SimulationCraft for Warlocks (4.0.3a numbers at 80 and 85) - Page 98

Nadagast · 05/11/09, 8:35 PM

Originally Posted by Demi9OD

Assuming your mistake is to always cast additional fillers rather than refreshing dots on time, as I believe would mirror actual ingame play, you need merely look at the filler DPET to determine which specs maintain high performance with sub-optimal play. Destruction builds suffer the least due to the high DPET of Incinerate, with mistakes on refreshing immolate causing more harm. Demo comes in the middle, and affliction suffers the most with additional shadowbolts not only being low DPET but haunt mistakes quite costly to overall damage.

I think this would flow naturally from what I described, I wouldn't make the mistake refresh dots early, just cast the next best thing in the priority system instead of the best thing. This would mean a lot of filler instead of refreshing dots, and sometimes casting a lower priority (DPET) dot instead of the 'best' choice

Marco · 05/12/09, 1:01 AM

For destro builds, I haven't found much treatment of the following situation: immolate is going to expire in more than 1.5 seconds but less than 2.25 seconds (adjusted for haste), conflagrate and chaos bolt are both on cooldown, and you don't need to life tap or refresh your curse. On live, you probably just cast incinerate because the damage difference isn't very large. But in 3.1.2 you'll miss a 15% multiplicative bonus on incinerate, and the dilemma applies even if chaos bolt is cooled down.

The options I can think of are:

1. Fire the nuke for lower damage.
2. Recast immolate and clip the last tick.
3. Cast corruption, assuming it's not already up for whatever reason.
4. Wait until it's time to recast immolate.

It appears that the Simulationcraft runs in this thread are doing (1); it would be interesting if the report could convey how many incinerate and chaos bolt casts are unboosted as a result, and of course it would also be interesting to see if (2) works out better. Waiting is very rarely the right answer, so I doubt (4) is the winner.

Abominatus_DMF · 05/12/09, 2:39 AM

It seems unlikely to me that 2 would be the correct answer either in a live environment, simply because you run the risk of hitting the "A more powerful effect is already active" issue when you try and clip Immolate. The DPS loss from that far outweighs what you will lose from the reduced efficacy of one Incinerate.

Of course, if you are able to determine that the new immolate will overwrite the old then the option is intriguing, but I don't think that anyone can reasonably expect to do that with certainty in a busy fight.

tkoreaper · 05/12/09, 3:30 AM

Originally Posted by Abominatus_DMF

It seems unlikely to me that 2 would be the correct answer either in a live environment, simply because you run the risk of hitting the "A more powerful effect is already active" issue when you try and clip Immolate. The DPS loss from that far outweighs what you will lose from the reduced efficacy of one Incinerate.

Of course, if you are able to determine that the new immolate will overwrite the old then the option is intriguing, but I don't think that anyone can reasonably expect to do that with certainty in a busy fight.

The game does in fact let you clip Immolate currently on the PTR, even if the one already applied is stronger, so #2 is the best option. I'm so glad they fixed this, it was rather frustrating in situations like this.

Hidden · 05/12/09, 4:16 AM

Originally Posted by Marco

For destro builds, I haven't found much treatment of the following situation: immolate is going to expire in more than 1.5 seconds but less than 2.25 seconds (adjusted for haste), conflagrate and chaos bolt are both on cooldown, and you don't need to life tap or refresh your curse. On live, you probably just cast incinerate because the damage difference isn't very large. But in 3.1.2 you'll miss a 15% multiplicative bonus on incinerate, and the dilemma applies even if chaos bolt is cooled down.

The options I can think of are:

1. Fire the nuke for lower damage.
2. Recast immolate and clip the last tick.
3. Cast corruption, assuming it's not already up for whatever reason.
4. Wait until it's time to recast immolate.

It appears that the Simulationcraft runs in this thread are doing (1); it would be interesting if the report could convey how many incinerate and chaos bolt casts are unboosted as a result, and of course it would also be interesting to see if (2) works out better. Waiting is very rarely the right answer, so I doubt (4) is the winner.

Option 4 is probably worth it if you're waiting less than ~0.2 seconds usually (depending on your total DPS and Immolate ticks).
Otherwise I'd do option 2 in 3.12 because raidbuffed the Incinerate bonus damage should be more than a tick of Immolate. (Assuming the overwrite bug is fixed)

@Above poster:
How do you test if the bug is fixed when Immolate ticks currently don't scale with SP on PTRs?

Asmodaeus · 05/12/09, 5:33 AM

Originally Posted by tkoreaper

The game does in fact let you clip Immolate currently on the PTR, even if the one already applied is stronger, so #2 is the best option. I'm so glad they fixed this, it was rather frustrating in situations like this.

I doubt this is the best option. The Immolate dot is one of the spells with the highest DPET and a clipped Immolate only does 4/5 of the dot damage. Not only that, but if you clip an Immolate when it's 1 second away from expiring, you're doing the same thing as reducing Immolate uptime by 2 seconds.

Now that we have a new PTR builld with the announced changes to F&B, it would be really interesting if someone could determine how exactly F&B works, how it stacks with talents like Emberstorm, etc.

Asmodaeus · 05/12/09, 8:26 AM

Originally Posted by Midra

As an aside, how long has S&F multiplied onto coefficients, rather than added? Tooltip still feels misleading to me.

Since 3.0

Asmodaeus · 05/12/09, 11:05 AM

Originally Posted by haldol

I guess my question is, if I spec 3/13/55 and do my best to only LT during movement/transitions, do you guys think it would be higher dps to glyph for LT or Immolate in ulduar real situations?

The sim gives you an upper bound on the value of Glyph of Life Tap.
You can run the sim, changing the actions list for 3/13/55 so that it's equal to the actions list of 00/13/58. That would give you a lower bound.
The real value would be somewhere in the middle, probably closer to the upper bound than to the lower bound as long as you try to spread Life Taps.

And btw, another difference is that 00/13/58 has an Imp capped for spell hit, while 03/13/55 doesn't. It shouldn't matter much, though.

Robi · 05/12/09, 11:28 AM

Fire and Brimstone: Now increases the damage done by your Incinerate and Chaos Bolt by 3/6/9/12/15% on targets afflicted by Immolate, and increases the critical strike chance of your Conflagrate by 5/10/15/20/25%.

Originally Posted by Midra

Incin hit damage:
Avg damage without immolate down, just 20 hits:
Base: 2028
w/F&B: 2191
w/F&B, ember: 2333
These are pretty close to what we'd expect, no surprises there.

Shouldn't the damage without immolate be the same with and without F&B? Why is there an increase in damage? Did you spec the same just without F&B/ember?

Demi9OD · 05/12/09, 11:46 AM

Originally Posted by teddypicker

the three points in affliction are not as effective in deep destro in my view becuase if you went for 11%hit and had 3% in suppression you will still be taking a chance on missing on your incinrate,chaos bolt and conflagrate as there do most of your damage going 13/58 is much more beneficial in your dps. also alot of items in ulduar are more hit based than naxxramas so it wouldnt be difficult to acquire the correct amount of hit you need

Suppression covers all spells now, not just affliction. You are basically trading 3% crit and the Lifetap Glyph for the Immolate glyph and less globals spent on Life Tap.

pfooti · 05/12/09, 11:46 AM

Originally Posted by teddypicker

the three points in affliction are not as effective in deep destro in my view becuase if you went for 11%hit and had 3% in suppression you will still be taking a chance on missing on your incinrate,chaos bolt and conflagrate as there do most of your damage going 13/58 is much more beneficial in your dps. also alot of items in ulduar are more hit based than naxxramas so it wouldnt be difficult to acquire the correct amount of hit you need

You do realize that suppression now gives you a flat +3% hit chance to ALL spells, not just affliction ones, right? And that the simcraft shows that you can re-itemize for less +hit on your gear overall in order to get gear that has more offensive stats on it? In the "real world", you're of course likely to be limited by the gear you have rather than the gear you want, but in terms of theroetical maxima, it's possible to get a high-hit and a low-hit set of gear.

nuibank · 05/12/09, 6:30 PM

Originally Posted by Tis I Lucifer

hmm. I have been researching this all day and it has been driving my nuts. Here is my armory... The World of Warcraft Armory

With my current gear and either spec I am able to use 11% as a hit base? I know there are many threads on all of this but the problem is, none of them agree. Some say 17% which means 14% with suppression. Yet some say 14% with suppression. ARGGHH! Anyways I raid 25 man ulduar and would like to be effective during the fight and not miss.

You need to get 17% to not miss a lvl 83.

Suppression = 3%
Improved Faerie Fire(Oomkin) or Misery(SPriest) = 3%

289 hit = 11%
368 hit = 14%
446 hit = 17%

Find a combo that gets you 17%

Examples:
Suppression + Misery + 289 hit = Good
Suppression + 368 hit = Good
Misery + 368 hit = Good
446 hit = Good

irharry · 05/12/09, 8:08 PM

There a few numbers in Zak's sims that don't match anything I can reproduce, and seem wrong. I'm wondering if anyone can clarify, or explain for me?

These numnbers;

Warlock_T8_00_41_30 Int=0.27 Spi=0.77 SP=1.23 Crit=0.69 Haste=1.32

Warlock_T8_03_52_16 Int=0.35 Spi=1.32 SP=1.18 Crit=0.34 Haste=0.73

My sims replicate Zaks numbers very closely, except these two in bold.

Spirit is of course ranked a lot higher for deep demo builds due to using glyph of life tap, but I cannot work out how the glyph makes this much of a difference? Im getting a value of 0.99 for spirit for the deep demo builds. Is there something I'm missing?

Also, I can't understand the huge drop in value for haste. I'm showing a reduction, but not to anywhere .73. Again, is there something I'm missing?

Darkmantle · 05/12/09, 8:59 PM

Originally Posted by irharry

There a few numbers in Zak's sims that don't match anything I can reproduce, and seem wrong. I'm wondering if anyone can clarify, or explain for me?

These numnbers;

Warlock_T8_00_41_30 Int=0.27 Spi=0.77 SP=1.23 Crit=0.69 Haste=1.32

Warlock_T8_03_52_16 Int=0.35 Spi=1.32 SP=1.18 Crit=0.34 Haste=0.73

My sims replicate Zaks numbers very closely, except these two in bold.

Spirit is of course ranked a lot higher for deep demo builds due to using glyph of life tap, but I cannot work out how the glyph makes this much of a difference? Im getting a value of 0.99 for spirit for the deep demo builds. Is there something I'm missing?

Also, I can't understand the huge drop in value for haste. I'm showing a reduction, but not to anywhere .73. Again, is there something I'm missing?

For a discussion of different scaling factors for the specs go to here. Try changing spirit to be 600 higher and then running new scale factors. That will help eliminate mana usage as a reason for spirit scaling.

Looking at the damage breakdown between the 2 specs doesn't give any clear signals as to why the haste scaling is so different. There is a little bit more felguard damage for 52_16 and 41/30 uses CoA which gets more benefit from haste than 52_16 which uses CoD. 41/30 also has better DPET on its main nukes with inc 4861 > shadow 4424 and soul fire 10617 > 10055. 41_30 also has a far stronger immolate 10766 > 7988.

More work will need to be done however to determine the exact reasons for such a large difference.

Marlucia · 05/13/09, 12:08 AM

Originally Posted by Darkmantle

For a discussion of different scaling factors for the specs go to here. Try changing spirit to be 600 higher and then running new scale factors. That will help eliminate mana usage as a reason for spirit scaling.

Looking at the damage breakdown between the 2 specs doesn't give any clear signals as to why the haste scaling is so different. There is a little bit more felguard damage for 52_16 and 41/30 uses CoA which gets more benefit from haste than 52_16 which uses CoD. 41/30 also has better DPET on its main nukes with inc 4861 > shadow 4424 and soul fire 10617 > 10055. 41_30 also has a far stronger immolate 10766 > 7988.

More work will need to be done however to determine the exact reasons for such a large difference.

It seems likely that insufficient mana is the reason for both discrepancies - Spirit for obvious reasons, and haste since I'd imagine you gain less per point of haste if, for example, you need to life tap once more near the end of the fight to get another spell off.