Thoughts on balancing

Discussion in 'Game Play' started by Candles, Jun 21, 2015.

  1. Candles

    Candles CAPTAIN CANDLES, DUN DUN DUN, DUN DUN DUN DUN.

    Messages:
    4,251
    Likes Received:
    10
    Trophy Points:
    0
    This has been something I've wanted to have a serious discussion about in a while because it cuts fundamentally to the heart of Empires.

    This is not a thread about discussing potential changes per se, and should not be taken as deciding the direction the game would go too. It is simply a discussion at the most, and more generally a collection of thoughts on the matter. I'm curious as to how others view balance on a more abstract level, ideally with some concrete examples paired with the concepts.

    Balancing in Empires has always been a complicated matter, and I certainly learned this personally when I tried my hand at balancing the scripts. As such, I want to list some things I've observed about balance, expectations of what balance is, and player's reactions to certain balancing methods. I do not intend to pass judgement and declare that any view is correct, but simply to try and foster some understanding of what balancing entails.

    The first thing I want to talk about is the concept of hard vs. soft counters vs. skill-based equipment.

    For those who don't understand the terms, a hard counter refers to anything that is almost guaranteed to beat another thing regardless of individual skill. For example, Absorbent has classically been a hard counter against Railguns because the Speed to Damage modifier drastically lowers the damage that Railguns do to the point that a tank with Absorbent with almost any weapon will beat another tank that's using Railguns simply because the second tank will do a comparatively negligible amount of damage. Another example would be Bio MG and Reactive when Reactive took 150% damage from Bio, causing Bio MG to tear a Reactive tank to shreds. A soft counter would be something that has an advantage but isn't guaranteed to win, or can be countered with skillful play. An example of a soft counter would be the Rifleman vs the Engineer at range, where the Rifleman would be at an advantage but the Engineer can still cinch a win. Another example would be 3-Phase vs. Homing on an open map, where 3-Phase can potentially just outrun the Homing missile, but can still be hit by, say, aiming ahead of the target instead of relying on the homing capability. Skill-based equipment refers to anything that isn't inherently good by default, but has some trait that makes it more effective with skillful use. Two common examples of this are Reflective and Regenerative.

    Empires has, historically, tended towards skill-based equipment. This is descended from its FPS roots, where hard-counters are rare and soft-counters relatively uncommon, where a pistol can be better than an assault rifle or a shotgun in the right hands. Fundamentally, Empires players tend to be FPS players, and as such don't like the idea of "Weapon X always beats Armor Y", because that takes a lot of the skill out of the player's hands. However, it is important to note that RTS's take the exact opposite view, where hard-counters are the bread-and-butter of gameplay and soft-counters form a support for this. Both of these views make sense within the respective confines of their genres: an FPS game that relied on hard-counters would almost be reduced to a game of chance, each player hoping that they pick the class that would beat a majority of the enemy, while an RTS allows a player to compose their own army and use a variety of units to emphasize strengths and cover weaknesses; FPS's rarely have the type of coordination required for such a system of balancing (If you imagine, say, an AoE-styled first-person game, it would be perfectly feasible that one team might be nothing but archers and the other team cavalry, while it'd be harder to imagine a single person managing an army deciding to put all his eggs in one basket).

    Certainly, one of the things I've seen is some people asking for hard-counters to be the way research is balanced in Empires. For example, Electric beats Bio, Bio beats Physics, Physics beats Electric, would be the classic rock-paper-scissors balance. Oftentimes, the reasoning for this is two-fold: Firstly, it would make balancing a helluva' lot easier; simply give each armor in each tree a resistance to the tree in front of it and a penalty to the tree behind it. Secondly, it puts a real importance on research and causes players to have to constantly be researching new items so they don't fall behind.

    However, there are two clear downsides to this approach. The first is that it takes a lot of the skill out of the game when it comes to encounters; a two players in tanks can only hope that they have the counter to what the other guy has. The second is that it moves a lot of how the game will go from the players to the commander. If the commander researches the wrong thing, then it makes it a lot harder to make a comeback than it would normally, and it's a lot more important to have a commander who knows exactly what they're doing. Another thing of note but not as important is how keeping the enemy radar down would be much more of a boon to stop them from getting the counter to whatever you have. As it's well known that it's harder for a losing team to afford a radar and keep it up, this would have the potential to make it harder for a losing team to make a comeback, by making research all the more important.

    Soft-counters would have somewhat of the same issues as hard-counters, but less so. I'd imagine that an Empires using a counter system such as this would tend towards soft-counters (say, 20% damage resistance on armor) rather than hard-counters (50, 75% damage resistance).

    Skill-based counters are another beast entirely. As I said earlier, Empires has generally tended towards skill-based counters due to its FPS roots. Like hard-counters there are upsides and downsides. The upside is that individual player skill becomes more important and it reduces reliance on the commander getting it right. The downside is that individual player skill becomes more important and it reduces reliance on the commander getting it right. I cannot stress how important it is that that cuts both ways.

    You see, oftentimes, commanders don't research the best upgrades, they research the most usable upgrades. That means Homing over Guided, HE over Ranged, Compo/Reactive over Regen/Reflective, Gas Turbine over AC/Fission. They can't afford to get Ranged if it just means the lesser-skilled players are going to get near-misses all the time, or get AC if players won't learn to stop moving their tanks, or Reflective if players won't learn to angle their tanks. When something is made more skill-based, players tend to want it more but at the same time, commanders want it less. The end result is that commanders classically have gone for safe research that doesn't require them to really weigh the skill of their team. This is also why Bio used to be such a popular tree with a team of new players; they don't have to worry about not being able to move when overheated, they don't need sustained firepower to keep the DPS of Bio up, and the armor means they don't have to worry so much about making sure they repair.

    I believe that the best example I have for this is Reflective. Reflective has been the best armor in the game for quite some time, by the numbers. Previously, if you could angle it by that 45 degrees, it become the best armor in the game because it had effectively double the health. But it was too risky of a maneuver to get Reflective because it was skill-based. So it got made safer, first by lowering it's weight, then increasing it and it's health, then by lowering it's health and it's weight to 80/15 respectively. By that point, it became safe enough for commanders to be willing to get it. However, that simple metric alone belies how much it had to be buffed to make it safe while keeping some semblance of skill mattering. When you bring the reflective modifier into account, it becomes extreme.

    At 10 degrees, Reflective has about the same health per weight as Reactive. At 30 degrees, it can match Reactive's health-per-plate and health-per-res, while matching Capacitive's health-per-weight. At a full 45 degrees, it smashes everything else out of the park at 160 health-per-plate and 10.66 health-per-weight, a full 25% better than the second-best with 8 health-per-weight. In other words, it was pushed to the point that players could accidentally make it as good as Reactive, and with a bit of skill make it outright overpowered. Certainly, the reflective modifier could be changed so it becomes less extreme, but then that just pushes it towards Reactive as an armor. As long as it wasn't passively as good as Reactive, it was too dangerous for commanders to get it.

    One of the more common complaints that I got from general players was how similar things were made to each other; Plasma was 3-slot HE; Railguns were 3-slot ER. There were certainly differences still, but most people boiled them down to something like that. But diversification necessarily requires specialization to reduce overlap between items. When I talked to pub commanders about their research (And believe me, I was doing serious 1-1 conversations over Steam with a lot of them), a lot of them were telling me that when things were specialized and diverse, they couldn't run the risk of players not being able to properly utilize the more specialized weapons/armors, so they opted towards the more generalized weapons/armor, the types that had the sole positive traits of having no negative trait and being easy to use. This leads to under-utilization of the specialized equipment, and in my perspective, there was no point in having things in the research tree if they never got used.

    Sometimes I wonder if that was the proper solution, but it was the solution I came up with at the time, and so it's what I decided to stick with. However, I never really explained my line of thought, so I decided that putting it on the forums might encourage a good discussion of balance in Empires.

    TL;DR, Read the damn post, I'm not writing a summary for a 1810-words-long wall of text.
     
    Last edited: Jun 21, 2015
  2. Candles

    Candles CAPTAIN CANDLES, DUN DUN DUN, DUN DUN DUN DUN.

    Messages:
    4,251
    Likes Received:
    10
    Trophy Points:
    0
    Reserved for potential future wall-of-text
     
  3. Candles

    Candles CAPTAIN CANDLES, DUN DUN DUN, DUN DUN DUN DUN.

    Messages:
    4,251
    Likes Received:
    10
    Trophy Points:
    0
    Reserved for potential future wall-of-text
     
  4. ImSpartacus

    ImSpartacus nerf spec plz

    Messages:
    8,598
    Likes Received:
    7
    Trophy Points:
    0
    Tl;dr, soft counters are muh nigga cus skill gonna skill and vehicle weapons aren't allowed to be unique because pubbers be derpin.
     
  5. Paradox

    Paradox I am a gigantic asshole who loses people's hard wo

    Messages:
    6,926
    Likes Received:
    148
    Trophy Points:
    0
    Can u like not joke around here. Like infract yourself for this.
     
  6. flasche

    flasche Member Staff Member Moderator

    Messages:
    13,299
    Likes Received:
    168
    Trophy Points:
    0
    spot on candles.

    thats about the point when i realized the benefits of an indivdualized research system. the hard counter thing only is bad when its team-wide ...
     
  7. Lazybum

    Lazybum :D Staff Member Moderator

    Messages:
    4,827
    Likes Received:
    190
    Trophy Points:
    0
    I just feel the need to correct you on regen, regen was the highest skill based armor before it regened from 0 plates. No commander would even think of touching it if they didn't have any good players.

    Anyhow, I agree with pretty much everything you said. I'll leave out my thoughts on what I would want for some future version of empires, but for current empires soft counters is really the only way to go if you stick to how research works. Actually I really need to finish my god damn scripts, it attempts to keep some skill based stuff while improving the soft counter aspect.

    Or one other thought is making it so when you research something, say regen, you get both a high skill version and a low skill general purpose one. High skill would have lower hp but higher regen, the more skilled player could keep driving and moving to maximize the potential of it, while the lower skilled version would have more hp but less regen, which is more forgiving to new players but won't be as amazing as the high skilled one. This effectively means commanders can get more specific stuff because there's an option for newer players who can still get something done. Which translates to having really unique equipment that can make or break games but doesn't equal an instant lose for the scrub team. Right?

    Slight downside is cluttering up the vf menu, but that feels like small potatoes. Actually now that I think about it, doesn't the vf menu have a category sort for weapons? Maybe I'm misremembering some older version... but it would fix the clutter menu problem.
     
    Last edited: Jun 21, 2015
  8. flasche

    flasche Member Staff Member Moderator

    Messages:
    13,299
    Likes Received:
    168
    Trophy Points:
    0
    your regen example, i know its only an example - why do you assume the high(er)-hp-low(er)-regen armor to be the less skill armor? i mean sure, i can see where you come from, but out of my own experience, i wouldnt judge so fast, maybe its the exact opposite, you shouldnt neglect the usefulness of regeneration ...
     
  9. Lazybum

    Lazybum :D Staff Member Moderator

    Messages:
    4,827
    Likes Received:
    190
    Trophy Points:
    0
    I'm not neglecting it. I won't get into numbers right at this moment, but we can simply look at regen and compo. Admittedly this example would work better if compo didn't have the extra buffs, but if it just had regen and the same hp/weight it's still a decent armor. Someone in regen could possibly live longer then the compo tank though, but only if they really understand to maximize their plates. If you had 2 new people fight each other in a compo and regen tank, the compo tank is going to win most of the time. You have vets in the same tanks the regen tank is more likely to win, though again they really have to know what they are doing.
     
  10. ImSpartacus

    ImSpartacus nerf spec plz

    Messages:
    8,598
    Likes Received:
    7
    Trophy Points:
    0
    It's less noticeable when we're talking about armors with HPs that most of us would consider "reasonable", but it's very obvious when you get closer to extremes.

    If you had an 80HP 18weight/plate ~3HP/s armor, then it would be a very nice all around armor. Very easy to use and forgiving, but not particularly amazing. It would operate much like a normal armor, but with a healthy regen to allow gren drivers to get back to full strength after chilling in a safe-ish spot for a minute or so (which is a long time).

    If you had a 20HP 5weight/plate 10HP/s armor, then it would operate very very differently. It would be very easy to die with that kind of armor, especially to grens, but it would dominate against certain kind of tanks. But that's the thing, it takes an experienced player to know when this kind of armor would be appropriate and how to properly use it.

    The vast majority of players (even what we consider "vets") would prefer the former armor because it's supremely versatile and very easy to use. The latter would feel useless in numerous circumstances, but would absolutely trash the competition in a handful of situations. Unsurprisingly, our regen armor looks most like the former because vehicle weapons, armors & engines generally need to be easy to use (for reasons that Candles mentioned).
     
    Last edited: Jun 21, 2015
  11. Lazybum

    Lazybum :D Staff Member Moderator

    Messages:
    4,827
    Likes Received:
    190
    Trophy Points:
    0
    Your second example only works on heavy tanks though, that's the tremendous flaw in it. Who's gonna use a light with 40 hp armor? Even it regens instantly it's still a bad idea, only good against vehicle mgs and some splash damage. IF you used my example this basically translates to getting a good late game armor at the same time as an armor that works through out.
     
  12. flasche

    flasche Member Staff Member Moderator

    Messages:
    13,299
    Likes Received:
    168
    Trophy Points:
    0
    experience != skill, though certainly necessary, its not the same.

    but i can see what you mean and i dont mean to argue about words.

    to make it less complicated, assuming a weapon dealing 99dmg, and you have a 100hp and very high regen armor, like fully regens within twice the enemies RoF or smthg.
    so its forgiving for a single, random hit and solely your movement-skills decide how good an armor would be (basically its the rail-situation in quake3's 2box4 map, you start with 100 hp, rail deals 100dmg, and have the chance for a few armor shards to get you to 110hp or smthg - its a pure rail skill map)

    totally see how this would be a "skill armor". but we will never have such extremes for the reasons candles described and thats where reality parts with idealized ideas. movement skills would make both armors better and its really depending on the final implementation which one helps your skilllevel more?!
     
    Last edited: Jun 21, 2015
  13. Lazybum

    Lazybum :D Staff Member Moderator

    Messages:
    4,827
    Likes Received:
    190
    Trophy Points:
    0
    This is where my suggestion comes into play, it gives options for the more skilled/experienced players while at the same time letting new players be effective. The main problem with balance in regards to vehicles beyond the simple faction stuff is research, everything hinges on research just as candles outlined. If you make it so research is like now, very forgiving, but at the same time offering soft or even some hard counters for the more skilled players to use you can really open up the diversity simply because the commander can't make or break the game from a dumb/unlucky guess.
     
  14. Paradox

    Paradox I am a gigantic asshole who loses people's hard wo

    Messages:
    6,926
    Likes Received:
    148
    Trophy Points:
    0
    The main question you need to ask yourselves is; do we develop the game with the idea that its balanced for high skill games or pub games?

    Thats the main problem in this entire discussion
     
  15. ImSpartacus

    ImSpartacus nerf spec plz

    Messages:
    8,598
    Likes Received:
    7
    Trophy Points:
    0
    Yeah, there would be numerous situations where you wouldn't want that style of armor. Even with six plates, you're still laid bare by a single rpg.

    But against something like du? You'd be nigh invulnerable with a little bit of effort.

    It's just an example of a hard counter (and a neat one because it doesn't cheat and use resists) and why hard counters would be really frustrating.

    To clarify,I would say that you need experience to know when to use such an armor, but a certain level of skill to to know how to use such an armor.
     
  16. Trickster

    Trickster Retired Developer

    Messages:
    16,576
    Likes Received:
    46
    Trophy Points:
    0
    I went balls deep into the skill-based counter system and pretty much found what Candles has said. Players will always take what's easiest to use over what's the most effective simply because you can't guarantee your entire team will be able to use it. It makes sense but it is sad.
     
  17. Grantrithor

    Grantrithor Member

    Messages:
    9,820
    Likes Received:
    11
    Trophy Points:
    0
    I enjoy skill based balance with a few things like armor traits, and some engine traits (standing still with AC, overheating with bio, driving with fission). As for weapons, I feel they should be mostly soft counters. Players should have a clear understanding which weapons are anti-infantry, anti-armor, and anti-building, and have slight overlap between those, but not have weapons that are just good against every thing (like UGL was, Plasma is, or HE has always been). Another problem with armor/engine traits is that they're not readily known, average players can't take advantage of them if they have no idea what they are. I have no idea what half of them are.
     
  18. Ikalx

    Ikalx Member

    Messages:
    6,210
    Likes Received:
    9
    Trophy Points:
    0
    To be honest, I thought it used to be that Empires had soft counters more often than hard. Yes, sometimes there were hard counter versions, but those tended to be patched out (ofc years in between patches sometimes) rather than the norm. If we go way back, though, the whole system was a bit fubar since heavies could only stand up to other heavies (not even counter them, really, just parallel them).

    I want to say that in earlier versions abs had a speed to damage modifier that countered rails, but it was pretty low. And the armour itself was pretty much the worst for a long time - mostly to balance out chemistry being incredibly overpowered. The speed to damage mod at that time just made it not completely buckle under a rail onslaught, is all.

    Similarly with the bio/rails thing. Sure that was there for a couple of years, but before, after and even during it wasn't really considered a good idea. Why? Because it mostly just made people avoid reactive. It's heavy, expensive, and takes a while to get. Bio weapons on the other hand are cheap, versatile and easy to use. A counter like that really isn't very logical at all.

    I feel like Empires should have soft counters. Some skill weapons would be nice, but not a lot as they're pretty confusing. I want to say also that physics used to be my favourite tree, and I saw a lot of teams work really well with reflective at that time. Yes, it's a bit skill based, but people would still make it work to at least be half as effective, so...it wasn't ever a waste. And it was relatively cheap and strong, so you wouldn't be punished too much for losing a vehicle.

    That being said, those were the days when people actually had to wait for team resources in order to get a vehicle. I think that actually taught people more respect for them, and they treated them better. Now it doesn't really matter if I suicide a tank. If it isn't a heavy then it's easily recouped in most games. Before, if you lost a heavy, it was bad. It had an impact, and people would call you out on it.

    Yeah, it's much better now that you don't have to worry so much about those things, and yet at the same time...resources start to matter less and less. It's only a problem now when you don't have them, and even then, you probably have a lot of wages stored up. We had enough wages for 20 apc's at the close of yesterday's game, just from wages alone. And vehicle salvage is a great mechanic, too.

    It's just really different now, but I don't think that's actually reflected in the game, or the way research operates.
     
  19. flasche

    flasche Member Staff Member Moderator

    Messages:
    13,299
    Likes Received:
    168
    Trophy Points:
    0
    its actually unimportant for what candles said, or the randomness due to hardcounters (assuming the same research system) might even have a worse impact on organized play. in empires it would be like you pick a single hero for the whole team in dota. with hard-counters you need way more items as only synergies can make it worthwhile in a team. if not its like rolling dice, whoever picked the counter first - and this pick is completly random as you have nothing hinting you in any way - wins the round, end of story.

    also you mix up skill with organization - less skilled organized players beat less organized skilled players. its the organization part which you need to neglect for balancing, its impossible to account for.
     
    Last edited: Jun 21, 2015
  20. Lazybum

    Lazybum :D Staff Member Moderator

    Messages:
    4,827
    Likes Received:
    190
    Trophy Points:
    0
    I know this isn't exactly related to the topic at hand, but I feel this needs to be mentioned somewhere because I keep seeing it everywhere. Currently armor costs are pretty much the same for all armors. Or in other words, their weight to cost is the same so the costs are the same.
     

Share This Page