FW bonus

Discussion in 'General Discussion' started by DiCEM0nEY, Jul 31, 2017.

  1. themacca

    themacca Master of Challenges

    I love reading kalasle arguments when they aren't disagreeing with me.

    That's my only contribution to this thread aside from the usual it's fine can we please drop this ridiculous topic
     
  2. DiCEM0nEY

    DiCEM0nEY I need me some PIE!

    Example of good FW play which uses the bonus to create value:

    1st deploy bone elemental, charge to mid font. second deploy tomblord.

    Bone elemental creates X lesser maulers, and deals Y damage to b an opponent. Tomblord deals supporting damage safely from behind. bone elemental dies, then is immediately redeployed when CD is off. Conditions are great for bone ele, there's no source of indirect damage, opponent is overwhelmed, you win.

    IS bonus: you take a bunch of spell damage, then retreat.

    SL bonus: enemy broken bones can not damage your skeezick resilient

    these are optimal conditions and optimal use of bonus.
     
  3. kalasle

    kalasle Forum Royalty

    Could you please elaborate on how what you described is a matter of playstyle -- it seems to me as though, especially in the FW and SL examples, you described runes being good in particular scenarios, rather than something about the bonuses more generally. What makes these "uses" of the bonuses in an active sense, rather than merely instances where the bonus passively affects the situation?

    I'd also love the other questions to get responses, if you have any.
     
  4. super71

    super71 I need me some PIE!

    Fw bonus is fine, i'm shocked this went 3 pages and surprise no side has found a solution. I think the problem on these forums is were quick to jump on an opposing idea to seem intelligent, but very rarely do either sides find a middle ground or a solution. @kalasle @DiCEM0nEY If you guys put as much time into talking about different bonus ideas as you do arguing about nothing we might have solved our problem already.

    Also @kalasle Do you think the current bonus is fine, if so then your not going to be swayed because your mind is made up.
    @DiCEM0nEY I don't see the fw bonus negating other bonuses being much of a bonus, I see it as just a nerf to other factions bonuses. What did you think of my death benefit bonus, what other bonuses can you come up with that you feel would fit in theme ?

    I'd like to see this thread actually go somewhere versus the typical argument and thread closed which is where we seem to be headed.
     
  5. kalasle

    kalasle Forum Royalty

    I'm not really looking for a solution to anything, mainly and especially because I'm not making a claim as to whether or not there is a problem. I figure deciding on problems and solutions is up to Sok and the greens. I'm talking to DiCE and other people in this thread to discuss how we think the FW bonus functions in its present state -- how strong is it, in what ways, and what various things does it affect. I have opinions about the FW bonus, but I not here with the intent to argue in their favor. I have disputed anyone else's reasoning vis a vis changes to the FW bonus to the extent that I think they misunderstand, descriptively, the current bonus.

    Basically, firk this goal-oriented idea that players are designers, I'm here to talk about the game.
     
  6. super71

    super71 I need me some PIE!

    Do you or do you not like the current fw bonus ?
     
  7. kalasle

    kalasle Forum Royalty

    I mostly like it, I guess.
     
  8. super71

    super71 I need me some PIE!

    Still not really a yes or no, but i'll take it.
     
  9. DiCEM0nEY

    DiCEM0nEY I need me some PIE!

    You can't give a definitive value of the FW bonus in every scenario, nor can you of any stat in every scenario. For example, the current costing of flight is X value. But if EVERY map has a cliff you would feel X is too small relative to its power. But if EVERY map was flat with no terrain modification, you would feel that flight was definetly over priced.

    You can speak as generally as you want about bonuses, but know that they ARE in fact based on the scenario. I was simply showing how to use the bonus in an ideal scenario.

    Of course, you can use the bonus in a less than ideal scenario. For an easy example, an SL takes a majority of his damage through loss of life. An SL unit takes 3 hits of Soul strike damage. A ranged FW unit is sacrificed because there is no better option present.

    The deck you choose is categorically what you describe as a playstyle. If both players play a deck to their full potential, they are utilizing their calculation abilities and risk management (play) with the runes they have picked (style).

    Most importantly, you are looking at the game as if a majority of it has not already been decided before the game starts. You have map, matchup, draw, and initative to consider right when the game starts. There is no "passive" component of the bonus to utilize when you are playing competitively (optimization of win rate). You simply use all tools at your disposal as best as possible to win.

    The fun part imo, is when we make plays that the other player didn't see coming. For example, prediction, positioning, or surprise, such as unexpected and underused cards.

    The reason why I push for the change, is because the change would allow for more fun and less gimmicky playstyles, while probably not completely removing any opponents playstyles.

    @super71
    Bonus would work better if the stat bonuses were based on nora cost, and applied y times max. Y being between 1-3 and based on unit nora cost.

    But the bonus would be as gimmicky in design as current. It would be an unnecessary change, I have no opinion as to whether I favor it the current, I just see it as different side of the same penny coin.
     
    Last edited: Aug 10, 2017
  10. kalasle

    kalasle Forum Royalty

    I am not asking you to do that.

    You still haven't given any explanation why ranged/melee is relevant to your understanding of the FW bonus, so this example doesn't make any sense to me in any way.

    See my earlier question about what makes something a "use" case rather than an "occurrence" case.

    Ok, so, this is an interesting and potentially substantive idea, so thanks (even if I am not immediately and whole heartedly going to agree). It does not, however, seem particularly coherent with respect to some of your other comments -- for instance, about the SL bonus or even your example for the FW bonus. Your examples about the SL bonus in relation to loss of life and runes like Broken Bones seem to have nothing to do with the SL player's deck choice, and instead much more to do with the opponent's. How is there any optimal burden of "use" in the active sense on a player when the important factors here seem to be totally out of their control.

    Ok, fine, we can break things down like that -- so going back to your earlier comments about...
    what part of play is optimal here? What is the player actually doing in any of your examples to use the bonus in an effective way? Your earlier language if "runes...picked" constitutes this "style" part of a playstyle then, once again, what is optimal play? Here's the basic claim about which I am trying to ask all of these clarifying questions: "In the event that a FW player 'plays optimally', the sooner that their champions die, the better." What does "plays optimally" even mean?
     
  11. DiCEM0nEY

    DiCEM0nEY I need me some PIE!

    You are missing the one of the more simple concepts: there is no such thing as a "passive occurrence " . Things are happening for a reason; the bonus will always use a cost, whether it is used or not.

    In an ideal design state, it will have a similar value to an opposing players bonus a majority of the time. Why? Because it would further incentive building a variety of different decks, and allow for more open ended playstyles.

    What I mean by both players playing optimally is that the runes were being used in the best manner possible. It's obviously easier to measure the value of something if you set controls which was the reason I gave such simple examples. Obviously, in a real game, there's ways around simply attacking a bone ele, or casting a bunch of damage spells on an IS unit.

    The FW bonus DOES give a strict (or array of) multiplier to the value of each rune. The amount of calculations needed to know this number, and the amount of data needed to predict this value, is definetly out of my paygrade. But I promise, SOK simply ball parked the value of immortality and adjusted.

    This is (again) because in a completely competitive (win rate optimization) your deck (or select few decks with equal chances to win) would be predetermined, and your moves and tactics would also be predetermined (what gives the highest chance to win).

    There are AIs that can do precisely this, through trial and error and repeating things over (many more times than we could) and over until they have 100% win rate optimization. I know it accomplished amazing and funny results for mario and starcraft 2, although I can't remember the aI name.

    So, that pretty much shuts your argument about use vs passive use of the bonus down I hope. As to whether or not it would affect design, I think that it would, since there would be fewer conditions needed to see returns.
     
  12. kalasle

    kalasle Forum Royalty

    To spell it out for anyone else who has bothered to read through all of this stuff, and because it will be a lot easier than trying to work you into saying, my contention about the FW bonus vis a vis its effect on play and deck construction is this:

    No, the FW bonus doesn't do much to directly affect the contours of champion tactics or strategy, contrary to your earlier assertion to that effect, because the very efficiency on which the fundamental power of the FW bonus is predicated depends upon each champion maximizing its value while alive. In that case, retreating, staying alive, and avoiding over extending all remain important components of FW tactics and strategy under the influence of the bonus. Let me explain further.

    "Use" of the bonus is a two-part process. First, it means to a greater or lesser extent building with the bonus in mind. It's not required that a deck be built to take maximum advantage of the bonus, but compared to other bonuses, FW's has a larger range of value and so there is more incentive to consider it. (It is possible to build with the respective faction bonus in mind when playing other factions, but most of these methods boil down to a variant of "play more summons and cheaper champions"). We've already talked in this thread about the kind of champions that work better or worse with the bonus. The second part of the use process is the in-match strategic and tactical part -- the player should play the right champ for the situation (this being more important in FW than elsewhere), and in those situations make sure that a single champion life provides as much value as possible. When I asked "What is this "optimal play" that transmutes "early unit death" into the appropriate use of the FW bonus?" this is what I meant. Champions dying earlier rather than later is relevant in the event that they have been used to their fullest tactical potential -- a requirement that runs contrary to running them into early deaths. In theory, the most efficient champion is the one that never dies and constatly generates value by gaining AP, dealing damage, and absorbing opposing resources. As a result, the FW bonus does not necessitate by extreme reward the immediate death of FW champions; rather, the way in which it operates depends upon the *opposite* of an early and purposeless death. This contradiction creates a tension in FW play between wanting to needing value through longevity and micro-tactical play, and wanting to amplify that value via the FW bonus by exploiting champion mortality.

    Now, there are some indirect effects of the FW bonus, namely, that because the bonus depends upon champions dying, you want to build a deck such that champions yield their efficiency by means other than focusing on longevity. This generally puts a cap on an individual champions value over the course of its life, or at least makes it much harder to stretch out their value, and as such may modify the circumstances under which one would choose to sacrifice a champion by virtue of the champions selected and the deck built, but it does not fundamentally make sacrificing champions a good decision. Basically, think of champion use as a set of vectors representing tactical decisions -- the FW bonus will modify the magnitude of the vectors without altering their direction.

    If you or anyone else has some substantive critique or response to this conception of the FW bonus, then we can keep talking, but until then I am done with this conversation.

    And two other things that don't matter that much by now:

    You say this, and then...
    in the very next sentence you incorporate the existence of the dichotomy you just suggested does not exist.

    The FW bonus had its current form before Sok was even an assistant dev under SOE, before a costing formula even existed. That I have assigned it a value within the current costing formula means nothing about its historical development. The earliest version of the FW bonus was -6/-3 to champion CDs, and the -60%/-30% was an adaptation of that bonus in response to critiques that the flat values disproportionately helped cheaper champions, and had very little effect on champions that cost >~80 nora.
     
    themacca likes this.
  13. DiCEM0nEY

    DiCEM0nEY I need me some PIE!

    Here is the point you are missing. In theory, a champion which could generate ap and consume all of the opponents resources would never exist, because that would be atrocious design. If you were more open minded, you would see that I am right.

    Picture a scenario where the game was completely balanced, and all runes were exactly the same. However, the other person starts with +4 damage on all of his runes. The longer the game goes on, the more the value the +4 damage gains. Pretty soon, you can't protect your shrine, because he has an inherent advantage.

    Now pretend you have the FW bonus. Opponent still has +4 damage to all units. If no units die, you would still lose the game, because the other player has more value on his side of the board.

    What the FW bonus DOES is allow you to use units in a manner which the opposing player can not. By having reduced coolbdowns, you are able to "suicide" a unit, then recover the unit and use it again. The real value gained only begins to occur when the turn your unit comes back into your runedock. And the value gained from the reduced cd decreases each turn your runes are off CD.

    So, there is very much a time component of the bonus you are not factoring in. And yes, the dynamics of the bonus do change once 30 runes have been revealed, but let's not bother to include this, since most games end well before this.
     
  14. kalasle

    kalasle Forum Royalty

    This is an interesting facet of the bonus, so I'm going to talk about it for a bit.

    It makes intuitive sense that the following would be true: you get X turns of CDR from the bonus, rendered abstractly as "X value", and every turn after the champion comes off its modified CD without being deployed, you get "X-1 value" because you are only cutting X-1 turns off of its expressed CD. The X-1 formulation of CDR value assumes that the value of a CDR bonus can be understood by comparing instances of that bonus to functional alternatives -- that is, a bonus of X in this situation could be substituted for a bonus of X-1, and is therefore only as good as a bonus of X-1. I do not think that this intuition holds up under scrutiny. What is relevant with respect to CDR is not whether a rune can be played immediately but whether it can be played at a time when it is valuable. Contrary to the intuitive formulation, I think it is better to consider the value provided by CDR as a binary, between [play] and [not play].

    One important thing that gets elided from the X-1 formulation is that, even though it would be possible to deploy a champion to use X CDR, it is better to deploy it to use X-1 CDR (or, framed as turns, better to play it on turn Y+1 rather than turn Y). The intuitive X-1 formulation would see this decision as inherently depreciating the bonus, but I don't think that this is the case; what is chiefly relevant is whether or not on any given [turn of valuable deployment Z] it is possible or impossible to deploy the particular champion. A [turn Z] means a turn on which, given either some set of options A, B, C ... or that set of options + Z, the player would choose Z. This sort of formulation of CDR (which for ease of future reference I'll call a Z binary formulation) helps resolve one of the apparent problems with looking at CDR bonuses in strict binary terms: namely, that the binary formulation doesn't provide an avenue for explaining how greater CDR is better than less CDR. (There is another method of explaining this, but I will get to it later, and I think it's weaker regardless.)

    In the Z binary formulation, you can define the gained value of a CDR amount as whether or not it makes the particular champion playable on a significant turn Z. The X-1 formulation can be thought of as a subset of the Z binary formulation, but one that is only relevant in the event that every turn through Y -> Y+X is a turn Z (in which case, the entire idea of losing value becomes nonsense, because the player will obviously just deploy the particular champion on Y, given that Y is also a Z turn). In the likely even that not every turn is a Z turn, how should we talk about the value provided by CDR within the Z binary formulation of value? Lets look at an example and talk about it.

    Over the course of its CD time, a champion named Alice has say 3 Z turns, and 1 Z turn after its default CD, on which it would be the best resource investment. One of these is on T2 (second turn after champion death), one is on T10, one is on T14, and the final is on T20. The champion has a default CD of 15, and benefits from 8 CDR, leaving it with a final CD of 7. Before CDR, its first functional Z turn comes at T20; After CDR, its first functional Z turn comes at T10.​

    In the above example, Alice moves two Z turns forward as a result of the CDR, and the difference between its original and new functional Z turn is 10 turns. Now, it's worth noting that the benefit afforded by the difference of Z turns in these examples is completely arbitrary. Similarly, the difference between Z turns as manifested in real games is going to have only a passing relationship to a champion's CD before and after CDR. A champion could benefit from a Z turn difference as small as 1 turn, or as large as literally thousands of theoretical turns. For this reason, I don't want to start getting into analyzing Z turn differences for effective value as though it's a substantive topic, because the idea of a "Z turn" is heavily abstracted and any more precise analysis would be mostly nonsense.

    What it the example does highlight, however, is when CDR becomes relevant, and when it doesn't. Having 3 less CDR, for 15 CD - 5 CDR, would make no difference to Alice, because she would still be deployed on the same Z turn of T10. Likewise, gaining 5 more CDR, for 15 CD - 12 CDR, would also make no difference; Alice's nearest available Z turn is still T10. Gaining more than 5 or losing more than 3 CDR, however, would mark a discrete shift from one Z turn to the next, something like a step function. This suggests that, contrary to the idea proposed in the X-1 formulation that a champion loses or gains value from the bonus continually and in proportion to the number of turns it isn't deployed, changes in CDR are functionally worthless unless they reach a Z turn, at which point the champion will be played sooner or later. The X-1 formulation is merely that statement, but with steps up or down at every single turn. It is within the framework of turn-by-turn option sets, described earlier as A, B, C... + Z, that CDR gets rendered as real value. In this sense, the Z binary formulation is binary is the [play]/[no play] dichotomy, but discrete and distinct when looking at value accrued on various turns.

    The end result is a conception of turn-by-turn importance that looks similar to that proposed by the X-1 formulation, but with a key difference: yes, in a particular case a bonus of X-1 CDR might be equivalent to a bonus of X, but this is because the extra turns had no value in the first place in relation to the Z turns, not because some hypothetical value has been lost. It is silly to assume a comparison in value between X CDR and X-1 CDR based on single examples, because what is significant is how frequently, on average, some amount of CDR will allow access to new Z turns for some particular rune.


    Now, this whole analysis does bring up one other thing about the FW bonus as-is, namely that it is possible to see no material value from it, unlike the other faction bonuses (excepting poor IS). By material value, I mean that no effective numbers within the game change as a result of its presence -- the bonus could have been absent, and the same sequence of plays would have resulted in the exact same game state. This is the alternative metric by which I suggested one could resolve the apparent problem of differing values of CDR in the binary formulation: lower CDR values make it more likely that you will hit the [no material value] state sooner, and thus will not see the [play] benefit within the binary. I think, however, that such a conceptual method of addressing the problems with the basic binary formulation is less robust than using the Z binary formulation of CDR value.

    I want to make one other point about this situation of no material value, though, a point which has more bearing on the IS bonus (and which is incidentally solved with respect to the FW bonus by using the Z binary formulation, and specifically using option sets as a way of talking about value within that formulation). This point is that, despite saying that the same sequence of actions would lead to the same numerical results, a change in options can influence what decisions people make in the first place. This is a point that I made obliquely about the IS bonus earlier in this thread, but it has been made most explicitly to me by @Markoth about the IS bonus: merely having the IS bonus means that people will be less likely to use AoE spells at all. If a player never uses a damaging spell, the IS bonus could numerically not exist; strategically, however, the presence of that bonus likely affected the pre-action calculations that led them to not use damaging spells. This is a big reason that, despite the main irrelevance of the point to the FW bonus and CDR specifically, it is worth talking about value and decision making beyond just the material value yielded in situations where all actions remain the same.
     
  15. kalasle

    kalasle Forum Royalty

    Maybe I'll talk about this later, but the above post was enough writing for now. I haven't thought much how the CDR actually interacts with Pox's reveal mechanics prior to full rune reveal, but I would intuitively agree that the matter is not significant in relation to other factors. As a matter of empirical fact, however, I would completely disagree that most games end prior to 100% reveal. People have often used 30 turn / 15 rounds as a standard for game length, which I have considered too short especially when talking about FW games, but even using that standard, each player would have full run reveal a few rounds prior to the game ending.
     
  16. DiCEM0nEY

    DiCEM0nEY I need me some PIE!

    Thanks for the post. I understand your theory, and I still believe it to be wrong, for a number of reasons. Obviously you know the difficulty of writing about these topics when we use very precise language, so it's going to take some time for me to respond. (I'll do a double post)

    A lot of the concepts that you mention I did take into account in my original understanding, but I feel we are getting closer to actual achieving new (hopefully) knowledge on both our ends.
     
    Last edited: Aug 13, 2017
  17. potatonuts

    potatonuts I need me some PIE!

    Even though i strongly dislike the backloaded nature of the FW bonus I think it's synergy with the faction makes up for it if you can make it into the mid-late game.

    With the current state that the game is in I don't think any major changes like this would be appropriate but in the future I wouldn't mind seeing it tweaked so it can be of some use in the early game or against decks which don't suffer from long cooldowns.
     
    DiCEM0nEY likes this.
  18. DiCEM0nEY

    DiCEM0nEY I need me some PIE!

    What do you mean by appropriate / inappropriate?

    I actually think that now is a good time. Players want a reason to play vampyres, I doubt 1 rune will be enough (nor should it be).

    Plus, even though I don't really think it is deserved, I smell bad design and nerfs coming towards many current fw themes.

    FW will probably get major nerfs this patch. There are just too few players, and too much visibility on FW.

    I have always said that things like skeletons should be nerfed, I only argued that it wasn't a good time to do it when it was done. Since medal of valor isn't in the game right now, skeketon decks have seen a major increase in visibility and viability. I promise sok doesn't realize this, and will throw in random skelly nerfs.

    Truthfully, this band aid philosophy is what causes bad gameplay. Hesitating to make major changes is what causes the game to become stale and dated.
     
  19. kalasle

    kalasle Forum Royalty

    Could you elaborate on what you mean by this?
     
  20. yobanchi

    yobanchi I need me some PIE!

    Does FW deploy more nora in champs then non-FW on average over a 20 round game in total?
    Do bgs run out of deployable champions over a 20 round game?
     

Share This Page