Asphyxiated

I have a question about haste from gear.

I did my own theorycrafting and it seems getting all the haste gear will yeild more DPS than the more dmg and crit in places, assuming still maxed hit.

My real question is: has any one else done any theorycrafting on haste? If so, is it worth it? From what I've seen you need ALL the haste gear (including staff from Illidan) to notice a difference in a 10~ minute fight. Anything less I would assume you won't notice it unless the fight is 15-20 minutes long, and fights that long are few and far between. I'm just wondering if gathering the peices would be worth it, or is it as I think: that its not worth it unless you have every peice of haste gear available. Thanks in advance

 

• Dochas

Well personally after doing some calculations with the mage spreadsheet haste is looking to be pretty close to spell damage as far as value per point, unlike crit which is closer to .4-.5 the value per point compared to spell damage. I have the ring and illidan staff already (which probably isn't better then arch sword/offhand but I could wait 3 months for that combo or go with the extra haste with the staff now) and plan on getting the crafted wrist/shoulders.

Waistwrap of Infinity doesn't seem to compare as well to Blasting or Anetheron's if your hit capped when you look at the comparison between Cuffs of Devastation vs crafted and T6 shoulders vs crafted (although the non-set options look a bit more appealing). As long as your hit capped already from other pieces in general its mostly a crit for haste trade which haste wins in an even ratio comparison.

The 4 pieces will make close to 10% haste. But honestly even if you only manage to shave .1 second off each cast you should see the results of an extra cast even in a 2 min fight. It will be more apparent with more haste obviously as you see yourself getting more fireballs off although the more haste obviously the more other stats you give up (mostly crit). If your only shaving .1 off then your only getting 2-3 more fireballs in a mid-length fight but with the 4 pieces I mentioned your looking at ~2.71 second fireballs. That ends up being ~8 more fireballs over the course of 4 mins of fireball spamming.

Lag still effects it though and shorter casts means more stopcasting and more room for errors or not catching it as fast as you could, hopefully their upcomming "fix" for stopcasting macros will actually be usefull so you can chaincast as fast as you should be able to.

 

Asphyxiated

Its just, we just downed Kael, we're looking to progress further so the way I see it, I'm at a cross-road: go for haste or more crit/spell dmg. Personally I'd like to avoid the illidan staff. I don't know how fast we can progress or if I'll have to quit at any time due to some random RL reason, and I would feel terrible for taking the staff and then have to quit soon after when some one else could have used it.


Thanks for the help, I may go for the off-set peices, think I might theorycraft a bit more before I make a final decision.

On another note: they're fixing stopcasting macros? I hadn't heard about that.


Got another question regarding Fire vs. Frost.

I'm 10/48/3, 1050~ fire dmg, capped hit, 32%~ crit (all unbuffed).

Does anyone know what kind of Frost dmg and crit I would need to get to match the DPS I do with fire?

 

Kavan

I finally have enough data to give an answer to this question with at least some certainty. According to testing pre expansion it was established that each level above caster level contributes 8 resist that cannot be countered for partially resistable spells. For +3 level mobs this resulted in 24 resist or equivalently to 6% damage lost to partial resists.

The question is how does this extend to level 70. Is it still 8 resist per level or is it 2% per level? The equivalent of 6% at level 70 is 28 resist. Preliminary results brought no conclusion as averages remained in the range roughly around 26.

Here is a collection of data collected from bosses with sample size of at least 1000 (for arcane spells), all of which were collected at the same level of penetration. For collection purposes I used the ResistTracker mod which collects data in real time taking into account current buffs/debuffs on the mob and player.

Collected data:

Hydross the Unstable: N = 3731, 29.52 resist
Leotheras the Blind: N = 2033, 29.34 resist
Morogrim Tidewalker: N = 4469, 27.72 resist
The Lurker Below: N = 2227, 28.67 resist
Gruul the Dragonkiller: N = 6749, 28.37 resist
Magtheridon: N = 3762, 29.08 resist
Prince Malchezaar: N = 5856, 29.47 resist
Shade of Aran: N = 2404, 28.16 resist
Nightbane: N = 2459, 29.60 resist
Void Reaver: N = 4913, 28.33 resist

Total: N = 38603, 28.77 resist

At this sample size we're still in the +/- 1 resist range, but it is enough to disambiguate 24 from 28. I think it is safe to assume at this point that partial resists are 2% per level difference.

 

Hate Monkey

Were you testing both AM and AB? Also your results are not totally correct in that you need to collect a sample size of much, much greater size, uptime of CoS/E, and of various Spell Penetration amounts.

You'll see that when I get enough information to release my Spell Penetration stats work that there are some anomalies that occur with some bosses, and some work into dps loss from when CoE/S drops off a boss.

Though from the work I've done so far, that 2% per level seems on a track of something else, not quite sure the correlation yet.

 

Kavan

It includes both AM and AB. All the samples have the same penetration ammount, the mod will track with and w/o COS as different parameters. I only posted the value for the combination of parameters under which I have the most data collected.

I don't understand what you mean by it not being totally correct. For this sample size the error range is under 1 resist for 95% confidence interval. Also I don't see what you would gain by collecting data for different penetration amounts. It has already been established that you can't go under level-based resists and it is obious that the bosses I listed have 0 natural resists. This is also confirmed by checking other parameter combinations for the same bosses, all are around the 28 value, -resist plays no effect at all.

 

paedur

A sidenote question about resists, I don't think I'll ever truly understand the hit-rating vs. boss/mobs. It seems to differ allover the place and just when you "should be" safe with enough +hit to see hardly no partial or full resists (except that tiny 1%), something like this happens.

Last night we did heroic ramparts, mobs in the elite 71-72 range and I got resists allover the place, even up to a point where 3k+ fireball hit for 1k or less (I'm 11/47/3). Maybe those specific mobs have big fire-res or something but I was frustrated to say the least. It was a quick run so I didn't think about changing gear at all before, silently thinking " ah nvm, I'll go with this +hit since it'll good enough...".

Had 120 in +hit, which I easily thought would erradicate any sort of resist in that amount, but nope. I usually go with less +hit for hero's/trash, but this time I didn't.

I just got the feel that having too much +hit for 1-2 levels above, will be to much 'over the top', actually making you worse than if I'd put on less.

Is this something that just happens sometimes, resist-spree to the fullest on trashmobs, or is there actually some crafting behind it that one should keep in mind? Like don't go overboard with +hit for less than 73+/??-lvl's ?
Another example to clarify. Say you, lvl 70, fight a lvl 70-mob with 16% hit. It should really be a small chance you'd get a resist right ? OR, would those %-hit over the limit needed for a 70-mob, actually give you higher risk to get resists ?

Would be a 180 turn from what i've gathered from reading all threads about it but, definite 'No' would be nice to get so I can just discard last night as a weird anomaly =)

edit* It was partial resists I got loads off btw, forgot to clarify that

 

• TGS

I might be off base here, but I thought +hit only affected the initial hit chance, not partial?

For fighting a level 70 mob, you only need a few percent of +hit (I forget the % exactly, but 4 seems to ring a bell), so most of the 16% is wasted. Everything will have at least a 1% chance to resist no matter what however.

Edit: 3% is the max hit you need against level 70 mobs (at 70)

 

Docjowles

For Fireball anyway, +hit only mitigates your chance of a full resist as I understand it. The chance to miss is only affected by your level vs. the level of your target.

Some mobs (perhaps the ones in ramparts) also have some amount of fire resistance. Spell hit has no effect on this. Spell PENETRATION, however, does. This resistance leads to "partial" resists where your 3k fireball crit only hits for 1k.

People say that spell penetration is worthless around here because it's a raid-focused board. In any raid environment you're likely to have a warlock casting Curse of the Elemnts which largely negates a mob's resistances. But in specific situations (like heroic ramparts without a warlock) rocking some spell penetration gear can be helpful.

 

paedur

Hmm yea , when I read TGS answer abit earlier suddenly spell penetration hit me as maybe the culprit since I changed some gear and now have less of it. And when I read yours it feels kinda clear that it prolly was that which gave me those resists, doh
On top of that I haven't done any heroic for some months and by reading this "raid-focused board" several times a day, and only raid, I guess I got kinda dulled in anything concerning outside of raid, ty both for your answers.

 

Pheroz

This is what I pulled up from search.

Partial resist cannot be countered with +spell hit.

IF a mob has specific resistances, you will see that by getting partial resists. Spell penetration will reduce those resistances, which will result in a decrease in the amount of damage lost to partial resists - you will get fewer resists.

Also, Raid Boss Level mobs also cause partial resists. This is not due to having specific resistances, rather is often explained through a level difference between the caster and the boss. The current theory is that penetration does not reduce these resists. I know Hate Monkey was specifically looking into this theory, but as it stands, that is the generally accepted coventional wisdom.

Spell penetration would have helped you in Ramparts, becuase you were fighting an nps that has specific actual resistances.

 

paedur

Yes, even though it wasn't as explicit on Nazan at all, but way more on the mobs and I afterwards couldn't find if the trash-mobs had any spell-resistance, I got blown off track totally and just forgot to think abit further on.

I've read up on spell penetration but since it's generally discarded as worthless for raids and I've never accounted any use for it nor bothered to notice if a piece has it even( think I've had maybe 30 tops, now 10), still feels weird that plain trash in one instance has it but ...
The ones you listed atleast has some logic on why they would have resistance.

I'll just accept the facts and move on, penetration has some uses atleast.

 

Hate Monkey

I would like to see the proof you have showing that the bosses you showed have 0 natural resists. If that were the case, then CoE/S would have no affect on test, but the thing is they do.

I would say else wise to this, but as to a huge gap in all the data I've collect so far, it prevents me from saying it.

And the main reason I haven't released anything yet is that there are very few mages running around with more than 30 spell penetration, just due to the fact that "its widely known that you can't reduce partials to lower than 5%" being thrown around so much with very little relative proofing being done.

Once I collect enough data from mages wearing more than 30 spell pen, along the lines of 40-50 preferably, I might be able to release the tables and data.

I will say this though, of the few collects of 28 spell penetration, and 48 spell penetration set, and I mean very few, there is a small difference from what I have so far. But it is along the lines of your 2% resist per level theory that you posed.

 

Kavan

One of the first proofs showing that most bosses don't have natural resists was for Onyxia. This was done by comparison of resists in 1.8 and after -resist nerf in 1.9. It turned out that distribution of partial resists in 1.9 with and without COS was the same and it matched distribution of resists in 1.8 without COS (there was no penetration gear at the time; Onyxia with COS in 1.8 was vulnerable which indicates her natural resists were less than COS penetrated). To verify it was not just a problem with COS similar tests were done later with penetration items such as Eye of Moam. This was validated on other bosses as well.

In particular for the bosses I listed before here are some numbers for those where I have most data with different penetration values:

Hydross the Unstable:

30 penetration: N = 1083, 28.43 resist
50 penetration: N = 1615, 27.49 resist
118 penetration: N = 3603, 26.57 resist
138 penetration: N = 3731, 29.52 resist

Gruul the Dragonkiller:

30 penetration: N = 6836, 28.28 resist
118 penetration: N = 4280, 28.71 resist

Magtheridon:

30 penetration: N = 2099, 26.91 resist
118 penetration: N = 3762, 29.08 resist


This clearly shows that COS has no effect.

 

Prod

So why do several cloth shoulders in bt/hyjal have penetration?

 

Maledict

Erm, because Blizzard don't itemise for casters very well? It's the same reason Tier 5 Gloves are appallingly bad compared to Tier 4, it's the same reason there's only one caster item atthe tier 5 level yet 3 pairs of caster shoulders and 3 caster belts. It's the same reason that all my gear has more intellect on it than stamina.

Spell penetration does nothing. It's also incredibly easy to logically test - simply don't put Curse of Elements on a raid mob, and see if your amount of partial resists shoot up. They won't, at all. Now you can extrapolate two answers from this :

Does anyone really think that Blizzard would have coded an entirely new mechanic for spell penetration when they introduced it, such that it affecst the inbuilt level based resist rate and Curse of Elements / Shadows doesn't? Or is it more likely that they work the same and spell penetration on PvE gear is merely a mistake or part of their PvP / PvE gear cross over?

I'd love to be proven wrong on this one, and shown that spell penetration has some value in PvE, but so far no-one has ever shown the numbers for it doing this, and logically it shouldn't given CoElements / Shadows.

 

Vhad

The rest I agree with, but do you really think it's a design flaw that mage gear opts for Intellect? As an arcane mage I love it.

What!?

 

Sancus

Spiking Int as high as it is on the mage set gear is bad, not good, because even an Arcane Mage does not get sufficient value out of int to justify the itemlevel cost.

Most of the time, the items would be superior if they had less Int and more damage/crit. That's not to say that they should have NO Int, but the amount on them is simply too much and it is a waste of itemvalue.

<Vontre> I removed the cooldown on evo
<sancus> and what happened?
<Vontre> DPS went down rofl

 

Roywyn

You just said it yourself, intellect is nice for arcane specced mages - mind mastery, +5/10/15% intellect from gnome/BoK/talents, some little crit to synergise with TLC/Ashtongue/4pcT5, and some more mana to burn. It's a competitive stat for arcane specs when you check numbers/spreadsheets.

But for a fire or frost mage, intellect is basically wasted item budget. Okay, maybe not wasted, you your returns from intellect (and spirit for that matter) on gear are very lacking and any other stat investment (dmg/hit/crit/haste) would net you about four times as much benefit.

T5 is simply an arcane set, if you want it or not. It's stat heavy (i.e. intellect), the 2/5 bonus is required for competitive* use of AB, the 4/5 bonus benefits fast cast spells like AB spam and AB/AM/Sc rotations, and 4 of the pieces have no spell hit at all.

Arcane and non-arcane gearing is just very different when it comes to intellect, and tiered gear has always been loaded with it for whatever reason.
Pre-BC, I used frost damage greens over T1/T2 because they gave more spell power. Now, Spellfire/Spellstrike beat T4 for fire specs, and T5 barely breaks even for them, but at least it's not back to greens for us.

The thing is that you can't make gear that satisfies everyone. An arcane set will blow for frost/fire. A fire/frost set will be subpar for arcane.
And however you design a set, you can always remove one point of intellect/spirit to get more damage stats.
I don't think there will ever be a solution for everyone.

When you want to hear about really bad itemisation though, talk to shadow priests. Or feral druids.

Edit: But yes, spell penetration on tiered sets (and PvE gear in general, but it stings particularly in tiered gear) is dumb. I'm just afraid that the blizzard solution is "more resistant mobs".

 

Sancus

Nope, but you could make multiple sets. I mean, hybrids have them, there is no good reason that pure classes can't have them too, especially when there is a clear value and need for them.

<Vontre> I removed the cooldown on evo
<sancus> and what happened?
<Vontre> DPS went down rofl

 

◊ manly

Has anyone went and tried the latest PTR patch AM spam ? I went to test, for fun, just to see. I could be wrong here, but it felt really strange. First off, I did 3-4 round of test before a warlock decided DR BOOM was his. On the last test, I procced no less than 7 focus in a row. Or maybe it wasn't consuming it all the time. (kind of hard to tell with no mods installed). Anyway, much to my surprise, using the same character copy I had from quite a while ago, without 4pc t6, I did get 1600 unbuffed DPS. Here's the results I had: 1300, 1050, 1000, 1600. That just using the ring of spell haste, nothing else (ashtongue/tlc tho). I'll try and run more tests, but if anything, I was god damn surprised to see that kind of dps.

Log on with different model:
1- Create a character of the desired model. Log on/off.
2- At character selection screen, select your actual character; mouseover the new, desired model character, and hold down left click; hit enter and release left click at the same time.
bug Arcane Potency only applies to the first Arcane Missile bolt.

 

Tyrian

And an awesomely designed set at that - the blizz set designers really got it perfect here - on crafting a set that suits the needs of an arcane mage.

 

Logun

I got 4-5 procs in a row a couple of times in my testing today. I don't know if this is just a statistical fluke but it did seem much closer to the rolling procs than any time since they fixed it.

 

Krevan

While Spell Penetration doesn't do much in PvE it's not entirely useless. It might help a bit on some encounters, as, for example, Illidari Council (+all school resist aura) or Supremus for a fire mage (he has very high fire resistance). Possibly other bosses have resistances too.

Although I do agree that its use is very situational and thus it's a waste of item badget (just like +Spirit on mage tiers -- It's not entirely useless, but it's still a waste)

 

Shalkis

Before dishing out gold for Darkmoon Card: Crusade, I'd like to know whether any deep frost mage has compared it against The Lightning Capacitor or Quagmirran's Eye.

Granted, TLC shines in AoE and might perform better in fights with distractions, because you don't lose charges while running/sheeping/whatever. But does DM:Crusade clearly outshine TLC or Quag's Eye in a straight-up tank n' spank fight? I currently have 13.83% crit (unbuffed), but I do have Winter's Chill and Empowered Frostbolt, which would put me at 28.83% crit.