Damage vs Variance

[GooberGrape1]

Damage vs Variance

Is there a point where the variance is just "good enough"? or is it best to tighten the variance so that you always hit for nearly the exact same damage?

I decided on two bows to compare the ends of the spectrum:

Cruel Diamond Bow - 130 to 157 damage
Rare Hydra Bow - 49 to 242 damage

Fighting Baal on /players 8, I noticed that he took very consistent damage from the Diamond bow. you could see it nock his health down in an even chunk every time, and he died quickly.

Same fight, same gear, this time with the hydra bow. Damage was very inconsistent, sometimes barely scratching him, other times nocking off nearly an entire letter of his name. He managed to get his clone off, and toss several other problematic spells, and I finally switched back to the diamond bow just to finish him before he caused any more trouble.

As an aside, both bows absolutely dominated WSK. They just hit so hard compared to everything else I've ever held with this character. I'd definitely say that either bow can do the job.

I'm not ready to give up my hunt for an SP Windforce just yet, I'm just starting to think that consistent damage is more important than MAX damage.

Thoughts? Suggestions? Snide Remarks?

GG

 

[Ravenforce3]

Sometimes average damage isn't all it's cracked up to be. That's why, given the choice, I'd take the Diamond Bow. Higher min damage means consistancy. This is one reason I DISPISE Windforce. While it is arguably a good bow, but the min damage is 35. 35! I know that lightning spells have the chance to only hit for 1-4 damage (depending on synergy/mastery), but their max damage being so much higher means that they can be a bit more consistant.

 

[sorcererbob]

For me, it's all about average damage. Sure, you could get nothing, or near nothing, but if you are an optimist, like I am (today), then you could score several million.

Ok, a couple of thousand then...

Potentially, that is a LOT of damage. Although, the other thing with lightning spells is their 100% piercing. That means that while you may scratch one or two monsters, you can potentially destroy everything onscreeen (if you're lucky).

 

[GooberGrape1]

Wow, forgot about piercing until you mentioned it...

During the test with the hydra bow, I was shooting at Baal and the arrows were piercing him and actually killed his clone (who was behind him) but real Baal was still at half life. I forgot that the damage rerolls. I'll have to take that into account, pierce is a big part of my strategy in this build.

 

[Nightfish]

What matters most to me is the base damage of the weapon. (and possibly the average of that) I always try to look for things that have "adds X - Y" damage as these things get modified by your ED bonuses. This doesn't really apply to bowzons all that much as they don't get a lot of Skill ED bonuses. Only dex, might and whatever you have on your gear.

 

[factotum]

I use average damage as my guide, because if you didn't, you'd never use lightning spells or flails, both of which are extremely useful. (A Werewolf attacking with a flail hits at an incredible speed even with no IAS at all!).

 

[DreamlessDude]

But i am pretty sure that a werewolf will hit with a phase blade fast too with both at very fast rates but not too sure about the fps. Never been too into druids. Also computer numbers are randomly generated so i dont really like flails and lightning spells(cept static and teleport of course)

 

[sunbearie]

I've always weighed the average damage against the speed of the weapon. Basically the faster you attack, the more likely you do your average damage over time. Since you're a zon, attack with both bows should be pretty fast anyway hence the lack of an overall difference.

If you have a super slow weapon or low ar so you have less hits per min, a high min damage makes a bigger difference.

 

[ULTIMATER]

Please dont say bad things about the WF, i never owned one, found one and its the one thing that keeps me playing D2. My Ama wants a WF..


*cover ears*

 

[flying_dug0ng]

*thoughts*
Same reason I never go for lightning charms. Imagine if you're out of potions and just one more Chain Lightning can 'potentially' kill Big D... and you deal 3 damage.


*suggestions*
If you can hit 7/2 IAS breakpoint, you'd be firing so fast it wouldn't matter. -edit- sun beat me to it

Rare Hydra Bow - 49 to 242 damage ... Fighting Baal on /players 8 ... other times knocking off nearly an entire letter of his name.

*snide remark*
Impossible even with Crushing Blow :teeth: You must have *lots* of elemental damage on your zon? Or too much coffee this morning

 

[csarmi]

Well if in the area monster HP's are all 10, I would rather do 1-15 damage, then 7-9.

Against big HP monsters and in the long run, variance does not pay much role.

 

[Andrew]

*snide remark*
Impossible even with Crushing Blow :teeth: You must have *lots* of elemental damage on your zon? Or too much coffee this morning

Maybe he was running nightmare baal? Normal =D.

 

[Ti-Ben]

I like having a lower maximum but a very high minimum. Since minimum gives you the assurance to hit that damage, while max only gives you the opportunity to hit high. It's up to anyone though...

 

[King's Specter]

I usually go off of average damage also, but there is something to what csarmi is saying. If a weapon with minimal variation lets you kill the critters in two hits all the time, then a higher-ranged weapon that lets you kill them in 1 hit 20% of the time, 2 hits 70% of the time, and 3 hits 10% of the time would noticeably speed up things like pit runs.

This would be very tough to model unless you ran the exact same monsters all the time.

-KS

 

[NSXdreamer]

I don't think it matters. I assume they use a bell curve to determine your damage, and that should yield same damage no matter what's the varience.

 

[burke615]

Is it appropriate to assume a bell curve? I guess it depends on what kinds of dice the gnomes inside the computer are rolling... :teeth: Seriously, though, does anyone know whether damage is calculated on a curve (and what kind) or if it is just "straight" randomness?

Cheers!
Burke

 

[Thrugg]

Damage specified as a range is distributed uniformly ("straight" randomness).
You have the same chance of getting 1 as 50 and as 100 in a 1-100 range.

However, sometime uniform distributions will stack - 10-20 damage + 10-20 fire damage, you have a smaller chance of both being 10, or both being 20, than you do of some mid-range total. The more stacked uniform distributions, the closer to normal (a bell curve) the damage will be.

I think csarmi brought up the most important thing - if you are doing enough damage to take the critters you're killing down in 1 or 2 hits, then a higher min / lower variance is very useful to make sure you don't take 5 hits sometimes. But if it takes you 100 hits anyway because you're running Hell Baal on p8, only average matters.

 

[NSXdreamer]

Ah crap, I always thought it would be a bell curve.

But it wouldn't matter on long term damage anyway, since I use player 8 most of the time, and just hold down the mouse button to wack away, while take a drink or somethingwith my other hand.

 

[GooberGrape1]

Lol yeah, I'm pretty sure it was on /players 8 normal. I didn't want to accidentally die to a pack of hell-gloams just because I wanted to test out some new toys.

 

[Bolinbrooke]

just hold down the mouse button to wack away, while take a drink or something with my other hand.

:lol: sorry, cant help but laugh at what that could be :teeth: