For those who main/semi-main a DPS (e.g. DNC, RDM) but like Healers...

[ForsakenRoe]

1) Every Healer in this game design does.
2) If one counts 1-2-3 as a single action for non-branching combos (since there's no case you don't use them as such and it's a "fatfinger" check, not a skill check), I wonder how many Jobs meet that criteria.
3) Why 10-12, not 3-5? What's sacred about that number?


With a 2.5 sec GCD, and 6 take by Dia and Afflatuses, that's 18 Glares, 4 Afflatuses, 2 Dias, 1.3 Assizes (oGCD), 0.5 PoMs (oGCD). Of course, if healing needs to be done, that cycles in more oGCDs, and eventually GCDs if it is insufficient. This is hardly "Glare-spam" in my book.


No one's ever quite been able to explain why it was good design in SB but not in ShB/EW to have the same rotational frequency of "Stone/Glare-spam". Arguing the DoTs made the timing different isn't an argument against the spam itself, especially when people like Ty routinely use the amount of buttons being pressed, not the pattern, in their arguments, as he did above.

Ironically, I'm the one that critiqued this by noting the difference was WHM had to rely on GCD heals more back then (more damage and less oGCDs), so some of those Stones were substituted for Medica/Cure spells by necessity. It wasn't just the oGCDs, encounter design - the thing people insist is immutable - was apparently mutable and required a different approach to healing, that is, MORE HEALING, at the time. And it's this I point to as a possible solution to the problem.


In a nutshell, though, my proposed changes would amount to:

1) Encounters that require more consistent healing. Low difficulties would still have this, but the values would be so low that novice players could cover it with GCD heal fallbacks. Harder fights would actually make this threatening where the healing has to be addressed in a consistent manner. No more fights with lots of damage and then 45 seconds of nothing at all happening. This change would reduce the ability to address all healing with oGCDs, while still allowing oGCDs to be somewhat powerful for the Jobs that would continue to rely on them.

2) More somewhat randomized damage, like boss random targeting of players or continued auto-attacks of Tanks while prepping mechanics/casting. This was done in HW and was also generally well received for what that's worth. Again, this wouldn't hurt Jonny Casual as casual content would have the damage low enough they could handle it with their fallbacks if necessary.

3) A general reduction in oGCD power for Healers...but with caveats. Specifically, some Healer Jobs would be designed to work with their oGCDs being a source of their healing, and so they would be tuned to that end.

where to start

Counting 123 as 'one skill' is an odd distinction to make but I'll go along with it, by that rule, NIN (barely), DRG (specifically because it's 123 combo is actually a 12345), MNK (if you were to count it's gameplay as 123s but I wouldn't), BLM, PLD depends if you count the Atonements after as part of the 123 or if they're a separate 123. Every healer fails the 'sub-30% check', SMN barely scrapes it with the 'needs 10-12 buttons' criteria. Other than that, every job I'm looking at passes the second two criteria (since they're not healers they can't pass the first). Even DRK passes, even GNB passes (unless you get really picky and say that Gnashing plus it's Continuations are all one combo). This was looking at highest ranked speedkills for this tier

Once again, this is a picture showing the current 1min rotation for WHM. You might not consider 75% of your GCDs being used on Glare as 'spam' but a lot of us sure do. While I'd consider Ty's 30% a little difficult to aim for (I'd accept maybe 40%, since BLM is roughly there with it's Fire4 cast rate), it is possible to reach said 30%. It's surprisingly simple to, designwise too. Though, maybe the fact Quake replaces Glare in my design bumps the Glare count up in Ty's eyes from 28% to like 32% and that means it fails the check too, idk


I don't think anyone's saying encounter design cannot change, I think they (at least I am) say that since encounter design HAS changed and we're now flooded with OGCD heals, to change it to a more GCD reliant system would require nuking those OGCD tools super hard, and people are never a fan of feeling like they've been made weaker. You try and convince the casual players that 'sorry, they had to make your Lilybell heal for 200 per pulse instead of 400, it's for the good of the game though cos now you'll be pressing Medica more often to cover it!', I don't think that's gonna fly well


Ok the meat of the meal:

1: Doesn't address the boredom of 'player who has any degree of competency while healing', as the 'lower values' would be so low as to be basically ignoreable. If it's an amount that novice healers can keep up with via 'GCD healing fallbacks', ie they're panicking, we could throw out Kerachole on CD and keep up on the healing because of it's regen effect. OFC you address this in point 3 though, so lets get there

2: Doesn't address etc etc. Same as point 1, the point of the 'rework' is that it has to address the engagement of all players, in all forms of content. If you keep stuff easier in lower content, then it doesn't solve the issue. We'd go from 'zero GCDs used' to 'Medica2 once per minute/per 30s, then business as usual'. If you're aiming for this random damage to be like... you need to spot heal that person asap, then it sounds like you're asking for those old style WOW raids where your job as a healer is to just whack a mole the HP bars, and people would argue about getting sniped by the Paladin using Flash of Light

3: To follow on from part 1, reducing the OGCD effectiveness is very unlikely to work out. You risk making certain old content (ultimates for example) unbeatable. The amount of 'reduction' you'd have to do would be so big, you'd almost be removing the skill entirely. Removal/heavy reduction in OGCD effectiveness also runs counter to the second point, of having more 'random damage' to react to, as OGCDs wouldn't even get to exist as your panic button. There's been times where I try to save a tank after a big hit, and cannot, because I'm trying to Solace but am GCD locked. I've said in the past, by my estimations, to get to a level of 'GCD healing is the main source' that would have a big enough effect on the game to change how we look at healing, you'd have to reduce the power level of every OGCD we currently have by at least 50%, potentially 75%. HW and SB to a lesser extent worked because we didn't have as many different OGCD heals. In HW, for example, AST had one 'OGCD heal', and it was by channeling the bubble. The buff did not stick to you back then, it had to be channeled. I saw a video by Momo about how he and his cohealer got down to 4 GCD heals in O12S and was impressed by it, because of how much tech they used (including a Super potion on every member after a certain raidwide). Now, 'clear the fight with zero GCDs used' is kinda commonplace, as a result of how many different tools we now have. Also, you say 'so they would be tuned to that end.' re: OGCD-based healers, so you've already set up for failure. Every healer except WHM is now an OGCD-based healer. If you only reduce their OGCD healing power by say... 25%, and WHM by 50%, then we get SB style WHM exclusion again

Incentivize GCD healing, by all means, but not at the cost of OGCD throughput. The idea of 'lets nerf OGCDs so they're forced to GCD heal' is not gonna work. Also, have you considered MP economy at all with any of this? We'll be dry in Savage after like 4 minutes if we're doing damage and AOE healing using GCDs. What's your solution to this problem? I assume 'stack piety' but that's a godawful 'solution' that we can see that everyone hates (look at 6.0 WHM complaints)

[ty_taurus]

Counting 123 as 'one skill' is an odd distinction to make but I'll go along with it...

There's definitely nuance when you start looking at melee DPS rotations especially, because it's not just a 1-2-3. Like Monk's six main rotation actions I would count as all separate because you are constantly shifting between your Raptor, Coeurl, and Opo-opo GCDs. Even the AoE 1-2-3 is a little bit unique because although there's no variation, there is a small amount of variation in using it during Perfect Balance. I count Gunbreaker's 1-2-3 as '1 action' because it's a fixed 1-2-3 that never requires any sort of deviation or decision-making. You can very easily do the PVP thing of having it condensed onto 1 hotbar button and nothing changes functionally about your performance, whereas you couldn't do that with Monk. Ninja, on the other hand, is a little bit different in that it has a 4 that it can swap its 3 for, so I'd look at that as "2 actions" for the purposes of criteria. I'm generally not a huge fan of 1-2-3 filler, but on Ninja I think it's more justifiable because Ninja has a lot going on elsewhere in its gameplay, but I digress.

Once again, this is a picture showing the current 1min rotation for WHM. You might not consider 75% of your GCDs being used on Glare as 'spam' but a lot of us sure do. While I'd consider Ty's 30% a little difficult to aim for (I'd accept maybe 40%, since BLM is roughly there with it's Fire4 cast rate), it is possible to reach said 30%. It's surprisingly simple to, designwise too. Though, maybe the fact Quake replaces Glare in my design bumps the Glare count up in Ty's eyes from 28% to like 32% and that means it fails the check too, idk

I think you were looking at the source of Black Mage's damage, not it's action usage. I'm not the most knowledgeable about Black Mage stuff, though so maybe this isn't a typical example of Black Mage gameplay, but this was the reference I was working with:



But something I said in my first post about those requirements is that those are soft requirements. I don't mean to imply that if you were to create a Sage redesign where Dosis III was cast 32% of the time even when performed at the most optimal level, that I'd treat that as a completely failed attempt. The actual feel of the gameplay matters too of course, but I gave those criteria to establish a ballpark objective to aim for. If something doesn't quite reach that goal, but still offers a lot of variation in the gameplay and feels good, than that's fine.

[ForsakenRoe]

stuff

I think I might have been looking at the damage page rather than casts, yes. Would explain some things, I guess. As for combo things, I was looking at each combo ender's percentage of uses, and then adding on the previous steps of the combo (for example on NIN, I'd look at Spin-Gust-Aeolian as one 'action' and then Spin-Gust-ArmorCrush as a second 'action') and see if either hit 30% of total casts. However, as pointed out, I was likely looking at 'damage source' for all of them, meaning a lot less of them which 'failed' would actually be fail-grades, because they have a load of OGCD uses that were not included. Though I figure the 'GCD loop being stale' would mean that excluding OGCDs would be of slight help to paint the picture. For example, a hell of a lot of AST's gameplay is in the 2min window, playing cards, doubleweaving etc. But that doesn't do anything to help the fact that outside the 2min window, it's Malefics and Combust refreshing all the way down. I guess having both charts ('damage sources' and 'total actions used') would help to illustrate the nuances of this kind of thing

Hmm might go make a pie chart of the SCH DOT design I had, see what kind of Broilspam it'd have, I estimate about 55% which is unfortunate but better than what we have now (75%ish). If we are including 'every action used' in that 30% number, then that makes it super easy to reach. No need to worry about 'Quake/Glare being one button and pushing over 30%' when Assize, POM, Benison/Aquaveil, potentially a 60s Cleric Stance CD, etc are involved

edit: My SCH idea ended up at 47% Broils over a 4min duration (lowest common denominator on the DOTs), when looking at just target dummy scenario, so Broil, 3 DOTs (30,24,15 duration), Aetherflow CD and spending all 12 charges on Energy Drain. Dissipation was not included

[ty_taurus]

I think I might have been looking at the damage page rather than casts, yes. Would explain some things, I guess. As for combo things, I was looking at each combo ender's percentage of uses, and then adding on the previous steps of the combo (for example on NIN, I'd look at Spin-Gust-Aeolian as one 'action' and then Spin-Gust-ArmorCrush as a second 'action') and see if either hit 30% of total casts. However, as pointed out, I was likely looking at 'damage source' for all of them, meaning a lot less of them which 'failed' would actually be fail-grades, because they have a load of OGCD uses that were not included. Though I figure the 'GCD loop being stale' would mean that excluding OGCDs would be of slight help to paint the picture. For example, a hell of a lot of AST's gameplay is in the 2min window, playing cards, doubleweaving etc. But that doesn't do anything to help the fact that outside the 2min window, it's Malefics and Combust refreshing all the way down. I guess having both charts ('damage sources' and 'total actions used') would help to illustrate the nuances of this kind of thing

Hmm might go make a pie chart of the SCH DOT design I had, see what kind of Broilspam it'd have, I estimate about 55% which is unfortunate but better than what we have now (75%ish). If we are including 'every action used' in that 30% number, then that makes it super easy to reach. No need to worry about 'Quake/Glare being one button and pushing over 30%' when Assize, POM, Benison/Aquaveil, potentially a 60s Cleric Stance CD, etc are involved

edit: My SCH idea ended up at 47% Broils over a 4min duration (lowest common denominator on the DOTs), when looking at just target dummy scenario, so Broil, 3 DOTs (30,24,15 duration), Aetherflow CD and spending all 12 charges on Energy Drain. Dissipation was not included

I think it's fine to include OGCDs in that balance, which is what we see when we look at the Black Mage example I shared, as well as others. While Bard's current state is generally revered as mid in contrast to what it was during Stormblood, in both cases a lot of Bard's GCDs are dedicated to Heavy Shot/Burst Shot because it does a lot of work weaving, and I think that's okay. It does ultimately come down to how it feels when you put it all together, but I just think using those metrics... no more than 30% time spent on any singular action, spending around 80% of your gameplay on at least 10-12 actions... Creates a healthy ballpark to shoot for when designing a job--that you'll have a greater chance for success with an idea when you try and aim for that range.

And I did also look at Astro because I was curious about how its card gameplay factored into its action usage. Where Scholar has a 58% usage of Broil, an Astro in that same fight had a 53% usage of Malefic, so not a particularly significant difference, and still Malefic is used more than every other action combined. And that's P12S data.

[Shurrikhan]

I think it's fine to include OGCDs in that balance, which is what we see when we look at the Black Mage example I shared, as well as others. While Bard's current state is generally revered as mid in contrast to what it was during Stormblood, in both cases a lot of Bard's GCDs are dedicated to Heavy Shot/Burst Shot because it does a lot of work weaving, and I think that's okay.

While I agree that oGCD CPM should be accounted for, Bard is atrocious even then.

Since there is zero reason ever to use Burst over Refulgent when the latter is possible (just as there is zero reason ever to hit any other non-prepped combo action), they together really just form a single filler action: Refulgent-over-Burst. That action then makes up almost 47% of Bard's CPM. That's near to Fire IV's portion even despite the hugely higher amount of oGCD casts on Bard.

And without those oGCD actions, it's obviously far worse: R>B would comprise ~86% of GCDs, or approaching 90% in sustained combat, because it solely competes with AA->BA and IJ(over SB->CS), whereas F4 makes up only ~38.5% of BLM's GCDs.

In short, I agree with the test, but if combo steps (rightly) aren't each considered a separate action, Bard would fail that exact test.