I'd like an accurate equation on haste or an accurate graph or some confirmation from SOMEONE in the form of an equation. I've heard three different exactly opposite things now, which is incredibly frustrating when one is trying to figure out how to rate something. Two people are blowing smoke and one person is right.
The three things that people are saying about haste.
Most recently we see that Haste is a linear spell. Which makes a simple sort of sense to me, but still was not supported by ANY sort of concrete evidence
Before that we say Haste as a sort of exponential thing, where it only gets capped at the GCD. (where the graph stopped) if there was an equation you used to form this graph I would love to see it, as it stands you could have just been drawing lines though?
And even before that, when haste was still new there was a post about how haste scales down as you increase it, this one even cited simple math. 100% to 101% is a greater increase than 101% to 102%. So the more hit you have the less stacking it will help.
Can someone please clear out cobwebs and form some sort of equation or make an enlightened post that doesn't feel like so many people talking at once.
There are two definitions of benefit that must be made clear. The first is absolute benefit--how much real DPS is added. The second is relative benefit--what percent DPS is added.
The use of a consistent DPS formula helps determine both. Roywyn posted one formula; I post another representation of that formula that is exactly the same save for notation.
D = (1+z)*h*q*(m+r*d)*(1+b*c)/t
D = DPS
z = haste (as a decimal)
h = hit chance
q = damage multipliers
m = average base damage
r = +damage coefficient
d = +damage
b = crit bonus
c = crit chance
t = ordinary casting time (pre-haste)
Again, this is exactly the same as Roywyn's formula; the only substantial benefit is that it makes pencil-and-paper algebra easy (if that). This is merely what I'm accustomed to using.
In terms of absolute benefit, which we will call ∆D, adding ∆z haste (which, again, is a decimal--if you added x haste rating, ∆z = x/1576, or rather x divided by the rating constant) yields...
∆D = h*q*(m+r*d)*(1+b*c)/t*∆z
Thus, in terms of absolute benefit, haste has no diminishing returns. It is absolutely linear--the next +∆z haste rating is as effective as the last, as long as other variables are constant.
In terms of relative benefit, which is ∆D/D, we have...
∆D/D = ∆z/(1+z)
Which, in terms of relative benefit, shows that haste's value diminishes as you stack it, again for the simple reason that 101% to 102% is less than 100% to 101%. Note that relative benefit depends on none of the other variables.
I already showed how to figure out the effect of latency on haste and how small that effect is, even though it's kinda moot come 2.3.
Note that the increasing effects on haste are hardly practical, as at any reasonable range of haste rating your "cuved" line is extremely close to linear (you can do the math to compare the first part of the graph to a Y=aX+B and see for yourself that you will probably get "R^2" very close to 1). The reason I assume haste increases its effect is that you end up casting more fireballs between scorches and thus adding more haste affects more of your DPS (as your fireball:scorch ratio increases). This is an extremely minimal to even non-existant (depending how your rotation works and how early if at all you refresh scorch) effect at any realistic haste value, though. Then again maybe I'm missing something else that causes haste to scale non-linearly, but either way it's nearly linear at the "realistic values" area.
Since the patch, i've a problem with Ice block. I'm not able anymore to cancel it before its full duration, either by reclicking my macro (/stopcasting /cast IB), or even by right clicking the buff.
Is it something i've missed in the patch note, or do you have an idea of what to do?
Maybe slightly off-topic but does anyone have any real justification for giving a ele shaman to mages over destro-locks? Recently I've just given it to the destrolocks though what I'd WANT to do most likely is give the top 4 casters the ele shaman which is actually split 2mages and 2locks. I find buffing the people that do the best, the best increase.
I find buffing the people that do the best, the best increase.
Ya bloodlusts really goes furthest when put on the right DPS'ers. The wrath totem actually should scale fairly equally amongst anyone who has the skill to use it.
Pressing once will set off iceblock and set-off a GCD which will in turn prevent it from processing the second command (and return an error message: ignore this) After the GCD is done, you may press it again, where since Ice Block will be on CD it'll go ahead and cancel the aura command. This will also return an error for failing to cast ice block. Again this is normal.
Righ-clicking macros is now slightly strange, though I suspect as usual you still can right-click off a buff, you just must make sure you press the button while hovering over it, and release the button without moving from it. Most of the time that's my problem with clicking-off buffs, in my haste I slightly move the mouse and the release of the button is off-target.
Dustwhisper: While both BL and Totems benefit both classes, you ignore two facts: If a destro lock does not have the pleasure of a shadow priest, he will be life-tapping. Each life tap is wasted Bloodlust time, ToW time and WoA time. (although WoA increases the life tap size, it doesn't contribute to dps as much). Given a lock will do any of the above, they are losing value from all these buffs. You might argue that scorch refreshing will not gain from Lust just like LT won't, but one scorch per 28.5sec is far far less waste than 2-3 lifetaps per 5 shadowbolts. And even if they tapped as often as we scorch only, scorch still benefits from ToW and WoA better. Lastly, given that locks consume mana at a much increased rate, mana spring is of less value to them: 4 mage 1 ele sham is a sustainable team, which in certain cases might even be able to destro-pot during the last 20% rather than chain-pot. subbing in a lock without subbing in a Spriest seems to me to be a lost cause: He will waste so much of his time tapping that any shaman benefits will be marginalised.
Though it is true that tapping will deteriorate the net-gain of bloodlust I would think a warlock could sustain straight out shadowbolt nuking while under bloodlust, atleast by using one manapotion and having lifetapped to full pre-bl. Though the question then remains if the lifetap done to get full mana pre BL gives a net-gain or a net-loss. I though noticed that our best destrolock just plain shadowbolting outdps all our mages except one on certain fights.
I'm just finding it rather hard to justify the ele to mages over locks seeing as a restoshaman has better manaspring which benifits mages highly, atleast me with manaproblems.
Dust, unless you're feeding locks they can't spam bolt alone. Plain SB spam will surpass mages, we know that already. And the more you have the better it gets (Imp SB up time). But even if they can sustain the duration of BL worth of pure spam they will waste resources and their own DPS if they aren't fed by a SP during the whole encounter. One bloodlust is one thing; 10m boss fight is another.
Reposting this here, thanks to those that directed me here
Four bosses in BT
This is just the report for me, from last night. When analyzing my performance, I see that my DPS, while acceptable, still lacks in comparison to the other casters, specifically the Warlocks and Mages. I know part of this is that my gear not up to par with most of the other casters, but I also cant help but think that i'm failing in some other aspect.
For the entire night, my miss % came up to 4.3%, and add that into a mitigated % of 6.6%, roughly 10% of my fireballs were not doing their full damage. On boss fights i run at about 150-160 +hit, on trash its about 130. I'm working on upgrading my gear, but it's been slow going so far. i stick with the usual spell rotation of 5 x scorch (im the only fire mage) and then fireball spam while keeping scorch up every 30 seconds or so.
Can anyone offer any advice? Is there something i'm missing. I hate feeling like i'm not holding my own weight.
Looking at your spec, you have 3 points in Imp FB, I wonder if having two of those points added to Incinerate would help. Getting additional crits off the Scorch's your applying to the target would add up, along with mana refunded from MoE, which add's to your uptime.
I am sure that someone has the actual details that would either back up/dismiss this thought.
11/47/3 deep fire spec. I just picked up the [Hex Shrunken Head] last night and discovered it stacks with the [Icon of the Silver Crescent]. With this in mind is it worth using the two trinkets together with combustion or am I better off replacing the [Icon of the Silver Crescent] with the [Darkmoon Card: Crusade]. The reason I ask is because Vontre's spreadsheet pegs the Darkmoon trinket as slightly better than the Icon.
To answer your question, both of them together are most definetly more effective than Crusade.
You'll have the 4 Combustion crits (3x Fireball + 1x Fireblast) do massive damage, and as both the trinkets have 2 minute CDs, they will be avalible for every third Combustion, in a fight a bit longer than six minutes this will boost your burst damage considerably, adding to the fact that these trinkets have no ramp up time and provide passive spelldamage.
Thanks for the input. I was thinking the exact same things. Plus I've been very irritated these past few months that the Crusade buff falls off frequently in many encounters.
I was very lucky to win the trinket last night. It wasn't the guilds first Malacrass kill, but it was my first and I rolled a 98. Even luckier considering there were 4 other casters in the raid that were after it.
After we finished the first thing I did was head into a battleground for some big PvP numbers. WHOO!
To answer your question, both of them together are most definetly more effective than Crusade.
After looking over Max's possible Item Combinations I would disagree, saying that he should keep the Crusade Trinket instead of the Icon. His base damage increase with the Hex and Icon is an overall constant (a)96+spell damage with a +366 damage burst every 2 minutes for 20 seconds.
However if he had the Crusade trinket aligned with Hex trinket he would have a base damage increase of (b)133+spell damage with a +211 damage burst every 2 minutes for 20 seconds. So the variable between the two comes with that of a 2 minutes time span (120 sec.) where he would have with option (b)vs.(a) a +37 base damage total for the 120 sec period span with a -155spell damage for a 20 second span.
If you look at the overall bonus of having that +37 dmg base in (b) in a ratio comparison to that of (a) +155 dmg burst for 20 sec in a 120 sec period, you would see that the 155~20sec divided over a 120 sec period is only equal to that of a +28.53 dmg base. Resulting in a stance where that +155 dmg burst in a 20 second period would have an overall lesser effect on his dmg output over a 120 sec period of time displaying only a +28.53 dmg vs. the Crusade which would provide a base +37 dmg base throughout the 120 sec. period.
Now this being stated with or without the variable of combustion, because if you triggered combustion at the same time with both options (a) and (b), and assuming you have the same variable yields for crits in that alotted time period, you would still be returned to the base equation of whether the burst or base line provides more dmg over the alotted 120 sec span, and you would still get the option (b) with the crusade trinket providing a larger overall DPS number.
So what I'm trying to say is that (b)Hex Shrunken Head+Darkmoon Card: Crusade>(a)Hex Shrunken Head+Icon of the Silver Crescent in terms of the 120 sec. span base overall DPS count. So for higher end overall DPS you should be using option (b) over option (a). However if your on trash mobs, and just want to see pretty big flashy numbers that hit hard and aren't concerned with overall DPS go with option (a) for the larger 20 sec burst dmg. Hope that helps a little.
(a) +155 dmg burst for 20 sec in a 120 sec period, you would see that the 155~20sec divided over a 120 sec period is only equal to that of a +28.53 dmg base.
This isn't entirely accurate, because fights are not a multiple of 120 second periods. A 140 second fight gives you 40 seconds of +155 damage, which would be a +44 damage base. Also, multiplicative effects like Skull of Gul'dans Haste and Bloodlust further increase the 'relative contribution' of the burst.
<Vontre> I removed the cooldown on evo
<sancus> and what happened?
<Vontre> DPS went down rofl
Unrelated to trinkets but related to mages. Went kaz'rogal for the first time last night and though we killed him in three attempts I could not survive long on any of them. Mage armor, manatide (no SP), medallion of karabor for the ocassional resist, manapots,managems, evocation. I even tried wanding myself up over the necessary mana and keep me inn a healthy zone but Inn the end i just decided on unleashing all the mana I had to atleast have spent my mana on DPS instead of blowing up :P How do you guys survive this fight if at all without SP?
Unrelated to trinkets but related to mages. Went kaz'rogal for the first time last night and though we killed him in three attempts I could not survive long on any of them. Mage armor, manatide (no SP), medallion of karabor for the ocassional resist, manapots,managems, evocation. I even tried wanding myself up over the necessary mana and keep me inn a healthy zone but Inn the end i just decided on unleashing all the mana I had to atleast have spent my mana on DPS instead of blowing up :P How do you guys survive this fight if at all without SP?
More SR (150-200ish range, buffed) can help quite a bit. Make sure to use pots / gems asap so you can get those timers up before the end of the fight. Reallly though don't think that going OOM means there's something you should be doing differently; if your raid isn't getting it done fast enough mana pools will suffer due to how fast the drain timer gets.
Yea I noticed. I really don't enjoy the gimping with SR but I guess I could try, especially as we just bought 80 hearts cross-server for our sharaz gear might aswell start crafting it... Ye i was using manapots/gems asap and evocation as well the second i dipped low enough.
I can't comment on the given boss as I've never been there, but given you're running high resistance, the question rears it's head: How often do you get full resists? If the answer is over 3-4 times per encounter, perhaps there's merrit in speccing at least 3 pts into Magic Absorbtion. Before you laugh, remember that each full resist replenishes 1,2...5% of your mana. Raid buffed, a 5/5 MA spec will give you upwards of 500 mana per resist. Ditching Incineration isn't that dramatic as far as I'm concerned (how many scorches will you do over a bossight? 30? That's one less scorch crit per boss-fight.) and that one stray point can go into giving you 300+ mana per resist.
Speccing into magic absorption for that one fight would be a huuugeeeeee waste of gold. If you resist you resist it fully and ye only with the neck i had like 3-4 resists in the fight but still just no need :P If I could fully resist 90% of them it'll never be a problem manawise anyways. But sure talent would help.
Actually I found that certain boss-fights scorch is better than fireball. Reliquary of Souls p2 for example, you'll never nuke yourself to death with scorch there (though I almost have :P). Scorching on manaintensive fights is usually a good option as well due to higher crit% on scorch for MOE returns and you don't spike your mana as much. Though a lot less dps and so forth.
Yea I noticed. I really don't enjoy the gimping with SR but I guess I could try, especially as we just bought 80 hearts cross-server for our sharaz gear might aswell start crafting it... Ye i was using manapots/gems asap and evocation as well the second i dipped low enough.
I liked to use evocation right before a mark would put me oom. The drain doesn't cause pushback, so you can evo during it. This would leave me with a near full mana bar after the Mark. Have not tried it with the new evo. as we kill him long before mana is an issue now. It should still give you slightly more mana than using before a mark though.
Another option would be letting yourself go oom and get away from group. You should survive the blast and can then evocate back up.
I liked to use evocation right before a mark would put me oom. The drain doesn't cause pushback, so you can evo during it. This would leave me with a near full mana bar after the Mark. Have not tried it with the new evo. as we kill him long before mana is an issue now. It should still give you slightly more mana than using before a mark though.
Another option would be letting yourself go oom and get away from group. You should survive the blast and can then evocate back up.
That's what I did, evoed at ~1500-2k mana or so. Or let myself blow up with some mana left so I didnt die then evoc. Still a bitch, this is where invis should friggin take you out of the fight so you could drink.
Guys, at the moment I'm arcane specced doing a rotation like: 3AB, 1AM, 1Scorch.
Now in 2.3 AM isn't as good as it used to be and fireball got buffed by ~7% so I was wondering if I should change to a rotation like: 3AB, 2Fireball or even 3AB, 2Fireball, 1Scorch
11/17/07, 4:53 PM
Even luckier considering there were 4 other casters in the raid that were after it.
