d2jsp
Log InRegister
d2jsp Forums > Diablo II > Diablo 2 Discussion > Strategy & Guides > The Gavel Demon: Reimagining Goldfind Barbarians > D2:R
Prev1109110111112113211Next
Add Reply New Topic New Poll
Member
Posts: 13,184
Joined: Oct 14 2021
Gold: 86,215.90
Mar 6 2022 06:06pm
Quote (babun1024 @ Mar 6 2022 07:02pm)
1 hand bug still persists, headstriker's average damage is doubled = 463, it always does double damage unlike grief = 926. Grief does 433 * 1.35 = 649 damage with guilli on. Main problem with headstriker is ias.


so what's your theory on why 1x headstriker preforms so poorly compared to 2x?
Member
Posts: 19,309
Joined: Feb 24 2018
Gold: 9,765.50
Mar 6 2022 06:31pm
Quote (daanno2 @ Mar 7 2022 01:06am)
so what's your theory on why 1x headstriker preforms so poorly compared to 2x?

1x grief performs worse and 2x grief as well but the damage increase isn't 100% with 2x grief, more like 50% or less so there are some unknowns to be explored. Maybe, the bug works in one hand only but not on the other one.
Member
Posts: 13,184
Joined: Oct 14 2021
Gold: 86,215.90
Mar 6 2022 06:42pm
Quote (Exokek @ Mar 6 2022 06:54pm)
The DPS calc being fixed would make it worse than the times you've given. I'll def mess with it. Might be also that a longer range sword (Eth CBlade) is better for swapping targets & chasing, etc.


Quote (babun1024 @ Mar 6 2022 07:31pm)
1x grief performs worse and 2x grief as well but the damage increase isn't 100% with 2x grief, more like 50% or less so there are some unknowns to be explored. Maybe, the bug works in one hand only but not on the other one.


Definitely just found something very weird by accident, not sure if related. Was re-testing the 1x 340% NON ETH cryptic sword 22-338, which previously had these numbers:

run7 64s
run8 70s


After swapping back from a 2x Grief PB, all of a sudden I got:

run9 82s ???

Testing again, I got

run10 67s

Now, having in mind there's a lot of weird main hand/offhand glitches in the game, I decided to repeat the sequence of events prior to run 9 and re-test:

1) manually equipped 2x griefs
2) replaced main hand with the Oath
3) took off the other Grief

Results:
run11 81s ???
run12 80s ???


It's consistently adding 10s to kill meph.

For comparison, 2x Grief was killing meph in 43-47s.

Edit: retested the 2xGrief PB setup as well

Previous results:
run1 2xGrief:43s
run2 2xGrief: 47s

run13 2xGrief: 52s (after swapping around)
run14 2xGrief: 36s
run15 2xGrief: 36s

Recommendation is to exit/reload if any kind of weapon swapping has been done; seems to have huge effects on results.

This post was edited by daanno2 on Mar 6 2022 06:55pm
Member
Posts: 20,213
Joined: Nov 3 2019
Gold: 720.00
Mar 6 2022 07:48pm
Quote (daanno2 @ 6 Mar 2022 19:42)
Definitely just found something very weird by accident, not sure if related. Was re-testing the 1x 340% NON ETH cryptic sword 22-338, which previously had these numbers:

run7 64s
run8 70s


After swapping back from a 2x Grief PB, all of a sudden I got:

run9 82s ???

Testing again, I got

run10 67s

Now, having in mind there's a lot of weird main hand/offhand glitches in the game, I decided to repeat the sequence of events prior to run 9 and re-test:

1) manually equipped 2x griefs
2) replaced main hand with the Oath
3) took off the other Grief

Results:
run11 81s ???
run12 80s ???


It's consistently adding 10s to kill meph.

For comparison, 2x Grief was killing meph in 43-47s.

Edit: retested the 2xGrief PB setup as well

Previous results:
run1 2xGrief:43s
run2 2xGrief: 47s

run13 2xGrief: 52s (after swapping around)
run14 2xGrief: 36s
run15 2xGrief: 36s

Recommendation is to exit/reload if any kind of weapon swapping has been done; seems to have huge effects on results.


I mean the sample sizes are kinda wack, no?

For instance the same setup did Trav in 26s for me on Gavel and 56s with a clunky run. Need way more data points.
Member
Posts: 13,184
Joined: Oct 14 2021
Gold: 86,215.90
Mar 6 2022 09:05pm
Quote (Exokek @ Mar 6 2022 08:48pm)
I mean the sample sizes are kinda wack, no?

For instance the same setup did Trav in 26s for me on Gavel and 56s with a clunky run. Need way more data points.


Well, this is vs Hell Meph, with a huge health pool. The external factors that exists with Trav council (spawn locations, travel time between fleeing monsters, randomly healing, random affix spawns) don't exist with Meph. The only RNG factors are

1) Chance to hit (should be negligable with lvl 25ish battlecry
2) Deadly strike (20% chance on grief)
3) Min-max dmg ranges (huge on cryptic sword, nonexistent on Grief PB)

Imo the large health pool and the couple hundreds of hits it takes to kill averages all this out.

Some rough math:

Hell Meph 94,320 health
50% block
20% phys immunity
25% dmg taken from mercs
94,320hp * 2 / .8 * 4 = 942,300 ehp

+400 Grief pb (I actually had a 396 and a 385)
avg dmg = 433
20 % deadly strike = 520 avg base weapon dmg

Off weapon ed
326% frenzy
249% strength
160% damage vs demon (grief)
200% dmg vs demon (CoH)
Total = 935% ed

520 * 10.35 = 5383 dmg per swing
Plugging in the ias calc, we're supposed to be getting 5fpa frenzy swings once it ramps up

5383 * 24fps/5fps = 25838 raw dps
952,300 ehp / 25838 dps = 36.4s <--actually spot on with my tested results

Going to do the same math with the oath cryptic sword and see how it works out to my previous results.
Member
Posts: 13,184
Joined: Oct 14 2021
Gold: 86,215.90
Mar 6 2022 09:18pm
1x 340% NON ETH Oath cryptic sword (22-338, 180avg)
850% off weapon ed
180 * 9.5 = 1710 dmg per swing
supposed to be getting 5.5fpa (not really sure though how having just 1x sword is being calculated)
1710 * 24fps / 5.5fpa = 7462 raw dps
952,300 ehp / 7462 dps = 127s

As a remider, here were the results for this one. Runs 9, 11, 12 was what I now consider "glitched" due to weapon swapping.

run7 64s
run8 70s
run9 82s ???
run10 67s
run11 81s ???
run12 80s ???

So the 127s expected TTK is roughly double what I got on runs 7, 8,10. Again, this is with a SINGLE sword equipped. Not sure whether this is due to double dps bug or something else.
Member
Posts: 20,213
Joined: Nov 3 2019
Gold: 720.00
Mar 6 2022 10:11pm
Quote (daanno2 @ 6 Mar 2022 22:18)
1x 340% NON ETH Oath cryptic sword (22-338, 180avg)
850% off weapon ed
180 * 9.5 = 1710 dmg per swing
supposed to be getting 5.5fpa (not really sure though how having just 1x sword is being calculated)
1710 * 24fps / 5.5fpa = 7462 raw dps
952,300 ehp / 7462 dps = 127s

As a remider, here were the results for this one. Runs 9, 11, 12 was what I now consider "glitched" due to weapon swapping.

run7 64s
run8 70s
run9 82s ???
run10 67s
run11 81s ???
run12 80s ???

So the 127s expected TTK is roughly double what I got on runs 7, 8,10. Again, this is with a SINGLE sword equipped. Not sure whether this is due to double dps bug or something else.


Man the weapon swap glitch is weird. Wonder if you can swap in the opposite direction? Basically retaining the IAS/DPS of the initial set when swapping into new weaps.

Would be wild.
Member
Posts: 14,331
Joined: Aug 20 2021
Gold: 49,720.00
Mar 7 2022 05:33am
I just watched video from german d2r streamer about a5 merc...

The attack value generally is a joke especially vs diablo he dont hit very often only when using fanatism aura its slightly better...

This guy tried dual grief, dual lw and dual plague for poison java.

This post was edited by Spartacus160 on Mar 7 2022 05:36am
Member
Posts: 19,309
Joined: Feb 24 2018
Gold: 9,765.50
Mar 7 2022 06:28am
Quote (Spartacus160 @ Mar 7 2022 12:33pm)
I just watched video from german d2r streamer about a5 merc...

The attack value generally is a joke especially vs diablo he dont hit very often only when using fanatism aura its slightly better...

This guy tried dual grief, dual lw and dual plague for poison java.

That's why we use maxed battle cry on Mephisto. I could do the test with a demon limb charge but in general, even real frenzy barbs don't get enough attack rating for the points spent in mastery and frenzy skill.
Member
Posts: 20,213
Joined: Nov 3 2019
Gold: 720.00
Mar 7 2022 06:36am
Quote (babun1024 @ 7 Mar 2022 07:28)
That's why we use maxed battle cry on Mephisto. I could do the test with a demon limb charge but in general, even real frenzy barbs don't get enough attack rating for the points spent in mastery and frenzy skill.


Yeah Battle Cry or some way to get Conviction (RIP) would be ideal.
Imagine 1-handed Infinity :rofl:
Go Back To Strategy & Guides Topic List
Prev1109110111112113211Next
Add Reply New Topic New Poll