PvE DPS Balance Initiative: Scaling and Midrange

Lol. Don't blame me, I was pretty vocally against GU36 to begin with. I though the powers were more or less in-line, with some exceptions both above and below par, and in most cases more experience with a power or a technique netted stronger results. I would love to go back to that era, but we've foregone that playstyle in the name of balance. With that in mind, I must respectfully argue - even with my forum pals - with this in mind.

Not just that but Sorcery can do immense Burst with the Pet SC. Even Baleful and Polymorph are still good for WM.
then we have Quantum's Temporal Vortex in its AM

I think it's fair to test both ways and let the facts speak. I would ideally like to say "this is how the core AM is performing and this is what players are able to achieve". I think it provides the evidence needed for the devs to either accept the difference, balance the game with the tricks the design enables, or address the design to prohibit those tricks. For example with Gadgets/Mental, I definitely wanted numbers without the super clip. But if I wasn't already aware the super clip was going away, I'd want tests with the super clip also to be able to empirically show the impact of it because that may be the information needed to get them to address it.

Another point you may want to consider is the rotation window. All the testing done to date has been gathered using a specific rotation performed over and over within a specific timeframe (30 seconds, I believe). Adding in a supercharge skews that rotation timing in 2 ways:

1. 50% and 100% SC would not accumulate enough damage in the 30 second window, so you'd be forced to lengthen the window to accommodate....changing the test baseline.

2. If you pre loaded your charge before starting the test, the test is still thrown off as the larger SC's would only be fired once in the 30 second window due to cool down or recharge or both. Again changing your baseline or throwing off the 30 second window.

Just something to consider

So what about consumables?

The testing is 10s samples for a sustained duration of 3 minutes. I'll admit I've dropped as low as 2 minutes and called it good enough. 30s would be insufficient in any event. SC does have those concerns in larger samples too though.

In my mind they fall in the "variables too difficult to control" category.

Perhaps we should actually consult the Devs about what they intend to balance around. Testing on our opinions is moot.

I have sent them a PM with a link to this discussion asking for a ruling.

Like that part alot. Also question regarding 8-targets: If I hit 8 targets with an (1) abillity does it crit on every single target (or none at all) or is it "mixed" ?

Lets just get the regular power balanced first, players are trying to add to much other stuff. Lets not jump out are bodies

^ Exactly what he's saying.

Mixed. More specifically it crits once and the damage is split.

I agree with pitbull. SC abd RSK should be considered variables outSide of control and should not be balanced around. That said, some possible wiggle room is accounTed for in the +/- 5%. Those powers with the room for it should be in the lower section. We should not balance around all "tRicks of the trade" because that id's not pay of the power.

Should pet classes be balanced for a gadgets troll with battle drone because the pets do insane damage when that is active? Or earth using movement SC Suprisibg how many don't know about that to begin with.

That has been bothering me for quite a bit now. My concern results from the amount of targets used in both tests (8).
Im under the impression that "ingame" there are most of the time less targets being hit at the same time. Anecdotal observation: usually 3-5 max. Therefore also crit-affinities/leaguehall might be overvalued.
Let me explain (If my math doesnt fail me- it does sometimes):
I assume "Tunso" uses a total of 23% crit while "Live" uses a total of 29% crit (plz correct me on that if wrong)

a1) "Tunso" 8 targets, 23% crit -> expectation ~1,84 targets are hit by a "crit" per ability usage // "atleast 1 crit %" ~0,88

b1) "Live" 8 targets, 29% crit -> expectation ~2,32 targets are hit by a "crit" per ability usage // "atleast 1 crit %" ~0,94

-> "expectation" due to 6% crit more, on 8 targets, goes up by ~0,48 "targets being hit by a crit per abillty usage"
-> "atleast 1 crit" difference: ~6% per ability usage

a2) "Tunso on 3 targets", 23% crit -> expectation ~0.69 targets are hit by a "crit" per ability usage // "atleast 1 crit %" ~0,54

b2) "Live on 3 targets", 29% crit -> expectation ~0,87 targets are hit by a "crit" per ability usage // "atleast 1 crit %" ~0,64

-> "expectation" due to 6% crit more, on 3 targets, goes up by ~0,18 "targets being hit by a crit per ability usage"
-> "atleast 1 crit" difference: ~10% per ability usage

Why am I posting this?
Imo the choice of doing tests on 8 targets COULD atleast to some degree (which Im not to sure about as im lacking knowledge) lead
to bias regardig results and scalars as it puts more emphasis on the 6% crit difference. 8 targets/29% crit basically lead to almost every 2nd ability hitting 1 extra target (of the 8) with a crit over "Tunso"(23% crit).
Compared to "only 3 targets" (which imo also meets ingame conditions better) 6% crit difference only leads to every 5th+ abillity usage getting an extra crit over "Tunso". I would consider % of "atleast one crit" less important here (close values) although them being in favor of 8 targets. After all I think "8 targets" could lead to an overvaluation of crit-affinities/mods as well as possibly also influencing scalars of different powers according to their respective indivdualities even more.
It could very well be that what I posted doesnt matter all that much (and that I miscalculated ofc ) and Im not to sure about that either (rather short thought-process with lacking additional information) but I decided to leave that here nonetheless.

It would be highly naive to exclude Robot Sidekick. There is no point in balancing the powersets if you do not include abilities that will absolutely be used and absolutely increase the damage-out of some powersets. You will only create further imbalance by excluding it.

Supercharges should be tested, but right now there are practical reasons to exclude them. I won't even consider touching SCs until the devs provide us with an instant SC soder or trinket on Test Server.

There are also logical reasons to exclude buff soders and trinkets. You should always assume everyone running these rotations at a competent level in end-game content are already using these buffs.

This logic can also be applied to consumables to an extent. At the very least, you can apply this logic until a rotation or trick is discovered using consumables that truly creates imbalance, and it can be addressed as it arises.

Interesting idea and great innitiative, but I think this solo testing of powersets is a little bit flawed. Take Mental/Ice/Munitions you can test them as a solo powersets, but bring in Electric or Nature to buff them, and the results will start to look different. Take this example: a buffed Electric is weaker than an unbuffed Mental. Put these two together in a group and Mental will demolish Electric even more. This is a major factor that cannot be overlooked imo. After all, this is a MMO, not a solo game and these things should be factored in.

That's one thing, another thing is that damage on adds skews results and balance in this game, that's mostly because of bad raid design: 2 adds here, 2 adds there etc ... The way a few leagumates and I were testing powersets is this: 3-4 of us go to the 3 sparring targets, bring them down to 20% each (200k HP per target) and then all of us would start attacking them at the same time. The idea was to try and emulate what would happen during a raid on trash adds, which sadly make up the larger portion of the total HP pool of a raid (in most cases anyways). At the end we measured how much damage each powerset did. The discrepancies were huge, but you wont get these results by doing solo powerset tests. That's why results on sparring targets dont match results that people get in raids.

Honestly some supercharges are highly impractical to test. The hardest one would be Temporal Vortex. Quantum already gets inconsistent results with the large range of Time Bomb's potential Damage, and Temporal Vortex would have the same inconsistency. You would have to spend time getting sc to get enough uses to give an accurate representation of it which means potentially surrendering any semblance of a personal life all to test one power.

Frankly I would like them to balance the pure AMs, then at a later date look at powers with lacking supercharges and help give them better ones while balancing the others. I would rather have Mental finally get a useful supercharge at a much later date then have Quantum, Celestial, and Rage get nerfed now even for those that don't actually use the supercharges.

Yep too much this will go wrong way lol I mean. If we take into account of electrics super charge and add that to the damage it looks better then it is ..you can only use the super charge like every 10 mins lol then you co
Pare that to ice without super charge ? Know electric is not that good. Wow this is getting crazy ! Base power sets !

If an Ice player is buffed because of my Carnage, that's my damage. I don't care if the scoreboard doesn't show it that way.