SimulationCraft for Warlocks (4.0.3a numbers at 80 and 85) - Page 48

Anyakfe · 03/15/09, 6:01 AM

I've noticed that the profile Warlock_T8_53_00_18 and Warlock_T8_53_00_18_DG are not the same:

warlock=Warlock_T8_53_00_18
(...)
actions+=/haunt,debuff=1/corruption/curse_of_agony/unstable_affliction/haunt

warlock=Warlock_T8_53_00_18_DG
(...)
actions+=/haunt,debuff=1/curse_of_agony/corruption/unstable_affliction

• Zakalwe · 03/15/09, 2:06 PM

Originally Posted by Anyakfe

I've noticed that the profile Warlock_T8_53_00_18 and Warlock_T8_53_00_18_DG are not the same:

Good catch - I fixed the discrepancy, reran the simulation, and updated the OP.

Migage · 03/15/09, 9:27 PM

Originally Posted by Zakalwe

Turns out simcraft was modeling an old mechanic where dot applications could be fully resisted from level-based resists. This did not show up as misses in the reports, but still resulted in dots having to be recast. I fixed this bug, reran the simulations, and updated the OP.

Correct me if I'm wrong, and this may not be the right post to quote, but if a CoD misses or is resisted you cannot recast CoD for the minute cooldown. I've only seen it happen a couple of times live where i've casted CoD, its been resisted or missed, and have had to switch in CoA for the minute of CoD's CD.

Illijilli · 03/16/09, 12:03 AM

Originally Posted by Migage

Correct me if I'm wrong, and this may not be the right post to quote, but if a CoD misses or is resisted you cannot recast CoD for the minute cooldown. I've only seen it happen a couple of times live where i've casted CoD, its been resisted or missed, and have had to switch in CoA for the minute of CoD's CD.

You are correct.

bhaltayr · 03/16/09, 12:31 AM

The user is at the hit cap in all of these simulations through gear and raid buffs meaning that CoD will never miss or be resisted and therefore will never be on cooldown without being on the mob.

dunwich · 03/16/09, 12:31 AM

Has there been any oppertunity to run the deep demonology specs with the Meta Glyph? I was wondering how that little bit of extra time might play out on the DPS numbers.

duhwhat · 03/16/09, 12:46 AM

For a fight length of n+30 seconds, is it worth the GCD for destro specs to cast a final CoA after the last CoD?

Cydos · 03/16/09, 1:40 AM

Ok, for the 0/41/30, and 0/40/31 builds you are running CoA according to the profiles, but on all the other non-Aff builds you are running CoD. Is that an oversight, were the profiles just not updated, or is CoA really better for the hybrid builds for some reason? Seems like if they aren't spec'ed for CoA then CoD might be better for the boss fight.

Morrigan · 03/16/09, 1:45 AM

Originally Posted by Cydos

Ok, for the 0/41/30, and 0/40/31 builds you are running CoA according to the profiles, but on all the other non-Aff builds you are running CoD. Is that an oversight, were the profiles just not updated, or is CoA really better for the hybrid builds for some reason? Seems like if they aren't spec'ed for CoA then CoD might be better for the boss fight.

It's because the the 0/4x/3x builds rely on CoA for proccing Molten Core.

Cydos · 03/16/09, 2:20 AM

Okay, that certainly makes sense. I know all fights were tested using Doomgaurds, but what about summoning an infernal during the last minute of the fight? I know that's a trick that is suggested for affliction locks, and seems like it couldn't hurt the other specs, so long as it was limted to the last minute of burn time. I tested in on simcraft using the 0/40/31 build at 1k interations, with a max time of 300 and it came out to a 15 point dps gain. If that's correct it seems like a worthy investment of time and mats.

Abominatus_DMF · 03/16/09, 2:25 AM

15 DPS on a spec that outputs about 7000DPS is approximately 0.2%. That's not a meaningful value.

Cydos · 03/16/09, 2:43 AM

considering that some of the specs listed are seperated by less then that?

supplicium · 03/16/09, 2:45 AM

Might want to toss in 00/38/33 into your spec list it may end up scaling better, not sure. But I can't seem to find where to add in/change gear #'s for t8 to test this.

It's currently sitting in the high 6800's so fights like hodir it would end up scaling better as you often sit crit capped ( I'm aware this is a situational event BUT if you are going for a 3min kill every bit of dps counts)

Abominatus_DMF · 03/16/09, 4:12 AM

Originally Posted by Cydos

considering that some of the specs listed are seperated by less then that?

This is just my personal opinion here, but I think that if two specs are that close together, the choice between them is far more down to your latency, playing style and personal preferences than it is to the difference in DPS between them. This is especially true given that very few fights are equivalent to Patchwerk, and the mechanics of each specific fight will have FAR more relevance to your final output than 15 DPS from an infernal might have.

It is valuable to keep track of different specs which are initially close together in DPS because they will scale differently and because they will be affected differently by any upcoming PTR changes. Adding an infernal to one of the specs is a static change that is highly unlikely to change the relative scaling of the specs in question.

That's not to say that you shouldn't use an infernal when it's appropriate, but 15 DPS is an inconsequential change on fights that do not favor the infernal. You're talking about 4500 damage total in a 5 minute fight.

Naforce · 03/16/09, 5:11 AM

Originally Posted by supplicium

Might want to toss in 00/38/33 into your spec list it may end up scaling better, not sure. But I can't seem to find where to add in/change gear #'s for t8 to test this.

It's currently sitting in the high 6800's so fights like hodir it would end up scaling better as you often sit crit capped ( I'm aware this is a situational event BUT if you are going for a 3min kill every bit of dps counts)

0/38/33 will most likely end up with higher dps. Unless you're only talking about Hodir fight DPS, of course.