I made a thread about stats in the backer beta, and realized a poll would probably be better to get the opinions of the testers.

Old thread: http://forums.obsidian.net/topic/67059-stats-in-backer-beta/?do=findComment&comment=1482954

 

If I missed any opinion please tell me.

 

EDIT: I just want to add that I get the system Josh wants to make, with might helping magic too, as might is an aspect of how powerful your soul is, but I feel that the template solution on all abilities leaves a bad taste.

I also get that the system to a certain degree needs to be easy to understand, but I also feel that players can't know in what ways their stats will be useful anyway(since they unlock dialogue options).

Hopefully this poll will show how all others feel about this, and whether or not I should just shut up about it.

Edited August 20, 201411 yr by nilssab

I just want to clarify that the "whatever" options in question 3 roughly equate to positive vs offset zero scale, but not necessarily linear or anything.

Personally, if I wasn't a proponent of an complex original style, I'd go for yes to bad builds.. To me there is nothing as immersion breaking as a puny wizard having even 50% the intrinsic survivability of a tank character, even if the tank has tanky skills.

I like concept personally.  Choosing between offense, defense, and utility is good.  But the other 3 stats are kinda lame.  

 

Dexterity:  Maybe movement speed and recharge speed would be better for dexterity.  Accuracy = RNG, no real strategic value. 

Perception:  Free interrupt by just attacking is really lame.  Maybe something like DT penetration would be more interesting, so you spot the weakness of the enemy's defense.

Will:  Concentration... lol.  if i'm a caster, i'm going to avoid getting hit in the first place.   Instead, maybe reduce duration of CC cast on the character (basically a counter to Int)

 

Also the bonus feels too big imo.  Especially the AoE radius.  It's almost impossible to use some of those spells with friendly fire and max Int. lol  Maybe if we can adjust the size of the AoE at will, that would be great.

I like concept personally.  Choosing between offense, defense, and utility is good.  But the other 3 stats are kinda lame.  

 

Dexterity:  Maybe movement speed and recharge speed would be better for dexterity.  Accuracy = RNG, no real strategic value. 

Perception:  Free interrupt by just attacking is really lame.  Maybe something like DT penetration would be more interesting, so you spot the weakness of the enemy's defense.

Will:  Concentration... lol.  if i'm a caster, i'm going to avoid getting hit in the first place.   Instead, maybe reduce duration of CC cast on the character (basically a counter to Int)

 

Also the bonus feels too big imo.  Especially the AoE radius.  It's almost impossible to use some of those spells with friendly fire and max Int. lol  Maybe if we can adjust the size of the AoE at will, that would be great.

I was gonna add more questions to the poll to allow for those kinds of opinions, but it was limited to 3, maybe I should make question 2 into question 3 and add more options to the changes you want to see..

 

Would you feel that stats that feel less useful in combat could be balanced by making them more prominently useful in "adventure promts"/dialogue, as in finding some hidden switch with perception, or using will to convince a NPC of something?

Poll is too complex, or assume this is our answers, or that we could even have answers at this point. The game is really buggy, and it's hard to get a feel of every class and abilities when combat is barely working.

 

The only thing I would change, as far as I can tell, is that I'd remove the +Will bonus from intelligence so Resolve feels useful, and that I'd add a small %crit chance to Perception to make it less dumpy, and actually useful.

Poll is too complex, or assume this is our answers, or that we could even have answers at this point. The game is really buggy, and it's hard to get a feel of every class and abilities when combat is barely working.

 

The only thing I would change, as far as I can tell, is that I'd remove the +Will bonus from intelligence so Resolve feels useful, and that I'd add a small %crit chance to Perception to make it less dumpy, and actually useful.

I understand that the game isn't maybe really in the state were you can get a feel for exactly how the stats work.

If it's too complex, please elaborate on how it could be simplified while still allowing a large range of opinions, and if your opinion is not included, please tell me what kind of options you would like to see.

To be fair though I don't think it's possible to make a poll for what granular changes should be made at this point because it would have to have too many options that would probably become void really fast during changes in the beta.

My point is mostly that a poll isn't the best way to discuss this.

I found it really hard to figure out how to assign stats for my character.  The tradeoffs were completely unclear and there's no progressive point expense so there's zero reason not to run a couple of stats up to maximum instead of spreading your points around a bit.  What's the system here, higher = better?  Do you actually get a penalty for having a given stat below a certain point?  I dumped intellect and I was expecting my character to be a MORON in conversation, but this doesn't appear to be the case.  And how do the stats interact with skills?  DO they interact with skills?

It was really, really difficult for me to tell if there was going to be a functional build difference between different stat expenditure plans other than "you will get different conversation options".

My point is mostly that a poll isn't the best way to discuss this.

The counter argument is that no human can look at a 30 pages long threads of ppl making granular suggestions and get a feel for what the general opinion is.. Do you count how many ppl agree with a set of different statements? and what would that actually mean?

With that considered wouldn't it be better to try to make an as good as possible poll?

 

Atm. it seems like:

more ppl like uniformity of stats than dislike.

If it is made non-uniform, more ppl would like to keep it simple and only have differences between melee/ranged/magic, rather than make spell/weapon differences.

More ppl would like to see the stats make more difference, as opposed to keeping a simple positive linear scale.

 

of course this might be skewed by lack of proper opinion options, which is why I wanted feedback.

So the problem with these sorts of forum polls is selection bias. People who are happy with the way things work are unlikely to both reading the thread & or voting in the poll. While those who dislike the way things are will read and vote. Thus unless every one has to vote in the poll or you somehow manage to get a significant number of the registered 70K backers or the >10K backers with beta access to vote it's somewhat meaningless.

 

Also polls by there nature offer only limited choices thus providing another sort of bias.

 

The short of it is while polls are amusing they are generally meaningless and something developers should avoid paying attention to unless they can get most of the backers involved.

My point is mostly that a poll isn't the best way to discuss this.

This.

What poll?! I can't even play the game properly. Ask later. 

So the problem with these sorts of forum polls is selection bias. People who are happy with the way things work are unlikely to both reading the thread & or voting in the poll. While those who dislike the way things are will read and vote. Thus unless every one has to vote in the poll or you somehow manage to get a significant number of the registered 70K backers or the >10K backers with beta access to vote it's somewhat meaningless.

 

Also polls by there nature offer only limited choices thus providing another sort of bias.

 

The short of it is while polls are amusing they are generally meaningless and something developers should avoid paying attention to unless they can get most of the backers involved.

 

I have to disagree here.

 

First off, its a beta, you're not supposed to be happy with the way things are. You are supposed to find the negative things and help the devs make as good a game as possible. I for one read most of the different topics to see what people are talking about and if I can help replicate (if need be) the problem my self. We are supposed to be testing the game not "enjoying" it (all though I do enjoy playing the beta...).

 

And as to the vote it self, I find that the ones who are passionate (for lack of a better term) about beta testing will vote on these. Whether or not they are happy about the subject.

 

It is a completely other subject, whether a poll is the right way to go about such a large scale subject.

Edited August 20, 201411 yr by Totottoro

 

So the problem with these sorts of forum polls is selection bias. People who are happy with the way things work are unlikely to both reading the thread & or voting in the poll. While those who dislike the way things are will read and vote. Thus unless every one has to vote in the poll or you somehow manage to get a significant number of the registered 70K backers or the >10K backers with beta access to vote it's somewhat meaningless.

 

Also polls by there nature offer only limited choices thus providing another sort of bias.

 

The short of it is while polls are amusing they are generally meaningless and something developers should avoid paying attention to unless they can get most of the backers involved.

 

I have to disagree here.

 

First off, its a beta, you're not supposed to be happy with the way things are. You are supposed to find the negative things and help the devs make as good a game as possible. I for one read most of the different topics to see what people are talking about and if I can help replicate (if need be) the problem my self. We are supposed to be testing the game not "enjoying" it (all though I do enjoy playing the beta...).

 

And as to the vote it self, I find that the ones who are passionate (for lack of a better term) about beta testing will vote on these. Whether or not they are happy about the subject.

 

It is a completely other subject, whether a poll is the right way to go about such a large scale subject.

 

Thank you, when posting a poll I really didn't expect that the thread would be 100% derailed by ppl claiming polls are useless. o_O

I just don't see what other kind of posts/activity on the forum could ever provide any more useful feedback to the developers. Just discussions in a forum thread is simply too many posts to read for ppl that actually have to do work as well, so we need to give feedback in a way that is good for the developers. Obviously polls will have selection bias and option bias, but FAR LESS selection bias than a discussion thread.. (if you think the threads on here are discussing things to the point that there is actually a consensus valid to the devs, you are delusional) Anyway the developers know about all kinds of bias all kinds of posts have, they are not stupid and can form valid deductions from the activities on the forum. And I believe polls would be best.

 

Anyway, to get back on the topic, the thing I wanted to see in the poll was 3 things:

1. Are most beta players content with stat effects being uniform?

2. If not, what parts of stat effects do beta players feel should be non-uniform?

3. what kind of stat balance are people out for? Here, specifically I wanted to see if ppl want a forgiving system were no build is catastrophically bad, or reverse, and if the system should be more complex than it is now, or reverse.

 

And since I know that there is option bias, and I cannot possibly have gotten the options right the first time, I would appreciate feedback.

What would be better questions to ask in a poll like this and what options should be better. If you feel you would have liked another option in question 1-3 please tell me.

 

It's not like I think this will be super important for the devs, but I really felt it was immersion breaking, and if others think the same, maybe it deserves a look at and see what ppl think. If you post it as a regular thread, as I also did, most ppl posting will be like-minded or straight opposed. You get no idea at all what most ppl think. 

I don't mind the idea that all stats are useful for everyone and that they have the same effects for all classes. I do think however that the stats feel slightly bland at the moment though, and could use more effects.

 

Another thing is that I feel the minimum points you can have in a specific stat should be increased somewhat, maybe to 6-8 (and of course lower the amount of points you can spend to match that). Currently if the balance between the stats isn't perfect, which it isn't at the moment, you end up having a stat range of like 18/18/18/3/18/3, which makes the characters feel a bit more like machines than people to me. Alternatively or in addition to that you could make it so you need more points to increase stats above 14-15 or so, because I think no matter how hard they tried balancing them, there's always a stat that gives more useful bonuses for your character than the others.

I don't mind the idea that all stats are useful for everyone and that they have the same effects for all classes. I do think however that the stats feel slightly bland at the moment though, and could use more effects.

 

Another thing is that I feel the minimum points you can have in a specific stat should be increased somewhat, maybe to 6-8 (and of course lower the amount of points you can spend to match that). Currently if the balance between the stats isn't perfect, which it isn't at the moment, you end up having a stat range of like 18/18/18/3/18/3, which makes the characters feel a bit more like machines than people to me. Alternatively or in addition to that you could make it so you need more points to increase stats above 14-15 or so, because I think no matter how hard they tried balancing them, there's always a stat that gives more useful bonuses for your character than the others.

I agree completely, I think it should be harder to get high stats, and they should be more rewarding. To be clear, going from 18 to 19 might only gives you 1.5% more damage(EDIT: for comparison, going from 18/00 to 19 in D&D with a D10+2dmg weapon would give you 14.5 average dmg from 13.5 so +7.4%), so you can basically suffice with around 12.. Also, I think it is an important point for immersion that stats feel both rewarding and hard to get.

 

I realize maybe I'm not clear enough in the poll though, what I mean with uniform stat effects is not that all stats are useful for everyone, not that they have the same effect for all classes, but that they have the the same effect for exactly everything.. maybe that's what you mean with bland. I mean, the point of question 1b is that then what kind of effects should differ? 

All stats should help all classes(I think we all agree here), but that doesn't mean that might has to do the same thing for all different kinds of abilities and combat mechanics. Personally I feel that it is bad for immersion that dmg with bow/spells/melee are all based on might, why not change it around a little? Maybe spell range from might? Maybe int could give melee crit etc. Why should wizards deal as much damage in melee as a fighter when they hit with the same weapon?

And some more effects could be added to each stat that adds just a bit more complexity. 

It seems like there is a even divide between complex and simple stats, but all I hope is that we end up with a stat system that at least has more dimension to it than the stats in diablo 2, and were not all characters have exactly the same stat builds..

Edited August 20, 201411 yr by nilssab