The change to SoE hurts a bit, but it was a bit too good for a glyph. 1% crit to the whole raid seemed a bit much compaired to most of the other majors.

This cant be stressed enough. I only have 15-17% spell crit, and ED's uptime is clocked at around 23% with Shock and Awe. Of course you should run it through the sim, but +hit(until the cap) and +crit are so multi-beneficial, that their EP values skyrocket.


I do have a question though. I cant check, being at work, but I remember the cooldown for LvB being under 10 seconds. With the Flame Shock Glyph, and enough spell hit, wouldnt that give us essentially 100% ED uptime?

Correct me if I'm wrong, and I'm by no means saying that once we get LvB, spell crit would be worthless, but it's interesting to say the least.

The apparent PPM is higher on live due to:
1) haste increasing the PPM
2) WF causing extra procs
3) Lava Lash causing extra procs
4) Stormstrike causing extra procs

It does not mean that the 10PPM is incorrect. The 10PPM is effectively the base PPM, and plenty of factors are contributing to increasing the apparent PPM.
This has been extensively commented on over the last 5,10,15 pages or so.

I know this is a few posts back, but how would I properly test using a Dragonstrike in my OH using enhsim? I don't see any options that would allow for that kinda thing(kinda like Yo's sim).

There is nothing different about how it works on life and how it was proven to work on the Beta. So unless you actually have testing confirming it to work differently please take that smug attitude and stick it somewhere else.

Go look at the testing and then repeat it if you think it is wrong and then come and report on it but to state "I think the community still has to figure out this PPM mechanic..." is just another way of saying "I am lazy and I want you all to do all the work testing this for me."

I spent hours testing that mechanic and there were many others who did the same if not more and you have the audacity to assume that just because it doesn't feel right to you that we should be forced to repeat all that testing?

Go do a 30 minute run on a live server with a single weapon and no haste from your gear or from WF Totem or from Flurry (spec for MW but without Flurry) and see what your PPM comes back as. If you get over 10 report it and then others will have a reason to repeat the testing they already did to confirm your results. Otherwise stop making assumptions.

Okay here is a scenario that I feel is valid to mention:
You just hit a SS that put you at 4 stacks of MW, your ES ready (off of CD) what do you do?

Do you wait for the 5th and hit a LB/CL, or pop off a ES?

I am at just under 30% (29.7%)unbuffed spell crit and Earthshocking. Here is why, MW is just to random to prioritize ahead of a chance to get a spell crit (9% melee crit). Refreshing ED is HUGE dps folks....

Personally, i'd hit ES. Due to the randomness of the procs of MW, you might not get a proc for a few seconds.

Not sure if this will carry over to live but on beta, LL glyph just replaced the SoE glyph in the interface.


Did 3 10 min tests at lvl 70 with WF/WF and with WF/FT, all 3 tests were done with only totems no outside buffs, Syphon/Mounting, DST/Shard and no passive haste on gear. All three tests came out with FT ahead by 20-50 dps, when the sim is updated can get some better (more reliable) testing done but looking interesting; before the glyph FT was behind for me in similar tests by around 30-50dps.

I come bearing napkin maths.

Since the announcement of the new glyphs, I became interested in the viability of "slam casting" (I propose this become the official name due to awesomeness) lava burst in an enhancement rotation. I always heard rumors of it still being viable once it was removed from maelstrom weapon, but the serious theorycrafting community (this one) did not regard it as such. However, the flame shock glyph and the lava burst glyph both increase the viability of slam casting considerably. In addition, the base damage of the spell was increased. How viable is it now then?

I used the base stats that Rounced posted in the ehnsim thread (tauren in 5 piece naxx10 enhance gear, all wotlk gems and enchants, etc.) for level 80, which came to just under 800 spellpower. I used 800 for ease of calculations. No buffs, consumables or totems were considered. I used the base spell damage for the highest possible ranks of the spells mentioned. I calculated the coefficients as follows: Lava Burst = 67% (base of 57% plus the 10% from the glyph), Earth Shock = 42%, Flame Shock = 21% for direct damage/39% for DoT (with additional 6 seconds of DoT from the glyph). All earth shocks included the 20% boost from stormstrike. 5/5 elemental devastation and 2/3 call of flame were assumed. I am also assuming that casting lava burst suspends the swing timer rather than reset it.

Before determining the viability of slam casting, we need to assess the DPS gain or loss from adding flame shock to the rotation:

Earth Shock:
Average noncrit: 1450
Average crit: 2899

Flame Shock:
Average noncrit: 1970
Average crit: 2638

Assuming a 33% spell crit rate (for ease of math), 1 in 3 casts would crit.

3 Earth Shocks (1 Crit) = 5799 damage
1 Flame Shock + 2 Earth Shocks (FS crit) = 5538 damage
1 Flame Shock + 2 Earth shocks (ES crit) = 6319 damage
Average of the two possible FS>ES>ES scenarios = ~5929

Thus, from a theorycraft standpoint adding flameshock to our rotation and letting it tick its full duration will deal 130 more damage over 18 seconds (3 shock cooldowns), or about 7 dps higher that earth shock spam. Basically, they are so close that it safe to say with these numbers that there is little to no statistical difference in damage by using the FS glyph and adding it into our rotation.

Now for slam casting!

The most recent base damage supplied for lava burst, plus an additional 10% spellpower from the glyph, plus the talents that increase its damage, I calculate a lava burst crit dealing 4122 damage. The cast time is two seconds, so lava burst does about 2061 DPS. So, if this theoretical shaman cannot do more than 2061 DPS by himself, completely unbuffed, slam casting lava burst is at least potentially viable and warrants further testing.


I realize that many factors will affect actual testing results, such as 2 second worth of absolutely no attacks that could potentially proc static shock, maestrom, flurry, WF, etc. However, you also get a much higher elemental devastation uptime with slam casting due to the auto crits, potentially near 100%. This is only meant as a springboard for those smarter than I to do more accurate testing.



Discuss.

I've got an initial stab at a displaying of a priority queue icon implemented in ShockAndAwe. ie: you will be able to define a list of priorities and it will work through the list and show the icon for the highest ability that is off cooldown.

At present this seems to be working - it will need testers However to keep things initially simple I've implemented a fixed priority MW5_LB, SS, ES, LL and hard coded that. Once I prove that it is coded right I can move onto the trickier bit of coding a user prioritised list.

What I was wondering is if I should include the puppies in this priority queue as an option. ie: should they be used whenever they are off cooldown? Would it be better to limit this suggestion to only when person is in a raid? When is best to use the puppies?

I am currently capable of doing well over 3k dps at level 70. Before the patch I was regularly doing 2200 on Brutallus. Wouldn't that make 2061dps a very significant dps loss at level 80?

People are reporting 70%+ elemental devastation uptime so how is "Slam casting" actually supposed to increase dps since it will be less dps to cast Lava Burst and it will barely increase Elemental Devastation uptime, to say nothing about using up 2 glyph spots to do less total dps.

Go and play with the sim or test it on the Beta but without some real results just accept that you made a useless post trying to push a concept that has less viability then an Arcane Mage in SWP before 3.0 (In case you were unaware Arcane Mages in SWP before 3.0 were laughed at regularly and maliciously and were forced to wear helmets to try and prevent their stupid from infecting others).

I wouldn't say that he made a useless post. Isn't the whole point of these threads to consider new ideas, and find out if there's something less than obvious that may be optimal. While the idea may not end up working out, at least something new is being considered, and reviewed by others. We should be encouraging this kind of open minded thinking rather than discouraging it and telling people who try new things to "wear helmets".

Glyph of Lava Lash - Damage on your Lava Lash is increased by an additional 10% if your weapon is enchanted with Flametongue.

It's give less self benefit than SoE glyph.
Lava lash do only 4% of total damage.
+10% additional to 125% is only 8% not 10%.
4% * 8% = 0.32% dps increase.
At 70lvl we don't even use flame tongue to get benefit.

This glyph is just bad desing.

Which is all well and good, and really, I do agree with what you're saying, I just really disagreed with how you said it. There are better ways to get your point across, (your latest post being a perfect example) without having to belittle and demean someone. When you start treating someone like they have a disability just because you disagree with what they're saying, you're no longer providing constructive criticism, or useful advise; you're just being insulting. That sort of attitude really bugs me, and I replied because I was ticked off. I just prefer to give people the benefit of the doubt, not everyone is in the Beta, and not everyone is able to get the Sim running on their computer. Anyway, not trying to derail the thread, I'll stop now.

5798 is the average :o

Still better than a non existant glyph :p
And not all glyphs have to shine at lvl70

wrong tab :[

Instead of going the easy way and continuing the personal flaming, I did the constructive thing and checked up on your claims. I will give you this much, I was perhaps a bit unclear, the thing that seems not tested enough at the moment is the haste component and how it relates, not the 10 PPM, which stands pretty strongly.

The PPM mechanic has not been tested since page 112 of this thread, in which it is stated by by Xoya:

I'm finding it a bit hard to follow the discussion regarding how haste affects MW. Do we yet have a full understanding of which types of haste increase MW's PPM? From what I've gathered so far, "increases melee haste by X%" increases our PPM by X%, but "increases attack speed by Y%" does not increase our PPM at all, nor does haste rating on gear. Is this correct so far?

Before that, there is a short test by Suryaa stating:

Did some longer testing with a fast dagger for anyone interested in slow/fast combos, since someone else was working on 2.0 speed.

Each test was run for 1 hour using a 1.4 speed dagger with no enchant, imbue, or instant attacks. The only haste that was used is stated in each test.

Test 1 - 603 procs 10.05 PPM (no haste)
Test 2 - 645 procs 10.75 PPM (WF totem)
Test 3 - 669 procs 11.15 PPM (flurry 65% uptime 30.07% crit)
Test 4 - 679 procs 11.32 PPM (haste from gear 10.02%)


Since then NO ONE comments on the PPM mechanic to figure this out, and nowhere is it figured out exactly HOW the haste effect relates to the PPM. As you can see, the haste % in itself does not clarify it, WF totem is 20% specced, and if that was a simple 20% extra PPM, then there should have been 12 PPM. So it's a little more complicated than that. I'm not good at this, and I don't know if I'm looking at this number right, if I'm not, clarify it and insult me again. It says more about you than me.

So my point was to draw attention to the fact that we actually don't know how exactly the PPM relates to haste, and thus, I still think the 10 PPM is not enough to model it in the sim (which does not include a haste component as far as I'm aware, but I might be wrong).

What you should be disputing is not the 10PPM part of the simulator, since you actually agree that the 10PPM is correct, but instead the additional PPM that the sim adds on for haste effects, and the additional procs for WF, SS and LL.

Yes, which is exactly what I'm saying in my second post

I only scanned it right now, but that's the only possible problem I see.

Blizzard likes adding coeficients differently, such as Def stance on warriors reduces damage done by 10%, (90%), and then the talent to increase our damage while using a 1h weapon increases it by 10%, leaves us with 99% damage in def stance.

If that's the case, then the coeficient's 62.7% instead of 67%. I dont know if that's how it'll work in this situation, but it's probably worth looking into.

removed message

Should be easy to find out since there are talents like Tidal Waves that are worded the same as the glyph (and I guess it is more correct they work the same then), while the Defensive Stance/1HWS issue is a little different, so however those talents work, that's the most likely way the glyph works too.

Bellante go read your first post on the topic, you state "As far as I remember (correct me if I'm wrong), the enhsim uses 10 PPM for 5/5 MW talent. It's much, much higher on live, as everyone has experienced first hand". That's why I responded the way that I did. You stated emphatically that the PPM has to be higher then 10 when it has clearly been shown to be 10 by the testing done by this community. That's insulting to everyone who spent hours doing that testing and that was why I responded in kind.

So now, after looking back at all the research, you are willing to accept that the 10 PPM is correct but you still need to find some way to stick the bee back into your bonnet. So now you suggest that your disagreement with the Sim's modeling of MW wasn't about the 10 PPM at all, it was about how the Sim models haste and instant attacks.

The PPM mechanic seems pretty straight forward but trying to test for the specifics has a very large margin for error, which is why trying to determine what is going on required so many hours of testing.

How the Maelstrom Weapon PPM mechanic works is it takes your base weapon speed and uses that to determine the proc rate per hit. That proc rate is a constant that is unaffected by haste or by instant attacks, which is why you get significantly more then 10 procs per minute.

A 2.6 speed weapon will have a 43.3% chance per hit to gain a proc.
A 2.5 speed weapon will have a 41.7% chance per hit to gain a proc.
A 2.4 speed weapon will have a 40% chance per hit to gain a proc
A 1.5 speed weapon will have a 25% chance per hit to gain a proc.
A 1.4 speed weapon will have a 23.3% chance per hit to gain a proc.

Instant attacks, including windfury hits, all use that same proc rate. Adding additional haste (windfury totem, flurry and probably bloodlust) also doesn't change the proc rate.

The most consistant testing (although still limited) of this is probably the work done by Polocabbit
http://elitistjerks.com/924998-post2726.html

Tukez then made this post displaying how the sim was modeling MW.
http://elitistjerks.com/925176-post425.html

That is the mechanic that the Sim is currently using. If you would like to spend the time to prove what is happening with specific forms of haste in greater detail, feel free to do so and we as a community will applaud you for your efforts and I am sure that Tukez would be more then happy to incorporate any differences from your results into the Sim.

I stopped testing at that point because the mechanic made sense and it fit well with the results from everyone's testing. Reverse engineering a game through testing is not a simple process and is, at best, a guess at the equations. Does the sim give results that are comparable to what is seen in game? Your own post implies that

"That said, my own dps was pretty much in line with what the sim says."

Your dps was in line with the Sim so what is the issue? If your dps significantly differed from the sim then you would have a basis for doubting it's modeling of MW. If you want to be more precise about the mechanisms then, please, go ahead and do the testing and show the community the results but keep in mind that Suryaa's "short" test was at least 4 hours long.

That testing (and all the others) was enough to satisfy me (and others) that we have a basic enough understanding of the mechanism to model it properly and those are the mechanics that Tukez used in the Sim. Sure we could use more testing to more precisely determine the specific mechanisms, but don't ask us to throw the baby out with the bath water because you don't think the results align perfectly.