Doora

You know that what's best for you might not be best for another shamans? EP values change according to different stats.

 

Mox

As of 2.4 it does actually show on the paperdoll. Highlight umm I think it was your hit rating, and it shows there. Atleast If I remember right. I'll try and log into PTR now and see if its been implemented yet.

 

Illundai

Yah, that's right. Hadn't noticed it myself, but just logged on to confirm it. Executioner stacks with armor penetration from your gear, but I don't have a Madness to check if that one stacks.

Yes, Death Knight tanks are/were too good in 3.0.x. Having 38k HP, 76.81% Avoidance (Miss incl.) and not gimping your threat in the first tier of gear is broken. (Yes this is clickable)

 

Patrik

Ignore armor stat is displayed with hit rating in the new character pane in 2.4 build. I was looking forward to that one.

And I believe that Blizzard thinks two executioner procs stacking will be just too powerfull (-1680 ignore armor and then some passive armor penetration = scary numbers).

edit: argh, being slow

 

Lumb

It seems like whenever I run YO's simulator for more than 10,000 hours it always freezes up on me. How long should it take to complete, or is that entirely dependant on my processor speed?

 

Toots Hepcat

Additional armor has almost no benefit. For one, you should never be hit by any mob that does physical damage; most of the damage you take will be AoE magic for which armor is useless. Second, the difference between a shaman in ALL mail and a shaman in ALL leather is something like 10-15% DR and the difference between swapping out a single piece of mail for a single piece of leather is like 1% DR. Not negligible, but from a healer's perspective hardly noticeable.

Stamina is extremely useful, up to the point that you have enough for your healers to be able to keep you alive. Beyond that, it has little value aside from slightly de-stressing your healers. I keep a set of high(er) stam gear around for helping out new healers, but my raiding gear has 7800 health and I'd gladly sacrifice 800-1000 of that for more DPS.

When it comes to EP: I will take an upgrade that is ONE EP better than what I have, if it's going to get burnt otherwise. I will actively seek any upgrade that's 10 or more EP better than what I have. 90 EP vs 100 EP doesn't seem like a lot, but it's 2.5 DPS. Make that trade off on every armor slot, and you've just lost 30+ dps. If you wear

I almost never look at the other stats. Some items -- [Gauntlets of Sniping] vs [Trickster's Stickyfingers] (a 5 EP difference for me) -- are so close, I can almost convince myself it's worth the negligible survivability to pick mail. But these are rare, and I am aware that I am gimping myself.

 

Toots Hepcat

Please stop. Shaman DPS is a feedback system and the "best" weapons will vary dramatically by your EP, which changes every time you add an upgrade or change buffs you receive in raid. I dislike the concept of a "set" EP for T4/5/6 because this implies a level of homogeneity that does not exist. These are (I guess) useful from a theoretical standpoint, but bad when dealing with reality.

Shamans need to be mindful of this. A static list does not help them get there, it just reinforces a false assumption.

As an example -- a while back, [Fool's Bane] was about a 20 point upgrade for my offhand. Last I simulated it, it was nearly dead even -- because in the meantime, I'd added a lot of crit, hit rating, haste rating & armor pen, and the value of offhand weapon DPS went up.

What's more, what is important is not the BEST weapon, it's "is this weapon upgrade I am presented with worth the DKP/Honor/Grind cost" -- a question which can only be answered for an individual's gear. Malchazeen may well be an upgrade for some. It should also be a cue to farm up a better weapon.

 

Malan

Highly dependent on your computer and what else you're doing at the time. 10,000 hours seems to take about 5 min on a pretty decent system. If all you want is a DPS check, then don't select the EP option as that adds a lot of extra calculations.

Shitting up every single thread on EJ since '06

 

Grundar

I have been thinking about this too, especially in regard to killing priests and pallys. Is there a way to tie a purge in to the stormstrike timer so that the purge comes right before the stormstrike?



Originally Posted by rava
It really chooses randomly. Yes there is a chance that it will purge instead of SS.

I'm not 100% positive on this, but the macro you want is conditional(if/then) and not allowed as of whatever patch was before TBC. Two buttons is some harsh work, though!

/cast Stormstrike
/stopcasting
/cast Purge

It's not that complicated.

 

Malan

Out of curiosity I just opened 2 LootRank.com windows and input our Entry, Mid, and High end EP values into each one, excluded PvP items and set it to display leather only. With the exception of a few slots (low end placed cloaks and necks slightly different) the ordering of the top 5 'best items' in each slot was identical across the board. (Each slot generally had 1 item shifted slightly above/below one other item, but all of them agreed on what the top item was in each slot)

I'm questioning if we're going about this the right way. Clearly these values are not changing so radically between tiers of gear so that the best items are radically different depending on what you have equipped at the time. Are we missing something here?

Shitting up every single thread on EJ since '06

 

Toots Hepcat

Well, the problem with this test is that the "best in game" stuff isn't what an individual cares about ranking. It's immediate upgrades. There isn't a lot of "best" gear, it all comes from the same two dungeons, and it's already been mentioned that at T5 levels certain upgrades (the Talon is what I'm thinking of) actually outperform certain T6 upgrades. The BEST stuff outdoes them both.

The problem with evaluating the BEST is that the best isn't what's accessible in most cases. You really care about what's better, right now, for me. Caring ONLY about the best leads to DKP hoarding and other antisocial, DPS-nerfing behaviors.

 

Weem

If I remember correctly, from editing my Loot Rank as my values changed, the order of the items were virtually identical, as you also witnessed.

 

Malan

But that's what I just described Toots - other than 1 or 2 items being listed differently, the order of upgrades using each of those values was identical.

Leave DKP and loot grabbing behaviors out of this. I'm not talking about how a guild decides loot order here at all. I'm talking about how we're deciding to evaluate gear stats and apply weights. I'm questioning the model that we're using.

The funny thing is that MaxDPS.com is producing the exact same lists.

Shitting up every single thread on EJ since '06

 

Stopokingme

The new Sunwell weapons that are becoming available, and the seeming lack of an axe to them has got me mulling over the actual impact the orc racial has on our weapon choice, or in other words, what is the actual value of Weapon expertise.

The best way seems to be calculating them into Weapon EP, considering it only affects that actual hand you're wielding the weapon in. The formula's I've come up with are these:

MH W-EP = 2/3 * 20 * Exp-EP
OH W-EP = 1/3 * 20 * Exp-EP

The 20 times expertise EP speaks for itself, that's just the EP value of the 5 expertise the racial gives. The 2/3 and 1/3 are based on the fact that the main hand does double the damage of the offhand, so the damage spread is 2/3 and 1/3 respectively.

Enhancer does the awesome job of already calculating W-EP, so just tag on the calculated added EP from the orc racial and you should be able to compare them with non axes for a quick and dirty check if something is actually better or not.

Anyway, my math has been proven to be wrong in the past, so I wanted to run it past you guys.

Another thing that sprung to my mind is because of the unique mechanics of the racial, is how much of an impact having a non axe would have on increasing main hand windfury procs. To my knowledge dodges can't proc windfury, so in essence with your offhand getting 1.25% more dodges you should force your main hand windfury ratio up a small notch, I doubt this will have a huge impact on dps, but it might actually be worth considering. I have no idea how to go about calculating this though.

 

◊ Rob

Well, we kind of already went over this discussion a while back. Disquette, Tornhoof, and others had introduced a way for us to obtain AEP values, and Yo! came in and said that hey, these aren't really static values, you guys shouldn't use this system. As I recall, I and Sebudai both said something like "well, it doesn't matter because the best in slot items are the same for everyone regardless", but since Yo's sim makes it so easy to get customized EP values, it makes sense to use that and then you will choose the right upgrades along the way as well.

 

Malan

Yah I know we've touched on it before... what I'm getting at though is that I think we're dancing around the solution here. If just about any (valid) result from the sim is going to produce (nearly) the same results in gear choices, doesn't that imply an underlying relationship that we're missing? It seems to me that we're about 1 step away from the 'grand unified theory of enhance shaman' that would give us 1 set of relationships that would work for everything.

Shitting up every single thread on EJ since '06

 

◊ Rob

Your math is right (I posted the same answer a few pages back when someone posed the question).
Interesting idea, sim it.

 

◊ Rob

It is not surprising that the sim gives similar results in gear choices given different starting points. The "underlying relationship" that you refer to is, simply put, shaman combat mechanics. The sim is using those to give an answer.
Well, we had a lot of people trying to find a "closed form" model that could accept parameters like AP, Crit, Hit as inputs and give DPS as output. You can do that *reasonably* well, which is what MaxDPS.com is doing. Tornhoof also produced something, and we had someone else coming in here who wanted to use a Cobb-Douglas function. There are currently a few problems with this approach. First, it's not easy to add extra procs like Mongoose, Stonebreaker Totem, or Madness of the Betrayer to this equation. Each variable introduces another level of complexity. Second, the best models produce something that appears to be less accurate than the margin of error in the simulation. Given the choice between the two, it appears that for now, simming is the way to go.

 

Malan

What I'm getting at is that I think the sim can provide us with what we're looking for.

Shitting up every single thread on EJ since '06

 

Low Life

While I was checking Lootrank with my own values compared to the T6 level values presented in OP, I noticed a trend there. While a lot of items were at the same spot on both rankings, items with either haste rating or armor penetration varied immensely. Best example would be shoulders, where with T6 values [Swiftstrike Shoulders] ranked as the second best shoulders currently in game (4th overall), they were only 8th best with my values. A few posts back someone also mentioned [Gauntlets of Sniping] being slightly worse for him than [Trickster's Stickyfingers], while for me they're actually 17EP higher.

This made me think about items with these stats and choosing between two possible upgrades. If Item1 has 150EP with pure Agi/AP/Crit and Item2 has 140EP with AP/Crit and Haste rating on it, would Item2 actually turn out better in the long run, when I manage to find upgrades for my other slots? This would probably change my current "final" paperdoll with upgrades from ZA/Kara/Badges by quite a bit.

edit: My Lootrank
T6 Lootrank

The 2.4 items mess things up a bit there..

 

◊ Shalas

Using the T6 item values and lootrank, the order is almost always exactly what I would expect it to be with absolutely no knowledge about enhancement shaman at all. It's mostly sorted by item level, with exceptions exactly where you'd expect them: items with sockets appearing above items without sockets, items with more stats appearing above items with fewer stats, and items with caster stats appearing under items without caster stats. Take this lootrank search. If you sorted the top 5 items in each slot by ilvl with sockets giving +15 ilvls, one caster stat giving -15 ilvls and two giving -20 ilvls, you'd get the same best item in all but three slots, and only 11 items even change position.

The theoretical optimal items change significantly as your gear changes, but the items available do not. The optimal actual items don't change because there's very few items that have the budget spent in a decent way on stats that do anything at all. Does it really matter if hit or crit is worth more when the item selection is so bad that the third best hat has 28 int on it? I don't think there's any magical relevation to be found other than that the itemization available is really so bad that you're more concerned with finding items that actually spend all of thier points on useful stats and sockets than finding items which spend all of thier points of the right stats.

 

Malan

Except I can get almost the same lists on LootRank if I input 1 (one) for all the values except Armor Penetration.

Shitting up every single thread on EJ since '06

 

◊ Rob

This result is not surprising, though. The primary stats (everything to do with DPS except armor pen) all have the same "cost" on "item budget". To item designers, 1 STR = 1 AGI = 1 crit rating = 1 hit rating = 1 haste rating = 1 expertise rating = 2 AP. There is a second-order effect based on an exponential penalty for items which "stack" a stat, but we would not expect that to affect our results. If the game is properly designed, all these stats should impact us equally. However, there are some peculiarities to enhancement shaman mechanics, so for example, hit rating is worth more if you don't have 9% and less if you do. Kings makes strength and agility worth a bit more. Agility confers some "survival" benefits in addition to DPS benefits, so it's worth a bit less than crit rating. Again, we are basically discussing something months old here, Yo! said "well, if you want an EP system Blizzard already gave you one, it's called the item budget". The issue is that the game isn't perfectly balanced and that enhancement shamans do have some peculiarities, so the interesting choices are where the base stat values don't quite work out.

 

◊ Shalas

Assuming you mean all melee stats, that's pretty much my point. There's so few items that you can equip that spend itemization points on anything useful that choosing an optimal item is never an option. Once you get to raid gear, most of the time you aren't choosing between an item with the right melee stats or the wrong melee stats, you're choosing between items with the wrong melee stats, items with no sockets, items with the wrong melee stats that are a lower ilvl, and items with int/mana regen. Which stats you want only matters when you actually have a choice, but most of the time items with the right stats aren't even an option. There are zero items in the top 5 for each slot which have the right stats at all, much less the right stats in the correct proportions, which makes worrying about what those correct proportions are somewhat pointless most of the time.

 

Stopokingme

Hmmm guess I missed your math. Good to know it's at least correct.

And was thinking about simming it when I was real bored yes, don't know enough about how the sim calculates to know if it really accounts for this situation though. Might have some results tomorrow night.