Do Weapons Need Damage Ranges?

[Mr. Magniloquent]

I was delving through the PoE Wiki, and a thought occurred to me. In past games, weapons dealt a range of damage to approximate the quality of the hit. Since PoE determines both whether the attack hits as well as the quality of the hit within the same roll, why do weapons have a range? What does that range simulate? Would PoE be better if weapons possessed a raw integer, rather than a range? It certainly might make determining damage and chance to bypass DT more simple. What do you guys think?

[Seari]

Variance is a good thing, no? You would only have 3 different hits, that would always be the same number with a set weapon.

 

edit: Well mostly 2 numbers, since crits aren't common?

Edited September 13, 2014 by Seari

[thelee]

I was delving through the PoE Wiki, and a thought occurred to me. In past games, weapons dealt a range of damage to approximate the quality of the hit. Since PoE determines both whether the attack hits as well as the quality of the hit within the same roll, why do weapons have a range? What does that range simulate? Would PoE be better if weapons possessed a raw integer, rather than a range? It certainly might make determining damage and chance to bypass DT more simple. What do you guys think?

 

 

Variance is a good thing, no? You would only have 3 different hits, that would always be the same number with a set weapon.

 

edit: Well mostly 2 numbers, since crits aren't common?

 

In the long run, averages dominate, so variance (unless we're talking *extreme* variance, like 1-100 versus a weapon that does 50-50) doesn't matter much. EDIT - I'm not really disagreeing with Seari here, just summarizing what I think OP's response to Seari would be. 

 

That being said, the use of DT as a core mechanic impacts the math in an interesting way versus other straight up variant-damage systems (mostly D&D - I know DR factors in at times, but it's far less common) - a low variance weapon may be worse than a high variance weapon with a lower average simply because against DT small chance of doing a some damage versus 100% chance of almost no damage might make the trade-off worth it. So there's some level of itemization strategy. (EDIT - an example of this scenario is going up against a guy with DT 10. Would you rather have a 10-10 damage weapon or a 1-17 if there is a minimum damage of 1 after DT? I think most people will ultimately go for the latter, even though in the general case it is weaker (average damage of 9 versus 10) since in this contrived high-DT case the 1-17 actually gives you a chance of doing significant damage).

 

Adding random ranges also helps in balancing the game, theoretically. Your granularity is to fractions of damage without needing to actually use fractions. (Though the game already shows fractional damage, so I'm not sure how much this actually factors in.)

 

Lastly, there's just the visceral pop up of it. In fairly kinetic games (shooters, shooter RPGs) static damage works well because you can approximate variant damage by using physics - i.e. headshots, glancing shots, shotguns that miss some of their pellets, etc. In more macro-scale games (like tactical party RPGs like this), the variance helps capture the "flow" of the randomness of combat. IE if a kobold has 8 health and you have a 10-10 damage weapon, you will always kill it in one hit; whereas if you have a d20 weapon, that is far from certain.

Edited September 13, 2014 by thelee

[Mr. Magniloquent]

Variance is a good thing, no? You would only have 3 different hits, that would always be the same number with a set weapon.

 

edit: Well mostly 2 numbers, since crits aren't common?

 

In the long run, averages dominate, so variance (unless we're talking *extreme* variance, like 1-100 versus a weapon that does 50-50) doesn't matter much. EDIT - I'm not really disagreeing with Seari here, just summarizing what I think OP's response to Seari would be.

 

That is what I was alluding to. If quality of hit is already determined by the user's accuracy, then a damage range for the weapon would be redundant. Averages would be suitable, as variant comes from character ability--all things considered equal, of course. With balancing heavy on every-beta tester's mind, the passing notion seemed to lend to aiding that pursuit.

 

I think there is merit in the notion--particularly with the prevalence of DT and the uncertainty it brings, but not wholly necessary. I've been trying to think of ways in which PoE might be improved by further shedding D&D conventions (which run very deeply), and that thought cropped up.

Edited September 13, 2014 by Mr. Magniloquent

[Matt516]

 

I was delving through the PoE Wiki, and a thought occurred to me. In past games, weapons dealt a range of damage to approximate the quality of the hit. Since PoE determines both whether the attack hits as well as the quality of the hit within the same roll, why do weapons have a range? What does that range simulate? Would PoE be better if weapons possessed a raw integer, rather than a range? It certainly might make determining damage and chance to bypass DT more simple. What do you guys think?

 

 

Variance is a good thing, no? You would only have 3 different hits, that would always be the same number with a set weapon.

 

edit: Well mostly 2 numbers, since crits aren't common?

 

In the long run, averages dominate, so variance (unless we're talking *extreme* variance, like 1-100 versus a weapon that does 50-50) doesn't matter much. EDIT - I'm not really disagreeing with Seari here, just summarizing what I think OP's response to Seari would be. 

 

That being said, the use of DT as a core mechanic impacts the math in an interesting way versus other straight up variant-damage systems (mostly D&D - I know DR factors in at times, but it's far less common) - a low variance weapon may be worse than a high variance weapon with a lower average simply because against DT small chance of doing a some damage versus 100% chance of almost no damage might make the trade-off worth it. So there's some level of itemization strategy. (EDIT - an example of this scenario is going up against a guy with DT 10. Would you rather have a 10-10 damage weapon or a 1-17 if there is a minimum damage of 1 after DT? I think most people will ultimately go for the latter, even though in the general case it is weaker (average damage of 9 versus 10) since in this contrived high-DT case the 1-17 actually gives you a chance of doing significant damage).

 

Adding random ranges also helps in balancing the game, theoretically. Your granularity is to fractions of damage without needing to actually use fractions. (Though the game already shows fractional damage, so I'm not sure how much this actually factors in.)

 

Lastly, there's just the visceral pop up of it. In fairly kinetic games (shooters, shooter RPGs) static damage works well because you can approximate variant damage by using physics - i.e. headshots, glancing shots, shotguns that miss some of their pellets, etc. In more macro-scale games (like tactical party RPGs like this), the variance helps capture the "flow" of the randomness of combat. IE if a kobold has 8 health and you have a 10-10 damage weapon, you will always kill it in one hit; whereas if you have a d20 weapon, that is far from certain.

 

 

Took the words right out of my mouth. Especially the bolded parts.

[Matt516]

 

Variance is a good thing, no? You would only have 3 different hits, that would always be the same number with a set weapon.

 

edit: Well mostly 2 numbers, since crits aren't common?

 

In the long run, averages dominate, so variance (unless we're talking *extreme* variance, like 1-100 versus a weapon that does 50-50) doesn't matter much. EDIT - I'm not really disagreeing with Seari here, just summarizing what I think OP's response to Seari would be.

 

That is what I was alluding to. If quality of hit is already determined by the user's accuracy, then a damage range for the weapon would be redundant. Averages would be suitable, as variant comes from character ability--all things considered equal, of course. With balancing heavy on every-beta tester's mind, the passing notion seemed to lend to aiding that pursuit.

 

If the quality of hit were a continuum, I'd agree with you. But it's not - it only has 4 discrete values, one of which is no damage at all. And in any given fight, it is very likely that only 3 of the attack resolutions will be applicable (due to ACC-DEF needing to be between -5 and 5 for all 4 to be possible). So removing variance from the weapons would lead to a fairly boring experience.

[Roarkkk]

I'd rather have my char deal 9/12/8/11 damage than 10/10/10/10. +

I like to see different numbers even tho for fights longer than a couple of rounds its probably the same due to average damage.

IMO it adds more color to an otherwise too static approach.

[PrimeJunta]

I previously suggested flat base damage on a weapon, modified by the quality of a hit. Instead of graze/hit/crit, have a curve going from zero to double base damage depending on accuracy. That idea got shouted down fast though, and on further reflection perhaps it wasn't all that great to start with as it would make accuracy give the same type of damage bonus as might, which would kind of defeat the purpose of having two attributes governing damage.

 

I.e. as long as we have an attribute system where one stat governs damage and another accuracy, I think we do need a degree of randomness in weapon damage as well. I still think it could work well in a different type of system.

[Sensuki]

Weapon damage ranges are good, but the current ranges are wrong. This is the next topic I'd like to tackle with Matt - I have an idea already.