Mhhh might be offtopic:
Assuming you would record your runs , would it be possible to write a script in matlab for automatically count the average mob-density of a pull?
For differnt type of runs i mean like speedruns etc.
If not would it be possible to write a script that counts pierces? I mean the HAs have a brighter color than most other stuff, so maybe if youd tone down images to greyscales and then somehow count every colission it makes?
I am not very good at coding but would that be possible with a feasable amount of time?
And you know that because of a field analysis ? I have a PhD in applied Physics and play videogames. Yet i might be not as smart as some homeless dude on the street with no PhD or anybody else. My post where i critique the intro was removed anyhow now ^^
Yes, it is possible to do something like this, but it would take quite some time to write. The hardest part isn’t the counting or density but rather the computer vision code needed. Specifically, you would need some video parser that could distinguish hungering arrows from other things, detect hit events, and also track all monsters (of varying types) to note density.
With the backend of the code, it would be more feasible since one could insert a counter in the code that tracks the number of pierces but Diablo 3 is a closed-source code. If anyone would want to do this, I would imagine only the Blizzard development team could do it in a clean way.
@Iria
My math sucks, and I’m forbidden from doing it on the Wizard forums( I never agreed to the DH Forums!! = LOOPHOLE) if they were to institute a 210% maximum multiplier on Devouring Arrow, how many GR nerf would that translate to be? The math is also confounded by the seasonal buff. I just have no idea what kind of numbers are fair.
Mhh , i know its possible in general. But i think you might know somebody maybe and/or are good in Matlab yourself.
I just used it a few times. When i was working in a lab i worked with an OCT that could do 3D-Scans …of teeth with holes in it for example xD.
To determine the volume of the hole, we did Split that Scan into thousands of Images, turned them into greyscale , wrote a code for matlab where it determines the area and then did matlab loop that for each picture.
Thought if you are a Matlab-Crack you could do sth like that with a Video and then determine an average pierce amount in diff Scenarios like Speedruns, Pushes, Groupplays etc.
From my own experience in a normal run the amount of more than 20 pierces per HA seems unrealisticly high, if you are not Solo-Pushing anyhow. And i do think that the x2 Multi on the Setbonus would be very overtuned. I mean i understand the idea and where your coming from but im unsure if this might not go off in the wrong direction
Oh really? 
You’re trying to tell me that you can’t clear a 150 GR in less than 5 minutes in a 4 man META group?
Season EU server:
https://i.ibb.co/VWBcTjh/L-S-EU-150-GR-4-Man-Group.jpg
…or in less than 4 minutes?
Non-Season EU server:
https://i.ibb.co/2ZDdxCb/L-NS-150-GR-4-Man-Group.jpg
Anyway allowing to clear 150 GR only by META groups is the dumbest thing possible. Just allow players to have a choice… If I want to do 150 GRs in group I’ll play META groups if I want to SOLO let me SOLO a 150 GR for the love of Gawd…
Geez
1 Like
Well if its +70% per pierce and you really assume that Irias numbers with 40+ pierces are correct.
Wolfram Alpha comes up with a sick number.
https://www.wolframalpha.com/input/?i=1.7%5E40
If you want HA do max of 290% weapondmg you would have to lower the amount of possible pierces to 2?
No. That’s not what I want. I simply want them to cap the multiplier. It won’t increase damage by 28x anymore. All the arrows will still pierce like they normally would. I thought it was like ~24 GR nerf, which is a pretty huge hit. Doubling the six piece would make it more consistent, though. I just can’t compute the math.
Hehe if i did no superdumb error above with 40 pierces its more of a 1.65 billion multiplier lol xD
Devouring Arrow adds 0.7 to the multiplier per pierce starting with 1.0 for the first hit. It’s not a multiplicative exponential growth of 1.7x each pierce. Thus if the first hit does 100 damage, the second does 170, the third does 240, and so forth.
The damage per hit increases linearly with the number of pierces, but the total damage increases quadratically. If you include area damage on top of this, the damage grows cubically if you assume all targets are stacked on top of each other (number of hits = number of targets x number of pierces).
In a real situation, this almost never happens but my scaling argument was to illustrate how Devouring Arrow and area damage combine together.
1 Like
Ahh already thought im too tired already…so the base rune itself is completly linear 1 pierce 70% , 10 700% , 20 1400% and so on?
But ahhm then capping it to 20 and increasing the Setbonus by x2 might be a bit op 
Well lets just see what blizz comes up with for 2.7
I am honestly quite happy with my first DH and i do think balance-wise he is in a nice spot
The base rune deals constant damage each pierce. So with the first hit and 4 pierces you deal a total of 5x the damage of the base hit. With Devouring Arrow, you deal 1 + 1.7 + 2.4 + 3.1 + 3.8 = 12x the damage of the base hit. This grows out of control when you pierce a lot of things on top of each other.
Well I admit the numbers can be adjusted. The intent was to taper off the top end of the pierces where you get incredible density and can clear trash even in GR150 quite quickly. The Wudijo video showing a missile dampening elite in GR150 showcases a key part of the problem.
My nerf/buff idea was to hamper those top GR clears while buffing the lower end of the spectrum.
1 Like
Haha…its not like they farm on 150 and for this ladder its already too late. As mr wudjo already reached 150 with his 5500 keys spent on a perf grift xD
This isn’t my observation at all. It is only when you get to those higher pierce counts does trash actually die. I ran clips in slow motion and saw tickling for 2 seconds, then boom, the trash pack shrinks. (No CoE, no other variance like elite, zeis or oculus, all passives and legendary gems were defensive) and I was still making progress in GR130.
Referencing the above chart, When you compare the sum damage of two attacks: 20 pierces vs 40 pierces, the latter attack is 3.66x more damaging.
And when you do coordinate huge pulls with Cold CoE cycle and oculus ring, it is clear what is happening, those 40+ pierce hits range between 1-2Q in correct conditions, the low pierce attacks are 10-20T in worst conditions.
So in Non-seasons, DH has no spot in Meta, barb WW is a superior less annoying build, and saders still have +5 GR advantage, and you want to nerf it?
Finally get a decent DH farming build for GR 120s in NS and the fun police says no go?
Ok, but I can think of a lot better use of your precious time…
1 Like
The problem is you can’t get party with DH in non season
you have no idea what you are talking about non-season:
- all 150 clears are with zDH since the WD got nerfed (the fastest zDH clear is currently 5:54)
- all 120-125 rat runs are with zDH
- 120 xp meta is DH as well
I do agree dmkt, i mean i made the same observations when i pushed.
But i was talking about what i play mostly: Speedruns or runs that are at least < 4 Minutes. In this Szenarios i do ofent not even get the density for 20+ pierces.
In a group i find it to be even more unrealistic. And i think those are the cases where the proposed changes would be buffs, that could be, in my eyes, a tiny bit too much
1 Like
I really don’t understand this desire to nerf the DH…
There is clearly a bug with the way Hungering Arrow interacts with Missile Dampening, where it is actually Missile Enhancing, to the point where encountering a Missile Dampening elite is even better than getting a Conduit Pylon! (Drag all elites, and trash, on the map into the missile dampening field, pepper with HA’s and watch them all pop within 30 seconds.) I’m pretty sure that most classes could get numerous 150 clears if they were also getting the equivalent of 2-3 conduit pylons per run. Fix this bug, and only this bug and see where that puts the DH.
1 Like
Y actually this would be one of the easiest routes to go. Remove that affix^^
So i thought to myself: Ok…i cant write a script to determine how big the impact on those 110 speeds i do mostly really is, i wont record it and count every arrow frame by frame for an entire video.
How to see the impact of the piercing HAs in what is for me normal circumstances (solo 110s in <5 to 4 min)
Well…this morning i just switched the rune to Scatter Shot and after all the theory-crafting as you can imagine, i expected it to perform horrible on 110. Or at least noticable slower than HA DA.
But to my surprise it wasn’t. And in all honesty with a 20k Buff for 6P this set would just obliterate any other build for solo-farming GRs in a way that the call for nerfs from other classes would be justified.
But all in all…i mean the proposal is not bad. Maybe cap Cirri to 10 and the Setbuff to 15. Sth somewhere in the middle
What data did you use to reach the conclusion that crusaders are 5 GRs above DH?
The worldwide top solo clears in non-seasons show that DH are similar to or potentially exceed slightly the power of pre-nerf crusaders and pre-nerf witch doctors.
See:
http://ranks.zeroempathy.org/e13
Note: # = Class nerfed in the current patch, but leaderboard clears remain since the era leaderboard was not reset
| Class |
Number of Solo GR 150 Clears Worldwide |
Top GR 150 Clear Time |
| Demon Hunter |
34 |
11m45s |
| Witch Doctor# |
30# |
11m22s# |
| Crusader# |
26# |
10m21s# |
| Necromancer |
16 |
11m47s |
| Barbarian |
0 (Top GR 148 @ 12m53s) |
N/A |
| Monk |
0 (Top GR 145 @ 14m33s) |
N/A |
| Wizard |
0 (Top GR 143 @ 14m11s) |
N/A |