I've noticed that Omen and KTM don't calculate threat with real high accuracy on VR. You can ride the top of the threat threshold by a goodish margin without pulling. The one time I saw a Mage pull, they were some ~18% ahead of the highest tank on threat. It's definitely still an aggro-sensitive fight, though....I force myself to wait until 70% to Soulshatter the first time, seems to work best for the duration of the fight total for me.
Actually they do adjust for the knockbacks pretty accurately. A simple search on thottbot shows the threat reduction amount/percentage for each threat reducing spell in the game. Parsing the combat log picks up that knockback ability and adjusts the current tanks threat accordingly.
So unless thottbot is wrong and his threat reducing knockback changes or the amount of -threat isn't listed correctly, one can say that Omen/KTM are probably pretty close to accurate.
In fact the only inaccuracy I find is with the seed of corruption spell on your current target. It seems Omen/KTM isn't smart enough yet to understand it.
My simulation consistently says using Immolate will increase Destro Warlock dps. If the Warlock has equivalent Fire/Shadow damage and all the +fire talents...... it is always better to use Immolate. Which is obviously contrary to what everybody is saying...... Grrrr.....
Up until recently I had straight +dmg. The reason for this was as destro I was balancing all three of my stats and keeping +hit high. There is not much gear that is +shadow dmg and +hit. I was at ~1100 +dmg buffed, 32% crit buffed with destro spells, and 15% hit. I also have all of the +fire spell talents. I also found that adding immo in my spell rotation could not keep up with the dps I could put out with straight SB spamming.
How valuable 1% spell crit is, is GREATLY increased in a raid environment with consumables, a Shadow Priest, etc. and I don't see that being included like.....ever, when people are figuring how to gem and such.
...
The results didn't lie.
The sheet posted with the raid ISB model accuratly averages in all of the ISB uptime you get from your crits as well as every other ISB speced warlock in the raid. All you need to do is put in your buffed stats and it will produce accurate proportions between the stats in the next stat section.
I think your results are actually lying. Any time you take a random event and try to quantify real differences at the order of magnitude you're considering you absolutly have to take hundreds of thousands of samples to prove anything to any degree of statistical confidance. I am willing to bet that, at best, you are working off of perhaps a dozen or so of your own personal parses to make the comparisons you are making which boast, at most, a couple thousand observations.
It is completely possible and well within the limits of probability that you are simply observing a 'lucky streak' or even that the difference is all in your own perception. The most likely explaination is that something in your raid makeup changed that you're not taking into account.
My simulation consistently says using Immolate will increase Destro Warlock dps. If the Warlock has equivalent Fire/Shadow damage and all the +fire talents...... it is always better to use Immolate. Which is obviously contrary to what everybody is saying...... Grrrr.....
I know. That's what I meant in my earlier post -- I don't understand it, but I'm seeing it too, very consistently. I ran with immo+corr+SB for a very long time, and only recently switched to just SB -- and my dps has noticably gone up (which was the opposite effect I expected.)
And it's really bugging me, because it makes me wonder, if there is indeed something going on (even if it's just a human element), then, say, +crit might be more valuable than we think.
If someone has two combat logs of stand-and-shoot fights (e.g. mag and Al'ar, or parts of FLK), maybe those could be analyzed if they are from the same person, one with immo+SB, one with just SB, just to see if there's something obvious.
Or look at what the simulators should predict for a particular gearset, and look through a bunch of combat logs and see if that matches.
Maybe it's something as simple as, most people, when they are JUST solely spamming one spell, do far far better with stopcasting macros, etc., and when they add in another spell, minute hesitations add up over the time of the spell.
Or, as someone suggested, it's something about losing some of your own crits because you cast immolate every 15 seconds instead of a followup SB.
Or maybe we're all seeing what we expect (although I really didn't expect SB spam to win, but maybe I'm a terrible observer.)
As far as I can tell most differences between theoretical calculations to practical results is time in which the player is not casting - either moving around, not using /stopcasting optimally (at least enough to match the latency entered on the sheet) or just wasting time thinking "what do I do next?". When dead I try watching people low on DPS and I always see these sorts of delays.
Something I have been curious for a while, since I'm almost at 16% hit, I was wondering if you really do need 16% hit instead of like 15%? Don't mobs always have a 1% chance to resist a spell no matter how much hit you have and their level?
Something I have been curious for a while, since I'm almost at 16% hit, I was wondering if you really do need 16% hit instead of like 15%? Don't mobs always have a 1% chance to resist a spell no matter how much hit you have and their level?
Yes, and that's why the miss rate is 17% and not 16% for level 73 mobs.
Lately we've been hitting the debuff limit with 4 warlocks + normal raid stuff.
So right now we're trying to determine which debuffs to prioritize etc..
My question is: Has anyone here encountered this issue and how did you deal with it?
It's almost making me wonder if switching to a Nuke build would be advantageous.
SnF or even a FG build (Since we need someone to tank Leotheras anyway) are looking interesting at this point. Partially I'm bored with affliction and staring at dottimers, and partially I don't see affliction scaling as well with incoming gear (more crit/haste on it)
If your hitting the debuff cap with 4 warlocks then I'm assuming at least 3/4 of you are affliction. One of you going destruction is probably a good idea since dots wont be pushed off and more ISB uptime for everyone. Or I guess felguard if you want that.. though soul link is really not needed for leo, just need good FR gear
If your hitting the debuff cap with 4 warlocks then I'm assuming at least 3/4 of you are affliction. One of you going destruction is probably a good idea since dots wont be pushed off and more ISB uptime for everyone. Or I guess felguard if you want that.. though soul link is really not needed for leo, just need good FR gear
Everyone's affliction, only 2 down to UA.
The more I've been reading the more I've been contemplating destruction over affliction. Felguard looked interesting because.. well some people seem to make it work for them. It can be a fun build in pvp also; but I wouldn't spec to it if it would make my pve dps suffer noticeably.
The problem is my gear is not quite hit capped. I can get up to ~11.7% without sacrificing much (CE ring instead of Kara one). However this leaves my crit relatively low. I could replace 2 crit/dmg gems with hit/dmg, but that would only gain me .5% hit taking me to 12.25% roughly.
That would put me at 95.25% to-hit with current gear options. Still a bit shorter than I'd like to be if trying SnF.
Any suggestions? I have most of the +hit options I can get, short of replacing the +20 dmg on my gloves which isn't something I especially want to do.
Keep in mind that in many fights (at least in SSC/TK) you will be dpsing lvl 70/71 adds as much as the boss himself, so not being hit-capped will be less of an issue than you might think. Even less so if you have access to an elemental shaman.
Keep in mind that in many fights (at least in SSC/TK) you will be dpsing lvl 70/71 adds as much as the boss himself, so not being hit-capped will be less of an issue than you might think. Even less so if you have access to an elemental shaman.
The minimum requirements I used when I respec destro was the following:
1000 +dmg Self buffed
12% +Hit
20% +crit without talents
With these stats I did quite well. You should be able to get to these stats with kara, crafted, and rep gear.
As for your spec, if you have dedicated SPs in the raids then go with the shadow destro spec mentioned a couple posts ago. However, if you only have fire mages, take a look at my spec. Also, if you plan on tanking Leo it is nice to have the fire talents to build treat really fast to maximize damage.
Fire damage won't help all that much with Leo if you're able to time doom to land on each demon phase (pretty easy to do).. You should basically spec based on what gear you can get (there is nice +Fire gear, but most of the nice gear is shadow early on).
Anyways, I'm at a point where almost all my gear is just +damage, so my options are open, but we raid with 3-4 spriests (4 these days) and 3 locks (shadow), so my choice is obvious.
Fire damage won't help all that much with Leo if you're able to time doom to land on each demon phase (pretty easy to do).. You should basically spec based on what gear you can get (there is nice +Fire gear, but most of the nice gear is shadow early on).
Anyways, I'm at a point where almost all my gear is just +damage, so my options are open, but we raid with 3-4 spriests (4 these days) and 3 locks (shadow), so my choice is obvious.
Maybe I was not clear.... I do not recommend getting +fire dmg gear. I do recommend getting the fire talents in the destro tree to help with tanking Leo. If you time it right, you can have the entire raid go all out within a couple of seconds of demon phase. If you time CoD, Incinerate, and conflag you will do ~10k of damage right when he changes forms. Also having the bonus damage for searing pain helps with holding aggro.
I ran with immo+corr+SB for a very long time, and only recently switched to just SB -- and my dps has noticably gone up (which was the opposite effect I expected.)
And it's really bugging me, because it makes me wonder, if there is indeed something going on (even if it's just a human element), then, say, +crit might be more valuable than we think.
Maybe it's something as simple as, most people, when they are JUST solely spamming one spell, do far far better with stopcasting macros, etc., and when they add in another spell, minute hesitations add up over the time of the spell.
I tried an experiment with my simulator that you may want to try with your spreadsheets.
I composed a supporting cast of Aff-Lock (CoS), Aff-Lock (CoE), Destro-Lock (ISB-help), FireMage, ShadowPriest.
I added to the raid another Destro-Lock (including Fire talents) and examined his performance in two different experiments:
(1) Include Immolate in rotation. Set lag to a changing random value between 0.25 and 0.35.
(2) Skip Immolate. Set lag to a changing random value between 0.15 and 0.25.
Essentially: Removing Immolate from the rotation and reduce lag by 0.1sec.
Your DPM goes down when using immolate, regardless of fire talents in destro. Perhaps in real world situations this, coupled with cataclysm, leads to enough extra lifetaps to offset the higher DPCT of Immolate. Couple that with the added difficulty of stopcasting around the global cooldown (much easier to fail to start casting exactly at the end of GCD than during regular bolt cast) and you can begin to see why pure bolt spam can come out on top.
The minimum requirements I used when I respec destro was the following:
1000 +dmg Self buffed
12% +Hit
20% +crit without talents
With these stats I did quite well. You should be able to get to these stats with kara, crafted, and rep gear.
As for your spec, if you have dedicated SPs in the raids then go with the shadow destro spec mentioned a couple posts ago. However, if you only have fire mages, take a look at my spec. Also, if you plan on tanking Leo it is nice to have the fire talents to build treat really fast to maximize damage.
I actually wouldn't advise destro honestly if you can't get hit capped or close to it. You lose too much uptime, damage and mana on fully resisted spells. You might be fine with >15% but at 12% I bet you were resisting alot and your actual crit rate according do the DM's were more like <20% with talents.
Even now with 99% to hit, I still get 1 or 2 resists every now and then and it kills my DPS.
(much easier to fail to start casting exactly at the end of GCD than during regular bolt cast) and you can begin to see why pure bolt spam can come out on top.
This is a big reason that no spreadsheet currently explains as to why Shadow Bolt spam beats SB + DoTs. With 1.5 cast times, once in a while you will stop cast too early or too late, but I have never stop cast Shadow Bolt too early.
Potential Destro users, spell hit is really important, since you lose more time than an Affliction lock loses by missing your spells.
I actually wouldn't advise destro honestly if you can't get hit capped or close to it. You lose too much uptime, damage and mana on fully resisted spells. You might be fine with >15% but at 12% I bet you were resisting alot and your actual crit rate according do the DM's were more like <20% with talents.
Even now with 99% to hit, I still get 1 or 2 resists every now and then and it kills my DPS.
I do agree that it is ideal to have hit maxed but I would not discourage a person if they wanted to go destro with the stats I mentioned above. I found that even with only 12-13% hit I was still able to perform well. And if you got that "lucky" crit streak you would make up for some of the resists.
I'm thinking Bolt spam is actually doing more than Felguard spec would(Anyone else have logs of 7/41/10 +3 v/s 0/21/40 with same gear/same fight?). I know I lose out on some bolts between ground slams, but I also lose time running around for Karathress.
Prince was a good one, no movement on my part, not sure how much further I could push the dps out of 0/21/40. No shaman present, no spriest, no boomkins. For prince I decided to swap in different gear. I had 40 less shadow damage than normal, 2% more crit, and .5% less hit. I dont know if this helps with the "value of crit" arguement at all.
The Gruul run I had a SP in my group.
I'm not sure when we are doing Gruul again, but I'll log it with 0/21/40 next time we go for comparison. I have more logs but I just haven't uploaded them. I don't know if it's the downfall of pet survivability or the greatness that is bolt spam, but I'm sold on 0/21/40.
I wish I had the gear to support back to back equal fire/shadow damage logs on the same boss, one using immolate/corruption +bolt and one with just bolt spam.
Well I'm trying out felguard now. Just respec'd yesterday and I've only done Kara since. VR/Mag on Sunday so that will be a more valuable test.
I did 0/44/17, taking DReach over Intensity. 1 point in Mana Feed, 3 in Demonic Resilience; other stuff pretty obvious. As this build is basically pet + nuke; lacking Intensity troubles me. I don't want to be without reach talents either, so I am thinking of dropping 2 points in Demonic Resilience and taking Intensity as well. I am losing a lot of DPS on extended cast times with any fight that has aoe style damage to interrupt my SB casts.
In terms of damage.. well there was plenty of that. Easily topped meters and on pet-friendly fights the 'fire and forget' nature of the felguard was much simpler than recasting dots. I felt I had a lot more mana overall and didn't miss Dark Pact as much as I thought. This does rely on heals of course.
Maiden was pretty much impossible. Not a pet-friendly fight at all. Next time I would just take imp duty.
Everything else went pretty well. I had him die once on Shade and once on Netherspite. Not surprising considering I was taking the blue beam. Shade, basically Aran decided to kill him and of course he got no heals. No big deal really; I just instant-summoned him again in both cases.
Maiden I wasn't surprised that he sucked; but Aran I thought he would do better. However between the Blizzard and his random blasts I had to micromanage the pet a bit more. Overall I think he did well though.
Prince was pretty easy when I realized he was immune to the shadow nova back (At least I never saw him do it). He died once when we had to move due to an Infernal and I didn't manage him quickly enough. Instant-summon; no big deal.
Nightbane took a bit of effort also, but was pretty easy once I figured it out.
I definitely see how the new T5 2piece and the void star talisman would help out a LOT.
In summary, this spec is a lot of fun if you don't mind managing your pet. The enhanced survivability from Soul Link was beneficial as well. On one Nightbane transition the tank was off and I ended up getting hit by him twice. Just barely survived; no chance if affliction.
I encourage other people to try out other specs. I was an affliction fanboi for far too long. While I still recognize the strengths of that build; there are certainly a lot of factors not easily measured by spreadsheets. I believe a big one with affliction is your ability to maintain a perfect dot rotation - apart from being boring, this is a lot more difficult in practice.
07/19/07, 5:05 PM
Similar Threads
