Community-Agreed Parameters

[Galaxy Brain]

41 minutes ago, UltraAlt said:

The game sets the Parameters.

The Community is always within those Parameters.

The Community doesn't set the parameters; the DEVs do through game changes.

 

If parts of the Community want to limit themselves or create little boxes for themselves, then that is their version of game play and their own personally imposed parameters.

I am opposed to anyone that is trying to set parameters on others that are not set by the game itself or Code of Conduct.

What I mean by this is agreed upon parameters for testing stuff out as it seems discussions about certain topics vary wildly depending on the point of view, and it ends up with the discussion / balancing point (even when testing in beta) being lost in the fray. 

 

I'm not asking anyone to "play this way or else", but rather to find a way to have structure when it comes to balance/etc talks given we do have thousands of ways to play the game. If we can get to like "here are 5 concrete tests to go off of" that represent dozens of styles and approaches, then great! Otherwise, we could have any anecdotal evidence come in and be unsubstantiated such as "Oh, I bought 10 super inspirations and was able to do X, so Y is fine." which does not point to Y being fine as a very big variable was tossed into the mix.

 

The ultimate goal would be to have community-agreed standards that we can weight against. If sets A/B/C can all do X/Y/Z thing, but then Set D (which is in the same group) can only do X/Y and not Z, then there may be something to look at with D. We just need to determine XYZ.

 

[Yomo Kimyata]

19 minutes ago, Galaxy Brain said:

 

 

The ultimate goal would be to have community-agreed standards that we can weight against. If sets A/B/C can all do X/Y/Z thing, but then Set D (which is in the same group) can only do X/Y and not Z, then there may be something to look at with D. We just need to determine XYZ.

 

I hear you, and I admire your scientific approach.  I question, however, if there is a need for a community board on such things.  If Set D is so terrible, people vote with their feet (and their complaints on this forum).  If the devs feel it is necessary to equalize sets, then they can.  If they don't, then they won't.  It's not game breaking in the slightest if set D is subjectively worse than other sets; it's not game-breaking if it is *objectively* worse.  There aren't quotas here that I am aware of.  And the game is forgiving enough that even the worst sets can succeed with flying colors.

 

My two inf

[ScarySai]

I don't really see these restrictions being applied outside benchmark scenarios, like pylons.

 

And that's only because it's necessary to form a baseline for set performance. A claw scrapper rocking assault has a pretty good shot of beating a better single target set that's not using it, for example. 

[UltraAlt]

3 hours ago, Galaxy Brain said:

What I mean by this is agreed upon parameters for testing stuff

You are talking about testing by solo players I'm assuming as nothing you have posted seems to indicate group size, sidekicking/mentoring, group structure/mix, or player skills at teaming.

Teams are generally going to be a mix of the categories you listed even in beta.

 

Group tactics, the "gel" state of a group, and the experience and skill level of players goes a long way.

[Bill Z Bubba]

24 minutes ago, UltraAlt said:

You are talking about testing by solo players I'm assuming as nothing you have posted seems to indicate group size, sidekicking/mentoring, group structure/mix, or player skills at teaming.

Teams are generally going to be a mix of the categories you listed even in beta.

 

Group tactics, the "gel" state of a group, and the experience and skill level of players goes a long way.

This goes back to the argument about insps usage.

It has no relevance to trying to get an understanding of powerset performance within an archetype.

[Galaxy Brain]

37 minutes ago, UltraAlt said:

You are talking about testing by solo players I'm assuming as nothing you have posted seems to indicate group size, sidekicking/mentoring, group structure/mix, or player skills at teaming.

Teams are generally going to be a mix of the categories you listed even in beta.

 

Group tactics, the "gel" state of a group, and the experience and skill level of players goes a long way.

 

11 minutes ago, Bill Z Bubba said:

This goes back to the argument about insps usage.

It has no relevance to trying to get an understanding of powerset performance within an archetype.

 

It is also too big a variable with what others bring to the table and how they interact. With the note on insps, we could say "test with 1 purple insp on at all times" and that could emulate a support character on the team with defense boosts.

 

In fact, a quick eyeball between Corruptor, Controller, MM, and Defender versions of Cold Dom > Ice Shield, the average value (unenhanced) is 12.19.... very close to a small purple's 12.5%. Enhanced, this would be just under 20%, which we could fudge and say 2 purples / 1 medium purple on at all times could be the same as having a defense oriented support character with you (if they also have Maneuvers, etc).

 

 

[Bill Z Bubba]

1 minute ago, Galaxy Brain said:

With the note on insps, we could say "test with 1 purple insp on at all times" and that could emulate a support character on the team with defense boosts.

I don't think you should bother unless you're actually attempting to emulate a teaming experience which is just as ridiculous as attempting to take into account SetIO bonuses or Insp usage.

[Galaxy Brain]

10 minutes ago, Bill Z Bubba said:

I don't think you should bother unless you're actually attempting to emulate a teaming experience which is just as ridiculous as attempting to take into account SetIO bonuses or Insp usage.

I kind of want to out of pure academia, like redo the tough tests or run in with a blaster with just some added inspiration to see how impactful they really are, but it'd be more like "how impactful is an insp in general" vs redoing whole swaths of data

 

Edited March 9, 2021 by Galaxy Brain

[Bill Z Bubba]

Ain't nuthin wrong with curiosity.

[TemporalVileTerror]

Even if the tests don't include all variable conditions, they can still provide valuable data.  And -knowing- that particular variables exist and are not inherently represented in that data is also valuable.  It's the basis of the scientific method.

Establish parameters.  Conduct test.  Observe.  Remain skeptical.  Analyze data.  Ask questions.  Rinse.  Repeat.

 

Of course, much like I said before, we do have to be aware that these tests and benchmarks aren't necessarily going to answer the question "what's the best way to balance the game."  Rather it'll be closer to:  "Assuming players are interested in playing the game under the conditions expressed by Galaxy Brain's thesis statement of 'Playing CoH = Designing a highly customized character in order to use super powers and defeat hoards of enemies in specially instanced missions.' then how do various Power Sets perform under specific conditions?"

 

And within that framework, setting up a series of spreadsheets and formulae to create a predictive model that is future-proofed for balance passes gives the community and the Devs a potential tool to meet at least some players' expectations.

 

It's ambitious, but if @Galaxy Brain and other interested community members can pull it off, and share their findings, I say:  All the power to 'em!

Though I will encourage them to be mindful of introduced errors and biases potentially corrupting the data, and account for that when they perform their analyses and conclusions.

[Troo]

On 3/8/2021 at 10:11 AM, Galaxy Brain said:

Starting with difficulty, we have the ability to tweak the enemy's level from -1 to +4, the team size from 1-8, enable or disable bosses, as well as enable or disable AV's. off the bat, I feel that AV's should be disabled as honestly beating them down solo is more of a gimmick than an expected task. Likewise, I feel Bosses should be enabled as normal content seems to be built around boss-tier enemies being standard. The next step is a doozy with a possible 40 combinations... but I feel we can narrow it down to 6:

Whittling those down further makes sense. Then reflecting milestone levels when players spend time would dial it in to give a glimpse of what players experience along the journey.

 

It does seem like there should be an opportunity to get some baseline metrics, it just doesn't seem like that big of an ask to take a peak at what levels do player play at and at what difficulties.

The numbers may be skewed by factors such as multi-boxing, power-leveling, etc. but some measures could be used to eliminate easy data points such as AE missions and fire farmers. (maybe not)

 

Without knowing, it's just guessing. Certainty is not required but reality should be a prerequisite.

[America's Angel]

"How should the game be played?"

 

I've given it a bunch of though this week, but hard testing parameters are impossible in PVE, sadly. The PVE side of this game is just played too differently by too many different people. There is no such thing as a hard baseline in such an environment. A baseline for DPS and HPS are almost impossible to work out. (Unlike PVP, where everyone plays pretty much the same way and balance is pretty easy to figure out, comparatively.)

The best approach to PVE balance is just to have the players, each of them with a varied playstyle, provide their own, unique feedback based on how they play. And then collate all of that feedback get an overall "feel":

 

For example...

 

Galaxy Brain likes to stand still in Cimms using only SOs - this is their feedbackl.
America's Angel likes to run missions with insps, P2W, and everything else - this is their feedback.

Infinitum likes to run missions without insps/P2W, but uses incarnates - this is their feedback.

Werner likes to run missions with enemies debuffed, no insps, alpha-only, etc - this is their feedback.

Uniquedragon likes to run missions with red number SOs - this is their feedback.

Etc etc.

 

And then from all of this feedback mixed together, an overall consensus emerges. (Which the devs can then compare against their own knowledge, datamining, and number-crunching to conclude the best course of action.)

 

What this means is that Focused Feedback threads, and the debates within them (to help sort the good ideas from the bad) are already the best way to test armours in this game.

[Galaxy Brain]

3 minutes ago, America's Angel said:

"How should the game be played?"

 

I've given it a bunch of though this week, but hard testing parameters are impossible in PVE, sadly. The PVE side of this game is just played too differently by too many different people. There is no such thing as a hard baseline in such an environment. A baseline for DPS and HPS are almost impossible to work out. (Unlike PVP, where everyone plays pretty much the same way and balance is pretty easy to figure out, comparatively.)

The best approach to PVE balance is just to have the players, each of them with a varied playstyle, provide their own, unique feedback based on how they play. And then collate all of that feedback get an overall "feel":

 

For example...

 

Galaxy Brain likes to stand still in Cimms using only SOs - this is their feedbackl.
America's Angel likes to run missions with insps, P2W, and everything else - this is their feedback.

Infinitum likes to run missions without insps/P2W, but uses incarnates - this is their feedback.

Werner likes to run missions with enemies debuffed, no insps, alpha-only, etc - this is their feedback.

Uniquedragon likes to run missions with red number SOs - this is their feedback.

Etc etc.

 

And then from all of this feedback mixed together, an overall consensus emerges. (Which the devs can then compare against their own knowledge, datamining, and number-crunching to conclude the best course of action.)

 

What this means is that Focused Feedback threads, and the debates within them (to help sort the good ideas from the bad) are already the best way to test armours in this game.

 

I like this train of thought! Each one of those examples is something a single person could do in theory as each is a parameter, its just a matter of boiling down which would cover the most bases.

 

 

Galaxy Brain likes to stand still in Cimms using only SOs - this is their feedbackl.

For mine, it is somewhat unrealistic as in practice people do not stand still and just take it on the chin. It is good for isolating the (armor) however, but on reflection there can be a better way to track this though it will require more time. In the interest of boiling it down, this could be classified as "Run the test on SO's only with what'd be assumed basic slotting", which would be a category for anything, not just armor.

 

America's Angel likes to run missions with insps, P2W, and everything else - this is their feedback.

This is valuable info! But we need to be careful how far it goes when it comes to repeatable results. Sure, you can run content with a mix of buffs which is realistic as many folks do this. The trouble is that due to RNG in the game, you're likely going to need to run the same task 3 times or so to get a good average result. If we were to boil this one down: "Run the test with X loadout of P2W buffs / X set of inspirations available to use" in order to be consistent.

 

 

Infinitum likes to run missions without insps/P2W, but uses incarnates - this is their feedback.

This one is much more straightforward, you could tag this as "Run with full incarnate abilities" or "Run with a 'top tier IO build'", or both combined!

 

 

Werner likes to run missions with enemies debuffed, no insps, alpha-only, etc - this is their feedback.

Werner is the man, but I feel like this one gets a little too "specific"? We have other runs that would be 'no outside factors' as well as 'use incarnates only', Outside of the tweaking enemy stats (which is another bucket of variables), this one is sort of the same as Infinitum's example except limited to Alpha instead of a full suite. That said, it could be worth at some point breaking down the incarnates per tier to show their impact, but that'd be one of those things that would just need to be done once or so. As for what this boils down to, the specific run we could imagine would just be somewhere in between the base / full incarnate run logically as that is what these parameters provide.

 

 

Uniquedragon likes to run missions with red number SOs - this is their feedback.

So this one does puzzle me a little, is this not the same as just having no enhancements?  I can see value in just going at it with powers set to their raw base values... but this goes far against the grain of what is expected. While there is no expectation that everyone uses IOs/Top end IOs, or even Incarnates, the game is set up to where it does expect you to slot basic enhancements. I can see this being a very focused parameter if something is say, suspect of being overtuned and comparing it even without enhancement to something with enhancement, but as a general "we are trying this out" test I feel it'd be something you have to go out of your way for and would normally come up short compared to the rest.

 

 

Regardless, we will ultimately need multiple people running things to get the best data possible. But like you said, unlike PvP not everyone plays the same. Boiling down several criteria though where there are "brackets" of testing would be the next best thing where we can at least control for "they just used SO's" or "they went full incarnate" or "ok, they used all the P2W buffs and had 5 purples on hand to use".

 

[MTeague]

I would be disappointed to lose options for +2/x1 with Bosses. I find that a good mix when soloing on a Defender or Controller.  A character who can get overwhelmed by numbers, but wants a moderate challenge when soloing at lower levels. (not flashbacking... I mean when the character actually is lvl 12, doesn't have many slots, can only just start to use DO's, etc). 

 

That said, I'm with on "Bosses should be on."  I don't care if I'm soloing a Defender in First Ward, bosses are always on. Period

 

Nevermind. Re-reading, I misread what you were going for. Carry on.

Edited March 18, 2021 by MTeague

[DougGraves]

I play almost exclusively low levels (1-35) with almost all characters having just SO's.  I do use KB>KD on my AR character and my ice/time controller has a miracle +recovery.  Here is my experience with power.

 

Normal content a team of level 1-20 characters has no problem with at +0 and can do at +1 with some difficulty.  Hollows teams, radio teams, etc should be at +1 even at level 10.  KR story arcs are more difficult and can be tough at +0 but if everyone has SOs and is the right level +1 is very doable.

 

By 20-30 +1 is easy for a team.  +2 is not challenging but has some danger.

 

30-35 +2 is easy, +3 is routine.

 

35+ +4 is not challenging for normal content (radio missions, etc.)  Teamwork still matters, blasters are soloing half the mission.  But as a group you just steamroll everything.

 

45+ I just stop playing.  The team approaches an enemy group and a judgement wipes out everything except the lt's and bosses who go down seconds later.  It is beyond steamrolling and is just kids stepping on ants.

 

IO set bonuses do impact lower play.  But they generally just make a character invulnerable.  A blaster will tank on a team because nothing hurts them.  They do not steamroll the mission but it does make the mission trivially easy.  I quit those teams.  The worst is an IO'd character on a TF who just skips through the enemies to complete the mission solo.  TF's become a lame joke and are no fun with such a player.

 

tldr; for normal content even with SO's the characters are overpowered at higher levels at steamroll everything.  IO sets matter but not as much as level.  Judgements just make the game a trivial joke.