Sunday, March 01, 2009

Bloodlust Timing

Here is a very interesting theorycrafting discussion on when a raid should pop Bloodlust/Heroism. Should you Bloodlust early in the fight, or should you wait until 35%, when a lot of the extra damage Execute-style talents kick in?

Surprisingly, the answer turns out to be that it doesn't really matter! So the final conclusion is that the best use of Bloodlust is:
  1. If there is no other factor, use Bloodlust early to maximize the number of people and cooldowns available.
  2. If one particular phase needs to be burnt through in the shortest possible time, use Bloodlust then, and match cooldowns to Bloodlust.

Honestly, this result seemed a little counter-intuitive to me at first, but the math looks good. If you're interested, check out the algebra at the link.

Here's a "real numbers" example. Let's say there's a boss with 1,000,000 health. You can do 1000 damage-per-second. Below 35% health, let's say you do an extra 20% damage for 1200 dps. Further, let's say that Bloodlust boosts your damage by 30% for 40 seconds.

Case 1: Bloodlust Late

If you Bloodlust at the end, you do 1000 * 1.2 * 1.3 = 1560 dps for 40 sec.
  1. Normal mode takes 650,000 Health / 1000 DPS = 650 sec.
  2. Bloodlust for 40 seconds, doing 40*1560 = 62,400 damage.
  3. Then you do the remaining 287,600 damage in 287,600 health / 1200 dps = 240 sec.

Total time to kill the boss = 650 + 40 + 240 = 930 seconds.

Case 2: Bloodlust Early

If you Bloodlust at the start, you do 1000 * 1.3 = 1300 dps for 40 sec.
  1. Bloodlust for 40 seconds, doing 40*1300 = 52,000 damage.
  2. Boss has 598,000 health until 35%. It will take you 598,000 health / 1000 dps = 598 sec.
  3. Then you do the remaining 350,000 damage in 350,000 health / 1200 dps = 292 sec.

Total time to kill the boss = 40 + 598 + 292 = 930 seconds.

Conclusion

It doesn't matter when you Bloodlust during the fight if you just look at the bonus Execute-style damage and the extra damage from Bloodlust!

There are two advantages to going early. First, everyone can line up all their cooldowns like trinkets or Avenging Wrath and use them for maximum effectiveness. Second, it is more likely that everyone is alive at the start of the fight. By the time you reach 35%, you may have lost one or two people.

However, if there is a specific phase of the fight that needs to be shortened, then you should Bloodlust at the start of that phase. For example, even though the total fight lengths are the same, in Case 1 you only spend 280 seconds in the last phase, while in Case 2 you spend 292 seconds in the last phase. If a phase does increased damage, or has a separate enrage timer like Phase 3 of Prince Malchezzar or Lady Vashj, you'd be better off waiting for that phase before you Bloodlust.

To sum up from the linked article again, the best rules for Bloodlust/Heroism timing are:
  1. If there is no other factor, use Bloodlust early to maximize the number of people and cooldowns available.
  2. If one particular phase needs to be burnt through in the shortest possible time, use Bloodlust then, and match cooldowns to Bloodlust.

18 comments:

  1. I tagged you for a screenshot meme!!

    http://retribution-paladin.blogspot.com/2009/02/my-screenshot-meme.html

    I love heroism! Its awesome! I usually wait to pop Avenging Wrath till its up. (our raid leaders usually try to pop it early so we have a change to pop it again towards the end on long fights)

    ReplyDelete
  2. I also am a huge heroism fan (isnt everyione) but i tend to go for the begining of the fight to pop it out when im raid leading.

    ReplyDelete
  3. Using heroism at the beginning also starts the cooldown so it'll be ready sooner. You know...if shaman are ultra rare on your server and your guild only has one. =)

    ReplyDelete
  4. This comment has been removed by the author.

    ReplyDelete
  5. Ack, that's very counter-intuitive but it's true. I had to read thru the original math a few times.

    I think a way to make it more intuitive is to think of a boss kill on a videotape. Heroism is the fast forward button. You get to hold down fast forward for exactly 40 seconds at any point in the tape you like. Obviously you can't make the tape finish any faster no matter when you push it.

    ReplyDelete
  6. Yeah I like to have it about 1/3 into a fight, that way most of the raid is alive (hopefully), passive-on hit buffs are active ( demonic pact, the blood DK atkpwr buff--forget the name, elemental oath, etc), and the tank has enough aggro that it'd be pretty difficult to pull of him/her.

    ReplyDelete
  7. 2/3s in is what I think is best. People are still alive, and alot of the on hit, stacking has occured, and should help maximize it best.

    Also, with the next patch you'll only get one bloodlust/heroism per fight.

    ReplyDelete
  8. The one place I found a late heroism to be actually important, was on 3m Patchwerk attempts.

    We needed the first two min to gauge if we were on track for a kill (~33-36%) before poping it.

    ReplyDelete
  9. The math fails to take into consideration the haste coefficient above and below the execute phase.

    Affliction Warlocks and Fury Warriors (at least at a certain gear level) both receive an added benefit from the haste inside of execute range, that doesn't occur outside.

    For fury warriors at least, bloodlust/heroism during execute phase means dumping the excess rage into execute. Outside of the phase, you're dumping into the much less-efficient heroic strike

    ReplyDelete
  10. Maybe I missed something, but in your "bloodlust late" example, you divided the health by 1200 instead of 1000.

    It should be 287 seconds, not 240.
    I think so at least, from what I see there.

    ReplyDelete
  11. Nevermind, I see where I didn't get it. I can't edit my comment though, sorry.

    Feel free to delete.

    ReplyDelete
  12. I don't agree really. At the moment the fights are so trivial that noone should be dieing on the majority of bosses and therefore there is no argument for using heroism at the start "since people wont survive."
    Samownall - WoW Blog

    ReplyDelete
  13. It's completely dependent on the encounter. There is no universal "moment."

    ReplyDelete
  14. We normally wait untill the last 30% ish , but its definitely a valid point about using it earlier, in case people die earlier on, and losing the extra dps/healing ( and who knows that extra healing might have saved that person.. )

    ReplyDelete
  15. The flaw with your math is that you are oversimplifying. Bloodlust does not just plainly increase your damage by 30%, it increases your haste by 30%. An increase in haste has a whole slew of different effects for certain classes, most notably warriors who will see an incredible increase in DPS due to their near infinite rage generation. You should recalculate using more intricate specifications.

    ReplyDelete
  16. I see you have the same sort of posts that I did.
    Never fear, there's a TTT entry coming up soon with more math, more reason, and simcraft graphs to back it up (hopefully then people will stop arguing that we're wrong)

    ReplyDelete
  17. There's also the issue of popping cooldowns while inside BL vs outside. Personal cooldown contribution to overall DPS increases while inside bloodlust.

    Also, sated wears off upon death so if anyone dies and is resurrected again can be bloodlusted again for additional DPS.

    ReplyDelete
  18. The math seems right, but how about in PvP?

    When I play arathi basin, hero is so important to take blacksmith. Same thing for the initial fight in WSG. In arena, however, the usage of heroism was a huge question. The best thing is heroism scales with crowd controls.

    Now that they took hero away from arena, we need to do the math on how it helps BG teams. Does hero help cap flag timers?

    ReplyDelete