Imp. Fireball and Imp. Frostbolt damage coefficient changes

[Pantalaimon]

For that set of circumstances. But note where he says that other sources of mana regeneration would likely completely remove the difference, leaving the talent with its full DPS increase via casting speed change.

The problem I have here is that the argument is about the talent decreasing their DPM. As I see it, that means a mage that doesn't take the talent will receive more benefit out of things such as evocation, BoW, mana stream, mana gems and mana potions - since their spells are more efficient, the damage equivalent (how much pain they can deal with the mana restored) of, say, a major mana potion is larger. Even mana/5 will benefit a nontalented mage more. Slight, yes, but given the direction that blizzard seems to be taking the mage itemization (more regen, less bam) it's a valid point.

The math that was run ignored all that - indeed, even assuming it would work towards the talented mage's advantage.

But if running out of mana is at all a concern, I think the difference will actually increase. The only time the whole "DPS trumps all" argument holds weight is when both hypothetical mages can spend the exact same amount of time casting, and neither will run out of mana. For example your basic cast - evo - cast fight. If the fight ends before the less efficient mage runs oom, then he came out ahead. But if you did math for the situation quark posed using evocation, gems, etc. and they were still running oom towards the end then I'm pretty sure the nontalented mage will come out even more ahead.

Edit: I haven't done math, but the way they implemented this talent implies to me that there's some sort of +damage break-even point where the talent becomes absolutely useless. Has anyone done any back of the envelope math to see if that point is attainable given itemization in the near future?