GU50, Tunso Testing, and Player Testing

I didn't execute that test. The person that did may read that comment and give that a shot. Otherwise, I'm looking for anyone familiar with Mental with PC Test access to step up and run a couple tests. (I'll take as many as I can get).

Is there gonna be a Rage Tunso Test.
Frenzy got a nerf... not sure that makes sense at melee.

For a Tunso test, it just hit 98% baseline with that 20,239 DPS result. (I welcome more tests)

#FUNFACT: Right now we have 73 Tunso Tests and 41 Live Tests feeding these charts.

Here are the Live Tests that have come in so far.
Melee: TK Bolt, Phantom Flames, Psychic Shock, Mass Terror, Menace, Psychic Resonance
Range: Pyrokinesis, Resonance, TK Bolt, TK Push, Psychic Shock, Mass Lev

That hit 99% of the Live Test baseline for Melee and 104% of the Live Test baseline for Range.

Here's some Live Test Standings

Melee


Range


Deviations


I haven't forgot that MIDRANGE is something that needs a decision. It is a big deal. That said, the speed at which the outliers for pure melee and range are coming in line is pretty fast.

I'll do tests later
My fingers hurt and I'm sleepy

Nice to see rage doing better now though. And Mental.

Menace is still bugged.

Using Menace with WM and a harmful ability at 0m resulted in a 74 damage hit:



Using it with pure AM at 0m in a group of sparring targets resulted in something interesting. I thought Menace was a single target ability but here it hits twice with the first hit being normal and the second hit being only 49 damage:



Unfortunately, sometimes both Menace hits would be bugged. Here they only hit for 131 and 44 damage:



I recommend dropping Menace for now. Just replace it with Pyrokinesis. Menace may be skewing results, which could result in an overbuff once it's fixed.

quick question - do you guys testing have direct feedback access to the devs? or just hope they read it

cause they really need to read this stuff.

seriously.

They're reading it, and making adjustments.

I'm wondering if its some strange interaction with burning from phantom flames here? Those look like ticks of burning damage.

I didn't use Phantom Flames for that test. I didn't want to clutter the combat log with a lot of little DoT ticks.

Above was my Tunso test on the 8-target set. Below is the same test, but with the 3-target set:

17273 AVG 16008.81
17003 MEDIAN 15992
14991
11796
13297
15705
16308
18526
15593
18130
16096
16887
20267
17953
15434
17227
15813
14214
18380
15992
14728
15119
13261
15301
13642
16844
16858
16066
13627
14148
19794

You'll notice it's about 2500 DPS lower. This is due to the unique nature of PcDL. Each glyph that spawns (max 4) will split it's 5-tick DoT damage (not the initial burst) after 2 targets. That means up to 8 targets, if spaced properly, can receive the full DoT damage from PcDL glyphs. Most every other attack in the game will simply split after 2 on the whole. For this reason, going from the 8-target set, where I was consistently hitting 7/8 targets with the glyph DoTs, to the 3-target set, where only 3 were able to receive the DoT damage, dropped the effective damage of those DoTs by about 57%. When you see how large a chunk of damage those glyphs are in this rotation, it makes sense why the DPS changed with target number.



The reason I'm bringing this up is that I don't want Celestial to be adjusted based on the relatively uncommon 8-target scenario. Any other powerset, would demonstrate no significant difference in DPS between the 3-target set and the 8-target set (that includes Gadgets with it's broken Sticky Bomb that splits after 3, rather than 2). To drop Celestial's damage output, based solely on the 8-target results, would leave it underpowered in the more common situations in which there are 4 or fewer targets.

FWIW, I also tested the 3-buff loadout: DL/Anoint/DP/Benediction>ScH/RcW. It was inferior to the rotation I used above. Interestingly, it was also inferior on single targets by a little under 1k DPS.

Have the devs made any comments about possibly changing their testing methodology? The testers have clearly shown that damage does not scale equally between power sets from the Tunso spec to the “Live” spec.

Balance needs to be achieved at the high end of game play. Balance at the high end then helps to ensure skill is rewarded for competitive players. If the high end is balanced, scaling factors then have a smaller magnitude of effect on lower stats. But if you balance at a lower point, those stat and ability differences are greatly amplified at the top end. That is why we see scaling differences going from melee Sorcery at 1.25x to melee Quantum at 1.66x the Tunso tests. Both the Tunso and “Live” tests were done at CR 126 but the differences come from stats, mods, proficiencies, and how the power abilities magnify them.


Just looking at those two powers in melee, the Tunso test shows that Sorcery (18,253 dps) does 13.8% more damage than Quantum (16,043 dps). But if you change the spec to the “Live” test, now Quantum (26,711 dps) does 16.8% more damage than Sorcery (22,863 dps). That is a shift of 30.6%. And they are both on the low end of the melee spectrum.

Has anyone confirned that the major descepency in scaling is due to stays our just the addition of gone turf mods. Obviously escalating might will benefit quantum more than sorcery. Is the scaling more consistent of you tunso test with all white mods?

Quantum and sorcery are the cannonical examples.

Quantum doesn't use weapons. Core Strength buffs powers loaded into Time Bomb and buffs the strength of Time Bomb itself. Quantum gets stronger with Escalating Might. Quantum gets stronger with crit affinities. Quantum gets stronger with more might stats in general.

Sorcery has some of their damage get stronger because of all those things. Their abilities and weapon attacks benefit from stats and white mods and affinities. Fury does not get stronger with stats. Fury doesn't get stronger with white mods. Fury doesn't get stronger with league proficiencies or crit affinities. A notable portion of Sorcery's AM damage will not increase at all.

Meanwhile there's a few powersets that seem to scale together at around the same rate around x1.4. But odd things happen. Nature has a TON of loadouts that all do the same damage as each other in Tunso Tests but even they have a crazy amount of variation in the live results.

I suspect other factors affect the results but the math involved is more than I have time to prove.

I kinda agree. I can see why you would take your time to make the right decision.

No doubt there will be tears from someone, regardless, just as there are tears now. Ultimately, though, the issue can't be ignored.

Can you plz explain why? A cursory answer would be sufficient atm.