Please keep in mind that the default rawr settings assume a 300 second fight (5 mins, ie. short to mid length), zero movement, no target switches, and a long list of other things. Changing any of these may have an effect on your result. For example, fire suffers greatly on target changes due to ignite mechanics.

The "fire vs arcane" debate can't be solved by comparing two artificial numbers like this. Also, if anyone tried to write out all the detailed math derivations explaining why one was better than the other, it would be like a log of Rawr's computations. It wouldn't fit even in a 20 page document.

Just incase you forgot or have been busy or something, the fireball glyph still shows as 5% crit on rawr instead of the reduced cast time, that might change things a bit also i guess.

Enable 3.3.3 mode, it's changed. The tooltip will not change, the effect will.

You are right.
A significant part of ignite damage can be lost when the target dies. My example did not include this damage so the results are inaccurate.
Also I agree Marrowgard is not the best case to study fire single target DPS, even when focusing on the boss, because of accidental living bomb damage on bonespikes.

To try to answer your question I will use another example: a normal mode Festergut log

TOTALS:
Damage done: 2,205,679
Fire critical damage: 1,354,489
Expected Ignite damage: 541,796
Recorded Ignite damage: 446,244
Missing Ignite damage: 95,552

The 3 components of missing Ignite damage are:
- Resisted Ignite
- Dead Ignite (i.e. lost when boss died)
- Munched Ignite

Total resisted Ignite appears directly on log and equals 23,559

To find out the dead Ignite value we need to look at the original log file. In particular: the sequence between the last fully expired Ignite and the boss death. This sequence appears between the last two SPELL_AURA_REMOVED Ignite tags (in this example, it is the last 21 seconds of the fight).
Ignite munching also appears during this sequence, part of it can be identified by looking at the ignite ticks values. Dead ignite can also be munched but it doesn’t really matter here: it’s lost damage.

In this sequence:
Fire critical damage: 147,811
Expected Ignite damage: 59,124
Recorded Ignite damage: 24,768
Missing Ignite damage: 34,356
Munched Ignite: 2,214 (identified)
Resisted Ignite: 1171
Dead Ignite (missing - munched - resisted) = 30,971

Now with total resisted and dead Ignite, we get:
Total munched Ignite = 41,022

Damage loss as % of total damage done:
Total resisted ignite: 23,559 = 1.02%
Dead Ignite: 30,971 = 1.35%
Total munched Ignite: 41,022 = 1.78%

This is just an example, the values will always change.
I did not mean Rawr fire modeling should include the parameters that make Ignite so... unpredictable.
But people should know that with a 40% ignite coefficient the model might be a bit optimistic, which is another reason not to compare it to Arcane.

You don't lose ignite damage on target switches, except when that target switching involves the target dying. Swapping from Putricide to green or orange won't 'lose' you ignite damage. The damage still ticks on Putricide. Furthermore, rolling ignites doesn't mean ignite damage is increasing. Rolling of ignites just means that the ignite wasn't able to tick due to applications of other ignites (from additional critical hits), so the damage is added to a new ignite. You aren't gaining ignite damage by letting it roll.

What you do risk is losing damage due to ignite rolling and munching a huge ignite tick. IE, let's say it rolls up to 45k per tick (which is very, very high for a fight like Putricide -- 45k per tick is more common on a fight with inflated damage, such as Hodir). Once it reaches 45k per tick, you run the risk of a tick being 'eaten' and replaced by a new tick: like one for 4k or whatever.

From my calculations though, normal ignite damage loss (due to munching) runs in the range of 0% to 2% of total damage done, but is usually more close to 0% than 2%. The percentage is much higher when taken out of ignite damage, but that's not really relevant, since ignite is only a small part of your total damage.

Other classes suffer as well from target switching too -- think of DoT classes. But really, ignite damage is not the reason Fire suffers on target switching. The reason Fire is bad on target switching is because of upfront damage vs damage spread out, travel time, RNG, etc. Living Bomb, Pyroblasts, etc. are all a factor in your "total damage." When you're only able to get maybe one or two fireballs (which may not even be critical hits) onto a target before it dies, that's why target switching is bad for Fire, not because of ignite damage loss.

I can't seem to come up with a filter in wowhead to import the LK weapons into rawr. Every time I try it bombs out before it is done and nothing gets stored.

Does anyone have a working filter or a work around to get the weapons in the cache?

I've even tried to just use a name filter for one weapon, but to no avail.

Tools -> Edit Items -> Add

Get the item ID from wowhead or the armory, type it in the box.

If you have them checked on the buffs tab, and the talents selected, it is valuing them appropriately. If you're still thinking it's giving you 17% hit, you're probably simply not realizing that that 17% includes talents/buffs.

I have the latest Rawr 2.3.14 and i've used this quite a bit in the past (before my 3 month break). For some reason when I set everything up and it looks perfect, i then save it. Once I open it up, i looks fine and i'll flip through the slots by selecting SLOT: GEAR HEAD. I'll see my gear in the 5th slot (for example). If I switch it to Neck, the neck looks fine, but if I switch it back to Head again, it rates it as 2nd!?!? Does this all the time. If I keep doing it, the look of the graph's even change. Is this a bug or am I missing a setting? This is from a fresh download. Thanks!

There is a bug in 2.3.14 if you're using a 32 bit operating system, I'm not sure if that applies in your case. If so either use an older version or Rawr 3 until next release.

You can copy over the xml files in the Data folder.