Kazanir

That sounds like it's "intended behavior" in that it makes sense based on the structure of the spell.

1) You channel a spell. While channeling, your target receives a debuff called Mind Flay. The debuff has a 3s duration, ticking once a second. But if you break the channel, the debuff goes away.

2) Spell haste affects only the channel, making you only be casting for 2.x seconds. The third tick of Mind Flay (from the DoT debuff) will never go off because of this.

Sounds like this needs to be fixed.

'War' is too small a word for what I'm fighting. Like a candle in front of the whole burning Sun. I told you. This is bigger than a war. Now, I am not going to die today. I have other projects, and other options.

You can come with me. I can protect you.

 

Metadeth

edit : nevermind, was wrong -_-

 

◊ Caffeine

I didn't verify the exact timings of the ticks but at least the debuff text goes from 3 seconds to 2.95 seconds with the Warlock spell Drain Soul and 32 passive haste (channeling time is 14.77 instead of 15 seconds), so it seems this is taken care of.

 

gargosch

Edit: Sorry I had calcalated with wrong nummbers for Improved CoA bonus. Noticed right after posting. With the correct nummber it is not possible to have enough bonus spell damge that you can reach the point where CoA is better than CoD om a singel target fight longer than 1min.



Math to prove it
------------------

Source for damage multipicator Dr. Damage addon. Calculation for hit cap or no points in suppression. Calculation with 10% crit chance and without consider Improved Shadow Bolt uptime. For higher crit rate you need ofc more Spellbonus damage. Lag also incise x.

CoD + SB need 96 mana more per min than just keep CoA up. The time to LT/DP is not included in the calculation.

Terms
--------

Base Damage = BD
Spell Bonus Multipicator = M
Cast Time = t
x = the amount Bonus Spelldamage you need so that CoA do more damage then CoD

CoA (with Improved CoA)
-----------------------------

BD = 1356 * 1.1 => BD/min = 3729
M = 1.2 *1.1=> M/min = 3.3
t = 3.75s
Mana/min =662.5


CoA (with Contagion and Improved CoA)
------------------------

BD = 1356 * 1.1 * 1.05 =>BD/min = 3915
M = 1.2 * 1.1 * 1.05 => M = 3.47
t = 3.75s
Mana/min =662.5

CoD
-----

BD = BD/min = 4200
M = M/min = 2
t = 1.5s => time to cast SB 2.25s
Mana/min =380

SB
-----

BD = 576 * 1.05(crit/2) => BD/2.25s = 544
M = 1.057 * 1.05 => M/2.25s =0.999
Mana/min = 378

CoD+SB
----------
BD/min = 4744
M = 3
Mana/min =758

Now we can see that the Base Damage is higher on CoD+SB but it has the lower Multipicator. This means there is a point where CoA is better than CoD+SB.

3.3x + 3729 > 3x + 4718

0.3x > 989

x > 3297

If you have Contagion you need less Bonus spell damge but still more than you can get atm.

3.47x + 3915 > 3x +4718

0.47x > 803

x > 1867

If you don't have Improved CoA than its match easier:

CoA (without Improved CoA and Contagion)
-----

BD/min = 3390
M = 3

Lesser Basedamge but the same Multiplicator => CoD is better than CoA independent on bonus spelldamage

 

Rach3l

Can we please eliminate the controversy surrounding Bloodlust and the GCD? After many raids, many 5 mans, and many weeks stopcasting scorches in a raid setting, bloodlust definitely does NOT lower the global cooldown.

If you don't stopcast then you cannot test this! I read someone posting that, "Oh well, it seems that the GCD is exactly the same but we didn't use stopcast macros--"

STOP. When I use /stopcasting with scorches and I am bloodlusted, I always cast "too fast" for the GCD (by a fraction of a second, to be sure, it's not a lot of time but for a seasoned raider it is DEFINITELY MORE than noticeable!!).

Bloodlust does not decrease the global cooldown for party members.

http://armory.mmo-champion.com.nyud....30871BLiqS.png

 

Kirion

Except bloodlust does lower GCD. Bring some proof before posting.

42.

 

Rach3l

Where's the proof that it does lower the GCD? You haven't presented any yourself and as such, your evidence is empirical at best (and fallacious at worst). I sit and chain /stopcast scorch 20 times in a row without a single self-interrupt--using scorch to time the latency interrupts--then get bloodlusted, and all of a sudden my stopcasts (although executed at the same time relative to latency according to castbar distance on the red latency bar of Quartz) are going off too quickly and scorch is not ready yet.

The original post in this thread said that bloodlust's lowering of the GCD was still controversial, so clearly there is no consensus on the matter in your favor. In my mind there is no controversy because, in my experience, it simply has not happened. Can you provide evidence to the contrary of my own personal experience?

Also allow me to say that bloodlust does not lower MY GCD in a party. I never said anything about it lowering yours. You're a shaman and bloodlust may lower your GCD but it does not do so for party members. Or maybe it just doesn't do so for level 70 mages named Roger on Maelstrom.

If you want to argue this point you have to do more than simply blindly contradict what I'm saying.

 

Kirion

This thread is about accepted and prooved theories in this community, if you bothered to search this forums, you would know this. Bloodlust lowering gcd was first prooved in hunters thread by chaincasting steady shots with ~1.19 casting speed. And thanks god we have thousands of WWS parses as proof.
You came here with empirical evidence, with no combat logs and we don't know you latency, and your last posts went to dung heap so we for sure cant trust you. Yet you claiming you are right.

42.

 

Rach3l

Searching for "bloodlust" doesn't bring up any posts that appear to discuss its effect on the GCD (http://elitistjerks.com/search.php?searchid=407722). Now that you mentioned it being tested by hunters, the hunter thread in that search offhandedly mentioned that BL lowers GCD. I'd be curious, though, to know if quartz is reliable when bloodlust is applied to 1.5s cast times because this conclusion is not at all comparable with what I've found to occur in-game. There is definitely a mismatch between quartz's allotted latency space in the castbar and whether or not the spell actually goes off when immediately followed by another /stopcast--but only with scorch, and only while bloodlusted.

 

Kirion

Quartz are not ideal, but until blizzard give us something to get rid of /stopcasting its best that we have. With around 150 latency i can barely cast lesser healing waves with less than 1.5 cast time. Also to see gcd lowering effect you need combat log with milisecond time stamps.

42.

 

Rach3l

My situation is: about 60-100 ms latency on average. In situations where I chain cast scorch it's generally only 5 times in a row for maximum debuffage. But what I've noticed is that, when bloodlusted, I press my scorch macro as soon as the cast bar gets to quartz's little red latency padding. The scorch I just cast will go off, it always goes off, but the macro doesn't cast the next scorch and I get a "not ready yet" message. It's like I pressed the button a microsecond too early, but the scorch *still goes off* *without casting the next one* so I have to hit it again to cast the next scorch. I mean, the point of hitting a stopcasting macro is to stop casting the previous scorch and start the next one. If my previous scorch went off but the next one isn't ready yet, the only logical explanation for that is that the spell is shorter than the GCD. And the GCD is too long and is inhibiting me. And given that scorch's normal cast time is exactly the length of the GCD, this conclusion really is natural. If someone could provide an alternative reason for this occurrence I'd be more than happy to hear it. I get irritated when I have to apply 5x scorches while bloodlusted because I know it's going to feel clunky and I cannot trust quartz for the next 7.5 seconds.

I would like to test this out later with Dr. Boom, fraps, and my friendly neighborhood shaman, so I can look over the results from a third person perspective.

Oh yes and the macro I use is simple:

#showtooltip scorch
/stopcasting
/cast scorch

 

Touf

Try this:

40 second test periods.
No stopcasting.
Let's assume you have 100ms latency.

1st period: Jam the button, see how many scorches you get off.
2nd period: Get bloodlust, jam the button, see how many scorches you get off.

You should repeat once or twice, but there should be a noticeable difference between the two.

It doesn't solve your stopcasting problem, but to be honest I have no idea what's wrong. It could be a problem with the client GCD check screwing up.

 

Roywyn

Try using Quartz, and make sure you have the GCD ticker enabled, use /stopcating macros.

Tried on the PTR with the premade shaman and rank 1, 2, 3 healing waves (-0.5s specced).
Looks exactly like there is no GCD reduction to me. Which is what I also experienced in raid (cannot spam my 1.5s cast time spell during Heroism, the GCD takes significantly longer to complete than the 1.15s hasted cast time).
Link is http://elitistjerks.com/462426-post79.html - it's not going to convince you of something you don't want to believe. But it's a guide to test and see it for yourself (just copy a shaman to the PTR).

Edit: Test idea - use a 1.0s cast time spell without stopcasting. Under bloodlust, they should be timeable even without stopcasting.

 

songster

This has come up in the Benefactor's Bar. People (like hunters) who test stuff without needing /stopcasting macros do seem to notice a lowering of the GCD. People who test using /stopcasting macros don't.

There's been a suggestion that there may be a hardcoded 1.5s limit on the /stopcasting function itself.

So really, what we need to do it test it both ways - once with scorch and /stopcasting, and again with arcane explosion (or some other spammable instant). And for the icing on the cake, make up a /stopcasting macro for arcane explosion and see if it alters anything.

 

Rach3l

haha, amazing. Such a simple solution that I wouldn't have thought of myself. However, would putting a /stopcast before an instant spell actually make any difference in macro implementation?

Glad to know I wasn't just smoking something crazy last night

 

Kirion

Really interesting suggestion. Too bad i have crap latency at this moment, and cant help you with testing =/

42.

 

Nite_Moogle

The problem with this logic is that speeding up cast times using /stopcasting is an unintentional mechanic. We don't know exactly how /stopcasting works other than it'll cancel your current spell -- if you trick the server by doing it, that's great, but we really have no idea what else it does. As mentioned it may also forcibly cause your next action to take place at least 1.5 seconds later than the previous action.
Hopefully once they introduce the replacement for the current system to get rid of /stopcasting we'll be able to determine a definitive answer.

 

◊ Shalas

/run hooksecurefunc("CooldownFrame_SetTimer", function(_, start, duration, enable) if start > 0 and enable > 0 then DEFAULT_CHAT_FRAME:AddMessage(duration) end end)

C&P that, cast bloodlust, then cast anything without a cooldown. If it spams "1.5", the UI does not think Bloodlust effects the GCD. If it spams anything else, the UI does.

 

Roywyn

PTR 2.2, alliance shaman (Heroism), spamming talented Healing Wave rank 1 (1.0s cast time).
No stopcating used, cast staright from the spellbook.

The string above spams 1.5 normally and 1.5 under heroism.
(And it spammed a 600, the 600s for heroism I guess.)

Did it again with a log: Wow Web Stats
Cast 21 HWs, popped heroism, cast 22 HWs, then heroism ran out.
Got a handful of "spell not ready" messages in between.

 

Nite_Moogle

Good enough for me. Thanks Shalas.
My apologies for not getting a definitive answer sooner.

 

Baruk

So if I understand the results of the /run script and the wws parse correctly,
neither bloodlust nor heroism lowers the GCD? Or is it really just connected to stopcasting?

Per Aspera ad Astra

 

• Kyt

Got a few of my shaman friends together this morning and got about 5 minutes of bloodlusted testing in with this spamming steady shot with no macro. Plenty of spell not ready messages and all 1.5s. I think we can conclude that Lust/Heroism does not affect the GCD.

[10:42] <BrTarolg> trying shahraz wearing lvl 60 blues would be like, fucking a hot girl but with aids and the wrong kind of condom on.

 

Rach3l

Halle-fuckin-lujah.

Thanks for testing, I can't copy any more characters to the PTR so I couldn't do testing of my own with a premade shaman ._.

 

Leveret

Is it still an agreed-upon fact that spell penetration does nothing in PvE? The Ghastly Haunts in Karazhan, for example, have very high shadow resistance and will partially resist most shadow school spells until it is afflicted with CoS, at which point the partial resists will for the most part go away. Doesn't this mob indicate that penetration is useful in PvE?

 

Apate

Paraphrasing Slick Willy: It all depends on what you mean by useful.


One could have a spell-pen set and swap out to it for high-resistance trash. To date, no good list of boss resistance has been made from tested data (that I know of). Offhand, I can't think of a boss where you'd benefit greatly from a spell pen set, but I'd never discourage someone from investigating.

See you, auntie.