Glyph of Exorcism bugged or just sucky?

[Ununoctium]

"Your Exorcism causes an additional 20% of its damage over 6 sec."
I never really used it, cause the first time I used it, the dots didn't do the 20% it said. I never thought more about that until I read on Elitist Jerks that it is supposed to be better than Glyph of Crusader Strike.
So I went to a dummy and tested it. It's 20% of the base damage, not the actual damage. It doesn't matter how big a crit you get, you still get the sucky 404 ticks (or 525 with inquisition). How can Elitist Jerks see that and say "Hey, this glyph has a cool tooltip, so it must be better than Glyph of Crusader Strike"?

[Ryme]

They didn't, what they actually did was realise this on day one and math everything out and get to the correct conclusion.

The actual difference between them often falls into acceptable margin of error anyway, even more so after 4.0.6.

[Ununoctium]

They didn't, what they actually did was realise this on day one and math everything out and get to the correct conclusion.

The actual difference between them often falls into acceptable margin of error anyway, even more so after 4.0.6.

I just read today on Elitist Jerks forum that Exorcism is better than Crusader Strike, and that they earlier on said the opposite. That's just not right. Sure Exorcism is better at very low gear level, but already CS is proving better for me, and the better gear you get, the worse Exorcism glyph becomes, as it doesn't scale at all. Also, with CS being buffed in the next patch, the CS glyph will become even more desirable. And don't forget that Glyph of Exorcism doesn't stack if you cast exorcism before the last dot ran out.

I really just don't get how the tooltip can say "20% of it's damage" when it's more like 7%. They should really clarify that - or change the glyph mechanic.

[Evilmonkey]

if it was actually doing what the tooltip describes then it would be better, since its only doing 20% of the base damage, its just sucky. all of that of course unless major changes have happened since i was naked doing the same damage with exo glyph as full geared.

[Ununoctium]

if it was actually doing what the tooltip describes then it would be better, since its only doing 20% of the base damage, its just sucky. all of that of course unless major changes have happened since i was naked doing the same damage with exo glyph as full geared.

Exactly my point, hence my last statement.

[Fektoer]

I just read today on Elitist Jerks forum that Exorcism is better than Crusader Strike, and that they earlier on said the opposite. That's just not right. Sure Exorcism is better at very low gear level, but already CS is proving better for me, and the better gear you get, the worse Exorcism glyph becomes, as it doesn't scale at all. Also, with CS being buffed in the next patch, the CS glyph will become even more desirable. And don't forget that Glyph of Exorcism doesn't stack if you cast exorcism before the last dot ran out.

I really just don't get how the tooltip can say "20% of it's damage" when it's more like 7%. They should really clarify that - or change the glyph mechanic.

The difference in damage between glyph of exo and glyph of cs is -that- small that any deviation could very well be RNG. So yeah one time you may be doing 12k with glyph of exo and 12.1k with cs on a dummy, the other day you get one more crit on TV and your latency is a bit higher and suddenly exo pulls ahead. In a perfect world (perfect rotation, no RNG spikes, 0 latency, exo is better at the moment.
You're right that CS might pull ahead next patch, but then again the EJ post is about this patch. It's of no use to speculate what a new patch might bring since it's still on the PTR, so everything can change.

[Ununoctium]

The difference in damage between glyph of exo and glyph of cs is -that- small that any deviation could very well be RNG. So yeah one time you may be doing 12k with glyph of exo and 12.1k with cs on a dummy, the other day you get one more crit on TV and your latency is a bit higher and suddenly exo pulls ahead. In a perfect world (perfect rotation, no RNG spikes, 0 latency, exo is better at the moment.
You're right that CS might pull ahead next patch, but then again the EJ post is about this patch. It's of no use to speculate what a new patch might bring since it's still on the PTR, so everything can change.

Still I see no reason using a glyph that scales with... Nothing... And if you look at raw numbers, CS is ahead. CS is doing almost twice the damage of Exo overall on the sheet, and a 7-8% (at my gear level, which is 351 - less at higher gear levels) buff to exo (that doesn't stack) just cannot compete with 5% crit on your main ability.

[Rivin]

Still I see no reason using a glyph that scales with... Nothing... And if you look at raw numbers, CS is ahead. CS is doing almost twice the damage of Exo overall on the sheet, and a 7-8% (at my gear level, which is 351 - less at higher gear levels) buff to exo (that doesn't stack) just cannot compete with 5% crit on your main ability.

The question is, have you actually done the math to show that 5% crit to CS is better than 8% damage to Exorcism? Because you can be assured that if EJ says something, they did the math. They're not infallible, of course, but they generally have mathematical proof behind them.

[belfpala]

The question is, have you actually done the math to show that 5% crit to CS is better than 8% damage to Exorcism? Because you can be assured that if EJ says something, they did the math. They're not infallible, of course, but they generally have mathematical proof behind them.

You don't have to use a simulation. It's easy to test directly, which is nice because it accounts for your gear. Just equip the exo glyph, go do a nice long dummy session (5 minutes or so), then:

1. Note the damage actually done by the exo glyph.
2. Look at your CS damage and convert it into non-crit damage. That is... total non-crit damage + crit damage/2.
3. Multiply the number generated in step 2 by 0.05. This gives the damage you'd expect to see from the CS glyph.
4. Compare the results of step 1 and step 3.

Last time I did this, the Exo glyph was about 40% ahead of the CS glyph (but still small contributions from both). But that was when my ilvl was like 335.

[Deleted]

You don't have to use a simulation. It's easy to test directly, which is nice because it accounts for your gear. Just equip the exo glyph, go do a nice long dummy session (5 minutes or so), then:

1. Note the damage actually done by the exo glyph.
2. Look at your CS damage and convert it into non-crit damage. That is... total non-crit damage + crit damage/2.
3. Multiply the number generated in step 2 by 0.05. This gives the damage you'd expect to see from the CS glyph.
4. Compare the results of step 1 and step 3.

Last time I did this, the Exo glyph was about 40% ahead of the CS glyph (but still small contributions from both). But that was when my ilvl was like 335.

This is a pretty crude (= flawed) way to compare the two. You won't get an accurate picture of the damage contribution without being in an actual raid where both raid buffs / debuff scaling is accounted for.

Best way to do it, is run with exo glyph for 1 reset, and check each boss to see how much damage the glyph contributed, and then compare it to what a theoretical extra 5% crit on crusader strike would have granted you instead on the same boss.

On a fight like atramedes and nefarian, where you are likely to spend some time using exorcism during airphases, it is expected that exorcism glyph may beat out crusader strike in the majority of cases, however this might not be the same on a fight like chogall.

[Ciah]

Non-raid buffed trials on a dummy for 5 minutes tells you the following: absolutely fucking nothing.

I'd try going the course of a while 4 hour raid with exo glyph or even multiple raids before comparing, and look at each fight individually. I'd bet on the fights with bad melee uptime, exo glyph will pull much farther ahead. But, CS scales with gear. So it will vary depending on fight and gearing, the best way to know is to test it in the long-term.