Dying too often is not going to help your raid. And many 25 man raids nowadays are much more unforgiving. You may get by with one or two deaths if you already outgear the encounter, or if your raid is just good. But if you are doing progression content, or content you only recently killed, its highly likely that the margin of error is going to be low. While you might be able to shrug it off as just one unlucky death here and there due to being instant killed (healers couldnt do anything). But did your low hp contribute to that?

And dying once or twice might not be a big deal. But what if your death also contributed to the raid wiping because not enough dps, or not enough AOE, not enough cc, etc ? Throughout the 25 man raids, when you are progressing through them, AOE dmg tends to be high.

Take even Gruul for example. Not a difficult fight now. But when you are starting on him, people tend to clump and do damage to each other. That extra bit of hp might mke the difference between you dying and you surviving if someone just happens to run too near to you. And Gruul is on a timer. Too many DPS dead = too many growths = the raid wipes. Sure, when you are majorly overgeared and can down him in 7 to 9 growths, it probably doesnt matter even if someone killed himself early. But you aren't going to be doing that at first when you frist attempt him.

Same thing for many other fights in the T5 instances. Also, if you are relying on your healers to heal you, you might be putting unnecessary stress on them. Especially if you hp is so low any kind of damage puts you a hair whisker away from being dead. Mages are never supposed to be top priority in terms of heals. If you need a healer to keep his eye on you all the time and babysit your hp, after a while, the healers will get fed up.

I think its much more emphasized in my raid because my raid leader is also a healer. I must confess that initially, I was very against this myself also. But ask yourself this. Already, we are trying our hardest to do good dps, aoe, cc and justify our raid spot. If we still want to add stress to healers on top of that and making dying often a feature of our class, then I think it would make raid leaders even more wary of taking more mages in raids.

They would be thinking "hmmm, I get a lock, I get good dps, AOE, survivability, dun need healers to babysit him, dun need any special group arrangements. If I take this other mage, yeah. I get maybe good dps (if he doesnt die within 3 minutes of the fight), but I need to have healers babysit heal him, then I need to maybe try and put him in a SP group too. Wow, so much effort, I think I will just go with a lock."

A little less DPS for some more survivability is not a bad thing. Right now, I actually wear one piece of S3 gear for raiding permanently. And sometimes, on certain fights, I may even put on more.

I've recently really noticed this. I used to be a kara/gruul man, and it never really seemed to matter (especially given my blue boots and bracers and certain other items, which have close to zero stamina). However, I just moved into a fast progression guild (in our first three raids, maulgar, gruul, mag, hydross, VR, solarion down) and all of a sudden i've noticed the hp hit.

Although i can't take any specific examples - in ZA where we are on dragonhawk, if i get hit by flamebreath + adds (im on AoE duty with a warlock) then thats me dead, and its a raid wipe. So instead of wearing spellfire, I actually replace the whole lot with arena gear. Now all of a sudden I feel that arena gear hasn't been a waste at all, because it still has decent damage but has the huge amounts of stamina I need for certain fights.

At the cost of spell hit and regen(spirit). While the spirit can be argued down, especially without arcane meditation, the way i see it, the difference between tiered gear and arena gear is mana-based endurance (spirit/mp5) and spell hit (vs of course stam/armour/resil).

While for your specific example (aoe adds, not 73) arena gear does have its benefits over true PVE gear, it must be noted that on a straight nuke kind of fight that features a lot of aoe damage or burst raid damage, wearing PVP gear to stay alive is.. well, the in game equivalent of driving a tank to work because you want to survive a car crash. Sure, it works, but your fellow commuters and probably you yourself will benefit more from you learning to play the game with a focus on survivability in the intended gear.

That said, i find it absurd that PVE gear is SO bad in arena, a playing field based around survivability, and then bliz also see it fit to make season 2/3 gear WORKABLE for learning new encounters where survivability is an issue. Specific examples of this are ZA fights while learning, and najentus/archimonde. Thats very off topic though, sorry. First post also

Don't get me wrong - it's not like im saying that my PVE gear was a waste :P

But I do feel it's handy that I have the option of swapping out my gear for more stamina, usually on AoE (especially for example, on solarion and dragonhawk, which were our last two raids)
I also feel its rediculous just how bad some PVE gear can be, especially in terms of the ones which are loaded with completely useless spirit instead of say (warlocks - stamina?, esp with the 0.7 i-level cost). Considering my blue boots and wrists atm, which are completely DPS and nothing else, it's embarassing that silk vindicators are actually *better* dps and stamina than they are, sort of inclining me to do PVP to get them.

Yeah, I have the voodoo-woven belt, and t4 gloves, so the only thing I would really need is the spellfire robe, as well as some new shoulders. As I suspected, I'm going to have to respec my tailoring to make going fire worthwhile at my gear level; I'm just hesitant to do so, as I'd hate to spend all that gold and then change my mind...

But you are trying to compare rare gear with epic gear which are nearly same ilevel as BT epics. Rare gear may have more stats distributed towards DPS, but with the different in quality between rare and epic, plus the massive difference in ilevels, there really shouldnt even be a comparision between the two.

Well, T4, T5 doesnt have that much stamina, to be honest. But they have a ton more spirit, int than spellfire or spellstrike set. And I think they are still much more suited for heavy raiding than spellfire and spellstrike. That extra bit of DPS is not worth the big reduction in stamina, int and even spirit that tiered gear gives. The comparision is a lot closer when we compare arena gear with Tiered gear. In that area, I think swopping in one or two pieces is fine.

But I really wonder if a mage can raid well into T5 instances without replacing his/her spellfire set and spellstrike set. You could proably keep one or two pieces, but at some stage, you will start replacing most of them very quickly.

I had Spellfire set until I got the Vashj Robe and Belt of Blasting, so 6/6 SSC and 4/5 TK down. Replaced my Spellstrike set with T5 Pants and Cowl of the Grand Engineer, well into T5 content. I was pretty successful with my crafted gear, but I will admit the survivability was very low, with about 6-6.5k Health unbuffed.

I reworked my table a little bit.

Now, the Damage increase is calculated in percent.
Added the other Crit-Lines (150% (unskilled) and 200% (Frost-Skilled))
I left out the 175% from the Arcane tree. No need for another colored line. It's exactly between the two other lines.

Got a question:

Rep rings like the Kara one and Hyjal one, if I enchant the first version, when I upgrade it, does the enchant carry over? or do I need to re-enchant it again.

I have a question regarding [Band of the Eternal Sage]. Did anyone make analysis of how much the proc is worth in spell damage and how does the ring compare to others?

At the moment I sport [Mana Attuned Band] and [Ring of Ancient Knowledge]. Another notable ring is [Ring of Captured Storms] although I don't have it. I have the Band of the Eternal Sage however and I simply wonder which would be best.

The proc rate on the ring is 10%, lasts 15 seconds (as opposed to what says in the tooltip) and has a 45 second cooldown after the effect ends (60 seconds total).

Probably % increase in per-cast damage (or overall spell dps--the percentages should be equal) from having 0 rating, which, in all truth, I don't find very useful, because who has 0 rating? Gaining more of a stat changes the curves.

Ah thanks. Yes going from 0 to x instead of x to x+10 does make the visual less meaningfull.

It does not carry over.

Any observation? =)

No, it does not. The Increase in damage is linear. The only non-linear increase (haste pre-patch 2.1 (?)) was fixed by Blizz and is now linear too. And the percent increase in per-cast and dps is the same for theoratical analysis with constant Parameters.

Let me be clearer: the graph looks at percentage increase in damage or DPS compared to some single base value. I suggest it is better to look at percentage increase over the current value. The graph as it is gives us only a little information--it tells us when other stats outperform a commensurate amount of hit rating due to the cap, and it tells us which stats are "best," as they have the highest slopes. Everything else on the graph is pretty meaningless.

Showing how stacking a stat reduces the percentage benefit of the next +X of that stat can be useful.

Given how these ratios can fluctuate based on spec and gear, I would only emphasize that the overall benefit of the items can fluctuate.

Haste buffs should not affect the comparison, however, as these are known to stack multiplicatively (that is, they don't add like extra haste and thus don't decrease the value of haste compared to other stats).

That is why I listed exactly what stats i was getting the ratios from.

And getting a bloodlust/IV during the period where the ring can proc will incease the overall value of the Band of the Eternal sage since it will have less proc downtime. Just so everyone understands how it works, the buff duration lasts 15 seconds, afterwhich it is followed by a 45 second cooldown. After the cooldown, each of your spells have a 10% chance to proc the ring...thus 5 casts will be your average time it takes to have the ring proc again. So any sort of haste incrase during this time period will increase the value of the ring.

Granted.

...10 casts, I think you mean?

10% proc rate with 5 casts would give me a 50% chance of getting the proc during that time. It could take more or less casts...it's simply an average.

Using 10 casts would be another valid way, but it would lower the value of the Band even more when compared to the ring.

I'm afraid you're mistaken.

Chance it procs on the first cast: 10%
Chance it procs on the second cast: 90%*10% = 9%
Third cast: 8.1%
Fourth: 7.29%
Fifth: 6.561%

And so on. You must consider that for it to proc on the second cast, it cannot have procced on the first cast, etc.

The expected number of casts to proc is going to be 1/.1 = 10. If you must, Wikipedia's page on the negative binomial distribution can help; the negative binomial describes the number of failures expected for the given number of successes required. That number of failures, for a probability of success (proc chance) p and number of successes (procs) r is given by r*(1-p)/p = r*(1/p-1).

Since that is the number of failures, we add the number of successs (which is just r) to get r/p. r = 1, p = .1, in this case.

ah, I see. I always did fall asleep in my college stats class =)

This change...which I'll update sometime on the chart, would give the Ring of Ancient Knowledge an even larger lead over the Band as it would add an additional 14 seconds to the rotation (assuming you're only spamming fireball). Any encounter over 2 minutes would benefit more using the Ring of Ancient Knowledge as far as a pure dps increase.