sorry i've probably missed something but with the new casting mechanics of 2.3 why is timing required at all? can't you just mash the spell button?

I would still like some response back on the people who have the simulators on the difference in dps between the gear set I posted with 200 haste vs the one with 100 more dmg/crit. Including raid buffs, LO, 2 and 4 piece T6 and if possible the TLC. Daidalos said he'd do it a number of days ago, but he's not got back to me yet. All we have by looking through this entire thread is some conjecture on haste but no actual theorycrafting, at least none that has been posted here. Why it hasn't been posted by people with the sims I'm not sure, but I think that is what everyone wants to know, especially since there is quite a bit of haste gear in both T6 and on our badges gear. We have people here with a haste setup, people with pure damage, but we lack any solid theorycrafting. I can work out some numbers by hand, but it would be far more accurate for the people with the sims to show it, especially if they can factor in TLC. Thanks.

Ya was busy with work before getting ready to take off for the holidays. Maybe I'll get some time on my laptop in the next day or two.

Thank you. That would be much appreciated. If the person who posted the info on his sim can post his sim that would be great too. I mean if you look at some of the shamans in this thread you can see quite a variety of gear setups. We have some with pure dmg/crit gear everywhere. We have some with things like the Illidan staff and roughly 200 spell haste rating. But we have absolutely no idea which setup is better, haste with LB spam or dmg/crit with CL/LB.

Just wanted to say that I agree with Daidalos on this. Spell haste averages out as the same increase in DPS as seen by theorycrafting in the long run, depending on at which time you need to move.


Interesting. With the 4-piece T6, anyone know where the break point is that one LB does more damage than one CL?Does it come close to doing more DPS? The spell damage value used to be very high and unreachable, but with the set bonus this might change? Haven't done any math on it myself, just wondering.


I agree that DPS is very jumpy depending on which fight, but I have noticed that after I started to get more T6 gear (found very little usefull gear in T5 instances) I have started to climb DPS again. I have started to see myself on top 3 basically all of our Illidan tries in Phase 1 and 2, which is just tank and spank. Of course we get tons after in any fight where there is any AoE involved. Same situation was when we were doing T4 content (always on top with Shadow Priests). But overall I would say around place 3-7.

This I have seen as a very large factor, since I used to place higher on Supremus than other similar fights.


Have been thinking about this one as well. Currently I have about 8-9% haste, and even BL I don't notice getting a bunch more "Spell not ready" messages when spamming LHW, comparing to spamming it without. I know theres a script you can run, but maybe it is bugged, who knows?

Hmm, did a few other tests and they conviced me that BL doesn't affect haste.

1. Counting number of LHW casts I could get off with BL in 40 sec. Versus the amount I could get off without BL. Numbers turned up to be about the same.

2. Watching the GCD bar versus the cast bar in Quartz with and without BL. Without BL they finished about the same time. With BL the cast bar finished way before the GCD bar.

PS. I know I might be repeating some of the stuff said before, but I wanted to give my opions on it, have been away fro m the forums for a while.

I dont know if you guys have talked about this before, but what would be higher DPS for a shaman [Quagmirran's Eye] or [The Lightning Capacitor]

Over the last two weeks I have been checking out how often each procs and what the dmg would be in comparison. My theory is that the flat dmg of the Capacitor(which does not get bonus from + to spelldmg) does not out-weigh the 320 haste on a normal LB, which is effected by + to spelldmg. And on top of that, they are guarenteed to cast, you dont have to get the 3 charges, which u can only get 2.5s of the time on each crit. and on top of the haste from Quag's, you have a 40% chance each time to have it OL, which does not apply on the Capacitor.

I didnt know if there was anything else out there that would support this theory.

Any input on this question is greatly appreciated.

Thanks!

Yea it really depends on the conditions of the fight. Out of all our Illidan kills I've only been beat on the DM twice, 1 warlock 1 spriest (at the begining it was mostly due to my outgearing the avg dps, now Im staying competitive with haste). I sustained like 2300dps on Akama the other night, first time Ive beat the rogues there. But I find Im competitive with locks/rogues/token mage on most fights but like stated above, always top 5 rarely #1.
I find it really depends on my god damn trinket procs and how quickly I can get my sextant up, some fights it refuses to proc until a good minute into the fight.

This is wrong. The 2.5s cooldown only kicks in AFTER you release the charge. You can instantly charge it with EM CL, or you can charge it very quickly with LB + LO crits, or just back to back crits. It takes 2.5 seconds after the trinket fires a bolt before it can begin getting charges again.

Edit: I can't answer the question asked, because I've never seen quag's eye drop.

my question was which was better? although i thought it was 2.5s everytime u get a charge, so thats new to me. But does Quag's Eye trump the capacitor?

I want to know if anyone has tried it out. So far, quag's eye is putting out more DPS for me, but I dont know if Its because I am just getting lucky procs off it or what.

Any chance someone else has tried?

I'd go with TLC over Quag's... but that's just me. I haven't used Quag's, but it's not like I've done any math on it or anything

I opted against Quag's due to the fact that its usefulness is mitigated by Heroism, and the proc is worthless on the CL part of our spell rotation. I'd have to dig up the math I came to, but given worst case scenarios with TLC, and best case with Quag's eye, they just barely came close to breaking even, and neither situation is likely. All that said, you could do much worse if you don't have a TLC lying around.

For Pure Damage, not DPS:
Equation:
(((599 + 684) / 2) + (X *0.794)) * 1.05 * 1.05 = ((770 + 880) / 2) + (X * 0.641)) * 1.05

Simplify:
707.25375 + X * 0.875385 = 866.25 + X * 0.67305

And Solve for X:
X(0.875385 - 0.67305) = 158.99625
X(0.202335) = 158.99625
X = 785.807

For DPS the whole thing is:
((((599 + 684) / 2) + (X *0.794)) * 1.05 * 1.05)/2 = (((770 + 880) / 2) + (X * 0.641)) * 1.05)/1.5
(707.25375 + X * 0.875385)/2 = (866.25 + X * 0.67305)/1.5
353.626875 + X * 0.4376925 = 577.5 + X * 0.4487
X = 20338.25

Obviously not counting crits and hits, LB cannot equal CL's DPS without haste rating.

Sorry took so long but I was busy with work then was gone for the holidays.
Here are some numbers that Kasi requested.
Note: I just ran these in the sim. I make no gaurentees about the correctness of my sim. Bugs or incorrect theorycraft would mean the results are wrong. Please don't assume it's correct just because its a sim. I don't know of any current issues with the sim, but that doesn't mean there isn't any.
Also note even with 500 hours or so there will be some variance due to randomness. So small varations should be treated as basically equal. I would typically see variations of about -/+ .1%

LB spam with skycall:
Enter total hit percent (capped at 99) 99
Enter crit percent 36.76
Enter total spell dmg for Lightning Bolt without relic 1671
Choose relic: 1)skycall totem 2)Totem of the Void 3)Totem of Ancestral Guidance 4) none
1
Enter total spell haste percent 0
Enter spell rotation: 1) LB spam 2) 3 LB / 1 CL 3) 4 LB / 1 CL
1
Enter number of simulated hours 500
total dmg = 2821182825
total LBs = 932038
total LB LOs = 182315
total CLs = 0
total CL LOs = 0
total num of skycall procs 140351
Percent chance for skycall proc 15.058506198245137
total casts with skycall buff 582504
Percent of LB casts with skycall buff 62.49788098768505
average dps = 1567

LB spam with Ancestral:
Enter total hit percent (capped at 99)99
Enter crit percent 36.76
Enter total spell dmg for Lightning Bolt without relic 1671
Choose relic: 1)skycall totem 2)Totem of the Void 3)Totem of Ancestral Guidance 4) none
3
Enter total spell haste percent 0
Enter spell rotation: 1) LB spam 2) 3 LB / 1 CL 3) 4 LB / 1 CL
1
Enter number of simulated hours 500
total dmg = 2820590019
total LBs = 900000
total LB LOs = 175926
total CLs = 0
total CL LOs = 0
total num of skycall procs 0
Percent chance for skycall proc 0.0
total casts with skycall buff 0
Percent of LB casts with skycall buff 0.0
average dps = 1566


3LB 1 CL with Ancestral:
Enter total hit percent (capped at 99)99
Enter crit percent 36.76
Enter total spell dmg for Lightning Bolt without relic 1671
Choose relic: 1)skycall totem 2)Totem of the Void 3)Totem of Ancestral Guidance 4) none
3
Enter total spell haste percent 0
Enter spell rotation: 1) LB spam 2) 3 LB / 1 CL 3) 4 LB / 1 CL
2
Enter number of simulated hours 500
total dmg = 2971708069
total LBs = 720000
total LB LOs = 140700
total CLs = 237565
total CL LOs = 47029
total num of skycall procs 0
Percent chance for skycall proc 0.0
total casts with skycall buff 0
Percent of LB casts with skycall buff 0.0
average dps = 1650

Hasted LB with skycall:
Enter total hit percent (capped at 99)99
Enter crit percent 32.235
Enter total spell dmg for Lightning Bolt without relic 1576
Choose relic: 1)skycall totem 2)Totem of the Void 3)Totem of Ancestral Guidance 4) none
1
Enter total spell haste percent 13.26
Enter spell rotation: 1) LB spam 2) 3 LB / 1 CL 3) 4 LB / 1 CL
1
Enter number of simulated hours 500
total dmg = 2964875450
total LBs = 1053926
total LB LOs = 206456
total CLs = 0
total CL LOs = 0
total num of skycall procs 157468
Percent chance for skycall proc 14.941086945383262
total casts with skycall buff 654480
Percent of LB casts with skycall buff 62.09923656879136
average dps = 1647

Hasted LB with Ancestral:
Enter total hit percent (capped at 99)99
Enter crit percent 32.235
Enter total spell dmg for Lightning Bolt without relic 1576
Choose relic: 1)skycall totem 2)Totem of the Void 3)Totem of Ancestral Guidance 4) none
3
Enter total spell haste percent 13.26
Enter spell rotation: 1) LB spam 2) 3 LB / 1 CL 3) 4 LB / 1 CL
1
Enter number of simulated hours 500
total dmg = 2963909742
total LBs = 1016950
total LB LOs = 198469
total CLs = 0
total CL LOs = 0
total num of skycall procs 0
Percent chance for skycall proc 0.0
total casts with skycall buff 0
Percent of LB casts with skycall buff 0.0
average dps = 1646


Kasi def picked an interesting set of gear / rotations that all ended up being very close in dmg. I double checked the skycall vs ancestral a few times being so susiciously close to each other. I don't see anything wrong with my program after a quick once over (I have recently rewritten some of it so bugs are still rather likely). I guess that is about the point where the haste and dmg values for those totems meet. I'll do some deeper digging on it when I get the time to doublely make sure there isn't a bug.

If these are correct it certianly looks like a hasted LB spam with skycall is an easy way to achieve excellent results without the draw back of CL.

I good follow up would be to do a mana comparison of these I'll do that later today or tomorrow. I'm sure someone else could also to the mana consumption based on these numbers as well.

These results somewhat match my simulator. Inputs I used match those what you posted above (save I approximated 208 Haste Rating for 13.26% Haste since mine works on haste rating not a percent)
Test Set 1 = http://img220.imageshack.us/img220/1218/simrun1lo6.png Test Set 1 ||| 1611 DPS (+34 DPS)
Test Set 2 = http://img220.imageshack.us/img220/2209/simrun2sj9.png ||| 1564 DPS (-2 DPS)
Test Set 3 = http://img229.imageshack.us/img229/4/simrun3wm8.png ||| 1624 DPS (-26 DPS)
Test Set 4 = http://img231.imageshack.us/img231/9978/simrun4nv4.png ||| 1676 DPS (+29 DPS)
Test Set 5 = http://img229.imageshack.us/img229/7374/simrun5fa2.png ||| 1649 DPS (+3 DPS)

The only thing I notice that really separates our simulators is skycall modeling. My values when using Skycall compared to your values are more different, so either me or you might be missing something there.

Then again, my simulator takes a different approach than yours. I model DPS based on a 10 minute fight length and average together the results of 100 Simulations of those 10 Minute fights.

Yeah I considered doing that and making it multithreaded since all my computers are multicore but I wanted to really make sure I had a good flexible framework before worrying about speed. Also I was worried about accuracy if by splitting into too many segments I'd be cutting off buff times and procs etc.

Good info though I really need to just sit down and make sure everything works properly in my sim sometime after I lock down more of my code, but between working and actually playing wow I never seem to get around to it.

One thing that surpsied me was with ~200 haste rating the skycall hasted cast % did not really increase. I would have thought that given the fixed amount of time for the buff the % of skycall hasted casts would increase. Do you have any data from your sim on this?

With those stats the damage is a bit higher than is actually available in game, because those numbers you used did include the 85 dmg totem. However its a pretty good representation of what it will be like come sunwell, so that's pretty close. Regardless its only off by a little bit of dmg, and one can say a truely min maxed caster would be using things like drums, which would make up that difference, not to mention destruction pots. So that probably is roughly approximate to a true top of the line raid dmg perspective. I do want to double check that both of your sims do include the 4 piece T6 correct? Also for 2 piece T6 I included those stats in the actual dmg/crit numbers I formulated, so hopefully those weren't double counted.

As for the results themselves, I'm impressed. It does show that with enough spell haste spamming LB would be a good idea. Of course some fights CL is very nice, but a pure LB spam would be imo more mana efficient. So it does show that stacking haste will be good in the long run. Now of course the question for me is how to get all that gear, especially the staff. :p

Now one question here. At what number of haste rating will bloodlust push LB below 1.5 seconds? Just wondering if it would be good in such a LB spam rotation to use the 85 dmg totem when under bloodlust but skycall at other times, if Tej's data of skycall being better is true.

It's possible Tej's sim has that and the difference between your two in numbers is the 4 piece bonus.

My simulator is not currently modeling 4pc T6, although it can be modified to do so really easily.

I actually have more data being computed from the simulator than is being displayed on screen (such as Skycall uptime, number of TLC, LO procs, ability to change rotations at will to something like 7 / 1 if you wanted, etc). I can post a full screenshot with a bunch of data tomorrow. I dont know when I will get to it though, as the solution is on my work machine and I have some other things to do.

The only thing is that results can vary from simulation to simulation because of the random seeding. If I run that same test again, I'll likely get a different average DPS number. Fortunately, they dont seem to fluctuate more than 20 DPS in either direction.

Well I did some 'static' tests with the two trinkets back then on Dr. Boom. I don't have the exact numbers anymore, but Quags was slightly better in terms of DPS (did like 5 tests, Quags always won, could have been luck with crits of course, etc.). But don't forget that you can keep your TLC charges while moving, and they don't cost any mana while still increase your DPS, while Quags burns your mana a bit more, and if it procs at the 'wrong' time, the buff is wasted, and the internal CD is more noticable than TLC's. Plus, if you are using Skycall, and both proc, and you don't notice it for some reason, you can get stuck into GCD if doing 3/1 rotation and stand there surprised -> more attention is required.

Basically TLC is better on bosses like Leotheras and of course on trash, while I usually use Quags on Morogrim, Lurker, etc.

Oh, I have around 25% unbuffed crit (can't remember exactly, being PvP ele for a while now) which might effect my opinion of TLC.

Ok, added some new features to my simulator: 4pc T6 modeling, Quagmirran's Eye, and better reporting.

A new screenshot: http://img88.imageshack.us/img88/6849/simrun6ba4.png

Let me know some test data that you want me to run through, and I should be able to give you back some information about things.

Also as a note: right now the "Average # of casts" includes all casts done by the shaman (LBs, CLs, TLCs, LOs). If you want more granularity, I can get that too, but I just dont have it printing out to the screen.

Well since you asked I'll repost my data.

Set 1:

Damage: 1429 without both trinkets. Could we use TLC and Hex Shrunken Head as the two trinkets? HSH should be 88 passive damage which would bring 1429 up to 1517, and TLC well I'm not sure. This 1429 value does not include 85 dmg totem as well.
Crit: 503
Hit: 153
Haste: 0

Set 2:

Damage: (1334 without both trinkets, 1422 including 88 from HSH. Also does not include dmg or haste from totems equip slot.)
Crit: 403
Hit: 134
Haste: 209

So same thing as before.
Set 1 : LB spam with Ancestral Totem
Set 1: LB spam with Skycall totem
Set 1: LBx3, CL with Ancestral Totem
Set 2: LB spam with Ancestral Totem
Set 2: LB spam with Skycall Totem

Thanks.

A breakdown of LB/CL casts and their LO procs would be nice, so we can see the average hit/crit of each.