Not necessarily. Percentage modifiers may ignore caps, but the cap would still apply to a stat derived from ratings only. I believe Dodge or some other rogue mitigation stat works in that way. Dodge rating calculates out to be capped at some diminishing return, but percentage bonuses are added after the DR cap. So, without evidence I wouldn't say that crit is not in some way capped.

The gains from talents and raid buffs are what push us over the 50% mark in most cases, and these are flat percentage modifiers. Does spirit convert to crit rating? A good test then would be somehow stacking an obscene amount of spirit along with crit rating, at the expense of everything else, to try to get a calculated percentage well above 50%.

There is a difference between a cap and diminishing returns. Crit rating (and other offensive stats) do not suffer from diminishing returns. If you would like to see this for yourself, it should be fairly simple to demonstrate using your character sheet in game that crit rating and spell crit change are linearly related. The only cap on crit is when every attack is a critical strike.

It would be somewhat interesting to see a situation where a mage was at 100% crit rate consistently and see if one could generate a non-crit on a boss level target given what we've observed about crit deflation. Then see if a 102% crit rate returns the same result.

But the question of what the cap on crit is is silly. It's 100%, or very close thereabouts depending on when reduction effects are applied. We observed this during one of the PTRs in BC when the stat stacking bug came into play. You could get your paper doll crit rate to display above 100%, and nobody was observing any non-crits that I heard of.

This could probably be pretty easily tested on Loatheb. That's a boss level mob with a predictable and consistent crit buff. I think my paper doll has only displayed up to 98% crit, but with some gem swapping just for testing purposes I'm sure it can be done.

Those are flat percentage modifiers. Loatheb's debuff adds 50% crit, not a few thousand critical strike rating. As is the case for some other stats where diminishing returns apply, percentage bonuses are not subject to DR -- only rating is affected. A paper doll reading 100% crit means nothing in that situation, unless the underlying crit rating was high enough that hidden DR would bring the effective crit rate below 100% as an above poster states.

It would be hard to tell the difference between DR and a cap if the DR are severe enough.

All the current DRs (except for chance for enemies to miss you from defense) in WoW are reflected in paper doll percentage numbers. See large parts of http://elitistjerks.com/f31/t29453-c...gs_level_80_a/ for details.

An immediate recast of the HS pyro would be the highest dps.

Pyroblast causes a global cooldown. In a perfect situation you, the stellar mage with low latency, would be able to make that split second decision and recast pyroblast, but it's not the end of the world if you start casting fb and just tag it at the end of the next fb - certainly not worth cancelling your fb cast.

OK, that's what I thought but I wasn't sure. After some practice on the dummies I discovered that the 4piece proc can occur quite a few times in a row. Just now I had the opportunity to cast 4 instant pyros in a row.

Ideally a mod like scorchio(dotimer,nugrunning, etc) will show your pyro procs still up(4t8) when you cast your current instant pyro, giving you its GCD to decide to pyro or to start casting something else. Pyro>LB.

Scorch dropping is @2k dps loss vs fireball chaining so assuming no scorch glyph might want to cast it once rather then 5x even with a pyro proc but I am not as sure on how close the math is between the two. I would recommend using pyro procs from scorch stacking before getting to 5 generally.

We're going mimiron hardmode this week on 10man and I was wondering if it was worth picking up molten shields instead of 2 points in playing with fire?

Not really as the majority of fire damage is avoidable except Heat Wave and as an AoE spell, it will not trigger a reflect.

What are the relative stat values rawr is giving you now that you have 4 piece?

If it's dramatically different I'd be curious as well. I too picked up 4 piece last night and noticed it went off quite a bit actually.

Another mod which worked great for tracking hot streaks btw is debuff filter. Icon shows up for your pyro, and it won't disappear if you have chained another. Works for me anyways.

When I rerun rawr with my new setup (4 piece, offpiece is chest) my relative stat values for crit and haste are exactly the same 1.53

Optimization didn't have me regem anything from haste to crit.

In a raid environment you have alot more crit due to raid buffs/debuffs which gives the opportunity for more 4pc "procs" making it much more valuable then it would be against a target dummy. This also means the more crit you have on gear the more dps the 4pc bonus gives you which results in the jump you see in rawr on relative values. Once you get 4pc crit will be a better stat then haste in most cases until you upgrade your gear to the point that your total haste drops very low. Notice in the BiS list that Enthorn made the amount of haste is very low compared to crit and spellpower, ~400. That is the only reason it ends up being high in relative value again.

Not that Mana is really a concern for FFB mages, but there are released 3.2 PTR notes:

WoW -> Test Realm Patch Notes

at 3268 base mana, we have an absolute maximum benefit from this of 3268 * 0.01 / 2 * 5 = 81.7 MP5

That number will go down as you push ticks back with successive crits, but under absolutely ideal conditions for the talent its worth 81.7 in this form.

Except with the replenishment nerf of 20%, fire-based mages have less mana regen in 3.2 than they do in 3.1.

Example numbers from XT this past week for me: I'd have gained 4096 from emp fire, but simultaneously lost 5180 from replenishment nerf.

True indeed about replenishment.

Something else I didn't think of initially is that AoE effects will benefit from this greatly. A flamestrike that generates 15 ignite debuffs will be free after four seconds. Not that this really should be considered to be an incredible buff, but it will reduce AoE downtime slightly.


Edit: Doing some quick simulation on this (someone want to check my code? http://zaldinar.bounceme.net/sims/ignitemana/ignite.txt ), the pushback effect is a bit ridiculous from that perfect world 81.7 value. Here's a sample output of the script, which simulates a weeks worth of casting a normal FFB rotation and calculates the MP5 return:

http://zaldinar.bounceme.net/sims/ignitemana/output.txt

So in normal crit rates we can expect somewhere around 18 to 31 MP5 out of this. It scales negatively with crit along with the usage of burnout, as opposed to MoE that scales positively.

Unless this is accompanied by a change to ignite that makes it not push back ticks, I don't forsee this being significant.

Double the numbers from my simulator, its still not that impressive. Research will tell how replenishment will work out under the new system, but its still penalized as your crit rate rises.

So double the numbers. I'll gain 3012 mana over a 8 minute 20 second fight. 30.1mp5.

That's the gain from a point in Arcane Concentration.


Are your mana issues gone when you add one point to Arcane Concentration to go back to 5/5 rather than the 4/5 often taken lately to put points deeper in Arcane?

Somewhat OT to the 3.2 discussion, but, if you're taking 4/5 Arcane Concentration you're better of taking 3/5 and another point in Meditation (assuming you take 2/2 Absorbtion).

If you only have 4 points in the third tier (3/3 spell impact and 1/1 focus magic), you need 6 points in the second tier to reach the fourth. This is why many builds have 2/2 MA and 4/5 AC. The fourth point, unlike the fifth, is generally a matter of Concentration vs Student of the Mind, not Concentration vs Meditation.

Edit in response to below: Ah, sorry, yes, you're right. Kyth has, for some reason, decided basically to move a point from Meditation to SotM instead of Concentration to SotM. Meditation is definitely better than Concentration for most common values of haste and spirit.

For Kyth's amount of haste (including +8% raid buffed haste), a typical rotation (time between scorches) would be 14 spells cast (3x LB, 8x FB, 2x Pyro, 1x Scorch) over 27.0842 seconds. For Kyth's spirit (not even including raid buffs to spirit), 52.7mp5 from 1/3 Meditation over that period of time would be 285.47 mana gained. Contrast with 2% proc rate for a point of Concentration over 14 spells, at an average cost of 629.357 mana each, would be 0.28 procs for 176.22 mana saved.

If you aren't taking one point of SotM then yes, I was talking about having one point in SotM basically Kyth's spec.

My spec was a quick "oh shoot, that's right, I need MA" misclicked respec before Algalon and since I haven't raided since then, I haven't bothered to fix it.

You are correct about where the points ought to be.