Request for developer comment on dynamic scaling

Unknown

This is more a suggestion than a discussion, but who reads the suggestions subforum?

After two acknowledged "substandard" PvE events in succession, I would be much encouraged by a brief statement along the lines of the recent "Response to Recent Changes" D3PCS news post discussing Demiurge's assessment of and goals for the recently introduced dynamic difficulty scaling, whose initial outings proved a tad contentious with a portion of users.

Any communication would be gratefully received, but I presumptuously believe the two questions below adequately frame what some significant portion of your users would like to know.

What is dynamic PvE difficulty, when working as intended, aiming to achieve?
We ask you for many things; we are a petty and demanding bunch. To my knowledge, no one asked for this. Presumably, the intention is to tailor the challenge offered to each player. Certainly, there seems to have been a much reduced correlation between roster strength and achievable rating lately in PvE and PvP, in direct contradiction of your lead designer's previously stated goals, though anecdotes from brackets of 200 are hardly solid evidence of anything. At the moment, new players are complaining that established players and people with the temerity to give Demiurge money are holding them down, established players are upset that Johnny-come-latelys are making off with their rightful Daredevils, and many in the middle seem to believe both extremes are being promoted at their expense. It seems unlikely they are all correct. Could you give us some indication of what you are trying to achieve with the recent changes, perhaps before we need an entire subforum dedicated to accusations that two out of three players are obviously cheating?

How is dynamic PvE difficulty supposed to work?
At the launch of Oscorp, we were told that it is not: carried over from previous events' PvE "MMR", roster-dependent, working as intended, susceptible to PvE "tanking" by grinding unused characters into dog food, keyed to the lunar calendar. Fair enough. So what is it supposed to run off? I understand you might be very reluctant to answer this. Whatever you do, I and just about anyone else here will try to game it to our advantage, and I won't insult your intelligence by pretending otherwise. But if it isn't doing what it's supposed to and we don't have the first idea what it is supposed to do, tinfoil millinery and witch-hunts abound.

tl;dr: uh, guys, **** you doin'? And then, of course, there's PvP. Maybe another day.

Unknown

I think IceIX posted something about it in the other thread.

In an idealized environment, think of it like this:

New player gets into the game. Sees a level 10 Soldier to battle with his level 6 average 1* team. They can play and get through the content. If they push, they may even get enough Iso rewards that they can build their team to place in the top 50%-ish. Who knows? Maybe they'll hit an easy bracket and score a 3* reward and majorly improve their team in one fell swoop. If not, they were able to pull several thousand Iso, a bunch of Recruit tokens, and some other stuff out of the event.

A 2+-ish player goes into the game. Sees a level 30 Soldier battle with a team of 65s on average. They rock the first bit of it until the difficult ramps up. They'll have an easier time at an equal "difficulty" compared to the new player due to a larger roster and the ability to choose who is best in a given fight. Facing Countdown tile madness? Use board control. The newer player doesn't have this flexibility. This team would be hitting the 25-50% range with peaks in the top 10% of a bracket. If they place "average", they pulled a bunch of Iso out, more than enough to raise a few heroes a few levels to put them in a better position to hit the next event. The Recruit tokens may help there too, although some are just added Iso due to "useless" extras.

A 2/3* player does the same with a team of 85/100s. They are seeing harder matches level-wise than the mid 2* player, but the added utility that the other characters bring means that they're able to handle the situation more easily and thus win battles faster and with less downtime. This player would be hitting the 10-25% range with peaks in the top 5%. The extra Iso will be pumped into bringing up some laggards in the team that are Boosted for the event and giving better results. They're more focused on the meta at this point than the climb.

Continue from there. We want players to be able to experience all the gameplay that's available and to be able to continually take steps to become more powerful. That being said, everyone else is taking the same steps, so it'll be up to the player to play better or smarter to move faster than their opponents.

Unknown

Cheers, I thought I'd caught all IceIX's drop-in comments of late, but evidently not. Ought to use the search more.

So, about what you'd expect, and nothing whatsoever like what's actually happened so far, but they're working on that? Ok.

Unknown

IceIX wrote: Maybe they'll (new players) hit an easy bracket and score a 3* reward and majorly improve their team in one fell swoop.

That is a fallacy though. You do not improve your team at all by pulling a 3 star cover. Even 2 or 3 covers (spread out among all abilities) makes little difference to a person running with leveled up 1 star characters.

What it DOES do is force you into buying another cover slot to keep him on your bench.

Unknown

In my opinion, this point of view its valid only if their goal its that new players jump right in to PVEs.

I don't think this is the right move, the single player campaign should do this. Your progress depends on some characters, and to win some fights you have to level them. It looks like their focus now its PVE and not Story mode, saddly.

It would be awesome to have previous Chapters available to play whitout time limits, it would help new players to level their heroes, get some new ones (some of those events would give them 2* ou 3* chars) and bring back the casual aspect of the game, but... It not depends on us.

Unknown

Just for the reference, the idea of scaling was introduced on the following lines (in context of old-style pve; adding elements explicitly that were just considered obviious):

- initial state same for everyone
- state only depends on your work, nothing else
- more missions repeatable
- repeating mission adds to mob levels of that mission (only)
- that replaces the "reset every X hours" system
- get rid of rubberbanding too
- open issue: point of the repeated mission stays or decays a little

That set is supposed to address all the timing-related issues slanting the field: doesn't matter how and when you play the result is the same. Winners will come from the combination of good team and more effort, nothing else really. You build your own wall until you find it no longer feasible.

kensterr

The scaling still needs to be worked on. The only 3 star in my roster were all below level 30, with just 2-3 covers. My Storm was level 70, IM 50 and BW 40. I don't think I'm considered top tier, nor really mid tier since my Storm can't really tank mobs with levels between 100-150. I think the way the code looks into the roster and deciding what scaling to do is really messed up - there's no way to sugarcoat it.

Unknown

I'd rather they did away with dynamic scaling for the moment and just went to some static scaling instead. Every mission in an event has a set difficulty, from easy level 10 matches for the first mission, to very hard level 200 opponents for the final match. Every time you beat a particular mission, the level for that mission, and that mission only, increases by a set number of levels. After 8 hours, it resets back to the beginning difficulty, and the rewards reset as well.

Doing this would allow newer players to farm the fights they can actually beat. They would gain a load of ISO (comparatively) which would allow them to increase the quality of their beginner roster. It would also allow the more experienced players to get a greater challenge, and to also collect more ISO. You could say this is a case of the rich getting richer, but the simple fact is that for more experienced players, it takes a lot more ISO to level up. Winning 5k ISO might not do much for me except give me 5 levels across a few characters, but for a new player that 5k might allow them to level up several characters by several levels each. Also, the rich will always get richer in a game like this. The people with the best rosters will always be able to place at the top of the leaderboards for any event, if they put the time in. The only way to stop that is to artificially inflate the points the newer players get, and if you do that, you are effectively punishing your long-term players, which is something you should never do.

Unknown

I don't see that it's an objectionable concept, in the least. There were plenty of people commenting - I was one - on the older events that getting people invested in a week-long narrative, no matter how throwaway, then telling them the end of the story isn't for them because they didn't pass the insane dedication threshold yet is nuts. The usual suggestion, at least that I picked up on, was to reserve the high level stuff for side missions.

I do instinctively think Demiurge is digging itself into an extremely deep hole and forgot to secure its escape rope before heading down. I'm sure they're a fine bunch of right clever chaps, but they're a small shop trying to push updates at what must be a fairly exhausting rate. We keep hearing about their metrics, and I'm sure they do work to a large extent, but there's scant outward evidence they're more use than an ashtray on a motorcycle. When a simple option exists, I'd be taking it in their shoes.

Unknown

I think all of this comes from the fact that developers what everyone to see content they made. This is something that struck some of popular MMO's. But only thing that it leads to is lack of balance and plenty of unhappy people. This trend of devs wanting everyone to see everything is something that came recently into world of games. It used to be that you had to work hard, look hard to see everything. Now everyone is so afraid people wont like what they made that they do everything so "new" players can see all of it. I think it is bad idea. It rewards lazy people and punishes those who try hard. I just simply dont like this idea. Game should not be on same level for everyone.

Unknown

First, if you want everyone to see all the content, then you don't restrict access from it by requiring a certain character.

Second, you can do static scaling and allow everyone to see all the content. Have players choose what level they would like to start at. If they choose easy, they will get the baseline rewards and points. Then each mission is moved up to the next level. If they choose medium, each mission they beat nets them the rewards and points of the baseline and current level. Basically, if you know you can jump to medium, you can save a little time and health with the same rewards. I think difficult mode would be obvious. You can still have the levels reset every 8, 12, or 24 hours. I think the distribution would be about the same. If that doesn't work, then you can have people choose on the mission screen if they want to bump this mission to a harder level. The scaling is still static, but it isn't boring at higher levels and it isn't inaccessable at lower levels.

Unknown

nadarath wrote:

I think all of this comes from the fact that developers what everyone to see content they made. This is something that struck some of popular MMO's. But only thing that it leads to is lack of balance and plenty of unhappy people. This trend of devs wanting everyone to see everything is something that came recently into world of games. It used to be that you had to work hard, look hard to see everything. Now everyone is so afraid people wont like what they made that they do everything so "new" players can see all of it. I think it is bad idea. It rewards lazy people and punishes those who try hard. I just simply dont like this idea. Game should not be on same level for everyone.

I find that hard to believe. The Ares Hunt had Invisible Woman as a required character. She's 4*, available only to tournament winners, meaning .5% of the people that participate, or at the utterly impossible 2400 progression rank. Most of the rosters I've seen with her only have 1 or 2 covers, so for almost everyone that has her, she's just a hindrance in a level 230 fight. These throttles and restrictions are clearly meant to ensure most people don't see every fight, so I have a hard time believing they've done a complete 180 on that.