Thread: The potential problem with low level Roulettes in 6.0

Originally Posted by PyurBlue

Skill durations aren't changing, the longer duration comes from the PLD having more defense skills than GLA. PLD with halved or thirded effectiveness Rampart, Sentinel, Sheltron vs GLA with normal Rampart for example.

Your clarification here makes what you said previously make more sense in terms of the duration - the way I read it before it sounded like you were meaning the duration of all skills when combined, but rather on a per skill basis.

That being said - I still don't think it would feel good to play, assuming I'm understanding your suggestion here, that when playing on Paladin and being below a certain level, the damage reduction amount would be reduced to account for having more skills to use versus a Gladiator?

So Paladin would have Rampart, Sentinel, etc., which would last the same length, but do less damage reduction (like 10%, 15% below a certain level? And Gladiator's would do the normal intended amount, also at the same duration?

So assuming that is the correct interpretation...you still have the issue that you're asking the Paladin to play differently to survive the same packs...as it's ultimately the duration of the cooldown is only a portion of the consideration of when to use it, and honestly it's not at all the main consideration. Rampart is ultimately used not because it has a 20 second duration, but because it has a 20% damage reduction - that's the case in a trash pull and versus a tank buster. Most trash pulls at lower levels aren't going to benefit from being able to use more cooldowns in succession.

Instead it would almost certainly result in the Paladin feeling like they need to stack those cooldowns to achieve the same or similar damage reduction they had when they were lower level or they would feel squishier as a tank than the tank that comes in as a Gladiator with just a couple that do their intended full amount...which would still result in a different in how the job is played at higher levels, when the defensive cooldowns do their intended amount. You still end up introducing variations in how the job plays that make going into lower level content as a higher level player feel less good.

Your assumption that, in a same size pull, a Gladiator with a 20% Rampart and a numbers-synced Paladin with a 10% Rampart are both only go to press Rampart is faulty. The Paladin is, almost certainly, going to press an additional skill alongside that Rampart so they feel like the pull is like it was a Gladiator with a 20% Rampart.

That doesn't sound like it would be much fun to deal with, even if (again) the numbers look good on paper. I'd be having to press more buttons for the same effect, just because I'm a higher level going into a lower level dungeon.

In what gaming world does that sound fun?

Originally Posted by MariaArvana

The thing is, time is their mortal enemy. So regardless of the feasibility of things, they have quotas to meet and they will do that which they perceive will help the health of the game more and weigh the development costs & time spent vs projected gains. They're developing the game all the time, just not in the way you want.

This isn't new information to me. The thing is you're taking general facts and trying to apply them to a specific situation while lacking some critical information. Unless you know how long it would take to redesign level sync, what resources SE have available, etc, you can't rule it out as a possibility. You can point to difficulties that may arise and you have pointed out valid ones but you're in the same boat as me when it comes to knowing what is actually possible.

I'm putting forth an idea, not with the expectation that it is implemented into the game just because I want it but with the hope that the devs see it and consider it based on what they are willing to do.

I'm also not arguing against more moderate changes like redistributing skills for each class, but I don't have much reason to mention that idea when it's already popular in this thread.

Of course it's not. But again, they've made it clear time and again that they desire balance in their level sync system, and so they're not going to do anything until they have a system that works exactly the way they want while making sure there's no issues, weighing all the cons & pros, all the things that would need balancing, create expected deadlines & time spent vs projected gain numbers. If I, a simple player that has only taken a couple courses on rudimentary software design because she was bored outside her normal courses in university, can point out all the glaring flaws and the sheer amount of things they'd need to modify/change/edit/add/remove/etc in your system just to make it work compared to their current extremely simple system, their actual experienced game designers will have long since either tossed it out, or be looking elsewhere for a simpler idea.

I don't recall anything indicating that SE is perfectly happy with the level sync system. I do know that they've reworked skill progression, combined tiered skills in the past due to issues of button bloat, reworked cross class abilities, and reworked role abilities. To me it seems like skill design isn't totally set in stone.

I'd also point out that I've been responding to your criticism (and I don't mean for that to be taken negatively, I appreciate that you would critique my idea) this whole time pretty rationally. You are absolutely pointing out common general issues in software design but I'm not willing to use those as rigid constraints for a specific project that I cannot see the inner workings of.

Except that simply divides the queues; which need I remind you, are designed to fill parties as fast as possible; another reason why it'd never see DF use without being mandatory. And I seriously doubt they'd ever do such a giant overhaul in the first place just to make it a PF only option. As far as the first bit, again; I'm simply pointing out every possibility, which by their job, a game design has to account for. There's a reason the #1 rule in software design is "plan assuming your end user will do anything and everything to break you system, whether through ingenuity or ignorance." Without taking in EVERY factor, EVERY possible outcome, any system they implement will just as easily crash down on them.

A few things were conflated here. It would be a PF option initially for testing purposes, similar to how Eureka was used to test world visits. If they wanted to add an incentive to make it look like its own game mode (like Eureka) they could do that too. "Unreal dungeons" or "NG+ dungeons" with extra tome rewards or something. After the players have had a few months or however long doing that and trying to break the system, SE fixes issues and then adds the option to DF for regular use.

There is no need to split queues either as the option could be player specific. That I will admit is an assumption because I don't know how the level sync function works from a technical standpoint. We know the that the game has coding issues, but that doesn't preclude significant changes (like world visit).

[/quote]At least in their current system, High level player's skill is kept in check by the limited toolkits.[/quote]
In my system player skill is kept in check by math. We know what players can do at ilvl 505. It's about 20,000 DPS for the DPS classes. Impose a damage down on the best player that you can absolutely find that multiplies their damage by .00025 and that player will deal less damage than the average level synced Sastasha tank. There is no getting around that.

Which again, comes back to the MNK scenario; Given that jobs learn different parts of their toolkit at drastically different levels, you're essentially asking for them to go through every job at every level to determine what part of their toolkit they have access to, and then hit them with % reduction based on that. Since -20% on a BRD, who barely gets anything pre-50 will be vastly different than applying that same -20% to a SAM who gets a lot of powerful potency attacks at the same level.

If your solution can't be solved with a universal multiplier for every job at every level to create the same balance as currently, it's already less precise and far more complicated than what they have now.

However class skills were designed with the game's dungeon progression in mind. Classes while leveling are never perfectly balanced, your example even shows that, as you're saying that a level synced SAM is stronger than a level synced BRD. So when we take a level 80 SAM or level 80 BRD and apply the same multiplier to them, they don't necessarily line up exactly with their level synced counterparts. Where they do land however in on the average DPS for that level sync. You might end up with a ranking like this:

50 SAM > 80 SAM > 80 BRD > 50 BRD and all of this within a few percent of average level 50 DPS

There is no issue. No one is kicking BRD's and hoping that a SAM queues to fill their party. It would waste their time and they would also have to be assuming that everyone who joins the party would be playing perfectly.

Go pull every pack up to the first boss of Tam-Tara with 5 completely random healers without dying once, take vids, compare that to pulling 5 packs in Mt Gulg and then tell me it's "no more than a GLD could do." Or every single mob up to the first boss of Halatali with the same requirements.
Also people kick for even less reasons than that. I've seen tanks get kicked for single pulling, you really think people aren't above the possibility of kicking for a better tank? Plus, unless you're queuing in off hours, tanks/healers almost always get insta-refreshed anyway. (I've personally never waited longer than 20s for a tank to come in during the couple of times a tank was kicked in my party.)

I can consider recording videos although I'm not exactly jumping to do it, and I don't think the test you're proposing creates an equivalent situation. Ideally you would compare GLA pulling the entirety of Tam Tara with and without HG available. I suppose this can sort of be tested if you find a low level GLA, go in unsynced with two healers, one at level and one level 80, shield the GLA with the level 80 healer to simulate HG, and then have the GLA click off the shield after 10 seconds. Those 10 seconds are only about 4 GCD's, which is far shorter than the duration of the pull. HG would be helpful aboslutely but it's not going to trivialize the pull by itself.

Occasionally people will kick for anything, but that's not the norm. You mentioned that it's only happened to you a couple of times. While their kick may be rewarded with a tank that has more skills, that doesn't necessarily mean that they will have a better tank. The system I'm asking for reduces mitigation to make up for higher levels having more CD's. HG could even lose its invulnerability.

^

Also, You chain CDs. Once the tank leaves Hallowed, he should have a 30% rolling on up so that he stays buff and not just instantly drop dead or becoming weaker. 10s of 0 damage is a huge advantage gained, because it gives the healer 100% uptime during the 10s; 10s of knowing that whatever the healer does, you will not drop dead.

But half of skills aren't being nullified nor are 50 traits being added. We're primarily talking about 1 skill, HG, that might need to be changed and a blanket mitigation reduction in everything else. Once the PLD leaves HG there wouldn't be a 30% mitigation option available because it would have been scaled by the blanket reduction. The PLD would have to choose between trying to match a GLA by using multiple CD's at once or having less mitigation spread out over a longer time by using CD's in succession.

The average mob does about ~28 damage in Sastasha out of a tank's ~450 health. In our current system (or about 84 damage per pack of 3), an AST can get that much uptime due to their OGCD alone and because most tanks don't pull much more than 2 packs in most cases. An AST isn't keeping anywhere close to 100% uptime in sastasha when the tank is losing 1/3rd-1/2 their health every GCD going past the 2-pack mark with their OGCD on a 30s timer.

Right I did simplify things, but look at the line I was replying to. You were saying that 10 seconds of HG would help the healer melt the pack compared to without. It's true, it does help, but it's not an on/off switch for healer damage. Packs that die quickly allow AST to maintain 100% uptime already. Those same quick pulls will be required to get the max healer DPS gain from HG because it's only 10 seconds. After those 10 seconds, we have the situation that you point out where a 450 HP tank is taking 28*n damage per mob GCD. You've offset the entire mitigation process for the party by 10 seconds. The ratio of 10 seconds to the duration of the pull could be looked at as a really rough measure of the impact of HG on a given pull. Then you can do the same and compare to the length of the entire dungeon for total mitigation. HG has an effect but also has limitations. From my own experience a lot of tanks "save" it indefinitely. That's certainly a waste but it doesn't really make dungeons what I would consider noticeably harder. And there is still the solution of just reducing the mitigation to less than 100%.

Now that I'm thinking about it, all the other invulnerability are bigger headaches. Reducing 1 HP minimum to .5 HP doesn't do anything and might not even be possible if it did. These would either have to be left alone, which could be weird for low level healers, or turned into something completely different by traits. That shouldn't be difficult but it would alter the feel of the skills.

Which again, requires way more investment than you realize, as from a pure math standpoint, is extreme more effort than other solutions which require immensely less effort.

The pure math standpoint is what makes it look easy because we're not dealing with tight balance tolerances. The only difficulty is the coding.

If you want a level 80 class to feel completely indistinguishable from a level X class at every level, then the math is hard, but there is no point in doing that. That would preclude different classes from existing at all, preclude role skills, and preclude dungeons allowing different levels to enter at all. We have more room for error than you're implying.

But a 20 GLD can't use either, so they do count against it, you can't simply ignore that mitigation difference when it is in fact a + to the PLD's toolkit. Clemency may be a DPS loss, but if it saves the pull, its a dps gain, hence why its a tool for the PLD. (Such as saving yourself on the Mt Gulg mega pulls if the healer falls behind a bit - staying alive and losing a GCD of damage is infinitely less valuable than staying alive to complete the mega pull which saves literal minutes.)

I'm discounting the role skills because the current system just ignores that they are an imbalance, because the tolerances for balance aren't tight at all. The game as it is now doesn't care that the tank may be missing 10% mitigation every 60 seconds and 20% mitigation every 120(?) seconds. As far as the level sync system is concerned having Rampart, Reprisal, and Arm's Length is the same as having just Rampart. That's obviously wrong. But we're not dealing with tight tolerance so it's fine. Fair point on clemency.

As for the last point, then, again: what is the point of halving/quartering/whatever these skills down and butchering them, when the whole purpose of your system is to allow 80's to use their full toolkits? If you have to edit and butcher skills just to give you system even the slimmest chance of maintaining balance, you're no longer playing the exact job you had at 80, merely a gutted version of it. I really can't explain this any clearer.

How are skills being butchered? What is the PLD going to do differently if Rampart is 10% instead of 20%? You just pull and press the button. The point is to have more to do like you do at level 80. At level 80 the numbers are bigger, but that's just a distraction. It has no relevance, just like the exact values of Rampart, etc, have no relevance.

The major difference here being that a 75 & 76 WAR have access to 99% of the same toolkit as each other, so the disparity is far, FAR less than a 20 MRD who not only has 6 less CDs than a 76 WAR, but also 0 lifesteal. And I could repeat myself on the 2nd part, but really. Even if you were to quarter the healing of it, it's still > 0, meaning the 80 WAR would still have literally infinitely times higher HPS than a 20 MRD...or any <76 WAR in AOE, for that matter. Also again, having to butcher a job's identity just to give your system an even slim hope of being balanced.

Does that infinite % increase in lifesteal change that the tank needs healing? No. Over the length of the dungeon the mobs will do X damage. The MRD will reduce this by Y with mitigation, and the WAR would reduce this by Y+/- with a combination of less mitigation and some lifesteal. You're still trying to make level 80 WAR identical to level 20 MRD, but there is no reason to.

Because they never had to factor them in early dungeons when designing them? Literally the earliest non-trick attack Raid buffs are at 50 (Divination, Battle Voice). All of a sudden in your system though, you've got 80's hopping in with all their raid buffs. let me tell ya how fast you can annihilate packs with BV + Litany + Divination. In your system, they are now an issue that has to be factored in.

But you're ignoring that the current system has essentially the same issue. AoE is a huge multiplier on big pulls. Fire II is a horribly weak skill. Pull 8 mobs and it matters a quite a bit, at least when one of the alternatives is having your DRG reduced to Vorpal Thrust.

It's not that raid buffs don't make a difference, it's that they are not significant enough to change the dungeon experience in a game that doesn't require tight balancing.

At this point it looks like that's all there is to debate, how tightly the game is actually balanced. Your points are all valid when it comes to making classes totally equivalent across the entire level range, but if that was a requirement we'd need to throw the game as it is now out of the window completely. Comparing WAR's to MRD's makes comparisons a little easier to see but that specific match up is no more important than WAR vs GLA or PLD vs GNB. Balancing to one specific class at one specific level doesn't make much sense. What's important is the balance of a given role as a whole.

For no reason? Because its literally the design the devs have been wanting to achieve with their sync system. you can't get much tighter than "everyone has access to the same skills outside of some scenarios in the early levels, which only really apply to tanks because healer & dps role actions are just fluff and not giving extra power to the healer's healing or dps's damage." With their system, a Dev can reasonably know that a group in halatali isn't going to be super mitigating enemy damage, throwing out raid buffs or massive swathes of multiple AOE GCDs & OGCDs to melt packs, or commit 100% of their healing to regens & OGCDs. In their current system, player skill variance is the only thing they can't account for; an issue that is magnified and needs to be much more directly addressed in your theoretical system due to the plethora of unexpected variance it can cause.

The role skills aren't all fluff. Reprisal and Arm's Length are full fledged tank skills. Lucid and Swiftcast are kind of situational but when they matter, like when it comes to raise, they are significant healer buffs. I guess for DPS it's less impressive. Bloodbath is strong but it's level 12 so everyone has it.

Outside of role skills, skills still aren't equal at a given level because skills aren't only tied to level. Some are locked behind job quests and it's entirely possible for someone to queue into duties while missing skills. This doesn't drastically upset the balance of the game unless it's done continuously. Then there is the balance between classes which is even less tight. AoE isn't uniform and OGCD's aren't uniform. The game can tolerate some variation. The system I'm proposing falls within those tolerances.

Let me answer that for you. "Server Limitations." "next!" Yoshi-P has long since realized making promises in public often ends in backlash, so pretty much any feature they're not personally doing will be met with a PR-friendly/neutral response; hence why it's easy to use past development cycles and their general response to suggested things to quickly answer your own question.

That's not an answer. You could say the same to any proposed change and shut down the forum.

You can have your opinions, but observations of their dev cycle and implementations give pretty good concrete facts about how likely a potential idea would be accepted. Just from my courses in software design and general observation, I'm giving an observed opinion based on those realistic facts. They have a very 'forward looking; only fix the old when necessary.' Approach to their game design.

It took them literally 6 years to fix ARR and they still have Hrothgar/Viera to deal with, along with designing all the usual content each patch, especially now with an impacted dev cycle due to COVID. The odds of them doing anything to the level syncing that isn't as simple as 'shift a single variable on the skills' is basically null.

Then you're making your best opinionated guess and you don't really know. That's fine, but don't take more from a guess than you reasonably can. You can say that something is unlikely and I won't necessarily disagree, but you have no way of knowing whether or not you're correct. COVID isn't a factor in this. What is the time frame for the sync system to be updated? There isn't one, the disease could be eradicated before work on this begins.

Originally Posted by Berethos

Your clarification here makes what you said previously make more sense in terms of the duration - the way I read it before it sounded like you were meaning the duration of all skills when combined, but rather on a per skill basis.

That being said - I still don't think it would feel good to play, assuming I'm understanding your suggestion here, that when playing on Paladin and being below a certain level, the damage reduction amount would be reduced to account for having more skills to use versus a Gladiator?

So Paladin would have Rampart, Sentinel, etc., which would last the same length, but do less damage reduction (like 10%, 15% below a certain level? And Gladiator's would do the normal intended amount, also at the same duration?

So assuming that is the correct interpretation...you still have the issue that you're asking the Paladin to play differently to survive the same packs...as it's ultimately the duration of the cooldown is only a portion of the consideration of when to use it, and honestly it's not at all the main consideration. Rampart is ultimately used not because it has a 20 second duration, but because it has a 20% damage reduction - that's the case in a trash pull and versus a tank buster. Most trash pulls at lower levels aren't going to benefit from being able to use more cooldowns in succession.

Instead it would almost certainly result in the Paladin feeling like they need to stack those cooldowns to achieve the same or similar damage reduction they had when they were lower level or they would feel squishier as a tank than the tank that comes in as a Gladiator with just a couple that do their intended full amount...which would still result in a different in how the job is played at higher levels, when the defensive cooldowns do their intended amount. You still end up introducing variations in how the job plays that make going into lower level content as a higher level player feel less good.

Your assumption that, in a same size pull, a Gladiator with a 20% Rampart and a numbers-synced Paladin with a 10% Rampart are both only go to press Rampart is faulty. The Paladin is, almost certainly, going to press an additional skill alongside that Rampart so they feel like the pull is like it was a Gladiator with a 20% Rampart.

That doesn't sound like it would be much fun to deal with, even if (again) the numbers look good on paper.

It looks like this post slipped in while I was typing. It's probably a sign that these posts are getting a little big.

You're right that there are a few considerations for selecting CD's, but ultimately within one set of skills you end doing the same mitigation rotation if everything is scaled equally. Rampart first unless you need more mitigation to survive since it has the shortest CD. This let you fill in that short CD with other mitigation and maximize uptime. The biggest difference will be felt at the lowest of levels where damage isn't high and PLD shouldn't really feel all the squishy regardless. Combining mitigation to reach GLA's level would be a slight difference but I don't consider it drastic and it's not like it doesn't happen at higher levels.

Ideally you want to rotate cooldowns but there are differences between the ideal situation and reality. The clearest one is probably Arm's since it usually doesn't work on bosses. If it's up during a pull immediately before the boss I overlap it with other skills. Reprisal can also work like this because it's only down for 60 seconds. You would definitely overlap Sheltron with other skills unless you were having trouble building gauge.

There's no way you could scale down abilities to make them balanced for low level play. For example, DRG/LNC has no oGCD damage skills before level 30 (since they changed Leg Sweep in 4.0 to deal no dmg and also cross role). But having an 80 DRG sync'd down to level 16 gives them Geirskogul, High Jump, Spineshatter Dive, Nastrond, Stardiver, Mirage Dive, Dragonfire Dive. Even if these skills did like 30 potency, it's still much more than a level 16 Lancer's True Thrust->Vorpal Thrust. Melee don't have AoE before the mid 30s so you'd have melee using their AoE combo in low level dungeons but new players not having any AoE. There's honestly no way to do balance the potency to make it fair.

Originally Posted by HyoMinPark

If they weren’t going to allow players to keep their full toolkits when synced down, I would think that redistribution of how skills are acquired would be the best option. It’s something I’ve actually suggested with a job like DNC to combat proc RNG at levels below 72 (pre-Flourish), and I think some jobs could benefit from getting skills earlier.

Content redesign will likely always be completely out of the question. Too much work, and the developers have rarely expressed interest in going back to old content and adjusting the scaling of it.

Pretty much this.

RDM/SAM/DNC did so much damage just because of how well their low level kits were designed.

I would advocate for letting us keep our skills.

1) Tanking is brain dead easy now, throw on tank stance and 123 single target or 12 for aoe

2) DPS damage will ALWAYS be determined by weapon damage which, news flash, gets lowered in synced content.

3) It's synced content. It doesn't really matter if it's killed quickly.