WotLK - Complete Mage Compendium (3.3.3 live) - Page 83

Morgannon · 09/25/08, 12:59 AM

Originally Posted by Roywyn

What's new?
Nothing really. Arcane Blast is still useless. Fire doesn't actually look that mana-hungry any more.

What changed that made fire less mana hungry?

Zaldinar · 09/25/08, 2:03 AM

The ignite stomping situation may be more complex than just losing ignite damage to the Fireball/Pyro combo. Using ZIT (
http://zaldinar.bounceme.net/zit/zitv04.zip ) to record the ignite gained / lost from the bugs, I got a summary of around 5.4% ignite damage lost to the double crit bug, and 2.7% gained from gained bugs. Upon closer inspection of logs and watching as it happens, this is the basic timeline I've been able to come up with (assumes a 1 second travel time for Pyro/Fireball)

T+0.0 - Fireball 1 cast starts
T+3.0 - Fireball 1 cast completes, Fireball 2 cast starts
T+4.0 - Fireball 1 lands, crits
T+6.0 - Fireball 2 cast completes, Fireball 3 cast starts, Fireball 1 ignite tick 1 ticks
T+7.0 - Fireball 2 lands, crits, hotstreak procs
T+9.0 - Fireball 3 cast completes, cast instant Pyroblast, Fireball 1 + Fireball 2 ignite tick 1 ticks
T+10.0 - Fireball 3 lands, crits, Pyroblast lands, crits
T+10.5 - Fireball 4 cast starts
T+12.0 - Fireball 1 + Fireball 2 + Fireball 3 + Pyro ignite ticks, may have lost ignite damage from T+10.0
T+13.5 - Fireball 4 cast completes
T+14.0 - Fireball 1 + Fireball 2 + Fireball 3 + Pyro ignite ticks
T+14.5 - Fireball 4 lands, may crit

T+14.0 and T+14.5 are the important events. If, through variations in missile time and haste, those events happen on top of eachother, you have a chance of critting and having a hiccup of ignite, gaining damage. The chance at repeating a 3 Fireball and 1 Pyro ignite tick will tend to produce a counter-effect to the loss seen if the Pyro stomps on its fireball.

Regardless, it should be fixed.

manly · 09/25/08, 2:16 AM

Ok well, some good news.

1- Living Bomb got a new graphic. It also looks awesome.
2- Living Bomb explodes from the target (!)
3- Living Bomb pushes even higher now. Its pretty absurd how high the explosion is. Were talking a huge jump here.
4- I don't know if this is new, but if your mob is frostnovaed as it explodes, it won't move. I know its not a discovery per-se, but I didn't recall that happening when I tried it last.

Also I started running ZIT full time. I'll give results after this weekend - this way I will have sensible data points.

Zaldinar · 09/25/08, 2:22 AM

Manly, it may be worthwhile to actively attempt two separate rotations. One where you follow a standard 8:1 Fireball:Scorch with pyros on hostreaks, and the other a Fireball spam, Scorch->Pyro on hostreaks, extra scorches as needed. I'm about 20 minutes into a test with the latter rotation, and am up about 10% ignite damage with no lost bugs recorded. Haste and distance to target will play a large part in this interaction however, so any testing is sketchy at best.

manly · 09/25/08, 2:29 AM

Good data Zaldinar. I know that this isn't like scientific or anything, but I remember that from my many early ignite bug tests I had came to the conclusion that the bug was caused by either one of the 2 following cases:

1- As expected, only travel time comes into play. This means indirectly that whenever the bolt lands, it applies the ignite on the spot. What is unclear here is that it is highly unlikely that the server actually tracks the spell (in the air) and calculates server-side when it lands. This is doubled by the fact that positional data is mostly handled client-side. Of course, this is theorical, but I doubt anyone coding that would go through the trouble of tracking every bolt in the air, the direction it needs to go, and whatnot.

However, there has been several hints towards all debuffs having some delays on them. This is either an artifice from cleint-server communication giving the impression there is a delay before debuffs are applied, or its likely caused because the code is multithreaded.

Which leads to 2

2- The bug happens if you land a ignite while another ignite is in the process of being applied. The problem here is that it is extremely non-obvious to test. If we just knew more precisely how bolts are tracked and how it knows when it lands it would greatly simplify the testing of the bug.

Even though I did pass several hours trying to hammer out that particular distinction, I haven't been successful. Maybe your data better matches the scenario #2 ?

manly · 09/25/08, 2:33 AM

Originally Posted by Zaldinar

Manly, it may be worthwhile to actively attempt two separate rotations. One where you follow a standard 8:1 Fireball:Scorch with pyros on hostreaks, and the other a Fireball spam, Scorch->Pyro on hostreaks, extra scorches as needed. I'm about 20 minutes into a test with the latter rotation, and am up about 10% ignite damage with no lost bugs recorded. Haste and distance to target will play a large part in this interaction however, so any testing is sketchy at best.

Yes that is precisely what I suggested earlier. That maybe living bomb/pyroblast or drums/pyroblast or scorch/pyroblast on hot streak procs as a way to work around the issue. Problem is, we don't really know deterministically which of the 2 ignites gets lost, and much less what specific case causes it to happen. It is only travel time, or is debuff application delay also come into this ?

I am curious about ZIT. Does it tracks the order in which the spells are actually cast and compares that ti whichever ignite is lost ? That strikes me as something that would be vital to know. I understand you would probably have to code in some extra stuff to track actual cast order (since you cant rely on combat logs for that).

Zaldinar · 09/25/08, 2:49 AM

Originally Posted by manly

Yes that is precisely what I suggested earlier. That maybe living bomb/pyroblast or drums/pyroblast or scorch/pyroblast on hot streak procs as a way to work around the issue. Problem is, we don't really know deterministically which of the 2 ignites gets lost, and much less what specific case causes it to happen. It is only travel time, or is debuff application delay also come into this ?

I am curious about ZIT. Does it tracks the order in which the spells are actually cast and compares that ti whichever ignite is lost ? That strikes me as something that would be vital to know. I understand you would probably have to code in some extra stuff to track actual cast order (since you cant rely on combat logs for that).

It can figure out which ignite got lost retrospectively. It saves a dump of your combat log before the spell crit, and after the ignite cleared, so we can go back and look at what timestamps generated what issues, so we really do have perfect documentation of bugged ignites. I'll work on writing a parse script up for it tonight to make it more productive past the totals it displays.

Cast order can actually be gleened from the combat logs, since we have a cast start event in the log itself, ZIT already records all this, just doesn't display it in game.

As for the way ZIT works, it watches your combat log for fire spells (excluding Molten Armour... and actually, come to think of it, won't catch FFB in its current state, hmm), when it catches a fire spell crit it calculates how much it expects to see in ignite damage. Any successive crits before the ignite finishes are added into this total (Total Expected on all displays reflects this value, for ones spanning multiple ignites its just a sum of expected).

When it identifies ignite tick events, it increments the Total Actual value. It can, theoretically, identify at that point if something has gone wrong, but it became more and more complex as I worked on it more and more to make that method work, so I opted for just checking when the ignite completes (5 seconds following the last fire crit) and comparing Total Expected vs Total Actual.

Any net imbalance between these two more than 2 (to give a buffer for rounding error) is considered a bug, and put in its respective slot as a gain or loss bug. Loss to death is calculated as Total Expected - Total Actual at time of death. If the active ignite on the target is ticking incorrectly it will not be able to be identified as bugged. This is a technical limitation because of the way I have it "clearing" ignites. I'm uncertain due to the nature of the delays in events whether or not it can actually be made to do it on a per-tick basis in any consistent way.

The addon itself keeps a buffer of the last 15 seconds worth of combat log entries that have happened, as soon as it identifies a fire crit it stops deleting old entries and keeps recording until the ignite clears, then takes all entries from 8 seconds before the crit happened, until 5 seconds after the last crit happens and puts it into the data file.

Edit: And actually, I know you've mentioned in the past that you have programming experience, just not with LUA. I taught myself to program using functional structured C, and as a result always end up trying to write my programs like C programs. ZIT is pretty well commented, the only real annoying part is remembering which structure you're playing with, and how the data structures are built (I absolutely hate LUAs table system, I would *kill* for a C structure)

But as far as how it works, the program is pretty easy to read as far as LUA goes, if you want to poke through it.

LiquidHAL · 09/25/08, 3:03 AM

Shattered Barrier is not functional again with the latest patch.

LiquidHAL · 09/25/08, 3:12 AM

Ugh, and Deep Freeze is now doing no damage at all (frozen root effect is still working with it though).

Shadout · 09/25/08, 3:54 AM

Deep Freeze getting changed next patch it seems. I'm worried, seemed like one of the better new abilities around

Malavas · 09/25/08, 5:12 AM

At least on PTR Frostbite on rootable targets has stopped consuming a charge of FoF.

Can someone test the proc rate of FoF? it seems nowhere near of the 15% stated on the tooltip.

Zaldinar · 09/25/08, 5:30 AM

Originally Posted by Malavas

At least on PTR Frostbite on rootable targets has stopped consuming a charge of FoF.

Can someone test the proc rate of FoF? it seems nowhere near of the 15% stated on the tooltip.

I had it pegged at 16% in build 8970 over a few thousand casts of Frostbolt with 0/3 Frostbite. I'll repeat that test at some point.

Edit: And for Manly, this is the type of data available from the ZIT output ( http://zaldinar.bounceme.net/zit/example/parse.html )

Arg 1 - Arg 2 - Arg 10 - Arg 12 - Arg 18
T+7.415 - "SPELL_DAMAGE" - "Fireball" - 3166 - 1
T+8.227 - "SPELL_AURA_APPLIED" - "Ignite"
T+10.212 - "SPELL_PERIODIC_DAMAGE" - "Ignite" - 633
T+10.296 - "SPELL_DAMAGE" - "Fireball" - 3108 - 1
T+10.647 - "SPELL_AURA_APPLIED" - "Hot Streak"
T+13.057 - "SPELL_PERIODIC_DAMAGE" - "Ignite" - 938
T+13.286 - "SPELL_DAMAGE" - "Fireball" - 2180 - 0
T+14.26 - "SPELL_DAMAGE" - 	"Scorch" - 1214 - 1
T+15.275 - "SPELL_DAMAGE" - "Pyroblast" - 3482 - 1
T+17.816 - "SPELL_PERIODIC_DAMAGE" - "Ignite" - 1408
T+19.729 - "SPELL_DAMAGE" - "Fireball" - 3193 - 1
T+19.731 - "SPELL_PERIODIC_DAMAGE" - "Ignite" - 1408
T+19.732 - "SPELL_AURA_REMOVED" - "Ignite"
T+19.822 - "SPELL_AURA_APPLIED" - "Hot Streak"
T+20.265 - "SPELL_AURA_APPLIED" - "Ignite"
T+22.259 - "SPELL_PERIODIC_DAMAGE" - "Ignite" - 638
T+22.791 - "SPELL_DAMAGE" - "Fireball" - 2182 - 0
T+23.765 - "SPELL_AURA_REMOVED" - "Hot Streak"
T+23.853 - "SPELL_DAMAGE" - "Scorch" - 1217 - 1
T+24.291 - "SPELL_PERIODIC_DAMAGE" - "Ignite" - 639
T+24.292 - "SPELL_AURA_REMOVED" - "Ignite"
T+24.68 - "SPELL_AURA_APPLIED" - "Ignite"
T+26.702 - "SPELL_PERIODIC_DAMAGE" - "Ignite" - 562
T+28.684 - "SPELL_PERIODIC_DAMAGE" - "Ignite" - 562
T+28.686 - "SPELL_AURA_REMOVED" - "Ignite"

So we expect:

3166 * .2 = 633.2 Damage Ticks at first, we get one at T+10.212, then we expect:

3108 * .2 + 633.2 / 2 = 938.2 damage ticks, which we get at T+13.057, then we expect

1214 * .2 + 3482 * .2 + 938.2 / 2 = 1408.3 damage ticks, which we get at T+17.816 and T+19.731

So the first round of ignites clear. Second round at T+19.729 starts it with:

3193 * .2 = 638.6 ticks, seen at T+22.259 and T+24.291, but from the crit at T+23.853 we expect

1217 * .2 = 243.4 ticks, but we get 562, which is consistent with:

1217 * .2 + 638.6 / 2 = 562.4.

A classic hiccup. Examples of both hiccups and double crits can be analyzed in this way.

I've got an upload and assimilation script written, http://zaldinar.bounceme.net/zit/parse3.php

It doesn't allow for display yet, just will assimilate into the database. Display will come with time.

dedmonwakeen · 09/25/08, 5:55 AM

SimulationCraft has been updated for patch 8982. The list of classes now supported includes: Druid, Mage, Priest, Shaman, Warlock.

Multi-player simulation that supports user-specified talent setups and action priority lists.

The config file has been enhanced to support MMO-Champion talent tree links.

Using Naxx-10 baseline gear:

Player=Mage_Arcane  DPS=3039.8  DPR=8.7  RPS=347.8/397.6  (mana)
  Core Stats:  strength=44  agility=50  stamina=845  intellect=1052  spirit=490  health=11648  mana=16996
  Spell Stats:  power=2267  hit=0.5%  crit=20.5%  penetration=0  haste=17.8%  mp5=91
  Attack Stats:  power=34  hit=0.0%  crit=4.0%  expertise=0.0  penetration=0  haste=17.8%
  Actions:
    arcane_barrage        Count=100.0| 3.4sec  DPE=4645|44%  DPET=3598  DPR= 7.7  Miss=10.6%  Hit=3799  CritHit=6632|8298|44.1%
    arcane_blast          Count= 93.7| 3.7sec  DPE=4901|44%  DPET=2496  DPR=10.4  Miss=10.5%  Hit=4004  CritHit=6990|12003|44.1%
    arcane_missiles       Count= 13.3|25.9sec  DPE=9110|12%  DPET=4582  DPR= 8.2  Miss=9.9%  Tick=1489  CritTick=2604|3267|44.3%

Player=Mage_Fire  DPS=3480.6  DPR=11.2  RPS=310.8/403.6  (mana)
  Core Stats:  strength=44  agility=50  stamina=845  intellect=915  spirit=446  health=11648  mana=16996
  Spell Stats:  power=2004  hit=0.5%  crit=19.7%  penetration=0  haste=17.8%  mp5=91
  Attack Stats:  power=34  hit=0.0%  crit=4.0%  expertise=0.0  penetration=0  haste=17.8%
  Actions:
    fire_ball             Count=117.5| 2.9sec  DPE=5725|56%  DPET=2409  DPR= 9.9  Miss=10.4%  Hit=4645  CritHit=8114|8773|45.1%
    ignite                Count=  0.0| 0.0sec  DPE= inf|20%  DPET= inf  DPR= inf  Miss=nan%  Tick=2641
    living_bomb           Count= 30.7|11.2sec  DPE=5346|14%  DPET=4076  DPR= 6.2  Miss=10.7%  Tick=1121  CritTick=3718|4032|8.5%
    pyroblast             Count= 15.2|22.6sec  DPE=7539|10%  DPET=5750  DPR=12.4  Miss=10.5%  Hit=4639  CritHit=8112|8764|42.4%  Tick=440

Player=Mage_Frost  DPS=3127.7  DPR=17.3  RPS=180.4/316.8  (mana)
  Core Stats:  strength=44  agility=50  stamina=845  intellect=915  spirit=446  health=11648  mana=16996
  Spell Stats:  power=2004  hit=0.5%  crit=19.7%  penetration=0  haste=17.8%  mp5=91
  Attack Stats:  power=34  hit=0.0%  crit=4.0%  expertise=0.0  penetration=0  haste=17.8%
  Actions:
    deep_freeze           Count=  8.6|40.1sec  DPE=4713| 4%  DPET=3609  DPR=22.6  Miss=10.4%  Hit=2728  CritHit=5457|5469|83.1%
    fire_ball             Count= 20.7|16.6sec  DPE=4093| 8%  DPET=3147  DPR= inf  Miss=10.9%  Hit=3778  CritHit=5667|5689|38.5%
    frost_bolt            Count=144.2| 2.4sec  DPE=5568|75%  DPET=2790  DPR=14.2  Miss=10.5%  Hit=3867  CritHit=7744|7780|54.3%
    ice_lance             Count=  7.4|46.2sec  DPE=3918| 3%  DPET=3004  DPR=21.5  Miss=10.7%  Hit=2271  CritHit=4541|4554|83.3%
   water_elemental
    water_bolt            Count= 85.8| 4.0sec  DPE=1360|11%  DPET= 544  DPR= inf  Miss=14.3%  Hit=1425  CritHit=2141|2146|19.4%

Player=Mage_FrostFire  DPS=3516.1  DPR=13.3  RPS=263.6/348.6  (mana)
  Core Stats:  strength=44  agility=50  stamina=845  intellect=915  spirit=446  health=11648  mana=16996
  Spell Stats:  power=2004  hit=0.5%  crit=19.7%  penetration=0  haste=17.8%  mp5=91
  Attack Stats:  power=34  hit=0.0%  crit=4.0%  expertise=0.0  penetration=0  haste=17.8%
  Actions:
    frostfire_bolt        Count=117.1| 2.9sec  DPE=5584|54%  DPET=2350  DPR=12.6  Miss=10.7%  Hit=3783  CritHit=8490|9181|46.9%
    ignite                Count=  0.0| 0.0sec  DPE= inf|22%  DPET= inf  DPR= inf  Miss=nan%  Tick=2816
    living_bomb           Count= 30.8|11.2sec  DPE=5400|14%  DPET=4119  DPR= 6.3  Miss=10.5%  Tick=1112  CritTick=3716|4032|9.3%
    pyroblast             Count= 15.9|21.6sec  DPE=7694|10%  DPET=5865  DPR=12.6  Miss=10.6%  Hit=4635  CritHit=8114|8764|46.6%  Tick=440

EDIT: I'd like to get a decent default setup for talents/actions in each Mage spec, so I welcome feedback. The config file can be found here.

Roywyn · 09/25/08, 6:09 AM

Originally Posted by PsyBomb

Roywyn, have you considered deep frost with a Fire subspec? Something like http://talent.mmo-champion.com/?mage...00000000000000 is what I'm favoring at the moment, since those Fireballs start to really sting when they Ignite (and crit rates are getting a bit stupid, especially if you can chuck one off a FoF)

Ignite is a ~20% boost to Fireball. Spell Impact is 6%. Difference is 14% boost on a spell that does 10-12% of your damage.
It does shut out the option of using Frostfire Bolt on your second charge though.

Arcane gives you 3%+3% universal crit and the option to go for Torment or Meditation.

Originally Posted by Morgannon

What changed that made fire less mana hungry?

Fire getting a damage boost in Arcane, trading Icy Veins (haste) for Focus Magic (crit).
It's not a recent thing really, I just wanted to make a point that "Fire = Mana Sponge" is becoming obsolete.
Fireball spam has more DPM that Frostbolt spam currently.
Frost has efficiency talents and Fire has mana dump talents though, so Frost is still more efficent, but it's getting closer.

Originally Posted by dedmonwakeen

SimulationCraft has been updated for patch 8982. The list of classes now supported includes: Druid, Mage, Priest, Shaman, Warlock.

Multi-player simulation that supports user-specified talent setups and action priority lists.

The config file has been enhanced to support MMO-Champion talent tree links.

Using Naxx-10 baseline gear:

Player=Mage_Fire  DPS=3480.6  DPR=11.2  RPS=310.8/403.6  (mana)
  Core Stats:  strength=44  agility=50  stamina=845  intellect=915  spirit=446  health=11648  mana=16996
  Spell Stats:  power=2004  hit=0.5%  crit=19.7%  penetration=0  haste=17.8%  mp5=91
  Attack Stats:  power=34  hit=0.0%  crit=4.0%  expertise=0.0  penetration=0  haste=17.8%
  Actions:
    fire_ball             Count=117.5| 2.9sec  DPE=5725|56%  DPET=2409  DPR= 9.9  Miss=10.4%  Hit=4645  CritHit=8114|8773|45.1%
    ignite                Count=  0.0| 0.0sec  DPE= inf|20%  DPET= inf  DPR= inf  Miss=nan%  Tick=2641
    living_bomb           Count= 30.7|11.2sec  DPE=5346|14%  DPET=4076  DPR= 6.2  Miss=10.7%  Tick=1121  CritTick=3718|4032|8.5%
    pyroblast             Count= 15.2|22.6sec  DPE=7539|10%  DPET=5750  DPR=12.4  Miss=10.5%  Hit=4639  CritHit=8112|8764|42.4%  Tick=440

Player=Mage_FrostFire  DPS=3516.1  DPR=13.3  RPS=263.6/348.6  (mana)
  Core Stats:  strength=44  agility=50  stamina=845  intellect=915  spirit=446  health=11648  mana=16996
  Spell Stats:  power=2004  hit=0.5%  crit=19.7%  penetration=0  haste=17.8%  mp5=91
  Attack Stats:  power=34  hit=0.0%  crit=4.0%  expertise=0.0  penetration=0  haste=17.8%
  Actions:
    frostfire_bolt        Count=117.1| 2.9sec  DPE=5584|54%  DPET=2350  DPR=12.6  Miss=10.7%  Hit=3783  CritHit=8490|9181|46.9%
    ignite                Count=  0.0| 0.0sec  DPE= inf|22%  DPET= inf  DPR= inf  Miss=nan%  Tick=2816
    living_bomb           Count= 30.8|11.2sec  DPE=5400|14%  DPET=4119  DPR= 6.3  Miss=10.5%  Tick=1112  CritTick=3716|4032|9.3%
    pyroblast             Count= 15.9|21.6sec  DPE=7694|10%  DPET=5865  DPR=12.6  Miss=10.6%  Hit=4635  CritHit=8114|8764|46.6%  Tick=440

EDIT: I'd like to get a decent default setup for talents/actions in each Mage spec, so I welcome feedback. The config file can be found here.

Cool, thanks! There are two issues though:

1) 0.5% hit on gear murders your DPS. Should be 10.5%? Or, honestly, just cap out spell hit.
I'd gladly take a 10 DPS loss for the safety of Scorch never missing (and dropping).

2) Frostfire Bolt shouldn't be able to beat Fireball.
You have average hit and crit for Fireball as 4.6k/8.1k, Frostfire Bolt as 3.6k/8.5k.
I have Fireball as 4.6k/8.4k, Frostfire Bolt as 3.6k/8.4k.

Using a 14/51/0+x spec for Focus Magic (3% crit) and Impact (6% Fireball damage).
Fireball Glyph is 5% crit, Frostfire Bolt Glyph is 2% crit, 2% damage.

So, something seems to mess up data somewhere.

grayrest · 09/25/08, 6:23 AM

Originally Posted by dedmonwakeen

SimulationCraft has been updated for patch 8982. The list of classes now supported includes: Druid, Mage, Priest, Shaman, Warlock.

Looks like you missed the hit on the gear so everything is 10% lower/more expensive than our spreadsheets are showing. Also, we're showing the current best Arcane build as 53/18/0 using fireball instead of AB as ABr fill.

Grai · 09/25/08, 6:35 AM

Something I never thought I'd get to test, but currently the 2 piece Frostfire 10 man bonus seems to be simply giving the spell damage buff for 15 seconds and not giving the +40% mana on mana gem.

As expected, it's using the same buff as Serpent Coil Braid, just without the extra mana from gems.

Prom · 09/25/08, 6:54 AM

They finally fixed the switching around of Mirror Image so it's not annoying anymore, however they still do the same dmg (around 15K total), which is pretty weak imo.

The 2-piece T7 is still bugged and won't give you the increased effect of mana regeneration.

Kandir · 09/25/08, 7:40 AM

Originally Posted by grayrest

Also, we're showing the current best Arcane build as 53/18/0 using fireball instead of AB as ABr fill.

Despite that build showing higher dps in theory, I'd suggest that in practice it'd be better to invest two points in flame throwing instead of incineration. This would result in both your arcane spells and scorch having a range of 36 yards. It's a situational advantage but makes sense given the increased range magical attunement offers. Not only that, but the dps gain from 2 points in incinerate, especially if you're using the imp scorch glyph and no AB in your rotation, would be trivial.

I'm also concerned about possible pushback issues with that build. Yes, they've changed the mechanics of pushback to be not as damning for us, and there's usually not too many fights where it's a serious problem, but we all know how annoying it can be to put deal with. Perhaps someone with experience with LK instancing can give some insigt regarding that though.

Samuel · 09/25/08, 7:48 AM

Originally Posted by Wiedmaier

With the frostbolt glyph removing it's "slowing effect" does it still provide a sort of chill debuff?

With FoF now proccing off a chill effect, and frostbite always being based on chill, it would seem counter-intuitive for the frostbolt chill to be removed totally, but I am curious just how that mechanic has played out.

My understanding from the PTR and posts here is:

Frostbolt, Ice Armor, Cone of Cold, FFB and Blizzard (when talented) all apply chill effects and proc FoF (and Frostbite).
The frostbolt glyph does not affect FoF (or Frostbite) but it does reduce the speed reduction from frostbolt to 0%.
FoF is consumed by a "hit" so one wave of blizzard or one volley of arcane missiles will consume one charge of FoF.

[edit]
Arch just confirmed that FFB procs both FoF and Brain Freeze, changed post to reflect that.

arch · 09/25/08, 8:01 AM

FFB procs both brain freeze and fingers of frost, just tried it.

Pasture · 09/25/08, 8:12 AM

Initially when AB was 75% increase in mana per debuff it was sustainable. Not that the mana penalty has been decreased to 200% does this mean we can afford to throw in an extra AB into rotations for the slight increased damage.

I was thinking maybe AB, AB, ABr. Or substitute ABr with whatever filler is regarded best at the moment (fireball I believe). Can anyone try a two-stack AB rotation rather than a one-stack just to see how mana holds up and whether it's any kind of significant dps increase.

dedmonwakeen · 09/25/08, 8:36 AM

Originally Posted by Roywyn

1) 0.5% hit on gear murders your DPS. Should be 10.5%? Or, honestly, just cap out spell hit.
I'd gladly take a 10 DPS loss for the safety of Scorch never missing (and dropping).

2) Frostfire Bolt shouldn't be able to beat Fireball.
You have average hit and crit for Fireball as 4.6k/8.1k, Frostfire Bolt as 3.6k/8.5k.
I have Fireball as 4.6k/8.4k, Frostfire Bolt as 3.6k/8.4k.

Using a 14/51/0+x spec for Focus Magic (3% crit) and Impact (6% Fireball damage).
Fireball Glyph is 5% crit, Frostfire Bolt Glyph is 2% crit, 2% damage.

So, something seems to mess up data somewhere.

Regarding Hit Rating: We were doing some scaling analysis and reduced it to play around...... but by default the config file should be capped. I'll get that changed

Thanks for the Fire Ball analysis. I'll dig in.

EDIT: I think I have Fire using the "wrong" talent spec. I have two in the config file and went with Fire/Frost instead of Fire/Arcane. There were two reasons for this originally:
(1) I was having mana issues with deep Fire and wanted to pick up Frost Channeling.
(2) In the sim, Focus Magic "really works"..... meaning it is not just a flat modifier..... I have to choose someone in the raid to give it to. Since I was doing some comparative analysis it seemed unfair..... but now that I understand better it is "unfair" to the Mage -not- to use it. Perhaps I'll have Mages just round-robin each other with the buff.

Originally Posted by grayrest

Looks like you missed the hit on the gear so everything is 10% lower/more expensive than our spreadsheets are showing. Also, we're showing the current best Arcane build as 53/18/0 using fireball instead of AB as ABr fill.

Excellent. I'll change the default config appropriately.

grayrest · 09/25/08, 8:58 AM

Originally Posted by dedmonwakeen

(2) In the sim, Focus Magic "really works"..... meaning it is not just a flat modifier..... I have to choose someone in the raid to give it to. Since I was doing some comparative analysis it seemed unfair..... but now that I understand better it is "unfair" to the Mage -not- to use it. Perhaps I'll have Mages just round-robin each other with the buff.

There was a report upthread that said the proc didn't stack with the 30 minute buff.

sornok · 09/25/08, 9:05 AM

I have a couple of questions for those of you who are in beta. (I hope they haven't been answered earlier in the topic)

1) About our crit debuffs. Do they stack ? (I guess not). Are they the same debuff or a different one ? When we have both in a raid to they push each other out of the debuff list ? What about focus magic ? Does it stack with the one of them ? I guess yes, but isn't it too much for a single class to buff with 16% crit the whole raid ?

2) About the frostbolt glyph, since it removes the slowing effect, does it also remove the coefficient reduction of the spell ?

dedmonwakeen · 09/25/08, 9:08 AM

Originally Posted by grayrest

There was a report upthread that said the proc didn't stack with the 30 minute buff.

The code won't let the buff stack with the proc..... I just wanted to make sure that the "benefit" from this talent (Focus Magic) was restricted to the Mage class for the purpose of comparative analysis.