Diablo® III

Corpse spiders: Best single target attack?

@Xarkar:

Yeah I do my own testing at MP7-MP8. In my tests, I'm running different gear combos as well and survivability is a factor.

That doesn't take away from MCP's results though. His numbers show us the maximum damage possible, assuming no defensive movements at all. From that we should be able to make better informed decisions about various builds and gear to support the builds.
Reply Quote
@MCP:

Ghom MP0 has about 4M HP. Splinters without PtV took 24 seconds on average, which equals 166.7K effective DPS. Dividing the effective DPS by 180% (each splinter does 60%) suggests your paper DPS is about 92.6K; but that's higher than you claim. Any thoughts on the discrepancy?
Reply Quote
My poison darts crit for ~250k and because of my crit chance, 2/3 usually crit per shot. I've hit for near 900k in 1 shot of poison darts. with LS and VQ mana regen, I can quickly burst my darts in between bears to keep mana full without sacrificing life regen.
Edited by bicen#1974 on 11/29/2012 5:40 PM PST
Reply Quote
OK, so I'm having a conversation with myself here, whatever.

I ran the Ghom MP0 splinters test for myself. It was quite informative.

Methodology - in an attempt to get something like repeatable results, I had to do these things:

1) Dismiss follower.
2) No pets.
3) No hellfire ring.
4) Keep left hand on stopwatch; start simultaneous with first attack; stop simultaneous with death.

Even then there was a lot of variation in times.

My paper DPS for this test was 111K. Kill times were 15.4, 15.5, 19.7, 17.4, and 19.4. Given effective DPS of splinters = 180% paper DPS, every single one of these results was too fast. It's the same discrepancy I asked about above.

Well, looking at the skill description again... unruned PD does 100% + 40% over 2 seconds. I now believe that splinters does 3 x 60% plus 40% over 2 seconds (stacks) for 220% total effective DPS. That fits both my results and MCP's.
Reply Quote
o_0!!!

!@#$ing splinters making us second guess ourselves.

Very interesting btw, I'll be starting that effective DPS now, expect results sometime tomorrow.

I got that scotch going now :).
(scotch == http://mymaltwhiskys.files.wordpress.com/2011/11/glenmorangie-lasanta.jpg)
edit:
I retested BS+HRS+PtV and SiW+HRS+PtV and found SiW + HRS does more DPS.
As can be seen, i replaced the originals and at the bottom of the 'Ghom MP0 Kill Speed Data Collection' tab (renamed), I added 'retest' and 'old test' just to quickly compare.

I must have recorded it wrong or gotten REALLY lucky with crits for blazing spiders.

Human error is definitely a factor while recording times. There is no stop watch for this and I just have to watch the wall clock time.
Edited by MCP#1477 on 11/29/2012 10:11 PM PST
Reply Quote


My paper DPS for this test was 111K. Kill times were 15.4, 15.5, 19.7, 17.4, and 19.4. Given effective DPS of splinters = 180% paper DPS, every single one of these results was too fast. It's the same discrepancy I asked about above.

Well, looking at the skill description again... unruned PD does 100% + 40% over 2 seconds. I now believe that splinters does 3 x 60% plus 40% over 2 seconds (stacks) for 220% total effective DPS. That fits both my results and MCP's.


Mmmmm, it's been a while since I used Splinter but I am pretty sure it doesn't do any DoT damage and DoT portion does not stack.

Not sure why it's doing a bit more than what's being advertised. Blazing Spider is definitely doing way more than it's being advertised at 154%.
Reply Quote
That doesn't take away from MCP's results though. His numbers show us the maximum damage possible, assuming no defensive movements at all. From that we should be able to make better informed decisions about various builds and gear to support the bui


Aye its just that with MP0, stuff like Crits make a huge difference in time.

IE i did 2 MP1 Ghom kills last night with only Poison dart (no rune).

One took 20s one took 25s.

I didnt move on either run, i just stood there and attacked.

There are only two things that could have messed it up that much.

1. I fail with a stopwatch.
2. I crit a lot more in teh first test.

Longer fights will balance out the crit over time.
Reply Quote
Well, damnit, now I realize there's a better explanation.

Both MCP and I are using Skorn => BLEED.
Reply Quote
Well, damnit, now I realize there's a better explanation.

Both MCP and I are using Skorn => BLEED.


That's where your "DoT" comes from. :P

It's nice to know that Bleed adds quite a bit of damage in MP0.
Reply Quote

There are only two things that could have messed it up that much.

1. I fail with a stopwatch.
2. I crit a lot more in teh first test.


Yeah short fights are less conclusive if you do tons of Critical Damage which is what SKorn does because if you get lucky and score 1-2 more CD than the other run, the results are quite different.

Longer fights should even that out more.
Reply Quote
Just a thought, but are you guys running any IAS?

What if IAS is speeding up the spiders attacks in one way or another skewing the results.

Or maybe it is just Crits screwing things up since the fights only last 20s

(IE id id Ghorm MP1 in 20s last night with poison dart no run).

Anyone know how many HP Ghrom MP1 has?
I tecnically should have been doing 100% weapon damage per attack (1.16 attacks per second) and 40% weapon damage per second, (DOT). Meaning I should have been doing 156% weapon damage per second (assuming IAS has no effect on the Dot portion of the spell).
Reply Quote
Aye its just that with MP0, stuff like Crits make a huge difference in time.

IE i did 2 MP1 Ghom kills last night with only Poison dart (no rune).

One took 20s one took 25s.

I didnt move on either run, i just stood there and attacked.

There are only two things that could have messed it up that much.

1. I fail with a stopwatch.
2. I crit a lot more in teh first test.

Longer fights will balance out the crit over time.


Yes of course it's true that more is better. Taking the average of 5 trials should be enough to balance it though. Given 1.0 APS even his fastest result (CS/MS+bears) included 45 attacks. And even then, the crit % is applied independently (to each bear or spider) so that's easily over 100 crit% rolls. That should get you pretty close to the mean most of the time.

If you were going to test a pure wall of zombie pile on build at high DPS, yeah you could have fantastic variation depending on crits.
Reply Quote
Just a thought, but are you guys running any IAS?

What if IAS is speeding up the spiders attacks in one way or another skewing the results.


That could be. I have not done tests myself. I've only done key runs with Blazing Spider + PtV and while that pushes my paper dps much higher, my regular WoS setup kills faster. :(

Maybe I should do one run with Splinter. I am not surprised that SB does more than Blazing (it better be since it costs more than 10x mana!!!) but I am surprised Blazing beats Splinter in various tests.
Edited by Jibikao#1131 on 11/30/2012 10:46 AM PST
Reply Quote
That could be. I have not done tests myself. I've only done key runs with Blazing Spider + PtV and while that pushes my paper dps much higher, my regular WoS setup kills faster. :(

Maybe I should do one run with Splinter. I am not surprised that SB does more than Blazing (it better be since it costs more than 10x mana!!!) but I am surprised Blazing beats Splinter in various tests.


This could be explained by Bleed also, because it is known to stack. How many times are you going to proc bleed on a single jar throw? Possibly quite a lot.

When I get some time, I'm going to pick up a cheap 2H weapon with the tightest dmg range I can find and rerun some of these tests. Hopefully that will tighten up the range of results.
Reply Quote
Rofl @ the whole Skorn -> bleed thing.

That makes perfect sense as to why its messing up results so much.

The Bleed on skorn stacks, so if your using a low power spell (like PD) bleed is going to do more and more of the overall DPS.

Where as if you use bears, the bleed wont do as much simply because you kill faster.

I guess if you really want to test you need to use a non Bleed weapon har har
Reply Quote
This could be explained by Bleed also, because it is known to stack. How many times are you going to proc bleed on a single jar throw? Possibly quite a lot.


Being that you only proc LoH when a jar hits something, i assume bleed works the same.

IE if your jar hits a mob, you get X % chance to proc bleed on what it hit. (depends on skorn %).
Reply Quote
Are you trying to standardize the test results so that you can compare between each other, or just normalize the dmg and not change gear so that you can say well if 180% skill dmg does X then 220% skill damage should do Y but test results show it doing Z?

IE
I killed
Ghom MP1 6,037,544HP

20s kill

301,877.2 DPS

Sheet DPS is 171,864
Used Poison dart no rune i think I would expect 116% per second from the dart (100% * 1.16 IAS), and 40% per second from the dot.

However i think you are just trying to say ok if a 100% weapon damage skill has an effective DPS of 301k then a 200% weapon skill should have an effective DPS of 602k?
Reply Quote
For low MP testing, I am looking for results such as this:

* CS/WM + bears in a 1+3 rotation => %XXX sheet DPS
* CS/WM + bears + PtV in a 1+2 rotation => %YYY sheet DPS
* SB/WoS + BR + RoE => %ZZZ sheet DPS
* etc

A straight splinters test is not particularly interesting except as a way to make sure that the test itself is repeatable and that other variables properly controlled. If I understand your last question... yes, it may also be useful to uncover any hidden mechanics in the skills.
Reply Quote
Sheet DPS is 171,864
Used Poison dart no rune i think I would expect 116% per second from the dart (100% * 1.16 IAS), and 40% per second from the dot.


That isn't correct, because your sheet DPS already factors in your attack speed. What I would say your test tells us is this:

301K effective DPS / 172K sheet DPS => %176 effective for unruned PD, at your AS

(or at least that's what it would tell us, if you didn't use a skorn, lol)
Reply Quote
100 Undead Warrior
10965
Posts: 1,814
Corpse Spider do 120% (w/o rune) WEAPON DAMAGE BEFORE DIE, which sounds to me like a dot.

And just follow any rule of DoT, 2hander(skorn) is better than 1hander+offhand.


But unlike a DOT, spiders don't overwrite with each other, they stack, so cranking out more spiders does damage more swiftly. Re-applying Locusts or Haunt or whatever just refreshes the DOT. That's why big and slow works for them, but doesn't necessarily work for Spiders.
Reply Quote

Please report any Code of Conduct violations, including:

Threats of violence. We take these seriously and will alert the proper authorities.

Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.

Harassing or discriminatory language. This will not be tolerated.

Forums Code of Conduct

Report Post # written by

Reason
Explain (256 characters max)
Submit Cancel

Reported!

[Close]