Beta 6 Runemage Feedback

Reborn Beta
#81

If Cam’s calculations are correct, this upsets me immensely as someone who has mained playing a mage since the last beta prior to EA. No talents, in any world, should do more damage than having talents. That’s like going into a regular flatscreen mmorpg and only doing basic/heavy attacks.

Looking at balancing the mage with the current talent structure (single target/mild group dot dmg vs debuff), I do think that only allowing one affliction/fire/frost dot per mage is the way to go. Similarly, we gotta decrease the aoe dps on the mage and I think that the only-spreading-one-dot thing works alright for that(despite how much I will miss it…).

I think that the end goal for balancing the mage should be to make it so that both dot and debuff mages are valuable in a fight rather than figuratively dumping one of the talents into the trash so that everyone picks the other. With a decrease in dps from the dot talent and an increase to the effectiveness of the debuff talent to a point where they are both valuable is where balance should lie.

Similarly, I don’t think that mages and rangers should be balanced against one another in the realm of damage. Mages, in my opinion, should do more damage than rangers, but rangers should be able to sustain a lot more damage than mages (to the point where they can take two or so hits from a boss whereas mages are two/three shot consistently by ads). Mages, in my opinion, should be easy to kill but should put out the most damage between the two original classes.

I haven’t tried either shaman or scoundrel so idk how those balance accordingly but yeah.

However, there is something else that we could look at when it comes to mage… low single target dps, but high group dot (not aoe) damage (previous installment of dot spread talent, but fireball/frost spells do significantly less damage). This could make group dot dmg useful for ads during boss fights or clearing groups in dungeons, but they would be trash against bosses unless they took the debuff talent.

Just thinking out loud.

P.S. I personally would prefer to not be forced into using arcane ray as a dot mage. Additionally, I will admit I’ve been very out of the orbus loop for a while (hence my word choices of “dot mage” or debuff talent etc.) so if I’m sounding like an old geyser that’s resisting change just let me know lol

3 Likes
#82

has anyone done this sort of calculation with the ranger precision talent? (like at what % you need to consistently get to on the bar in order to do increased damage).

#83

I’m a little confused about a few things in your analysis not matching what our expectation would be. What is the exact methodology you are using? You are doing a run and then recording it in the DPS meter and then backing out the numbers somehow? Or is this based on some kind of estimate, and if so can you share your underlying data for how you are calculating the graph?

2 Likes
#84

I’m averaging damage numbers against the practice dummy (no selfish streak or runic diversity), creating a model of a TA and no talent rotation with the spells per second as the dependent variable, and comparing the model’s output to ingame DPS numbers via the combat logs and a DPS tool.

dps = damage over one rotation / time of rotation completion

No talent rotation = affliction, fire, frost, fire, frost (repeat until affliction expires 12 seconds after original)
TA rotation = affliction, frost, fireball, arcane ray (wait 3.5 seconds), repeat (I can only use damage from 3 DOTs or less boosting arcane, until the numbers are fixed)

Both rotations include 1 affliction at any given time.

cast speed (spells per second) as the independent variable
i.e. spells per second = 1:

No talent rotation:

  • one spell per rotation is affliction (every 12 seconds), the rest are fire/frost alternating

dps = ((12 * [spells per second] - 1) * ([frost initial damage] + [fire initial damage]) / 2) / 12 + [affliction damage]
dps = ((12 * X - 1) * (4000 + 6750) / 2) / 12 + 1263
X=1
dps = 6190.1

TA rotation:

  • assuming the fire dot will wear off after the first 2 ticks of arcane ray
  • assuming there is never more than 1 affliction on target at a time, even though there would be a few seconds each rotation with 2 afflictions. In a simulated party setting with 3 mages and one affliction each, the other mage’s afflictions would overwrite your temporary second affliction.

dps = ([frost initial damage] + [fire initial damage] + [affliction damage] * min([time of rotation completion], 13) + frost dot * min([time of rotation completion,8] + [fire dot] * 4 + [ray damage(3 dots)] * 2 + [ray damage(2 dots)]) / time length of rotation

time of rotation completion = (1 / [cast speed]) * [number of spells cast] + [time using arcane ray]

dps = (720 + 1500 + 1263 * min((1/X) * 4 + 3.5, 13) + 316 * min((1/X) * 4 + 3.5, 8) + 948 * 4 + 5053 * 2 + 3158) / (1/X * 4 + 3.5)
X=1
dps = 4149.1

#85

Okay, I see. I think part of the deal is that the ideal rotation for Affliction is probably going to be something like Shadow -> Frost -> Fire -> Ray -> Fire -> Ray -> Frost -> Ray etc. I think your formula is kind of taking that into account, but I think it would be more useful to actually build a model showing the rotation because there are going to be gaps in it.

The other thing that I’m not sure on this is…this is kind of theoretical in the sense that, for example on the non-Affliction DPS rotation, it assumes you are always standing still. You are always (for example) able to cast a Fireball spell exactly every 0.80 seconds on the dot.

My assumption is that in actual boss fights, that is unlikely. The Arcane Ray talent for example has a built-in advantage in that once you are casting the Ray, you could move around the battlefield freely for 3 seconds and as long as you are pointing it at the boss, you are getting maximum uptime on the damage.

This kind of reminds me of when people set up those WoW simulators for the Patchwerk fight, and then the raw numbers show that Rogues are going to destroy everyone, and then people actually do the raid and find out that melee DPS needs a buff because they are spending so much time running away from stuff on the ground. That’s not to say this isn’t something we should consider or discuss, but I do think I kind of want to see what happens in actual fights before we end up over-compensating and creating another situation where Affliction is just too overpowered.

I think the next step is we will probably look at some of the durations of the buffs and the ray itself and look at smoothing some of that over, since as I said looking at a 12 second rotation of this I see some gaps I’d like to eliminate to make it a bit smoother, which I think should help a bit.

#86

The other thing that I’m not sure on this is…this is kind of theoretical in the sense that, for example on the non-Affliction DPS rotation, it assumes you are always standing still. You are always (for example) able to cast a Fireball spell exactly every 0.80 seconds on the dot.

My assumption is that in actual boss fights, that is unlikely. The Arcane Ray talent for example has a built-in advantage in that once you are casting the Ray, you could move around the battlefield freely for 3 seconds and as long as you are pointing it at the boss, you are getting maximum uptime on the damage.

Is that going to be a policy with all bosses going forward? Because it certainly wasn’t the case in the original Orbus.

#87

Is what going to be a policy?

#88

Is it going to be a policy that bosses will require you to move around a lot? Few bosses in original Orbus required a runemage to move around a lot.

#89

I think the boss designs in the new dungeons are fairly reflective of the approach we are going for this time around, which is that all of the members of the party should be ready to move around the battlefield. I’m not saying there won’t be some fights where you can get more time to stand around and just cast without being bothered, but there are going to be Raid fights where that is definitely not as possible.

#90

I think the boss designs in the new dungeons are fairly reflective of the approach we are going for this time around, which is that all of the members of the party should be ready to move around the battlefield. I’m not saying there won’t be some fights where you can get more time to stand around and just cast without being bothered, but there are going to be Raid fights where that is definitely not as possible.

Then that begs the question of how you balance True Affliction in mobile vs static fights. I get that it doesn’t have to be balanced 1:1, but I look forward to seeing how you manage that.

#91

Balance it against what? True Affliction vs Affinity?

#92

Yes. If True Affliction has an extra benefit in mobile fights, how do you balance True Affliction vs Affinity when some fights are mobile and other fights are static?

Maybe you just accept that one is better than the other in some circumstances and allow the difference, so long as the difference is not too big.

#93

This is sorta what I imagine a decent class balance in any game to be. Having different styles within a class perform better or worse in different fights I think is just normal and techniques/strategies would develop accordingly.

#94

I think I said this in another thread already, but yeah the design goal is not “this talent is the best in every situation.” If you are pushing the hardest content at min-maxing your character, you should be switching talents for different fights. I have no problem with saying, “If this is a mostly static single-target fight, you are going to want to run Affinity”, and “If this fight has a lot of movement and many adds, you will want someone in your group running Affliction.”

So part of this discussion is kind of moot because if what this is showing is, if you’re fighting one mob standing still you’re going to want to use Affinity, then yeah I don’t think that’s being disputed by anyone. I don’t like the idea that untalented DPS is doing better than talented DPS which is why we are working on ideas to balance that out while still allowing Affliction to shine in multi-target situations without hopefully ending up in another situation where it’s just too good not to take all the time in every situation.

I also think there’s something to be said for, Affliction may be a better talent tree in single-target fights for slower-casting, newer Mages, because the damage is about 90% as good and you have a lot more downtime and fewer casts to make which means less errors, which you are likely to do as a newer mage. So again, not every tree has to fill every need.

2 Likes
#95

To keep DPS up it seems that you really need to always have 3 dots through most of the arcane ray. The fire dot wears off after just 4 seconds, so you have to use it before every ray. Even with a rotation of just affliction + frost + fire + ray + fire + ray + repeat, the frost dot won’t stay up. Frost lasts 8 seconds and the 2 ray effects alone take up 7, leaving 1 second to cast 2 fireballs and 2 rays. With affliction + frost + fire + ray + frost + fire + ray + repeat, there are 5 seconds left to cast 2 rays, 2 frosts, 2 fires, and the next affliction before the first affliction wears off (though it’s not a big deal if the reapplied affliction goes on slightly after the original expiers). Most mages can’t cast that fast (1.4 spells per second or 1.2 if you don’t care about reapplying the affliction right away).

The model is just based on what seems to be the most efficient rotation across different casting speeds. I can take a look at the numbers with a rotation like that, but it shouldn’t add much DPS. The underlying balance remains pretty much the same and low skill mages will still have very similar damage to high skill mages.

Getting a break while using the ray and getting to move around is a good bonus for the talent, though it doesn’t save much running time when you would otherwise be taking a one click teleport to dodge something.

I still think that the initial damage of frost/fire should be increased from where it is now, and the dots can stay the same. Relying on dots and a fixed time period spell (no matter how fast they are, all mages take 3.5 seconds of waiting to cast ray) for the majority of damage just lowers the skill ceiling to be a high damage mage, because average mages are close enough to the damage of casters who are twice as fast.

#96

Yeah I agree on the first points, we’ll be making some tweaks to make the rotation work better.

I think what I would say is, I expect the high skill ceiling of the this tree to be found in the group damage portion. Like, I assume a high skill ceiling mage that has 4 or 5 targets to work with, between keeping DoTs up, spreading them, using arcane ray on the primary target, etc. is going to a) be able to do a lot of damage, and b) be doing much more damage than a low skill mage.

#99

@Asmund_T and I both observed that perfect casts are dealing less damage than non perfect casts
and arcane missiles is doing 2 ticks not 3.

#100

Did you mean it’s doing less damage on a specific spell? I just tested it Fireball 1 and it seemed to be working correctly, I did more damage when I got the “good cast” indicator.

#101

well, I just tested some more and we misinterpreted what was causing it, the damage of arcane ray and arcane explosion are not affected by how accurate you draw the spells, is this intended?

what was confusing us was that the non perfect cast ones we happened to draw were after we had gotten selfish/shared streak which increased our damage.

1 Like
#102

Oh I see. No it should be getting boosted as well, we’ll get that fixed.