|
Post by kaldar on Dec 7, 2016 7:29:44 GMT -6
Seitaarin, so your argument is that since you managed to beat Faile one time with a damage build, that means there isn't a problem?
Why not go try 4 or 5 times and see more of a win % pattern there ... you are good at math I can tell. You should know better than to draw statistical conclusions from a single data point.
And I agree that melee builds should not be unstoppable. I just also think that a melee build should not be only 2% effective against a build with nothing more than a few TAs thrown on garbage gear.
|
|
|
Post by kaldar on Dec 7, 2016 7:30:07 GMT -6
Seitaarin, so your argument is that since you managed to beat Faile one time with a damage build, that means there isn't a problem?
Why not go try 4 or 5 times and see more of a win % pattern there ... you are good at math I can tell. You should know better than to draw statistical conclusions from a single data point.
And I agree that melee builds should not be unstoppable. I just also think that a melee build should not be only 2% effective against a build with nothing more than a few TAs thrown on garbage gear.
|
|
|
Post by kaldar on Dec 7, 2016 7:30:41 GMT -6
dangit I hate this forum .. not even going to bother editing and I can't figure out how to delete duplicates
|
|
|
Post by Tim on Dec 7, 2016 9:27:46 GMT -6
Let me try to summarize this so we can all move on: - Seitaarin believes the balance is mostly fine, as health gain is serving as a counter to damage builds. He duels a lot and thinks he knows better than Ger and Kaldar how things work due to his own number crunching and build testing. - Ger and Kaldar believe health gain is way over powered compared to damage builds and should be MORE level (though they are fine with health gain/poison still having some advantage over straight damage. They have a lot of experience in the game and have done a lot of number crunching to figure out the balance of things, as well as quite a bit of dueling this age and think they know better. - I think you're all good players and I just want to hear all of your opinions, but am getting really tired of waking up/coming home to see this thread full of posts about "You are wrong because you haven't played/dueled/experimented/mathed enough" or crap like that.
I'm fine with data points but seriously guys this is getting ridiculous. One day or week or even month worth of dueling isn't enough to convince me that "Yes. There is a problem and I need to change things." But if someone has an opinion I want to hear it. I'll then weigh it against other opinions and most importantly my own (sorry, but that's how it works). But I promise to listen and look into anything someone raises as a potential issue.
So all of you need to chill out. I honestly don't care if you respect each other, but this forum is a place for respectful discussion opinions about the game and was we can improve it. At this point I don't want to hear anymore digs at one another in this thread (or other discussion threads). If you are petty enough that you feel you need the last word, do it via private messages.
So let's move this discussion forward. If you have something new to add, please do. Just be respectful and try not to retread old points. Again: You don't have to convince each other of your opinion. Just put it out there and we can try to discuss it. But if you guys start making this personal and discrediting the opinion due to who is making it, I'm going to have to put an end to this.
|
|
|
Post by kaldar on Dec 7, 2016 11:11:31 GMT -6
I could be wrong but the only person I see making personal digs is Seitaarian. Ger and myself are mainly trying to defend against points he is trying to make.
These are my points with regards to of Tinkers : 1) It has the highest health gain of any talisman in the game. It has 1 less dodge than the highest TA in the game. It has 1 less than the highest Speed in the game. The penalty it has for this is not a penalty for the builds using it. 2) I can't see any other TA in the game that is best in so many categories. The closest I can find is possibly of the Shadow. 5-10 damage vs. highest of 11-13; 10% damage vs. highest of 18%;10 poison vs. highest of 12 poison. However, this TA has no offsetting penalty so is , I believe, the most expensive TA in the game to use. 3) There is no +2 speed combination TA for offensive builds to counter it. 4) Look at another example of inbalanced speed options : Scouts - +3 Speed, +3 Dodge, +6% defense vs. of Toraken - +1 Speed , +2 accuracy, +4% damage. The Scouts is better in each and every category for defensive builds than of Toraken is for offensive builds.
Also, The Scouts works very well in the defensive builds because it can go in a prefix slot while the of tinkers can go in a suffix slot. For the offensive builds, of Toraken is not that great and would take the place of an of the shadow or of Kings ... not a very good trade off to boost speed by 1 point.
In other words, if an offensive build wants to boost speed with TAs, we can use Rapid for a prefix of +3 speed (we lose 5% damage) vs. the defensive builds option of scouts and +3 speed with some good additional combinations.
We can use of Foxes for +3 Speed , while as already noted defensive builds have the "what would I want in a TA if I designed one with every good thing for my build" of Tinkers
And don't get too cranky about the thread Tim ... prior to it I don't think anyone had posted on the forum in months. So generally, when people do bother to post on here it will be something they feel passionately about ....
I think the offensive counter to of Tinkers would look something like : Of The Dragon - + 2 Speed, +5 accuracy, +15% health gain, -30 Luck.
But of course, then it all gets a bit silly.
Maybe part of the discussion we should be having is why we spend all these class skill points and the sum of all of it is less than what we can get on a few TAs ?
|
|
|
Post by Tim on Dec 7, 2016 11:27:00 GMT -6
Yeah. I agree that Tinkers is somewhat OP compared to the other TAs. When I last added a bunch of TAs, Speed and Damage combo was already king, so I felt wary of adding any that would strengthen that too much. I also don't want to make more TAs that are obnoxiously strong. Really, if I do 'weaken' Tinkers, I'm leaning towards cutting the penalty from -30% dam to -20% dam and maybe cutting the health gain back from 15 to 12. Not a huge nerf, but it will make it more costly to use and I don't like that 5 of them let you max out your poison stat with no other health gain from your build. Depends on what other tweaks I make though.
One point Seitaarin has alluded to but hasn't flushed out though is that I'd guess 90+% of all players right now have builds more focused on straight damage than a defensive poison build that can take advantage of health gain. Few players focus on dueling (unfortuantely) and straight damage is the easiest thing to set up to win vs NPCs and Hordes. I definitely don't want to discourage players from mixing up their strategies and defensive poison builds are really only effective at countering aggressive damage builds (and they do so VERY well right now). But other than winning duels against those builds, there's not a huge advantage to going that route. Ideally, if I could make more diverse builds viable and less players used the straight damage builds, defensive poison would become even more of a specialty build than it is.
|
|
|
Post by gerdonat on Dec 7, 2016 12:12:47 GMT -6
In addition, I think Health gain should have a counter, could be taint, could be something like raw health gain instead a % of what the other does, with the possibility to go over your health max so you dont waste it when someone does no damage. That could open the possibity for a build to do not damage at all but win because it increased his health more than the damage they received. Or could be a lower cap, something like 60% max health gain. Right now, 2 of the most powerfull mechanics, health gain and speed are really hard to use with a damage dealer, and i m not saying that we should allow damage dealers easy access to health gain and/or speed, but we should give them some way to counter it, sacrificing damage... Maybe a way to revitalize wound? What if wound in addition to lower speed lowers max health gain? makes some sense, that way you dont touch the turn calculations but give some more punch to wound that is much weakier than speed.
Taint not being healable, or being partially healable? taint having a second effect of lowering max health gain as Seitarin said? with the counters I m proposing, if you are a damage dealer you need to diversify for dueling, opposed to someone that wants to do only damage to hordes and could get the bigger numbers.
Anyway, there are tons of possibilities, I really think the game its pretty well balanced for the most part.
@tim, I built a battle simulator, where I throw the numbers of two oposing builds, and make 100.000 duels, and gives me stats from the battles (% of wins of each build, number of attacks on average, average damage, etc), I m willing to give it to you including source code if you want to have a tool to test the tweaks you plan to make. Would improve the test a lot and reduce the chance of introducing some unintended consequences.
Anyway, my interest is to improve the game, not to advance my agenda.
|
|
|
Post by Tim on Dec 7, 2016 12:49:02 GMT -6
The battle simulator sounds quite useful if you don't mind sharing it with me. Not sure if I necessarily agree that health gain and speed are hands down the best mechanics, but the fact that I can't think of another stand alone mechanic that I can say is better off the top of my head makes me think they are at least towards the top. There are combinations that I think can deal with either fairly effectively (speed at least). In my experience, a lot of stun with a touch of first strike counters speed pretty effectively. Still relying on some luck with the RNG, but stun's a lot cheaper than speed. Health gain we've already covered as difficult to push through.
|
|
|
Post by Seitaarin on Dec 7, 2016 22:58:05 GMT -6
Here's a balance idea - randomly generate AC's for everyone at the appropriate levels. We can all then play the chips as they land. Maybe choose 40 and 80 to start with, but randomise 20 and 60. Within nation/other restrictions of course.
EDIT: You would, of course, have the power over where you allocate points within that AC, and itemisation. Having a generated AC could potentially be the pattern's events of your lifetime unfolding, nudging you in a particular path for it's own weave's purposes. Please read all below information before replying to this idea, there are some merits to the idea itself I think, even if this is not the perfect implementation.
|
|
|
Post by Seitaarin on Dec 7, 2016 23:10:26 GMT -6
EDIT2: downside of that idea is the natural terrible point allocation efficiency of the 'focus' classes. But I guess those could be all opened up as starter options and/or removed from the AC random pool ... the other focuses are underused in any case, might be a good thing?
On that ... can we also get a -shield% cost TA implemented
|
|
|
Post by Seitaarin on Dec 8, 2016 0:00:28 GMT -6
Before shooting it down offhand, let me give a brief explanation as to why this could be of benefit ... this is where it gets a little bit interesting I think. Being able to choose your classes part and parcel creates this idea that the character is purely yours to control, a fact compounded by the fact that NPC are murderously easy. Basically folks tend to be rather good natured about losing duels early in the piece, because they have not chosen classes, or put any of their own polish on the character as such ... even NPCs are a bit tougher at the start, so losses in general are expected. As the game progresses however, and choices are actively made, people identify more personally with their characters. This is aided by the whole npcs getting easier and easier thing. Now, some players see things a bit differently. They think to themselves 'what do I need to choose to give me the best chance at beating that build, that build and that build' ... Side note here, when all of those players have the same build, the choice becomes a ridiculous degree easier. These players aren't building what they want to build, they are building specifically against what YOU (royal you here, referring to whoever) build, so already we have an unbalanced situation - Simply put, every single build cannot be completely balanced against every single other build. Not only is it not possible, but the game would be no fun even if it was, and acting as if every build should be a counter to every other build is extremely counter productive. I'll sidebar here to speak in an even, rational tone aimed at the OP about the poison build ... The concept of 'shortsuiting' yourself in order to raise effectiveness in one particular aspect of the game by sacrificing effectiveness elsewhere should not be squashed, I'll explain why in a sec. The Tinkers TA is analogous to that almost perfectly. If you 'balance' things like Tinkers out of the game, what will happen is you are removing potential variation. Yes, it's the 'best' for a poison build ... it's still by no means unbeatable despite protestations otherwise, both past and present have shown this fact to be true, that is part of the argument that I will not ever let go. To the shortsuiting, no matter how it's sold, a poison build by nature sacrifices effectiveness at other vital parts of the game in order to be strong in duels. The game wincon is decided by the LB mechanics, the majority of points during which come from NPCs, not from players. Add to this the fact that in order to win in duels, a poisoner often does not win by a large amount of HP due to limited damage, and the point potential from winning those duels is also a pretty dry well. The only penalty for losing a duel is cash and pride - nobody is forcing you to hit back on a dueller and waste your turns, especially one that has built such that they are a probable 80 percent win rate against you if you don't want to go to the trouble of building in such a way to counter THEM. And if you don't want to make the changes necessary to beat THEIR build as they have done to beat YOURS, then you don't deserve to get a win point up on them anway, no matter how you spin it. These facts are not irrelevant. With that all in mind, back to the original idea I put forward, having some sort of randomness to the AC's that any given player is able to use during an age will remove a little bit of the pointed counter build option, making things a little more uncomfortable and unpredictable for everyone, sure, but making things a degree of difficulty higher for players who want to counter other players. RNG could be very unkind in some circumstances, but such is life. Cranky Seitaarin bit, stop reading here if you don't like inflammatory comments The RNG will also give whingers and folks that are just plain negligent or don't care about their builds something to point at and say 'it's RNG fault'
|
|
|
Post by Seitaarin on Dec 8, 2016 0:30:36 GMT -6
Fleshing that idea out a little further, an argument could also be made about certain classes ONLY being available via RNG, while maintaining lore adherence at the same time.
Some examples, but not limited to:
> Channeller - Actually this is probable the most interesting of the limitable options, because there could be an RNG assigned 'natural channeller' option, that perhaps maintains the +1 water/air or fire/earth, or a selectable AC of 'learned channeller' that is the same thing sans gender bonuses.
> Wolfbrother - You're either born a wolfbrother, or you're not, it's not a choice.
> Treesinger - as above
> Seeker +/- deathwatch guard - a bit different, you are actually chosen, or born into these roles if you like, but again really not a personal choice.
> Thief-catcher - could be rebranded or added as thief-taker for another accident-of-birth option
> a bit more of a stretch, but an argument could be made to include gambler (it could be considered a natural compunction as opposed to a true choice), wise one ... actually that's it that springs to mind at the moment.
As it stands, seems to me that the lvl 20 AC seems most appropriate (to pay tribute to birthright skill classes being awoken). Further expansion beyond that point though ....
|
|
|
Post by mogmiester on Dec 8, 2016 5:23:40 GMT -6
I've skimmed over the thread, and in my opinion health gain is too strong. It "counters" everything, because at the end of the day it negates damage from whatever source it comes from. Tinkers might be strong, but if health gain was altered it would be in a better place. I guess Tim needs to tread carefully (not to imply he doesn't normally!)! Something I've suggested before (but was shot down) is to change health gain to health gain rating. Stacking Health Gain Rating would offer diminishing returns to prevent things getting too silly. It would also remove the need to hard cap health gain. Formula would be health gain % = HGR*0.01/(1+0.01*HGR). As for a dueling calculator, I've also made one. I posted it to the forum a while ago. A new version can be found here: mogmiester.uwcs.co.uk/GosBattle%20v0.4.xlsmIt's not perfect, but it does let you mess around with different builds. I also expanded the spreadsheet to use a genetic algorithm to find the "best" build, but it's a bit of a state at the moment so that won't be released just yet.
|
|
|
Post by gerdonat on Dec 8, 2016 9:13:23 GMT -6
About the random generated AC's, the problem I see, is that most people want to customize their chars to their liking, and some randomness could give someone a perfect combination of classes, and someone else a stupid one with no synergia at all... so, the first one has a big advantage, find at lvl 60 that your char suck with respect to other, after you spent a lot of effort and time to level it, its not good...
But..., I could see some of the randomness as optional, so, you can select at the start of the char, if you want a fully customizable char, or a random one, why would someone go for random you say? well, because the random char will be better at the classes that he gets than the customizable, for example, if i get random wolfbrother, then i m better at wolfbrother than someone that picked it (one option: you have +1 to all skills of your random classes, that +1 not factored in the cost of raising a skill). And the randomness should be controlled in some way that someone should not get two totally different and incompatible classes, so lets say you were born with a knack for poison, then you wont get randomly assigned a pure damage class at some point, and viceversa. So, introduce some randomness, but optional, and with some benefit to compensate that you are not chosing all the details of your chars.
|
|
|
Post by Seitaarin on Dec 8, 2016 10:00:31 GMT -6
I hear you ... and this is not a dig ... but you are running around on a no-damage poisoner at the moment with a wolfbrother class ... I am using a skilled up channeller with a knife and a shield ... a bit of creativity makes stuff work But, having thought about this some more, I think I actually think I do like the idea of a single rng class, the 'birthright' class that kind of points you in a rough direction for your character at level 20.
|
|