Going Forward: Cataclysm Discussion - Page 3

Vyshe · 04/08/10, 5:47 PM

Originally Posted by Arawethion

There's nothing wrong with that--Healing Touch, Tranquility, and to a large extent Lifebloom were completely dormant for an entire expansion. They don't have to make "new" spells to greatly expand the functionality we have.

I agree - Nourish scaling with HoTs could probably be extended to HT and with some adjustments to cast times/costs we have our synergetic big heal. Changing Tranquility to a shorter CD and making it raid wide or "smart" targetted feels possible too, with shamans getting a hurricane-type heal.

But I almost expect to see a "big life saving cooldown" for Resto since they are moving out of niches a bit. Either way, I'm really excited to see what they've come up with.

(If they do give us a new heal I hope for something really crazy, like letting it refresh all hots on a target.)

Lamente · 04/08/10, 5:55 PM

I'm still not very clear on how they'll handle "haste decreases the time between tics, but doesn't lower the full duration" bit. Using the most basic interpretation, Rejuvenation (for example):

At 100% cast time (no haste), you have 5 tics in 15 seconds.
At 83.3% (repeating) cast time, you get 6 tics in 15 seconds.
Between 83.4% and 100%, haste functions like the current rapid rejuv glyph: faster tics, shorter duration.

With the above model, you end up with haste "benchmarks", where a min/maxing raider would shoot for specific exact numbers of haste, and avoid more haste until they can make a significant jump all at once, to gain the next tic.

So what if they try to make haste linear? I can think of two solutions for our Rejuvenation example:

at the 15 second mark, a "partial tic" happens, which is worth whatever percentage of a normal rejuv tic is left over due to haste. Example: your tics are 2.6 seconds apart, leaving 2 seconds between your second-to-last tic and the 15 second mark. At 15 seconds, a tic worth about 77% of normal happens.

But what happens when the HoT is reapplied early, assuming the tic cycle is unaffected due to the stated "no clipping" goal?
- Having a normal tic at that original 15 second mark means haste isn't scaling correctly when the HoT is reapplied.
- Adding the "partial tic" combined with a normal tic at 15 seconds seems like a spiky and inelegant solution that would lead our healer a bit back to our "haste benchmark" scenario for consistency.
Any "leftovers" are distributed evenly between all tics. Example: your tics are 2.6 seconds apart, leaving 2 seconds between your second-to-last tic and the 15 second mark. 77% of a normal tic is distributed between all 5 tics, so that each tic is over 15% more powerful.

This creates several more problems:
- From 100% to 83.4% (for example) you're looking at a shorter duration but harder hitting tics. That's slightly better than the current rapid rejuv glyph, but still a situational benefit.
- Reapplying with "no clipping" creates further problems: either the next tic doesn't happen until the original cast's 15 second mark, creating a time gap with no tics, or the tics always happen normally and haste scales unreasonably well with reapplications as haste nears (but doesn't reach) each "benchmark".
Either way, you'd ideally be gearing around those "benchmark" numbers.

I'm just thinking out loud here. Apologies if my math is screwy, but I hope you get the gist of what I was driving at. If someone can poke holes in my logic or propose a better solution I'd love to hear it.

Vyshe · 04/08/10, 6:11 PM

I'm just assuming they will do choose the simplest solution for both HoTs and DoTs: No partial end ticks, let us deal with the haste "tresholds" for getting an extra tick for non-refreshed casts, since the main focus for the change seems to be on HoTs/DoTs being refreshed continuously during an encounter and for those cases haste scaling will be linear.

kalbear · 04/08/10, 6:32 PM

I think they're going to have to do partial ticks for it to scale properly and meet their goal of being a simple mechanic that everyone understands. If they don't do this, refreshes are going to be the order of the day. The reason I say that they're likely to do partial ticks is that there exist enough external haste buffs or temporary buffs that you don't want to have to figure out what's best depending on whether or not you've got bloodlust, a haste pot, a haste proc or some combination of the above.

And partial ticks aren't that hard to compute. Simply always have it tick at the end and compute the remaining time. This shouldn't be that hard to code, and if it's for all dots/hots it shouldn't require too many special cases.

Murna · 04/08/10, 8:52 PM

They also said, that we can't clip dots (and probably hots) anymore.
I don't think Blizzard wants "haste breakpoints". They'll probably find another solution.

Imagine a 15 second HoT/DoT that ticks every 3 seconds.
When you refresh it after 13.5 seconds, it will look like this:

0:12: tick
0:13.5: refresh -> new duration: 15 seconds (until 28.5)
0:15: tick
...
0:27: tick
0:28: refresh -> new duration: 15 seconds (until 0:43) (or let it expire, then: )
0:28.5: partial tick (if not refreshed)
0:30: tick

and so on.
That would work with every tick rhythm so you wouldnt have any haste breakpoints at all.

And by the way please don't cry about tanks doing DPS like the Dmg specs - these numbers will be tweaked countless times till Cataclysm. It's just the general mechanics, we should talk about - all of the numbers known now are irrelevant.

Royalite · 04/08/10, 9:06 PM

Originally Posted by Fallenangel

Might wanna read the fine print:

"Whenever a tank gets hit, Vengeance will give them a stacking attack power buff..."

You can scrap that 20 tanks raid now.

Yea I was considering it...although not for all 20 but that it could be abused based on some sort of taunt rotation to maintain stacks. Or however you will... It is all up to speculation but the traditional role for a tank is high defensive in exchange for low offensive. The price you pay for higher offensive is lower defensive capabilities. That whole choice and tradeoff aspect that Blizz enjoys to implement in the game.

The example used was considering the pve abuses but considering pvp if a prot spec keeps its high defenses while keeping large offensive damage, you're going to see pvp dominated by tanks of any sort in cata.

The numbers may be higher for tanks then we're used to seeing but I surmise it is b/c dps will be seeing large increases as well. I do see the numbers they gave to show how they are closing the damage gap but I think it is still a big jump to say the gap will be filled completely.

Although Blizz has been accused of homogenizing the classes/specs, I doubt they would veer from their traditional role definitions. I just don't see tanks becoming as good as even a hybird dps.

Lamente · 04/08/10, 9:17 PM

Originally Posted by Murna

They also said, that we can't clip dots (and probably hots) anymore.
I don't think Blizzard wants "haste breakpoints". They'll probably find another solution.

Imagine a 15 second HoT/DoT that ticks every 3 seconds.
When you refresh it after 13.5 seconds, it will look like this:

0:12: tick
0:13.5: refresh -> new duration: 15 seconds (until 28.5)
0:15: tick
...
0:27: tick
0:28: refresh -> new duration: 15 seconds (until 0:43) (or let it expire, then: )
0:28.5: partial tick (if not refreshed)
0:30: tick

and so on.
That would work with every tick rhythm so you wouldnt have any haste breakpoints at all.

I think I get what you're saying. A free partial tick at every 15 second interval, essentially. For some reason my mind wanted to combine this with either the previous or next tick instead. This still makes the end result less useful since the timing of the extra tick is completely random - doubly less useful since that extra tick has a lower value than all other ticks. Even if they averaged out the tick amount between all ticks including the extra one, you'd still have ticks going off at irregular intervals every 15 seconds, which increases the RNG (and therefore usefulness) of the extra tick. The only time this wouldn't be the case is when "breakpoints" are hit, so the concern that breakpoints are ideal is still there.

It's a better solution than I came up with, but I still wouldn't call it ideal.

Vyshe · 04/09/10, 6:47 AM

Originally Posted by Lamente

I think I get what you're saying. A free partial tick at every 15 second interval, essentially. For some reason my mind wanted to combine this with either the previous or next tick instead. This still makes the end result less useful since the timing of the extra tick is completely random - doubly less useful since that extra tick has a lower value than all other ticks. Even if they averaged out the tick amount between all ticks including the extra one, you'd still have ticks going off at irregular intervals every 15 seconds, which increases the RNG (and therefore usefulness) of the extra tick. The only time this wouldn't be the case is when "breakpoints" are hit, so the concern that breakpoints are ideal is still there.

It's a better solution than I came up with, but I still wouldn't call it ideal.

There's no need for partial ticks as long as you keep refreshing, since that won't reset the tick timer. Hots and dots will simply be like buffs/debuffs that while they are active tick every x seconds, modified by haste.

Also, calculating a partial end tick isn't hard, sure, but I'm still not sure they will bother with it. Some classes have lived with "haste breakpoints" during WotLk, be it cramming casts in between skill cooldowns or getting full benefit of trinket proc intervals. Hopefully I'll be proven wrong though.

Deliverance · 04/09/10, 10:05 AM

Originally Posted by Royalite

Yea I was considering it...although not for all 20 but that it could be abused based on some sort of taunt rotation to maintain stacks. Or however you will... It is all up to speculation but the traditional role for a tank is high defensive in exchange for low offensive. The price you pay for higher offensive is lower defensive capabilities. That whole choice and tradeoff aspect that Blizz enjoys to implement in the game.

For PvE purposes, even if you ignore the Vengeance aspect and assume it can be kept up on surplus tanks, isn't tank dps when the tank is not being pounded upon in general sufficiently much lower than it is when he is being pounded upon that it would be distinctly suboptimal to use tanks in a taunt rotation instead of traditional damagedealers?

I certainly know that, as a bear, if I am dual-tanking a boss together with somebody else (standard debuff-taunt scenario), I deal significantly less damage when I am not the one being hit than when I am being hit as I gain significantly less rage in the first case.

I just don't see how the vengeance mastery talent will affect that situation at all - especially when we have also been told by Blizzard that rage gain will be decoupled from the actual damage inflicted (i.e. bigger hits will not cause a bigger rage gain).

• moz · 04/09/10, 11:47 AM

On a slightly philosophical note, I hope they don't dumb down feral dps to the point that it becomes like everything else -- the ability to do very competitive dps if you're good/careful enough is an aspect that I think multiple classes wish they had more of. Including the revised mangle glyph should help a bit more with an easier rotation that's still reasonable in terms of DPS however, depending on what they do with clipping and refreshing of bleeds it could remove some of the complexity to the point it becomes too diluted. Tanking on the hand could certainly benefit from some variation, though I'd be interested to see how they balance the Vengeance concept for bears.

aceofsween · 04/09/10, 11:52 AM

Do bears even need Vengeance?

For the most part, Bears put on melee DPS stats to go tanking in. If tuned properly, they wouldn't really need such a mastery.

Royalite · 04/09/10, 11:59 AM

Originally Posted by Deliverance

For PvE purposes, even if you ignore the Vengeance aspect and assume it can be kept up on surplus tanks, isn't tank dps when the tank is not being pounded upon in general sufficiently much lower than it is when he is being pounded upon that it would be distinctly suboptimal to use tanks in a taunt rotation instead of traditional damagedealers?

I certainly know that, as a bear, if I am dual-tanking a boss together with somebody else (standard debuff-taunt scenario), I deal significantly less damage when I am not the one being hit than when I am being hit as I gain significantly less rage in the first case.

I just don't see how the vengeance mastery talent will affect that situation at all - especially when we have also been told by Blizzard that rage gain will be decoupled from the actual damage inflicted (i.e. bigger hits will not cause a bigger rage gain).

Well currently DK tanks don't have to worry about getting hit at all for a rage mechanic minus runestrikes from dodge/parries which might be phased out anyway.

While my examples may not hold a lot of water, the underlying principle I think is still solid... bring tanks to dps damage level will make tanks tip the balance in pve and pvp content. Maybe the issues in pve aren't so large and perhaps game mechanics could balance it out, but if Blizz continues to keep pve and pvp tied togetfer, warriors with tank survivability and high dps are just going to rock any hybird/pure dps. For example a feral druid would never shift into cat for burst dps but would just stay bear. Why shift when the bear and cat do the same dps?

Other classes/spec pay the price of lowered defensives for high offensives, yet tanks will get the best of both worlds?
I just don't see anything within the anouncement that supports the conclusion that tanks are the new dps spec; only that the gaps between tanks and dps will be smaller.

• moz · 04/09/10, 12:09 PM

Originally Posted by aceofsween

Do bears even need Vengeance?

For the most part, Bears put on melee DPS stats to go tanking in. If tuned properly, they wouldn't really need such a mastery.

It's something that's planned... let's hope they get it right.

Druids typically have more damage-dealing stats even on their tanking gear, so their Vengeance benefit may be smaller, but overall the goal is for all four tanks do about the same damage when tanking.

Videl · 04/09/10, 1:35 PM

Originally Posted by Darian_TruBlade

The general assumption is that it works like current mechanics that refresh DoTs and debuffs to their full duration. That is, casting a DoT before it runs out resets the duration to its maximum without interrupting the DoT ticks.

In this situation you can refresh your DoTs anytime you want without extending the time between ticks, but the motivation is to wait as long as possible so that you waste less time refreshing.

I'm surprised the statements haven't been more widely interpreted this way. Until I see a clear explanation otherwise this is how I'm inclined to interpret them.

Saying that they won't be clipped isn't necessarily the same thing as saying that you won't lose any ticks, just like refreshing a 6-tic dot via talents right now doesn't mean the same thing as adding 6 more tics. It could just mean that you won't see a gap longer than the normal tic duration when refreshing early, like you would now. If you allowed the resulting dot to tick off it could still mean you lose a tic of the dot, and over the course of the fight you'd have wasted time casting your dot equal to the sum off all your clipping divided by the duration of your dot.

They'll have to work something out with hasted hots to make that effective. For a hot that only has 6 tics like rejuv there just aren't enough haste thresholds to ignore them. If we see haste % numbers similar to this expansion you'd expect to start out hitting that first threshold for an extra tic in a raid and by the end of the expansion you would only cross the threshhold to add a third in very top-end gear with haste stacking. It would be noticably less often than one extra tic per tier.

aceofsween · 04/09/10, 2:10 PM

I don't really see how this change really makes that big of a difference. Whether reapplying DoTs clip or simply extend durations, it's still more effective to wait until just after the DoT falls off to reapply it, is it not? In both cases, you lose the remaining damage on the DoT which is the whole reason to avoid clipping anyway. Unless this is implemented in a way that allows for you to tack on the full duration to the current duration, I don't see how this change is really that helpful.

Let's take Insect Swarm. After 12 seconds, there's 1 tick remaining. With this change, you can reapply the DoT and it won't disrupt the time between the ticks, meaning the 7th tick will still come 14 seconds after you first applied Insect Swarm. However, when you extend the duration, you will only extend it up to 14 seconds, meaning you lose the damage from any remaining ticks (in this case, you'd lose 1 tick).

For this change to really make an impact, it would have to work slightly differently. For instance: say there is 2 seconds left on Insect Swarm (one tick left). When you refreshed IS, you get an additional 7 ticks (14 second duration) tacked on to the end of the current duration. However, as some people pointed out before, this would allow people to essentially stack DoTs up endlessly giving them a nearly infinite duration. They'd have to employ some sort of threshold so that you can never have more than the total number of ticks +1 at any given time. In Insect Swarm's case you'd have if you refreshed with 2 ticks remaining, you would still only have 8 total ticks left. This gives you a window of opportunity between the end of the duration and the 2nd to last tick where you can refresh your DoTs without wasting any damage. If you were to refresh before this window, you'd waste damage already guaranteed. If you refreshed after the window, you would be sacrificing DoT uptime.