sigh... if you are looking at hitting 20 different targets then it is 20 x140% over 20 targets for... surprise, surprise... a stunning single-target shot of 140% percent.
multishot begins at 110% damage unruned. Give it the damage modifier and it's at 242%
Besides, this doesn't matter anymore... they apparently already updated the new demon hunter skills.
New stats for HA is 115% with a 60% chance to pierce targets when unruned and now generates 5 hatred.
HA Gold does not require 20 separate targets. It can hit the same target multiple times.
I don't think Golden Hungering Arrow is going to make it in the way it is now. Unless the damage goes down per pierce or there is a pierce cap, it's just way too powerful.
Most of the rune skills do appear to have benefits divisible by seven, so I think it's highly likely that's how they're supposed to work. Hungering Arrow just scales too well with a pierce attack, so it's not a good benchmark for comparison.
For what it's worth though, they don't have to make all rune levels scale linear. It's not important that a level 3 crimson rune is just as powerful as a level 3 indigo rune. It's not even that important that level 7 runes are equal, either. They just have to be close enough to provide the player with a choice. Which currently, Hungering Arrow Golden does not.
I don't think Golden Hungering Arrow is going to make it in the way it is now. Unless the damage goes down per pierce or there is a pierce cap, it's just way too powerful.
Most of the rune skills do appear to have benefits divisible by seven, so I think it's highly likely that's how they're supposed to work. Hungering Arrow just scales too well with a pierce attack, so it's not a good benchmark for comparison.
For what it's worth though, they don't have to make all rune levels scale linear. It's not important that a level 3 crimson rune is just as powerful as a level 3 indigo rune. It's not even that important that level 7 runes are equal, either. They just have to be close enough to provide the player with a choice. Which currently, Hungering Arrow Golden does not.
Going from 90% pierce to 95% pierce doubles your damage. Going from 7 shots to 8 shots on Indigo increases your damage by 13%. How could you possibly get those two anywhere near enough for there to be a choice?
I'm saying Pierce chance needs to go up by less and less to reflect the increasing returns built into it - just the same as Defense% or Armor%. So early on it would go up by 10% or so, and for the later rune levels by 2% or 3% so as to keep the increase constant, or at least closer to constant. Of course this is yet another case where it'll confuse the heck out of people who aren't mathematically inclined but the devs don't seem to have a problem with that so might as well confuse them some more.
If that's true then it's a moronic way to do things.
/snip
It's impossible to balance them at each level with that kind of formula.
I wasn't claiming yellow hungering was balanced. I definitely don't think its going to go live in its current shape. Maybe reduce damage on each subsequent pierce? Its stupid OP right now. Why even bother to use a hatred consuming ability when you have yellow hungering?
I was just saying the likely progression from rune level to rune level based on current values. Alabaster evasive fire probably adds 3 "yards" to the vault for each rune level.
Another possibility is that there is some flat amount added for the first level of a rune, then some value continues to scale linearly (on the rune, not the damage). Obsidian strafe could make the rockets do 19% damage at level 1, and 17% per level thereafter, to get to 120 at level 7.
We agree that Golden Hungering is too powerful and scales far too well. The problem isn't that it pierces or that it seeks a target, the problem is that it does both. Now that it's a Hatred generator, I think they should change golden to read each successful hit restores an additional (1-7) hatred.
Another option would simply be to reduce the damage of successive pierces.
I was also looking at your indigo damage and I came out with a different number. I looked at your calculations and I see where we differed, but I'm not sure why you calculated it the way you did.
Here's what I did for Indigo:
115% WD for the first hit
+60% chance for 8 new arrows. If the 8 new arrows are all seeking and piercing, then each of them is expected to do 1/(1-.6) = 287.5 damage. 8 arrows would then be (287.5)(8) = 2300% WD. (this is the part where we differed. Your equations showed (1+.6)/(1-.6), and I'm not sure what that equation represents.
That comes to 2415% weapon damage total, which is both more damage and more reliable than Hungering (the high expected damage of hungering comes in part from huge damage results from low probability events).
Now that I think about it, the expected damage is not at all the whole tale. How often you reach that expected damage matters too. Imagine you had a skill that did 10,000,000% weapon damage .001% of the time, and 1% weapon damage 99.999% of the time. Would you take that skill? It's expected damage is (10,000,000)(.001)+(1)(99.999)= 10099.99%, far more than any other skill. But usually it's not going to be worth firing. It's an extreme example, sure, but it's meant to illustrate the point. Hungering arrow does less than expected damage most of the time. In game, that matters.
We agree that Golden Hungering is too powerful and scales far too well.
/snip
It's an extreme example, sure, but it's meant to illustrate the point. Hungering arrow does less than expected damage most of the time. In game, that matters.
So maybe it's fine as is after all.
isn't the 20 pierces the average number of pierces per use?
It's not like the pierce chance is wildly variable from pierce to pierce, so I don't know if your example fits.
isn't the 20 pierces the average number of pierces per use?
It's not like the pierce chance is wildly variable from pierce to pierce, so I don't know if your example fits.
Yes, 20 is the average, but I still think it matters. The average case is not necessarily the common case. In fact, the chance of getting 20 pierces is only (.95)^20 = 35.8%.
Does it matter that a skill does less than average damage almost 65% of the time? I think it does. In a testing scenario it doesn't, because you can fire forever. But in game, when things are coming to kill you, reliability matters too.
It's true that Hungering Arrow's actual damage curve is smoother than my example, but I still think it's a point worth discussing.
Some breakpoints:
90% chance you'll get at least 236% weapon damage
75% chance you'll get at least 645% weapon damage
50% chance you'll get at least 1554% weapon damage
25% chance you'll get at least 3108% weapon damage
10% chance you'll get at least 5162% weapon damage
1% chance you'll get at least 10324% weapon damage
It may still scale too well, but in practice I don't think it's going to be as powerful as the math suggests it could be, due to the realities of gameplay.
I'll take that any day of the week. Especially since the minimum damage is 115%, and it has a non-small chance to do thousands of percent weapon damage.
But another part of the problem is not only does it do good (and regularly exceptional) damage, but it also gives you hatred.
I mean, aside from for a change of pace, the only time you'd use a hatred spender is for a massive AOE pack.
I'll take that any day of the week. Especially since the minimum damage is 115%, and it has a non-small chance to do thousands of percent weapon damage.
But another part of the problem is not only does it do good (and regularly exceptional) damage, but it also gives you hatred.
I mean, aside from for a change of pace, the only time you'd use a hatred spender is for a massive AOE pack.
I agree, it's probably still too good. Piercing scales too well on a skill that can seek new targets every time. It's not an issue for a skill like impale that travels in a straight line, for example.
If the maximum pierce was 75%, the skill would do 460% weapon damage on average. You'd have a 75% chance of doing at least 230% weapon damage. That's probably a much better place for it to be.
edit: I just realized a small error in my previous math for breakpoints. I'll have to fix it later. Basically I was only showing the damage from additional shots, and didn't add the guaranteed 115% damage from the first shot.
We agree that Golden Hungering is too powerful and scales far too well. The problem isn't that it pierces or that it seeks a target, the problem is that it does both. Now that it's a Hatred generator, I think they should change golden to read each successful hit restores an additional (1-7) hatred.
Another option would simply be to reduce the damage of successive pierces.
I was also looking at your indigo damage and I came out with a different number. I looked at your calculations and I see where we differed, but I'm not sure why you calculated it the way you did.
Here's what I did for Indigo:
115% WD for the first hit
+60% chance for 8 new arrows. If the 8 new arrows are all seeking and piercing, then each of them is expected to do 1/(1-.6) = 287.5 damage. 8 arrows would then be (287.5)(8) = 2300% WD. (this is the part where we differed. Your equations showed (1+.6)/(1-.6), and I'm not sure what that equation represents.
You've got your parentheses mixed up, but it's just another way of writing the damage multiplier from pierce chance.
1/( 1- 0.6) = 1 + 0.6/(1 - 0.6)
That comes to 2415% weapon damage total, which is both more damage and more reliable than Hungering (the high expected damage of hungering comes in part from huge damage results from low probability events).
Now that I think about it, the expected damage is not at all the whole tale. How often you reach that expected damage matters too. Imagine you had a skill that did 10,000,000% weapon damage .001% of the time, and 1% weapon damage 99.999% of the time. Would you take that skill? It's expected damage is (10,000,000)(.001)+(1)(99.999)= 10099.99%, far more than any other skill. But usually it's not going to be worth firing. It's an extreme example, sure, but it's meant to illustrate the point. Hungering arrow does less than expected damage most of the time. In game, that matters.
So maybe it's fine as is after all.
It does average or above-average damage 1/3 times. It does below-average damage 2/3 times. Personally I don't think it matters due to all the +APS bonuses on equipment. If you're firing 3-4 times a second, that's more than enough to make up for the variance.
You've got your parentheses mixed up, but it's just another way of writing the damage multiplier from pierce chance.
1/( 1- 0.6) = 1 + 0.6/(1 - 0.6)
I see where I went wrong. I missed the 60% factor for the 8 extra arrows. That brings it down to 1495 Weapon Damage.
It does average or above-average damage 1/3 times. It does below-average damage 2/3 times. Personally I don't think it matters due to all the +APS bonuses on equipment. If you're firing 3-4 times a second, that's more than enough to make up for the variance.
I still think it's important to consider. I also think the fact that the damage is spread out over time is a pretty important factor, particularly in a game like Diablo. If an arrow takes 1 second to exit a target, turn around, and hit it again, that means it will take an arrow 20 seconds before it even reaches the average damage case. 20 seconds is a pretty long time in Diablo combat terms. The point here is that it takes 20 seconds before you're doing 2300% weapon damage per shot. You'll be ramping up that whole time, but most AOE packs are going to be dead by that point. I don't think bosses are going to let you stand still for that long.
There's a lot we don't know about the skill yet, as well. Like if the arrow hits a wall does it disappear? I'd imagine it would. This means any combat area with obstacles can severely reduce the damage you'd otherwise get. It wouldn't work well at all in close corridors. Or what happens when a mob it's tracking dies? Does it seek a new target or does it keep going straight until it hits something?
Right now I think it's probably still too strong, but I'm just playing Devil's Advocate. Perhaps the limiting factors prevent it from becoming OP in-game. Not to mention that they literally just changed the system. Maybe Hungering Arrow will become a spender instead, or they'll adjust it's damage down now that it can be spammed endlessly. If only we could test with geared 60s in Inferno...
You've got your parentheses mixed up, but it's just another way of writing the damage multiplier from pierce chance.
1/( 1- 0.6) = 1 + 0.6/(1 - 0.6)
I see where I went wrong. I missed the 60% factor for the 8 extra arrows. That brings it down to 1495 Weapon Damage.
It does average or above-average damage 1/3 times. It does below-average damage 2/3 times. Personally I don't think it matters due to all the +APS bonuses on equipment. If you're firing 3-4 times a second, that's more than enough to make up for the variance.
I still think it's important to consider. I also think the fact that the damage is spread out over time is a pretty important factor, particularly in a game like Diablo. If an arrow takes 1 second to exit a target, turn around, and hit it again, that means it will take an arrow 20 seconds before it even reaches the average damage case. 20 seconds is a pretty long time in Diablo combat terms. The point here is that it takes 20 seconds before you're doing 2300% weapon damage per shot. You'll be ramping up that whole time, but most AOE packs are going to be dead by that point. I don't think bosses are going to let you stand still for that long.
There's a lot we don't know about the skill yet, as well. Like if the arrow hits a wall does it disappear? I'd imagine it would. This means any combat area with obstacles can severely reduce the damage you'd otherwise get. It wouldn't work well at all in close corridors. Or what happens when a mob it's tracking dies? Does it seek a new target or does it keep going straight until it hits something?
Right now I think it's probably still too strong, but I'm just playing Devil's Advocate. Perhaps the limiting factors prevent it from becoming OP in-game. Not to mention that they literally just changed the system. Maybe Hungering Arrow will become a spender instead, or they'll adjust it's damage down now that it can be spammed endlessly. If only we could test with geared 60s in Inferno...
From the videos I've watched, Hungering Arrow does seek out new targets after its current dies, and does keep attacking the same target if there are no others around, or it just feels like attacking one twice.
Although Hungering Arrow is like a damage over time skill, there's nothing to stop it from incidentally hitting several enemies in a tightly packed group. Therefore it will use up its pierces faster against groups of enemies who die quickly, and slower against bosses who die slowly, which is pretty much ideal.
I still think it's important to consider. I also think the fact that the damage is spread out over time is a pretty important factor, particularly in a game like Diablo. If an arrow takes 1 second to exit a target, turn around, and hit it again, that means it will take an arrow 20 seconds before it even reaches the average damage case. 20 seconds is a pretty long time in Diablo combat terms. The point here is that it takes 20 seconds before you're doing 2300% weapon damage per shot. You'll be ramping up that whole time, but most AOE packs are going to be dead by that point. I don't think bosses are going to let you stand still for that long.
I think you're under a misaprehension. The arrow may take 20 seconds to deal it's full damage, but as far as I know there is no limit to the number you can have flying around. So if one arrow takes 1 second to cast, and will deal 2300% weapon damage over it's lifetime, then you have a total DPS increase of 2300% by spamming HA.
Assuming you were non stop shooting for 20 seconds (the average time for one arrow to expire) you would deal, on average a total of:
56,000% weapon damage at roughly 695% weapon damage per second.
Again, this assumes that you stop shooting after those 20 arrows go off.
I wrote a simulator for this test case if anyone is interested in it, let me know and i'll post the code.
I still think it's important to consider. I also think the fact that the damage is spread out over time is a pretty important factor, particularly in a game like Diablo. If an arrow takes 1 second to exit a target, turn around, and hit it again, that means it will take an arrow 20 seconds before it even reaches the average damage case. 20 seconds is a pretty long time in Diablo combat terms. The point here is that it takes 20 seconds before you're doing 2300% weapon damage per shot. You'll be ramping up that whole time, but most AOE packs are going to be dead by that point. I don't think bosses are going to let you stand still for that long.
I think you're under a misaprehension. The arrow may take 20 seconds to deal it's full damage, but as far as I know there is no limit to the number you can have flying around. So if one arrow takes 1 second to cast, and will deal 2300% weapon damage over it's lifetime, then you have a total DPS increase of 2300% by spamming HA.
Assuming you were non stop shooting for 20 seconds (the average time for one arrow to expire) you would deal, on average a total of:
56,000% weapon damage at roughly 695% weapon damage per second.
Again, this assumes that you stop shooting after those 20 arrows go off.
I wrote a simulator for this test case if anyone is interested in it, let me know and i'll post the code.
What I mean is that your DPS doesn't immediately go to 2300% weapon damage/second when you shoot that first arrow. As you continue to shoot arrows and get more of them airborne, your actual damage/second goes up, but won't hit 2300%/second until you have 20 arrows in the air. Under those circumstances, that won't happen until at least 20 seconds into the fight.
The same thing is true of any damage over time ability. Each time you cast a new DoT, your damage per second goes up, eventually reaching the point where all your DoTs are applied. But until then, you're not doing full DPS.
What I mean is that your DPS doesn't immediately go to 2300% weapon damage/second when you shoot that first arrow. As you continue to shoot arrows and get more of them airborne, your actual damage/second goes up, but won't hit 2300%/second until you have 20 arrows in the air. Under those circumstances, that won't happen until at least 20 seconds into the fight.
The same thing is true of any damage over time ability. Each time you cast a new DoT, your damage per second goes up, eventually reaching the point where all your DoTs are applied. But until then, you're not doing full DPS.
I completely agree with that statement. I would like to point out that I ran my simulation for No Rune, Gold, and Obsidian. Given the changes to the hatred cost (there is none) obsidian comes out as the best damage dealing rune for hungering arrow. This also assumes a minimum amount of movement, because the more you move the less DPS means, and the more total damage per cast means.
These simulations ran firing 20 arrows each, and each one ran 10,000 times. Here are the results:
No Rune
-------------------------------
Average Total Damage: 6,999.88%
Average DPS: 308.16%
Gold Rune
-------------------------------
Average Total Damage: 56,058.66%
Average DPS: 690.20%
Obsidian Rune
-------------------------------
Average Total Damage: 17,375.03%
Average DPS: 766.85%
I haven't been following this for awhile, but you'll be able to get 20 HA's in the air far faster than 20 seconds. I do a lot of Shadow Power based builds, so, if we look at the top end gear where we are at for 20 arrows.
My understanding is that anything that's an arrow will go by weapon speed. That leaves out chakram, fan of knives, traps, sentry(if you can have more than 1) and grenades. Could be wrong though. If it didn't, then only the 2 hand weapons would be viable as the skill damage is based on the weapon damage, not weapon dps. Increasing the firing rate of the skill allows 1h weapons to balance with 2h weapons....
Except when you realize that you spend a lot more hatred shooting multiple times to get the same damage...
My understanding is that anything that's an arrow will go by weapon speed. That leaves out chakram, fan of knives, traps, sentry(if you can have more than 1) and grenades. Could be wrong though. If it didn't, then only the 2 hand weapons would be viable as the skill damage is based on the weapon damage, not weapon dps. Increasing the firing rate of the skill allows 1h weapons to balance with 2h weapons....
Except when you realize that you spend a lot more hatred shooting multiple times to get the same damage...
Based on my beta gameplay last night this is not the case. The spells seem to fire at the same speed regardless of the weapon that you're using. I tried with one handed crossbow, two handed crossbow, and a short bow. All were shooting hungering arrow at the same speed.
I appologize for my previous statement. I tested it again last night after seeing my roommate rapidly shooting HA. It appears as though the lag I was getting on my first night of testing was drastically messing with my casting speed.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
If that's true then it's a moronic way to do things.
For example, this is what the Weapon Damage would come out to for each level of HA gold.
Compare that to Indigo, which scales linearly
It's impossible to balance them at each level with that kind of formula.
Most of the rune skills do appear to have benefits divisible by seven, so I think it's highly likely that's how they're supposed to work. Hungering Arrow just scales too well with a pierce attack, so it's not a good benchmark for comparison.
For what it's worth though, they don't have to make all rune levels scale linear. It's not important that a level 3 crimson rune is just as powerful as a level 3 indigo rune. It's not even that important that level 7 runes are equal, either. They just have to be close enough to provide the player with a choice. Which currently, Hungering Arrow Golden does not.
I'm saying Pierce chance needs to go up by less and less to reflect the increasing returns built into it - just the same as Defense% or Armor%. So early on it would go up by 10% or so, and for the later rune levels by 2% or 3% so as to keep the increase constant, or at least closer to constant. Of course this is yet another case where it'll confuse the heck out of people who aren't mathematically inclined but the devs don't seem to have a problem with that so might as well confuse them some more.
I wasn't claiming yellow hungering was balanced. I definitely don't think its going to go live in its current shape. Maybe reduce damage on each subsequent pierce? Its stupid OP right now. Why even bother to use a hatred consuming ability when you have yellow hungering?
I was just saying the likely progression from rune level to rune level based on current values. Alabaster evasive fire probably adds 3 "yards" to the vault for each rune level.
Another possibility is that there is some flat amount added for the first level of a rune, then some value continues to scale linearly (on the rune, not the damage). Obsidian strafe could make the rockets do 19% damage at level 1, and 17% per level thereafter, to get to 120 at level 7.
Another option would simply be to reduce the damage of successive pierces.
I was also looking at your indigo damage and I came out with a different number. I looked at your calculations and I see where we differed, but I'm not sure why you calculated it the way you did.
Here's what I did for Indigo:
115% WD for the first hit
+60% chance for 8 new arrows. If the 8 new arrows are all seeking and piercing, then each of them is expected to do 1/(1-.6) = 287.5 damage. 8 arrows would then be (287.5)(8) = 2300% WD. (this is the part where we differed. Your equations showed (1+.6)/(1-.6), and I'm not sure what that equation represents.
That comes to 2415% weapon damage total, which is both more damage and more reliable than Hungering (the high expected damage of hungering comes in part from huge damage results from low probability events).
Now that I think about it, the expected damage is not at all the whole tale. How often you reach that expected damage matters too. Imagine you had a skill that did 10,000,000% weapon damage .001% of the time, and 1% weapon damage 99.999% of the time. Would you take that skill? It's expected damage is (10,000,000)(.001)+(1)(99.999)= 10099.99%, far more than any other skill. But usually it's not going to be worth firing. It's an extreme example, sure, but it's meant to illustrate the point. Hungering arrow does less than expected damage most of the time. In game, that matters.
So maybe it's fine as is after all.
isn't the 20 pierces the average number of pierces per use?
It's not like the pierce chance is wildly variable from pierce to pierce, so I don't know if your example fits.
Yes, 20 is the average, but I still think it matters. The average case is not necessarily the common case. In fact, the chance of getting 20 pierces is only (.95)^20 = 35.8%.
Does it matter that a skill does less than average damage almost 65% of the time? I think it does. In a testing scenario it doesn't, because you can fire forever. But in game, when things are coming to kill you, reliability matters too.
It's true that Hungering Arrow's actual damage curve is smoother than my example, but I still think it's a point worth discussing.
Some breakpoints:
90% chance you'll get at least 236% weapon damage
75% chance you'll get at least 645% weapon damage
50% chance you'll get at least 1554% weapon damage
25% chance you'll get at least 3108% weapon damage
10% chance you'll get at least 5162% weapon damage
1% chance you'll get at least 10324% weapon damage
It may still scale too well, but in practice I don't think it's going to be as powerful as the math suggests it could be, due to the realities of gameplay.
I'll take that any day of the week. Especially since the minimum damage is 115%, and it has a non-small chance to do thousands of percent weapon damage.
But another part of the problem is not only does it do good (and regularly exceptional) damage, but it also gives you hatred.
I mean, aside from for a change of pace, the only time you'd use a hatred spender is for a massive AOE pack.
If the maximum pierce was 75%, the skill would do 460% weapon damage on average. You'd have a 75% chance of doing at least 230% weapon damage. That's probably a much better place for it to be.
edit: I just realized a small error in my previous math for breakpoints. I'll have to fix it later. Basically I was only showing the damage from additional shots, and didn't add the guaranteed 115% damage from the first shot.
1/( 1- 0.6) = 1 + 0.6/(1 - 0.6)
It does average or above-average damage 1/3 times. It does below-average damage 2/3 times. Personally I don't think it matters due to all the +APS bonuses on equipment. If you're firing 3-4 times a second, that's more than enough to make up for the variance.
I still think it's important to consider. I also think the fact that the damage is spread out over time is a pretty important factor, particularly in a game like Diablo. If an arrow takes 1 second to exit a target, turn around, and hit it again, that means it will take an arrow 20 seconds before it even reaches the average damage case. 20 seconds is a pretty long time in Diablo combat terms. The point here is that it takes 20 seconds before you're doing 2300% weapon damage per shot. You'll be ramping up that whole time, but most AOE packs are going to be dead by that point. I don't think bosses are going to let you stand still for that long.
There's a lot we don't know about the skill yet, as well. Like if the arrow hits a wall does it disappear? I'd imagine it would. This means any combat area with obstacles can severely reduce the damage you'd otherwise get. It wouldn't work well at all in close corridors. Or what happens when a mob it's tracking dies? Does it seek a new target or does it keep going straight until it hits something?
Right now I think it's probably still too strong, but I'm just playing Devil's Advocate. Perhaps the limiting factors prevent it from becoming OP in-game. Not to mention that they literally just changed the system. Maybe Hungering Arrow will become a spender instead, or they'll adjust it's damage down now that it can be spammed endlessly. If only we could test with geared 60s in Inferno...
Although Hungering Arrow is like a damage over time skill, there's nothing to stop it from incidentally hitting several enemies in a tightly packed group. Therefore it will use up its pierces faster against groups of enemies who die quickly, and slower against bosses who die slowly, which is pretty much ideal.
I think you're under a misaprehension. The arrow may take 20 seconds to deal it's full damage, but as far as I know there is no limit to the number you can have flying around. So if one arrow takes 1 second to cast, and will deal 2300% weapon damage over it's lifetime, then you have a total DPS increase of 2300% by spamming HA.
Assuming you were non stop shooting for 20 seconds (the average time for one arrow to expire) you would deal, on average a total of:
56,000% weapon damage at roughly 695% weapon damage per second.
Again, this assumes that you stop shooting after those 20 arrows go off.
I wrote a simulator for this test case if anyone is interested in it, let me know and i'll post the code.
The same thing is true of any damage over time ability. Each time you cast a new DoT, your damage per second goes up, eventually reaching the point where all your DoTs are applied. But until then, you're not doing full DPS.
I completely agree with that statement. I would like to point out that I ran my simulation for No Rune, Gold, and Obsidian. Given the changes to the hatred cost (there is none) obsidian comes out as the best damage dealing rune for hungering arrow. This also assumes a minimum amount of movement, because the more you move the less DPS means, and the more total damage per cast means.
These simulations ran firing 20 arrows each, and each one ran 10,000 times. Here are the results:
No Rune
-------------------------------
Average Total Damage: 6,999.88%
Average DPS: 308.16%
Gold Rune
-------------------------------
Average Total Damage: 56,058.66%
Average DPS: 690.20%
Obsidian Rune
-------------------------------
Average Total Damage: 17,375.03%
Average DPS: 766.85%
http://us.battle.net/d3/en/item/natalyas-redemption
dual wielding this will result in 2.3 attacks per second. 4.6 with shadow power.
That puts 20 arrows in the air in 4.3 seconds.
Except when you realize that you spend a lot more hatred shooting multiple times to get the same damage...
Based on my beta gameplay last night this is not the case. The spells seem to fire at the same speed regardless of the weapon that you're using. I tried with one handed crossbow, two handed crossbow, and a short bow. All were shooting hungering arrow at the same speed.