Indeed we agree on this; the problem is very minor.
I was btw not fixed at a single haste number but my static haste is already getting very close to 1269, and having Scale of Fates and the new EM change really urged me to investigate how large the problem was as I will go structurally below 1sec GCD.
Incidentally I used SEIC to get the EP values and for me the haste EP per point is slightly higher then sp. I didn't look at ZAP yet but i'll check it out.
As for people making radical gear-change choices, I don't really get that.
If you simply get the EP values for crit/haste/sp/int from the above spreadsheets using your current stats, the only thing to do is evaluate what the total EP value of an item is (and ensure you are hitcapped). Whether that is coming from crit or haste is really irrelevant as long as the EP values are correct .... which we argued here are not really affected much.
In any case, keep up the good work with the spreadsheets, they are a great help
Since Reign of the Unliving can only proc every 2 secs does it mean that only every other LB will proc it since we have so much haste now? It seems its doing less damage then before I had all my BIS.
I should have said this earlier, but your argument is mostly a moot point, as spellpower will most likely be more valuable than haste for any given gear setup, so changing the value of haste to be slightly lower will have no impact on gearing.
I love(d) your spreadsheet, which gets this issue right even if you don't. You're correct, of course, that there is no magical "bad haste" number to be avoided. However, your statement that changing the relative value of a stat will have "no impact on gearing" is just horribly wrong (and I'm actually rather appalled that you would make it).
As an example (for which I have entirely fabricated the numbers), lets say we have a yellow socket that provides a socket bonus of 6 crit. Let's also assume EP values of 1.1 for spellpower and 0.5 for crit. Now compare the difference between setting the EP of haste at 1.0 and 0.9:
11 spellpower * 1.1 = 12.1 EP.
At 1.0: 10 haste * 1.0 + 6 crit * 0.5 = 13 EP. Socket a reckless ametrine and pick up the bonus.
At 0.9: 10 haste * 0.9 + 6 crit * 0.5 = 12 EP. Ignore the bonus and socket a runed cardinal ruby.
It's worth noting that it only takes a 10% drop in the EP of haste rating to make this shift occur and hitting the GCD cap seems to be at least that large. (DISCLAIMER: Vacuum assumed for the purposes of argument. Haste, crit and spellpower all affect the relative value of one another in a synergistic manner and therefore this exact example probably couldn't occur, but it's close enough.)
The same situation could occur when comparing items. As the value of haste decreases, a piece itemized with more crit or spellpower might overtake the value of another with more haste. Changing the relative value of any given stat will always have some effect on gearing. Whether and/or how much this matters is highly situational and requires taking your specific stats and buffs into account, so as you and Rahvin both agree, using a simulator with your own data is the best way to make these decisions.
This honestly is not the place to be asking gearing questions, but on a strict value of stats alone, IODS still tops the others. 200sp >= 84 haste+~118 SP and is > 87crit+125sp.
So while the difference is basically a wash between IODS and abyssal rune. On an EP value, IODS is still higher.
As an example (for which I have entirely fabricated the numbers), lets say we have a yellow socket that provides a socket bonus of 6 crit. Let's also assume EP values of 1.1 for spellpower and 0.5 for crit. Now compare the difference between setting the EP of haste at 1.0 and 0.9:
11 spellpower * 1.1 = 12.1 EP.
At 1.0: 10 haste * 1.0 + 6 crit * 0.5 = 13 EP. Socket a reckless ametrine and pick up the bonus.
At 0.9: 10 haste * 0.9 + 6 crit * 0.5 = 12 EP. Ignore the bonus and socket a runed cardinal ruby.
As you said it yourself, all stats have a high synergy. And your numbers are a bit wrong. 11 SP can't be found on any kind of gem. It would be 12 or 23. Now:
23 SP * 1.1 = 25.3
At 1.0 EP for haste: 12 SP * 1.1 + 10 haste * 1.0 + 6 crit * 0.5 = 26.2
At 0.9 EP for haste: 12 SP * 1.1 + 10 haste * 0.9 + 6 crit * 0.5 = 25.2
These numbers should actually prove your point, if, and only IF, spellpower would be evaluated so low. In practice, spellpower is evaluated much higher. Also, because of the synergy between stats, a decrease in EP for a stat should result in an increase in EP for the other stats. A decrease of 0.1 EP in haste should provide (in theory at least) an increase of 0.05 for each spellpower and crit. In such conditions, your estimated 0.9 EP formula would turn into:
and would offer a total EP value much closer to the initial one.
I am talking about a steady value of the sum of EP/stat because otherwise, with the EP/stat sum decaying, we could reach a point in which the EP sum of all stats on a piece of gear of higher iLevel would actually turn out to be lower than the EP sum of all stats on the current gear and thus an upgrade is not recommended. I was considering the 2 pieces of gear having the same stats, same proportions between stats, but with different values.
Overall, I go with Bink's statement, that a small change in the relative values of stats will not have a big impact in gearing. The only point in the gearing process where the change in the relative values of stats has a high impact on gearing is when a player reaches the hit cap.
The same situation could occur when comparing items. As the value of haste decreases, a piece itemized with more crit or spellpower might overtake the value of another with more haste.
I fail to see your point. How could an item have more spellpower when it has the same ilvl (provided you have the same number of gem slots) ? How could an item have enough crit to be better than haste ? Haste will still be far above crit. Ok these stats have a synergy, but not enough to overcome a 50% EP difference.
The only thing it can change is gemming, but gemming sp or haste to get the bonuses is hardly a decision that will make much difference in the end since gemming doesn't give a lot of dps in the first place and either choice is a decent one. Maybe you were considering two different items which had a difference of +3 haste -6 crit for example, but then again that's not something that will make a lot of difference.
I fail to see your point. How could an item have more spellpower when it has the same ilvl (provided you have the same number of gem slots) ?
Not sure where you got the idea you should only ever look at the item level when selecting gear, but I frequently find lower level items to have better itemization than their higher level counterparts (e.g., Chalice of Benedictus is better than Symbol of Transgression unless you need the extra 13/14 points of hit - drop the value of haste in proportion to either crit or spell power, however, and this will not always be so, illustrating exactly my point, even though the difference between them is not +6/-3).
For what it's worth, yes, we're discussing a relatively small issue and if you don't care whether it's better to socket an orange gem instead of a red one, you probably should have skipped the last several pages of this thread.
Originally Posted by Divis0R
As you said it yourself, all stats have a high synergy. And your numbers are a bit wrong. 11 SP can't be found on any kind of gem. It would be 12 or 23. Now:
23 SP * 1.1 = 25.3
At 1.0 EP for haste: 12 SP * 1.1 + 10 haste * 1.0 + 6 crit * 0.5 = 26.2
At 0.9 EP for haste: 12 SP * 1.1 + 10 haste * 0.9 + 6 crit * 0.5 = 25.2
These numbers should actually prove your point, if, and only IF, spellpower would be evaluated so low. In practice, spellpower is evaluated much higher. Also, because of the synergy between stats, a decrease in EP for a stat should result in an increase in EP for the other stats. A decrease of 0.1 EP in haste should provide (in theory at least) an increase of 0.05 for each spellpower and crit. In such conditions, your estimated 0.9 EP formula would turn into:
and would offer a total EP value much closer to the initial one.
I am talking about a steady value of the sum of EP/stat because otherwise, with the EP/stat sum decaying, we could reach a point in which the EP sum of all stats on a piece of gear of higher iLevel would actually turn out to be lower than the EP sum of all stats on the current gear and thus an upgrade is not recommended. I was considering the 2 pieces of gear having the same stats, same proportions between stats, but with different values.
Overall, I go with Bink's statement, that a small change in the relative values of stats will not have a big impact in gearing. The only point in the gearing process where the change in the relative values of stats has a high impact on gearing is when a player reaches the hit cap.
First: No, you won't find 11 spellpower on a gem. You will, however, find a net difference of +/-11 spellpower and +/-10 haste between the two gems I used for my example, so the 12 spellpower shared by both gems can safely be removed from both sides the equation.
Second: As I've already admitted that holding the EP values of spellpower and crit steady while adjusting that of haste is not a valid "real life" scenario, you just wasted three paragraphs of typing. Regardless of your differing outcome with differing numbers, my point is valid and remains true: Modifying the value of a stat will always impact gear decisions. We can dither over how much impact such a change will cause when supplying different numbers for eternity.
But while we're dithering, since the original issue of contention was whether the drop in EP of haste as a result of reaching the "soft cap" was significant, my decision to hold the other two variables constant is actually a closer approximation of a "real life" scenario than your adjustment. As confirmation of my point, using ZAP's default supplied values and moving haste up to and then across the "soft cap" results in narrowing the gap from my above example from 4.6 EP to 2.4 EP. Socketting a reckless ametrine here is still the better option, but not by the same margin (you lose almost half the benefit of doing so). Now try the same exercise with a socket bonus of 8 intellect, instead, and the runed cardinal ruby becomes the better choice after crossing the soft-cap.
And finally: I don't have any clue what you mean when referring to the height of the evaluation of spellpower. Please elaborate on exactly what you mean and what impact this has on the argument. If you're contention is with the EP values I pulled from the air, it might help to know that my actual EP values are 1.25, 1.13 and 0.7, which are not all that far from the ones I assigned above.
Of course Symbol of transgression is not good : it has hit/crit. Even if the extra hit points are needed, I still think it's not as good as Calice. I didn't understand what you meant with "drop the value of haste in proportion to either crit or spell power". Could you clarify ?
There's no point improving one's ilvl gear with items which don't have haste on them (unless your upgrade is several tiers higher). That's what Binkenstein, Divisor and I were basically saying. That was how things worked for 3.2, and it will still be the same for 3.3. For gemming, I agree that we'll need spreadsheets to be sure the bonuses are worth taking or not.
Your approximation of "nearly constant crit/sp" EP when you move through the haste soft cap seems good, as crit/sp scale continously but not haste at this value. But it's highly situational, you have to move trough the haste cap for it to be exact. Once you are definitely over the cap, there's no reason to take constant values for crit/sp.
We normally run two elemental shamans in our raids.
I do about 7.1k on Anub 25 Heroic and the other does around 8k, I know he outgears me but I took a look at a parse in the fight. here is what i came up with:
Just wanted input in my dps, and how to increase it.. Seems like I am not using CL enough in this fight.
We do run with a demo lock on this fight...
Not the place to flood about damage done...
Only thing that's clearly different is the trinkets. Anub trinket is a great improvement, which accounts for some of the difference between you two.
And finally: I don't have any clue what you mean when referring to the height of the evaluation of spellpower. Please elaborate on exactly what you mean and what impact this has on the argument. If you're contention is with the EP values I pulled from the air, it might help to know that my actual EP values are 1.25, 1.13 and 0.7, which are not all that far from the ones I assigned above.
Well, I guess it depends on the spreadsheet we're using to get EP values. And it also depends on the term of comparison. EP = Equivalent Points, but you need something to equal to. I was using SEIC and now I'm using ZAP! In ZAP!, the EP are actually DEP, as Damage Equivalent Points, and at this moment my EP values are: 1.56, 1.36, 0.87. I tried to meddle a bit with stats, but I can't seem to find a way to devalue SP down 1.1 or 1.2.
EDIT: White spaces + forums = not funny. Preparing paint work of art
Second EDIT: @ Ongor:
Well, the other guy has 270 more spellpower, 106 more crit rating, 9 haste less, and a trinket that has a DEP value of 327 over your IotDS (Reign Heroic = 640, IotDS = 313). Are there still questions lingering ?
As Long as there are adds keep CL on cooldown and make sure you always refresh magma totem. You can more than double the dmg of your magma totem, and the dpct is insanely high.
Use your first Thunderstorm right after your tank positioned the adds on the ice and always use it as soon as it is ready again; it times perfectly with the adds.
If you get Anub down to 22-23% without anyone dying, use your Fire Elemental! If it's on CD, keep refreshing your magma totem. A lot of DD's tend to focus on Anub and ignore the adds - this can turn easily into a wipe.
Other than that: Work on your movement! Your dmg uptime is quite low with ~92%.
You are first at dmg done on swarm scarabs, but you can do way more than mages.
Has anyone done the math yet to determine if its worth going after the 2 piece for tier 10, or are we stuck waiting for 4 piece to start using it again just like T9?
Has anyone done the math yet to determine if its worth going after the 2 piece for tier 10, or are we stuck waiting for 4 piece to start using it again just like T9?
that will depend largely on the final stats or our t10 (still placeholder atm), but with the nerf to t9 4set it looks like you can switch to tier10 once you get 2 264 pieces.
Also from what I understand once we hit Haste soft-cap any more Haste only applys to our LB. So since Crit sucks because it only helps LB because our LvB 100% Crit chance, wouldnt that make Haste and Crit to be worth about the same when were Haste soft-capped since they both only effect LB at that point?
Also from what I understand once we hit Haste soft-cap any more Haste only applys to our LB. So since Crit sucks because it only helps LB because our LvB 100% Crit chance, wouldnt that make Haste and Crit to be worth about the same when were Haste soft-capped since they both only effect LB at that point?
Crit doesn't "suck", it's just worth less DPS per point than haste or spellpower. It's not like we weep tears of blood when we see there's crit rating on our gear.
Lightning Bolt constitutes the vast majority of our damage, so hitting what you call the "soft cap" (50% haste) doesn't make haste rating very much less desirable. Point for point it continues to be worth significantly more than crit rating and becomes worth decisively less than spellpower. Off the top of my head, the normalised weights drop from something like SP 1 / Haste 1 / Crit 0.6 to SP 1 / Haste 0.85 / Crit 0.6. Crit and haste ratings don't start to get really close in value until somewhere above 70% haste.
Every one of us is, in the cosmic perspective, precious. If a human disagrees with you, let him live. In a hundred billion galaxies you will not find another.
Anyone test the new Black Magic enchant vs. the current weapon enchant? Purely anecdotal, but I would like to think it will be a bit better.
It's been discussed; while it yields slightly higher avg DPS then SP, it also gets almost completely wasted under Lust/Other haste effects .: Avoid On Proc haste effects and stick to SP unless your haste is low (500-600ish).
Alright, that probably means we will go with Helm, Shoulders, Chest (sp/haste/crit) and the gloves (sp/crit/hit) for our 4 set bonus, or potentially just skip 4 piece altogether.
Zap right now simulates 4pc Tier10 at ~50 DPS which is ~28 sp. So e.g. if we get offset gloves (e.g. Gunship Captain's Mittens) that have 90 haste instead of 90 crit that should already be enough to offset the lack of 4pc.
So in the end we might be the only dps spec that is better off not using 4pc Tier10. Sigh.
I imagine if that actually ends up being the case, they will improve the set bonus somehow, but given how it works I'm thinking they will go with a different one.
11/27/09, 7:42 AM
