I was looking at your Armory and your hit seems really low for your spec. Is that a trash set you're wearing or are you insanely lucky and getting an elemental shaman every raid?
Also, I'm currently the only fire mage in my guild. We've downed Brutallus once now and usually run a heavy caster comp on raids (2 full caster groups). We have 3 regularly raiding warlocks (2 demo, 1 aff) and 2 mages. Our other mage is arcane/frost. Synergy classes generally include 1 boomkin, 2 shadowpriests, and sometimes an elemental shaman (but usually a resto shaman since our main ele shammy is currently on vacation).
Is it worthwhile for me to remain fire and try to do my best with maintaining scorch by myself? Some of the destro locks have been discussing going to a fire-heavy spec and my scorch debuff would be crucial to their dps for that. Our arc/frost mage loves his spec and pretty much won't change it unless the officers threaten his raiding career. I've found that being the only fire mage is kinda rugged on my theoretical max dps. I could go arcane/frost as well, but I have personal issues with downgrading my gear (I worked my butt off to get out of T5 just to start wearing it again?) and having a total dependency on the presence or absence of shadowpriest in my group on raids. I know fire is supposed to be the best raid dps spec but when you're not getting help on maintaining the stack it suffers quite a bit.
Muphrid I keep rereading your math above and it seems a little off to me, also there is no mention of haste or casting speed so haw can it calculate dps?
Here is what I came up with although it wont look as fancy, I also left the stats in their raw number values not calculated percentages, I only left hit as a percentage.
b = base spell dmg
d = bonus spell dmg
c = crit rating (i.e. 254)
h = hit chance
D = spell dmg modifier
C = crit dmg modifier
m = bonus dmg modifier
H = haste rating
s = cast time
(h((1-(c/2210))(D(md+b))+(c/2210)(C(D(md+b)))))/((1-(H/1575))s) = DPS for a given spell, remove the cast speed section to get average damage.
*edit* I am at work so I dont really feel like solving individual variables in terms of others but if anyone else feels like it go for it.
Thegoodman, I'm curious which frost talents you dropped for your 18/0/43. I wasn't going to guess, so I went with the standard 10/0/51, loaded my character in Rawr, added moonkin, retri pally and totem of wrath to my Brutallus set up for a total Frostbolt crit rate of 54%. With this all, AtoI does indeed become 6 dps better than Icon. However, with such high crit, TLC is preferred, and moreover: Hex, Sliver and Skull are still way ahead. So I'm still not really convinced that the trinket is 'terrific'.
Originally Posted by Roywyn
Whether a spell crits or not is determined on the end of the cast.
If you look at WWS for Nature's Grace you'll see the buff is gained some time after the spell lands. It's not sent by the server at the same time as the damage for the spell. So because the druid will have already started his next cast, I believe the shortened cast time must be applied retroactively. In any case it's not really the same mechanic as with AtoI since it's a reduced cast time buff rather than a spell haste buff.
If MSD was 'fixed' as you say, to affect the current spell being cast, then I'm very curious why AtoI doesn't act this way also. Have you tested this? Looking at wowhead, it's the same rating modification (524288) for both procs. So very curious that one would act differently to the other. Same with skycall totem in fact.
Muphrid I keep rereading your math above and it seems a little off to me, also there is no mention of haste or casting speed so haw can it calculate dps?
Here is what I came up with although it wont look as fancy, I also left the stats in their raw number values not calculated percentages, I only left hit as a percentage.
b = base spell dmg
d = bonus spell dmg
c = crit rating (i.e. 254)
h = hit chance
D = spell dmg modifier
C = crit dmg modifier
m = bonus dmg modifier
H = haste rating
s = cast time
(h((1-(c/2210))(D(md+b))+(c/2210)(C(D(md+b)))))/((1-(H/1575))s) = DPS for a given spell, remove the cast speed section to get average damage.
*edit* I am at work so I dont really feel like solving individual variables in terms of others but if anyone else feels like it go for it.
I typically call the ordinary casting time t (or t_0) and the haste (decimal) as z.
Thus...
In the post I was elaborating on, haste was not being varied, so I felt it would be easiest to stick to average damage/cast rather than introduce the casting time term (when it was just going to cancel anyway).
In my variables, your math looks like...
But you can easily factor this.
Which is only slightly different from my formula. You seem to have assumed that haste reduces casting time by a fixed percentage rather than increasing speed by a percentage. This explains the discrepancy in the two formulas.
I nearly edited my post to say this exact same thing. AToI will be an interesting trinket to hold on to for the upcoming expansion. Even Fire builds are going to be very crit heavy so this could have potential there as well.
I strongly disagree. Going 70-80 will cost you severely on stats downgrade. Haste will be valued roughly at 28-29 ratings per %.
<Eej> YOU"RE GONNA PULL
<Eej> IF YOU SQUEEZE OFF ANOTHER ARCANE BLAST
<Spectear> You've obviously never played with Manly.
<Spectear> That's hardly a reason to stop DPS. Very Manly Staff
If MSD was 'fixed' as you say, to affect the current spell being cast, then I'm very curious why AtoI doesn't act this way also. Have you tested this? Looking at wowhead, it's the same rating modification (524288) for both procs. So very curious that one would act differently to the other. Same with skycall totem in fact.
Thats just how it is. Buff/debuff applications seem to have a queueing applied to almost all of them, with very few exceptions. Thats how rolling ignites work, thats how AB works, and thats how ashtongue works. See, you scorch, then theres a delay before you gain/lose a buff, so your next spell isn't yet affected by the ashtongue proc.
This probably has to do with the way the game is coded. I am still unsure about whether or not cast time is client side or server side, as crazy as it might sound. I've seen 'evidence' of both. It would seem foolish to let cast time be determined client-side.
<Eej> YOU"RE GONNA PULL
<Eej> IF YOU SQUEEZE OFF ANOTHER ARCANE BLAST
<Spectear> You've obviously never played with Manly.
<Spectear> That's hardly a reason to stop DPS. Very Manly Staff
Ah I see what you mean. This morning I quickly factored it out and couldnt get it to work, but I tried again this afternoon and it worked, I guess my brain doesnt work so well in the morning! But what do you mean it doesnt reduce casting time by a fixed percentage? If I have 5% haste my fire ball reduces casting time by 5%. It goes from a 3s cast to a 2.85s cast. If I reduce it by 20% it goes from 3s to 2.4s. Am I missing something?
If you have 20% haste, you should get 2.5 second casts, not 2.4 second casts. For small values of haste, the difference between the two methods is negligible, but as haste increases, it becomes more noticeable.
Hmm what is the algorithm for figuring it out? I guess I had never looked at it that carefully, what is the difference between casting speed and the way I am interpreting it? Is it (regular cast length/cast length*1+haste) and then that multiplyed against the regular?
Ah that makes sense, thanks for the clarification!
Something I have wondered, its generally assumed that a spells damage and crit y/n is determined as the spell is cast. But why is it then that if I cast one fireball and then apply Combustion, the combustion counter ticks up once when the fireball that was cast when combustion wasnt applied lands? It seems that if that fireball crits when it lands you are losing 1/3 of the benefit of combustion as that fireball sucks up one of the crits. But shouldnt that not happen if everything is decided upon spellcast?
Ah that makes sense, thanks for the clarification!
Something I have wondered, its generally assumed that a spells damage and crit y/n is determined as the spell is cast. But why is it then that if I cast one fireball and then apply Combustion, the combustion counter ticks up once when the fireball that was cast when combustion wasnt applied lands? It seems that if that fireball crits when it lands you are losing 1/3 of the benefit of combustion as that fireball sucks up one of the crits. But shouldnt that not happen if everything is decided upon spellcast?
You realise that shatter combos wouldn't exist without this feature ?
The spell knows if it crits or not upon cast. However, it doesn't knows if the target gains immunities before it lands, or if the target dies, or if it gains an absorb effect.
<Eej> YOU"RE GONNA PULL
<Eej> IF YOU SQUEEZE OFF ANOTHER ARCANE BLAST
<Spectear> You've obviously never played with Manly.
<Spectear> That's hardly a reason to stop DPS. Very Manly Staff
Shatter combos are entirely useless in raid situations and I would rather have my 3 crits from combustion than 2. It is possible that shatter combos could still exist if spells were decided when they hit, ice lance travels faster than frostbolt and when doing a shatter combo they generally arrive at the target at the same time. Whats to say the frozen state doesnt hang around for the .05 seconds it takes for the ice lance to land and crit after the FB? Or vice versa I have seen ice lance land before frostbolt as well.
*edit* Sorry I mean useless in single target raid encounters, there are several bosses with AoE elements it would be useful on
I think you've missed manly's point; his shatter comment must have been quite a curveball. What he's saying is that the crit is calculated at cast, but not whether the fireball successfully hits. Now imagine if a Combustion counter were deducted at cast time (when the crit is calculated, as you suggest), but the target dies while the Fireball is in midair. Now you've truly wasted a Combustion charge on a useless fireball! The solution is to count Combustion at impact, and as a side effect count it against a crit mid-flight that you cast before Combustion. I think the alternative is much worse for us.
Thegoodman, I'm curious which frost talents you dropped for your 18/0/43. I wasn't going to guess, so I went with the standard 10/0/51, loaded my character in Rawr, added moonkin, retri pally and totem of wrath to my Brutallus set up for a total Frostbolt crit rate of 54%. With this all, AtoI does indeed become 6 dps better than Icon. However, with such high crit, TLC is preferred, and moreover: Hex, Sliver and Skull are still way ahead. So I'm still not really convinced that the trinket is 'terrific'.
If you look at WWS for Nature's Grace you'll see the buff is gained some time after the spell lands. It's not sent by the server at the same time as the damage for the spell. So because the druid will have already started his next cast, I believe the shortened cast time must be applied retroactively. In any case it's not really the same mechanic as with AtoI since it's a reduced cast time buff rather than a spell haste buff.
If MSD was 'fixed' as you say, to affect the current spell being cast, then I'm very curious why AtoI doesn't act this way also. Have you tested this? Looking at wowhead, it's the same rating modification (524288) for both procs. So very curious that one would act differently to the other. Same with skycall totem in fact.
Now you got me interested in Nature's Grace! I looked up how it works: Wow Web Stats
"Nature's Grace" procs 50-100s ms before the spell crit in the combat log.
That explains the immediate effect on the next cast.
On the other hand, "Lunar Grace" from [Idol of the Unseen Moon] procs after the spell hit.
It just seems completely random which procs were programmed to work properly and which proc do not.
Not sure what you mean with Have you tested this?. MSD wasn't fixed, it has always worked like that.
Imagine the old MSD - "half-cast time on the next spell" and not affecting the next cast. Or Netherwind Focus (T2). Doesn't make sense at all, so they had to make them affect the immediately following cast.
When changing the buff to "320 haste for 4s (6s)", they just kept the proc mechanic.
Originally Posted by Aeoliana
Something I have wondered, its generally assumed that a spells damage and crit y/n is determined as the spell is cast. But why is it then that if I cast one fireball and then apply Combustion, the combustion counter ticks up once when the fireball that was cast when combustion wasnt applied lands? It seems that if that fireball crits when it lands you are losing 1/3 of the benefit of combustion as that fireball sucks up one of the crits. But shouldnt that not happen if everything is decided upon spellcast?
Welcome to odd mechanics with spell hits, casts and travel time. Look at the comments above, it's pretty unintuitive chaos at times.
It's not just spells though. Did you know that Windfury/Sword Spec procs eat flurry charges?
Different mechanics, but it's the same situation as Combustion, and makes about as much sense.
As for AToI in Rawr - Rawr takes the not-on-next-cast-proc into account, just like it does for Arcane Blast.
I didn't mean to say it was a terrific trinket right now - but before we found that it didn't work on the next cast, TC showed that it would be the best trinket for Frost specs before you can get Skull/Hex.
Originally Posted by manly
Going 70-80 will cost you severely on stats downgrade. Haste will be valued roughly at 28-29 ratings per %.
It still takes a lot of item budget to outclass 145 haste with 95%+ uptime for AoE spam.
From the screenies we had, level 80 blues would be extrapolated to be half a tier above K'J quality, it not a lot of budget.
No point in speculating, but it would require a really well designed entry-raid trinket to drop AToI for AoEing. Caster DST or something.
As for Shatter/WG/AToI - it totally depends on the numbers and how much it will be tweaked/nerfed.
Bear in mind that WG uptime currently only scales with haste and adds crit, which are optimal conditions for AToI to work.
Eyeballing, it's still not worth it with haste skipping the first cast.
We'll just have to see how things will turn out really. Numbers, mechanics and nerf cries, business as usual
I am saying its possible that crit or not crit may be calculated at impact. The point is that you are losing a crit out of the three if an unlucky fireball crits once you pop combustion. And your point about losing a fireball if a mob dies, that doesnt really matter unless you are worrying that much over trash kills. You missed my point when I said the frostbolt and ice lance land at the same time meaning the frozen state crit bonus is still applicable to both spells.
Lets pretend that you arent throwing a ball of fire at the boss but a ball of numbers, the range of damage your spell can do, the chance it will crit, the chance it will miss, the possiblity of a partial resist, all that. And when it collides with the boss it explodes into calculation with the end result being the actual damage you do. With combustion all its counting is whether a fire spell cast while its up crits or not. So if we go with my model it makes sense that you lose a tick of combustion however, if crit was counted before it actually lands you should still get your 3 crits with combustion. But you dont.
<Eej> YOU"RE GONNA PULL
<Eej> IF YOU SQUEEZE OFF ANOTHER ARCANE BLAST
<Spectear> You've obviously never played with Manly.
<Spectear> That's hardly a reason to stop DPS. Very Manly Staff
No im suggesting that spell dmg/crit/miss is calculated when it lands not when it casts, not a whine.
Feel free to read on this very page why that would be a bad idea.
I'll give you a cookie if you list more than a dozen abilities that would be broken/abilities then.
[Edit]: People reported that even with 1) Scorch - 2) Combustion - 3) Any Fire Spell, the Scorch would eat a charge. Even without a travel time.
So that change might not even fix combustion.
[Edit II]: I did read what you wrote, and the Combustion thing is pretty annoying.
Of course it would be great if it could be fixed, that it only adds charges for spell that were cast and hit while combustion was acive.
The scorch example shows that even that woldn't work with current mechanics.
"Luckily" you just lose 10% crit for 1 cast. Your gain of combustion drops from ~1.3 to ~1.2 crits.
I'd love to have a working combustion. And AToI and AB and Ignite and Water Elemental (0.5-0.6s cast delays on auto cast, like Searing Totem).
But Combustion isn't fixable without a complete overhaul of the combat system and the order in which actions are processed.
I feel that an 8% loss on a talent that scales negatively with your gear anyway is simply the lesser evil.
I'm curious, as someone who's been away from the game for a while (~8 months), how some effective some specs are now compared to back then.
When I quit, Deep Arcane was just getting to the point where it wasn't worth speccing over deep fire with T6 content and the buff to Imp Fireball/Frostbolt (or the removal of the -5% nerf, however you want to look at it). It seems that 2/48/11 is pretty much the most consistent build, as deep fire has been for ages. However, a mage just joined my guild yesterday, and was commenting on how he was out-DPSing the fire mages in his old guild as 40/0/21. The mage in question is in some of the best possible T6 level gear, and I got the impression that so was his competition.
I guess in short, I'm wondering, how can 40/0/21 stack up to 2/48/11 at a T6 gear level at this point? I loved playing deep arcane when it was still "new and exciting", and if it's raid viable, I'd respec and play that in an instant. I do have 3 pieces of T5 available for the set bonus as well (pants, shoulders, gloves), since I'm assuming that's still pretty much required to make the build "work" at a high level.
I think Aeoliana is simply saying that it doesn't work the way we think it works.
I also think that Aeoliana is missing the distinction between when chance to crit is calculated and when the crit occurs. The mechanics seem perfectly consistent--though I may not like them--for Combustion to simply count crits that occur while it's up, even if the chance to crit is calculated when the spell leaves one's hands (completes casting, regardless of any travel time).
Ok what I read from in a lot of prior posts was that everything about the spell was calculated when it left your hands. Including whether it hit or not, crit or not, and how much dmg it did and was just waiting to show you that number in your combat log when it hit the mob.
And roywyn I did read the abilities that would be messed up by spells being decided upon landing not casting. And I addressed that... Try rereading what I wrote and understanding it.
My point was if you have 30% crit normally and you cast a fireball, that fireball has a 30% chance to crit right? But if you use Combustion while that fireball is in midair and the fireball crits it will eat up one crit off your combustion at only 30% instead of 40%. Like I said the fireball is like a packet of numbers that resolve when it lands. In the end you only get 2 boosted crits off Combustion not 3.
The point is that you are losing a crit out of the three if an unlucky fireball crits once you pop combustion.
I usually scorch then use my macro with combustion after the GCD is finished so this doesn't happen often, and minimizes the need to scorch with my big damage cooldowns up.
Well, were talking here about a mechanic that has led us to many 'greater goods'. First, it allows shatter combos to exist. Second, it allows combustion->flamestrike->blastwave crit chains. It also allows you to get more than 3 crits per combustion. It also allows for much much better controlled burst. Would you really looking forward to have to cast a fireball, then pop a trinket/dmg cooldowns before it lands because thats when the dmg would be calculated ? It would totally throw off your rotations. It would be a major pain to play. It would be nice to have fireballs cast at 21% but landing at 19% automatically get molten fury, but thats about the only real upside to this. There would be several downsides.
All of this because you want to not get screwed on your first combustion tick ? C'mon now. Combustion is a fairly poor talent as it is. So I get that you don't want to get screwed out of +10% chance to crit for one cast every 3:15 if-and-only-if your previous cast was a crit. Fair enough. But can't you see the if you changed that you would also be unable to get 4+ crits out of combustion, and there would be several other far worse implications ? You could also do like everyone else, and get leatherworking. Use the drums GCD as a way to micro manage your combustion charges. You can also scorch -> (wait a bit here for ignite to register) -> combustion/fireball. You can also interstice invis before combustion if you had planned to invis during the fight. You can also wait on movements if the fight involves movements.
<Eej> YOU"RE GONNA PULL
<Eej> IF YOU SQUEEZE OFF ANOTHER ARCANE BLAST
<Spectear> You've obviously never played with Manly.
<Spectear> That's hardly a reason to stop DPS. Very Manly Staff
"Nature's Grace" procs 50-100s ms before the spell crit in the combat log. That explains the immediate effect on the next cast.
The log I was looking at has it 350ms after. Wow Web Stats As Manly said it does seem like there is a queuing system for data packets, so the crit and proc and aura may well happen at the same time on the server, and just reach the client at various moments.
"MSD wasn't fixed, it has always worked like that." - sorry I misread when you wrote "Nature's Grace ... was explicitly fixed". But my question was more that according to Wowhead the effect from Skycall totem and the effect from AtoI are the same type (Energized - Spell - World of Warcraft / Insight of the Ashtongue - Spell - World of Warcraft) so that's why I found it curious that you say the totem works correctly while the trinket doesn't.
07/17/08, 11:23 AM
