skill tree

[erosson]

implement a skill tree, similar to d2

• except most are passive skills - this is an idle game, after all
• way smaller/less complex than poe's passive tree - too intimidating!
  • still, should not even come close to maxing out your class in a single run
• can put multiple points in one skill, but they do max out eventually
  • maybe you can find the same skill elsewhere in the tree, like poe
• lore: skills are beneficial mutations to your entire swarm's dna
  • parallel to zerg evolution, but let's not use the word "evolution". prefer "mutation".

    remove most unit upgrades and make them passive skills

• keep expansions for sure. probably keep hatcheries.
  • some larva-upgrades are important for progression for everyone
  • territory has like no other use except for expansions (and gating >0 military for your first nest)
  • everything else probably becomes a skill
• obviously removing upgrades is a huge change and will require a reset

  skill points come from resource produced by a new set of units: mutagen

• buy mutagen directly with meat, or mutation chambers (mutagen producers) with territory
  • something else to do with territory is nice, yes, but the cost ramp of expansions means it's only a few levels' difference either way. bad idea? every expansion counts since they're multiplicative though...
• mutagen per skill-point increases quickly, just like exp per level in most rpgs
• eventually, you hit a max # of skill points
  • prestige will let you perm skill(s) after reset, bypassing the point cap
  • after reset, you can have permed skills from another class!
  • eventually with enough resets, you can perm all the skills - but this should take years, like kol. or at least months. absolutely no multiple daily resets.

    "classes": different skill trees with different specializations

• larva production, everything-but-larva production + cost reduction, twin units, (more...?)
• implemented as separate trees/skill-tabs
• not exclusive; ex. there will be twin upgrades in the larva production tree, and larva production in the twin tree, but fewer/lower quality than the class itself
• pick a tree at start, when you get first skill point - around nest phase (because that's when people are quitting today)
  • (don't pick immediately on character creation, of course: i want to start playing now now now)
  • you are locked into that tree for a while
  • late game, when multiple trees is less information-overloady for newbies, you can switch trees
  • or not - make skills-from-multiple-trees a perk that only comes from resetting?

    why do all this work?

• bursty upgrades lead to gameplay variety
  • today, as with most incrementals: you unlock all upgrades as you can afford them, as long as they're don't wipe out your bank
  • gameplay feels the same at every phase of the game. "going through the same motions."
  • with skilltrees: you'll unlock (ex.) 5 larva upgrades right next to each other, then 5 meat upgrades right next to each other, then some awesome uberskill those last 10 points unlocked, then... etc. all of these phases will feel different, discontinuous
• classes allow different playstyles
  • there will always be an op class/strategy, but "op class" ought to be easier to balance than "op strategy"
  • downside: newbies may feel stuck because they pick the wrong class. we'll have to balance carefully
• upgrades are boring. players must make interesting skill choices
  • skill points are a scarce resource, with many ways to spend it. think carefully
  • compare to upgrades today: just buy everything you can afford, no real choice
  • there is some choice in what units to buy today, larva limits force that, but upgrades? just buy everything, steep cost ramp allows/encourages it
  • surely there will be optimal strategies, eventually players will read other players' guides with optimal skill choices, eventually this becomes "no choice"
    • that's okayish. today's "buy everything" is way more dominant and way more obvious/natural than a few op skills could ever be (right?)
    • our game is so small that the "no choice" phase won't ever arrive, or will take years. poe/diablo/wow/sc2/... have millions of players pruning the decision trees/searching for optimal strategies, we won't ever be that big
    • nah... our game is so much simpler than those big games. it'll happen.
  • why not use wow's mutually-exclusive-talents system? way simpler, less minmaxable, they tried talent trees in earlier versions and scrapped them
  • "way simpler" is actually a negative - this will be one of our main forms of scaling/progression, we don't have levels nor loot
  • can't make that work with prestige
  • still, we could stick a few mutex skills in the tree
• prestige system is straightforward
  • it worked for kol

    potential problems?

• scaling/running out of skills
  • upgrades scale infinitely. skill points are limited to what I define, just like unit tiers
  • maybe we could have some infinitely-scaling skills, but "infinite" isn't unlocked until very late game? until late game, individual skills capped on # points spent elsewhere in tree, like torchlight?
• defining/editing the skilltree might need a custom editor
  • spreadsheet definition is awkward - we'll probably want a custom editor for node positions, at least
  • maybe we could hack inkscape into doing it
  • cell position -> node position would be enough to have a spreadsheet define a grid-skilltree, like d2, without a custom editor. couldn't do freeform poe-skilltree that way - that's probably okay
• players will feel nerfed after we remove upgrades
  • "why not have both upgrades and the skill tree?" because upgrades are uninteresting, just buy them all - part of this is forcing interesting choices and removing uninteresting ones
  • especially nerfy very late game, since you can buy all the upgrades but skill points are far more limited/force choice
  • deal with it. that's why this'll need a reset and why we have a beta
  • not sure this is a good answer... late-game players will be pissed that they can't buy everything anymore, beta or not
    • emphasizing prestige/resets might alleviate it - spend less time in lategame; improve lategame by playing more earlygame
    • the whole definition of "late game" changes once we add prestige/resets, doesn't it? I think this is our answer
  • skills can't compare lategame since choice is required, but early-game we should try to make them pretty similar or better in strength to feel less nerfy
  • we could never remove upgrades after v1.0
  • should it be a sequel instead of a reset + content-patch? maaaaaybe. it's an awful big change. let's keep thinking about that.

[erosson]

this is ancient. mutagen is the natural place for a skill tree, if I do something like this, and it'd be a big content patch.

also, if I did replace upgrades it'd likely be with some sort of equippable/craftable loot, not a skill tree - but really, after 1.0 upgrades are probably here to stay