Multiple heroism rotations: - Page 2

galzohar · 04/08/08, 4:59 PM

Actually the RNG variance from MoE and JoW isn't all that great, at least not compared to taht caused by the changing uptime% of JoW due to paladin slack (or lack of ret pally combined with busy healing) goes anywhere from 0 to 100%.

Optikalusion · 04/08/08, 5:20 PM

Originally Posted by galzohar

2. While melee has a lot of haste buffs, it does not matter. What matters is that their WHITE dps increases by 30% and their specials are increased by whatever rage/energy is generated by that 30% extra attack speed. Overall not hard to calculate, but just the fact they have a lot of attack speed multipliers don't mean anything.

Once again, not entirely that simple! A good portion of melee dps comes from procs like sword spec and WF procs. These things also proc off of each other. So with every % of haste given a melee there is more chances per second that any of these things will occur, and increasing the occurance of one also further increases the occurance of the other!

Rogues are in the area of 70% white damage and 30% specials (warriors far less at around 40% white). What remains to be mathematically proven is if the added gains from proc dps and extra rage/dps in that 30% area is equivalent to a much more easily calcuable caster % dmg increase.

Also regarding the casting speed buffs:

bloodlust (+30% haste)
icy veins (+20% haste/20s)
drums of battle (+5% haste)
combustion (3 guaranteed crits)
skull of gul'dan (+11.15% haste/20s)
destruction potion (+120dmg, 2% crit/15s)
flame cap (+80 dmg)
hex shrunken head (+211 dmg/20s)

Drums of battle and skull of gudan are rating increases, not percentual and will therefore be added together with passive haste from gear before multiplied by veins/bloodlust. Small nitpicking, sorry!

manly · 04/08/08, 5:29 PM

Yes, I'm well aware. The point still stands though, a full t6 mage with perfect pre-T6 gear does roughly 2x his normal DPS when using all the cooldowns listed (and molten fury). This means roughly ~4k+ DPS for the duration. Of course, in this case I don't have molten fury (+20% dmg), which hurt the numbers, but regardless the bulk of the other cooldowns makes any additional 'cooldown' stack on top of it immensely powerfull.

Vulcana · 04/08/08, 5:33 PM

Originally Posted by galzohar

Actually the RNG variance from MoE and JoW isn't all that great, at least not compared to taht caused by the changing uptime% of JoW due to paladin slack (or lack of ret pally combined with busy healing) goes anywhere from 0 to 100%.

True, but I've been screwed out of some ungodly amounts of crit due to the RNG system not being my friend. But yeah, JoW uptime as opposed to any rng-based effects is going to be a bigger factor than anything else.

manly · 04/08/08, 6:05 PM

Thinking about this, why not simply compare the bloodlusted DPS from all members of a group assuming a baseline of 'best pre-sunwell gear'. Give to whichever group has the highest numbers.

I know for mages we can only get good dps from bloodlust twice in a 6 minute fight. I imagine all classes have to deal with the same thing up to some level (ex: haste potion cooldown will play into this in addition of other cooldowns).

Anarkii · 04/08/08, 10:39 PM

One other small thing to note is that for melee, haste is independent of reaction and latency since auto attacks occur server-side, while for casters it is not. The 1.732s fireball with bloodlust up may practically be a 1.8 or 1.9s fireball.

Iluminati · 04/08/08, 11:36 PM

Problem with double heroism on mages/locks is two-fold:

a) threat
b) mana - 30% cast speed = increased mana usage. Most mages are pretty tight on mana, and warlocks dont want to be using lifetap anymore than necessary.

Melee has neither problem in my experience.

ohrion · 04/09/08, 10:10 AM

Originally Posted by manly

Thinking about this, why not simply compare the bloodlusted DPS from all members of a group assuming a baseline of 'best pre-sunwell gear'. Give to whichever group has the highest numbers.

DISCLAIMER: I am not in sunwell.

I'm not sure if I'm understanding you correctly, but your statement sounds similar to one I hear from many raiders and if you're saying what I think you're saying it won't actually maximize raid DPS, it maximizes those individuals' DPS. To maximize raid DPS you want to find the group that sees the largest increase in DPS.

For example (just random numbers)
Group 1 has a collective DPS without heroism of 6k.
Group 1 has a collective DPS with heroism of 7k.

Group 2 has a collective DPS without heroism of 5k.
Group 2 has a collective DPS with heroism of 6.5k.

By your suggestion (again, if I understand you) we would give heroism to group 1, in reality raid DPS sees more benefit from giving it to group 2. It is possible that the highest DPS group will also see the highest gain in DPS, but many factors play into it so it may not be the case.

Trickykid · 04/09/08, 12:41 PM

Originally Posted by Iluminati

b) mana - 30% cast speed = increased mana usage. Most mages are pretty tight on mana, and warlocks dont want to be using lifetap anymore than necessary.

With very low regen outside of life tap, the life tap frequency (relative to nukes) increases by a very small amount. Haste just increases the speed at which a lock can go through a nuke/tap "cycle", it doesn't make him less efficient. This is why locks start at just shy of 30% increased damage from 30% increased haste if they didn't change their life tap habits during that period. If you life tap prior to the haste, you can amplify trinkets or other buffs stacked on top of heroism/bloodlust more.

Mjollnir · 04/09/08, 1:58 PM

Originally Posted by Iluminati

Problem with double heroism on mages/locks is two-fold:

a) threat
b) mana - 30% cast speed = increased mana usage. Most mages are pretty tight on mana, and warlocks dont want to be using lifetap anymore than necessary.

Melee has neither problem in my experience.

I don't know about you, but I am an aggro FIEND. In fact, my system won't support dmg meters (dmg meters, recount, swstats etc have all caused lag and d/c issues), and my measure of monitoring my dps has instead become watching Omen

This question is more than just class limited: Trolls have zerking, which is wonderful when timed with Lust and 2min cds.
Haste pots are not limited to rogues.
White damage for a warrior might actually *decrease* with haste, due to tighter requirement to use rage burnoff (heroic strike) and a recent discovery there is no 'hastened' penalty to rage generation.
Can't believe the OP forgot about warriors with the 20% burn... unless I read it wrong. 'Execute Range' is the popular term afterall, not 'Molten Fury Range'...

alienangel · 04/09/08, 2:16 PM

Originally Posted by Mjollnir

Can't believe the OP forgot about warriors with the 20% burn... unless I read it wrong. 'Execute Range' is the popular term afterall, not 'Molten Fury Range'...

Correct me if I'm wrong, but mages get bonus damage on several hastable attacks while in execute range, warriors only get one extra attack (Execute). So heroism on a warrior is only useful sub 20% in terms of generating rage faster, leading to more executes. Does that work out to more or less of a gain than a hasted mage sub-20%?

And if anyone has threat issues, I once more submit the option of giving the double heroisms to the 3xBM hunter + enhancement shaman + feral druid group, once at 30s into the fight, and once at 2:30 into the fight. Our mana regen scales well with heroism too (assuming JoW).

Voxx · 04/09/08, 2:32 PM

The difference is how much dps that one ability that we gain puts out. Most Sunwell-ready warriors put out an Execute of about 900-1000 damage (non-crit) every global cooldown when Heroism/Bloodlust is up. That's a direct 600 dps ish increase not counting crits and the effect on white damage that Heroism/Bloodlust has. I'm not going to say this makes warriors the undeniable choice for Heroism sub 20% but it's certainly misleading to have someone say that we "just" gain one ability sub 20%. That one ability is a huge factor.

Mjollnir · 04/09/08, 2:48 PM

Originally Posted by alienangel

Correct me if I'm wrong, but mages get bonus damage on several hastable attacks while in execute range, warriors only get one extra attack (Execute). So heroism on a warrior is only useful sub 20% in terms of generating rage faster, leading to more executes. Does that work out to more or less of a gain than a hasted mage sub-20%?

And if anyone has threat issues, I once more submit the option of giving the double heroisms to the 3xBM hunter + enhancement shaman + feral druid group, once at 30s into the fight, and once at 2:30 into the fight. Our mana regen scales well with heroism too (assuming JoW).

Yes, it allows us an additional ability to use which as Voxx stated will hit for 900-1000 every GCD outside of BT/WW usage.
What perhaps I failed to hint at was: this is the time when a warrior pops Recklessness. It's not just the abilities, but as this thread has set out to study: the timing of cooldowns coupled with abilities.

Edit: As far as how it works out to a hastened mage... I'd go so far as to blindly (without math) argue there isn't a class or spec in the game that outdps's a warrior during those 15seconds of fame (considering any comparision is popping identical/similar cooldowns/pots).

galzohar · 04/09/08, 2:59 PM

I think sub-20% bloodlust for the melee and mage group is a no-brainer (although nothing is definite, but I doubt this can be beaten). The real issue is who is worthy of the bloodlusts beyond the 2nd?

On a side note if your melee group has a ret paladin, your bloodlust/heroism for melee just became that much less effective due to his lower personal dps, which probably includes lower white+seal dps.

At the end it's something that'll just need to be modeled on every class' spreadsheet, like said above. Take a rogue on the spreadsheet and see the DPS gain with 1, 2, 3... bloodlusts, then do the same for warriors, mages, warlocks, hunters... and see what you get for each. Of course this would likely have some dependancy on fight duration and the party support those classes have. Giving optimal DPS party is probably the most logical though as you would generally not blooslust a sub-optimal party (say 2 hunters+feral+leftover healers for example are obviously not a good use of a bloodlust).

Rhaegal · 04/09/08, 3:23 PM

Originally Posted by galzohar

On a side note if your melee group has a ret paladin, your bloodlust/heroism for melee just became that much less effective due to his lower personal dps, which probably includes lower white+seal dps.

That's not entirely true. For any secondary Bloodlust/Heroism given to that group, you have to switch someone out, why not make it the ret pally? And for even the first one, why not switch the pally out with someone who would benefit more, and isn't in a group that would be getting it at that time? Yes, switching a lot is going to take up a few seconds of someone's attention, but it's generally worth it, especially if the person doing it has a DPS rotation that can support a couple seconds' less attention, like a mage mid-fireball, etc.