertia means more energy input is required for similar results in change in velocity than that of a lower inertia object. Often this is associated quite directly with the mass of the object.


    Game Inertia is a similar phenomenon. For now, let's assume we have a pure slippery slope. Perfect play on both sides means that any advantage gained in the beginning will accumulate as the game goes on and eventually result in a victory for that player. Inertia then is the measurement of how fast that advantage is grown based on how far down the slope the player is. In other words, low inertia games would mean the fall-off from equilibrium is relatively slow. For a higher inertia game, the victory follows swiftly after slight changes.


    Many of you may cry out that this sounds quite similar to the slope of the slippery slope. Steeper slope, faster falling. However, that doesn't take into account imperfect play.


    Inertia is the measurement of the comeback potential. In a way, it's how easily the movement of the advantage moves with player interaction. For many games, there's always a comeback available. Sure you lost that first skirmish, but you're not beaten yet. If you can just win well enough in the next battle, you'll have recouped your losses. For a small inertia game, the balance is easily swayed by small actions. Have a big enough win and you can overcome your previous failures. For a large inertia game, you will have to win hard and fast before the opponent starts to accumulate their advantage.


    Now, what is Empires? Or, more accurately, what do we want in Empires?

    Some claim that Alamo sieges should be winnable by a decent margin. Fight to the last ticket! Others believe that eventually the advantage should become insurmountable. How important should those mid-game engagements be? How strongly do they depend on previous battles? If the comm doesn't quite place the Barracks in the optimal position, how much will that hurt the team?


    TL;DL: how much sway of the game advantage should each individual action have? Should the advantage be able to swing due to the result of the last battle? Or depend on all accumulating battles and then some?



    EDIT: Another way to look at it is how much of the advantage accumulation is from player actions or from game mechanics. More inertia, the more the game forces the slipperly slope and the less impact player actions are on the equilibrium.
     
    Last edited: Nov 10, 2009
  2. Beerdude26

    Beerdude26 OnThink(){ IsDownYet(); }

    Messages:
    7,243
    Likes Received:
    13
    Trophy Points:
    0
    I feel that game inertia should be largely influenced by the amount of teamwork of a certain team during a certain action, but that is solely my opinion.
     
  3. spellman23

    spellman23 Member

    Messages:
    861
    Likes Received:
    0
    Trophy Points:
    0
    So, perhaps teamwork should work as a kind of multiplier for individual actions?
     
  4. Sandbag

    Sandbag Member

    Messages:
    1,172
    Likes Received:
    0
    Trophy Points:
    0
    Taking the following points over from the slippery slope arguement, that you need a game that either:

    is quickly ended once the game has been decided
    or
    a game that allows a comeback at any point.

    now, I'm strongly in favour of the last option, because it gives a greater width of experience to people playing empires. Both of these options though require a middle - low inertia. It's the slow, huge inertia games which drag the game down.

    However, higher inertia generally benefits the teamworking team. Ignoring economy that is. having a low inertia means that individual units are far better at swaying the fight. Take the micro example of infantry combat. With low inertia, i.e powerful quick-to-kill guns, teamwork can be a bit of a non issue unless you have serious verbal co-ordination about where the enemy is and so on. see counterstrike. with weak guns, wearing down the enemy and working as a team is more important. In comp TF2, which i've dabbled in, unless you are going for a pick (take out an individual player all by yourself, e.g the medic) you'll generally just try to wear the enemy down to such a point that one of your team has an easy task of finishing them off. How does this relate to empires? If you have really low inertia, such that one player can swing the balance, then spreading all of your players out across the map, everyone as engineer and in a tank, is the best option for your team, even though it leads to dull 1v1 encounters and mindless teamplayless gameplay.

    Ideally what we want is an economy system that is low inertia, allowing a team as a whole to make a comeback at any time, and micro, infantry level gameplay which is middle inertia, massively benefitting teamplay instead of rambo, but not so high inertia that weapons are ineffective (and being a rambo is dire- sometimes you're on a team where you NEED to rambo)

    I can't rate the empires micro inertia but on the larger, economy side of things, we definately need smaller inertia. Being out-teched and then simply grinded down sucks. When the enemy is fielding 2 tanks for every one of yours, that sucks too.

    Oh, and tank combat at the moment is bad. It's generally so high inertia because of massive health of units that the weapons feel pathetically weak (especially the grenadier), yet at the same time fighting in squads of tanks as a team rarely seems to benefit directly. I think this is because of the speed of tanks, that they can drive away so fast, but i need to think more about it.
     
  5. flasche

    flasche Member Staff Member Moderator

    Messages:
    13,299
    Likes Received:
    168
    Trophy Points:
    0
    no sandbag i think you hit it spot on. tho revive lowers that effect of quick killing guns a bit in terms of teamwork ...
     
  6. ScardyBob

    ScardyBob Member

    Messages:
    3,457
    Likes Received:
    30
    Trophy Points:
    0
    I like this idea. Would need some tweaking to fit Empires, though.
     
  7. Ikalx

    Ikalx Member

    Messages:
    6,210
    Likes Received:
    9
    Trophy Points:
    0
    Such as adding res on top of that, which would mean the team with more control on the map (in a basic function res is a bit like territory control), would have a larger pool of res...although refinery resources would have to be expended, rather than conserved.

    Although...having a refinery could just mean a boost to your maximum renewable resource pool instead.

    This is a lot more like the thread about having the base have a higher resource node...which I dug up again recently: http://forums.empiresmod.com/showth...ighlight=barracks-dependent+victory+condition (discussion starts near the end of the page)

    (apologies for mild off-topicness)
     
  8. Sandbag

    Sandbag Member

    Messages:
    1,172
    Likes Received:
    0
    Trophy Points:
    0
    nah if the base has it then you basically fix base positions to two positions on the map. that would suck. the best games of empires are where bases switch frequently
     
  9. Ikalx

    Ikalx Member

    Messages:
    6,210
    Likes Received:
    9
    Trophy Points:
    0
    So you said before, of course, that usually results in the CV getting trapped within the destroyed base :p

    I was mainly looking for a way to represent that idea without having disappearing/reappearing res, but I guess we could call it "power" and thus have it reasonable for it to reduce with the amount of elements on the field, and come back up when they are destroyed.

    Not quite sure what I think of that idea as a whole though.
     
  10. Sandbag

    Sandbag Member

    Messages:
    1,172
    Likes Received:
    0
    Trophy Points:
    0
    Well in terms of reducing game inertia, one option would be to remove the cost of research with the exception of a couple of deliberately expensive items. This would at least mean that a

    the reason research makes the difference in money between the teams significantly imbalanced can be demonstrated from a thought experiement.

    imagine Brenodi have 50res/second and NF 40res/second. Quite standard. Now, each side requires 10res/second to spend on turrets, walls, buildings though, because NF is going to need to replace buildings and BE place more buildings as the front lines push forward. Actual spending res: BE=40, NF=30. But to stay at the same tech level as the enemy, so that you don't get totally wiped by medium tanks with HE while you're still getting armour for your light tanks, you need to be researching all the time. lets say it costs 20 res/second to research. So spending money left for tanks is: BE=20, NF 10. What looked like a fairly even match, once you take away the upkeep of just staying in the game away, actually allows BE to field twice as many tanks as NF.

    Okay, so the numbers are probably bogus, but the idea is sound. "Upkeep" which both teams must pay equal amount of to stay in the game increases the % difference between the actual spending money the teams have. Perhaps making most research free, so long as you have a radar, would decrease the upkeep and improve the game.

    More equal teams = lower inertia = less slipperly slope
     
  11. Ikalx

    Ikalx Member

    Messages:
    6,210
    Likes Received:
    9
    Trophy Points:
    0
    I think I understand what the problem is now - research is tied to resources, so not only do you not have enough buildings on the map when you start losing, but you also lose any kind of ability to research which means your team can never come back unless they're completely superior than the other.

    It's basic...but maybe all we need to do to fix the slope is decouple research and resources. That way the attacking team will always have more tanks on the field, but the defending team won't be crippled by the lack of technology for their one of two tanks - meaning if those tanks survive through skill and team effort, they have an opportunity to come back.

    Or does that just make the loss take longer?...Someone post what's obviously wrong with that, my brain isn't working atm.
     
  12. ScardyBob

    ScardyBob Member

    Messages:
    3,457
    Likes Received:
    30
    Trophy Points:
    0
    I also like Sirlin's explanation of game balance and its challenges. Empires does violate some of the principles he describes. Though, I'm not entirely sure how to apply his balancing concepts to an FPS/RTS, since he talks more about side-scrolling fighting games.

    Edit: I think this is a relevant passage for Empires
     
    Last edited: Nov 11, 2009
  13. spellman23

    spellman23 Member

    Messages:
    861
    Likes Received:
    0
    Trophy Points:
    0


    Something to note in terms of Game Inertia is that typically economies are very high inertia.

    If you have two sides and one has a slight economy advantage, even if no effort is put out by either side towards economy the advantage accumulated quite rapidly. Couple that with the ability to re-invest to gain other advantages such as more research and you have a very powerful force.

    Some games like World in Conflict try to avoid this using a capped but regenerating number of points you can spend. You're always capped out so you can't accumulate more troops, but lose enough at once and the enemy can advance with ease until you can redeploy the troops you lost, providing a temporary advantage. Decoupling a "research resource" has also been suggested several times and in general has had mixed reactions. I personally kinda like it, as long as there are small ways we can influence it. Say spend a bunch of res to get a couple extra points, but it has to cost quite a bit. Just another choice to give strategic diversity.


    I also heartily agree with ScardyBob's notes. If we can lock down some serious game time (and have intelligent balance changes in the first place) most balance changes can work out. We just first have to have proper balance choices and secondly fairly evaluate the changes in a series of tests.


    Something from the article linked that I always wondered about was how asymmetrical Empires intends to be. I mean, there's plenty of more subtle differences, but it's all the same tech and same archetypes, just with certain (and sometimes almost hidden at first glance) tweaks. Even the subtleties of their models. Well, that's a tangent for another thread.
     
    Last edited: Nov 11, 2009
  14. Ikalx

    Ikalx Member

    Messages:
    6,210
    Likes Received:
    9
    Trophy Points:
    0
    I'm not sure how regenerating points would work with regard to research. Kinda slipped through my hands there.

    Anyway, if we start by saying research costs nothing but time (not ideal, but as a base idea), then we could have a mechanism that allows research to be sped up if a certain amount of resources is directed into it - e.g. 1k res = 5-10s off each tech item.

    We could also rebalance the trees to be more progressive - instead of it being a wheel, we could make it more of a proper tree (a bit like Drag's idea). Then...we could have fast and weak trees and slow and powerful ones, so the losing team would focus on weak trees that get them the research faster, while the winning team would be free to invest in more powerful trees, giving them the edge in combat.

    This way, we could also make the research follow a default path if there was no one commanding (the fast and weak one), which would mean that team would still be able to carry on (forget buildings for now though).

    Just using armour as an example, the losing team would end up with something like absorbant heavies, while the winning had composite. As they both have heavies, the losing team still has a chance to win, but the winning team has much more of an edge.

    Er...should I take this discussion to another thread :confused:
     
  15. RoboTek

    RoboTek Member

    Messages:
    323
    Likes Received:
    0
    Trophy Points:
    0
    I think you are underestimating the values of various elements.

    The problem is not strictly one of research, in fact, it is not directly one of research at all (though that is the end result).

    Teams have a disproportionate focus on tanks, generally spending 80% of their income just on tanks. This is especially true in larger games. The solution to large-scale games is simply to scale resource production, which minimizes the value of research.

    Overall, you really don't spend 'that' much money on research. Instead, research is the end-character in this situation.

    You might spend 10 on buildings then 40 on tanks, then half as much as the enemy on research. In your attempt to maintain combat power against the enemy, you generally will be under-cutting your own research attempts. At the very least, because of the stacked-order system of research that we have, you will be delaying your research attempts when trying to squeeze out an extra tank. The problem is systemic rather than with any specific issue value.

    By making research cheaper, we undervalue its importance. The ideal for scaling larger games is a much weaker growth pattern, proportional to the 'ideal' % of income that we would have based on tanks. Every 10 players might be another 20% income increase, as the number of desired tanks increases.

    Increased cash-flow all around actually leads to less value on any specific tank and increases the slippery-slope by devaluing the infantry on a team. If we decrease tank costs slightly, decrease resource scaling, and retain approximate research values, it will create a system where it is more realistic to produce less tanks, because you already have a gameplay style that assumes most players will not be in tanks.

    If you have a more valid pricing structure for variations in tank weight, then you will end up with the ability to give players more, low quality tanks if that is your strategy.


    Game inertia is not a bad thing, it gives weight to actions. The problem is that adjusting strategies cannot apply much force, changing the nature of the game.
     
  16. Sandbag

    Sandbag Member

    Messages:
    1,172
    Likes Received:
    0
    Trophy Points:
    0
    It's hard to have a thread like this which just goes down the page one post at a time, because ideas really need to shoot off sideways and get individually discussed. I recommend that if anyone sees any sort of good idea, they

    Elitist as it is, I kinda wish there was more of a council, or at least large group of people to discuss suggestions instead of everyone. The suggestions in the suggestions forum get pushed down the page by silly suggestions and the threads themselves get clogged up with thoughts from people that didn't really think about the outcome of their additions to the suggestion, if that makes sense.

    Yeah, this is important. I think the crucial point of this thread is to eliminate where game inertia makes it difficult for a team to make a comeback.

    With the slippery slope analogy, sometimes you may not slip very far down the slippery slope, but the effort required to change yourself from slowly slipping down to being on equal terms again is large because of the "weight" of game inertia on you.
     
  17. Chris0132'

    Chris0132' Developer

    Messages:
    9,482
    Likes Received:
    0
    Trophy Points:
    0
    Why on earth is he talking about strategy in a fighting game? There isn't any.
     
  18. Ikalx

    Ikalx Member

    Messages:
    6,210
    Likes Received:
    9
    Trophy Points:
    0
    Isn't the strategy planning your moves?
     
  19. spellman23

    spellman23 Member

    Messages:
    861
    Likes Received:
    0
    Trophy Points:
    0
    And now you've lost credibility.

    Just read the bloody article.
     
  20. blizzerd

    blizzerd Member

    Messages:
    10,552
    Likes Received:
    60
    Trophy Points:
    0
    a big portion of empiresmod steamrolls is that the mappers have a wrong idea of how a proper resource point stack is

    lets assume a simple linear map with 5 to 7 refs

    commonly, this is done

    [NF base] |1x| -- |1x| -- |2x| -- |1x| -- |1x| [BE base]

    the green is the contested area, people think "oh gee, people will be fighting for the middle ref, so i have to make it worth to hold it, il give it 2x"
    the bases refs can be different amounts of refs spread out, or 1 ref giving all the res (as long as its a ref that is protected by the base/really close to it)

    thing is, some team is gonna put a rax next to it, and it will not change sides for half the game, so now 1 team has 2x the other teams res flow for half the game (extreme example)

    what is better is this

    [NF base]|2x| -- |2x| -- |1x| -- |2x| -- |2x|[BE base]

    now, capping the middle is a strategical coice, it could end up in ticket bleed, but you will get some extra res out of it, not enough to roundhousekick the enemy, but enough to roll out some extra"s that the other team will have to save on

    what i want to try, and might even be far better is this

    [NF base] |4x| -- |2x| -- |1x| -- |1x| -- |1x| -- |2x| -- |4x| [BE base]

    this effectively makes the battlefield large enough, and non determined enough to make it so if a team loses a point, nothing "really" has been lost, but if it falls back to its main base, the fight will be over soon (while still having "some" money for stuff so its not totally hopeless for a last ditch effort, 4* res should be more then enough to hold out while coms get du spawn apc's)
     
    Last edited: Nov 13, 2009

Share This Page