I've no idea what you are using for EP calculations, but a quick look at the stats of the items you are comparing should tell you they are incredibly wrong.
Was using the EP numbers from the Math section of the latest SEIC spreadsheet. I could certainly be wrong. But a "quick look", considering the EP values being ascribed to the T6 bonus, was what got me concerned enough to attempt modeling it to begin with.
I used my own spreadsheet, which increases the likelihood of error, but spot-checking with Bink's spreadsheet the numbers are pretty close from what I can see. It's a trivial thing to do. I'm eager to be persuaded of a different conclusion.
Quite possibly this belongs in a different thread, although I don't see one dedicated to "gearcraft" and I don't have thread-starting privileges. My initial thought was that the use or non-use of the T6 4pc bonus might have implications for the relative value of a LB vs. CL in a rotation. But with the new patch I can't imagine that being the case.
A 162 est for T6 4pc is still not ehough to give those set peices more value than replacing them with items from T7 instances, especially since they're not ideal for the new stat weights anymore.
[e]And if I drop 2pc I don't have to spam four totems to get the set bonus anymore!!!
Thanks Bink. But I'm confused. Maybe I'm using your spreadsheet wrong. I get the following DEP split for these items between T7.25 / highest ranked item and T6:
Gloves 23 DEP
Wrists 36 DEP
Boots 51 DEP
Chest 91 DEP
Total (wait for it) 201 DEP.
If T6 4pc = 162, and T6 2pc (35 crit, 45 spellpwer) = 62, there's a 224 EP additional cost of dropping those 4 pieces of Skyshatter, which is greater than the 201 gain.
If the T6 weights aren't favored by new content, shouldn't that be reflected in the EP? Am I using this tool entirely incorrectly?
Was using the EP numbers from the Math section of the latest SEIC spreadsheet. I could certainly be wrong. But a "quick look", considering the EP values being ascribed to the T6 bonus, was what got me concerned enough to attempt modeling it to begin with.
I used my own spreadsheet, which increases the likelihood of error, but spot-checking with Bink's spreadsheet the numbers are pretty close from what I can see. It's a trivial thing to do. I'm eager to be persuaded of a different conclusion.
Quite possibly this belongs in a different thread, although I don't see one dedicated to "gearcraft" and I don't have thread-starting privileges. My initial thought was that the use or non-use of the T6 4pc bonus might have implications for the relative value of a LB vs. CL in a rotation. But with the new patch I can't imagine that being the case.
The set bonus won't make you not want to use CL in a rotation, but it will increase the value of LB. With LB getting increased coefficient, the value of that bonus will increase, not decrease. I've also been looking at the values of the set bonuses, but I'm not coming to the same conclusion. I have a hard time giving stamina the same weight that Bink is giving in the spreadsheet. Stamina's value only matters if you are low enough that its getting you killed, losing the increased stats on 4 pieces of gear shouldn't put you in a danger zone. Once you remove the 0.1 DEP per stamina on those items, you can't get enough EP value from anything in Naxx to break both of those bonuses (even using the conservative 162 EP value for the 4pc.) Keep in mind that the two piece bonus has a very easily modeled value. (roughly 61, depending on gear, ignoring the mp5 on it.)
Using the best helm, bracers, wrist and boots from available content, I can come up with at max 173 EP value with stamina removed from the equation. Assuming pre-patch 4pc bonus this brings it far short of the 223 required to top the 2 and 4 piece bonuses.
In my original sheet, I included +hit, which I now see is what your DHP means, which includes hit and stam, correct?
Here's the differentials in DHP.
Boots ~ 47
Bracers ~ 36
Gloves ~ 69
Chest ~ 64
Total 216
Still not enough to clear the hurdle.
WOTLK gems can be applied to Skyshatter, which compresses the jump in stats. If I understand you correctly, Bink, you're saying that weighting stamina at 10% may be appropriate for comparing level 80 epic gear, but not when you're talking about the significant jump from level 70 gear.
That's what you're saying right? That the ~170 stam differential involved in these 4 pieces should be weighed greater than 10%?
Its not ignoring hit at all, I'm just getting hit elsewhere. I'm only ignoring stamina because of what I said, its irrelevant unless its needed and its value isn't scaling. Using the DHP values for every piece of gear ends up with you massively over the hit cap, when it doesn't take all the much to end up capped (especially for the large number of us that are JCers and can get a large portion of it from 3 gems.)
I'm all for giving our stats some amount of weight, it just seems silly to toss some valuable bonuses on existing gear to make up the difference of 1-2k HP when we are already running 20k hp. Perhaps this goes down to personal preference, but it seems we are already riding the mediocre dps line, I'd prefer to take all the boost I can get.
EDIT: Even going back and using the "With stamina" and DHP numbers, the best I can get is 204. (shoulder, bracers, gloves and boots as the pieces used for t6, the smallest loss of the 8.)
I know this is kind of going back a bit and my math muight be kinda basic compared to yours but here is what I did when going over my rotation. I currently am a little undergeared, but for now I iave 303 haste and came up with this rotation breakdown:
the issue is after that cycle I was finding ~0.83 seconds of "dead" casting time. which I think could use for something. This thread gave me the idea for ES so I tossed it in there as 0.17s isn't too much to worry about: I came up with this:
Now this means that I can end with:
FS 18 - 19
LvB 19 - 20.37
But this means CL will be off cooldown for nearly 2.53 seconds which I cannot tell if it is a loss of dps or not. This issue will lessen and lessen as I get more haste, but do I not have enough to toss in that ES now? Should I just handle the 0.83s of essentially nothing. I guess it's a choice between being a second ahead of yourself or a second and a bit behind.
e: I know this was discussed at length, but reading it came up with the perception of very little consensus and would help for the less excel savvy readers to see the situation at play.
In my original sheet, I included +hit, which I now see is what your DHP means, which includes hit and stam, correct?
DHP is Damage/Hit equiv Points, which grew out of the fact that I have Elemental & Elemental Hit pawn value sets. As you can guess, one has it, one doesn't, which makes it easy for me to compare classic "DEP" values, as well as those values including hit.
What I am not getting here is why the 4-piece Tier 6 bonus is estimated to be so strong?
From my own calculations, and based on my current gear level, it would need 186 spellpower to provide an equivalent 5% boost to the damage of Lightning Bolt. However, I then factored in my current rotation, which includes 7 LBs from 12 spell casts, and the fact that other spells in the rotation scale better with spell damage.
This gave me a figure of 104 spellpower as equivalent to the 4-pc T6 set bonus overall.
I am obviously working from my own calculations, but referencing Bink's as I always have done, and otherwise our values seem very close. Am I making some elementary error here, or just coming at this from another angle... or what?
I'm not entirely sure what method you used to determine your numbers. I used two methods, a TC method and a "real" method.
Using the "real" method, I took my Patchwerk parse, found my LB damage, divided it by 1.05 and found the difference. Take that difference, (total damage dealt difference) and divide that number by the time of the fight, and you have the total DPS value of your set bonus.
Now you have to figure out the same damage value equivalent based on the coefficient, which is essentially that value divided by the coefficient for LB adj.
This is going to vary based on your current gear levels, so we can't give a one size fits all answer, and it will gain in value when the new talents come in as well. I came up with a conservative value of about 120, but I didn't have the most solid rotation, using the TC method, with an idealized rotation, it came out quite a bit higher. The value of it is also likely to rise as you stack haste, which essentially increases the percentage of your rotation spent casting LB (because its the only spell we have without a cooldown, your total "free" time becomes LB time.)
I'm not entirely sure what method you used to determine your numbers. I used two methods, a TC method and a "real" method.
Using the "real" method, I took my Patchwerk parse, found my LB damage, divided it by 1.05 and found the difference. Take that difference, (total damage dealt difference) and divide that number by the time of the fight, and you have the total DPS value of your set bonus.
Now you have to figure out the same damage value equivalent based on the coefficient, which is essentially that value divided by the coefficient for LB adj.
This is going to vary based on your current gear levels, so we can't give a one size fits all answer, and it will gain in value when the new talents come in as well. I came up with a conservative value of about 120, but I didn't have the most solid rotation, using the TC method, with an idealized rotation, it came out quite a bit higher. The value of it is also likely to rise as you stack haste, which essentially increases the percentage of your rotation spent casting LB (because its the only spell we have without a cooldown, your total "free" time becomes LB time.)
The method I am using is pure TC, I haven't been back to WWS yet.
Basically, I am calculating the damage of an average lightning bolt - (base damage + spellpower*coefficient)*1.05 for concussion. I am then factoring in what % will crit, and allowing for the additional damage of crits (including meta gem), and finally factoring in LO.
I am running this through the speadsheet multiple times for both my current spellpower, increased spellpower, and also with an additional 5% damage increase for the set bonus.
From this I established the value of 186 spellpower as giving the same boost to LB damage as the 4-pc bonus.
I then analysed how much additional damage the 5% bonus would add to my rotation overall, and found that an equivalent damage boost would be gained by increasing my spellpower for all spells by 104.
Haste is obviously a tricky stat to model now, and as a result I do this by comparing increased haste vs. the overall damage output of my rotation, allowing for the fact that the rotation will change to include more and more LBs. This is a bit of a theoretical change from TBC, where I could model haste as the increase damage of allowing me to cast the same spell faster. Now I have to model it as the increase damage from having extra "lightning bolt time".
Overall, I for one won't be going back to T6. Even allowing for a complete re-gem, and both T6 set bonuses, according to my calculations at best I would gain 43 spellpower equivalent for the loss of 74 hit rating, and this is just compared to current gear. Although I do only have 7/8 pieces T6, and miss the boots, which are obviously one of the higher quality parts.
I will probably re-look at the calculations in the light of 3.08 though, when the buffs and new glyphs may change matters, and Shamanism may make me drop CL completely from my rotation.
I suppose I'm also looking at it from a DKP economy standpoint. If I need four pieces of gear to get a minor upgrade (at best) I'd be best spent looking at other upgrades first. I'm sure this is a factor for many elemental shaman in DKP dependent guilds.
I'm not entirely sure we can make any determination right now until the rest of the chips fall from this upcoming patch. Things are going to get complicated for us for sure, considering at this point we aren't even 100% sure how to best model things, adding in the new elemental oath is really going to toss us for a loop. I'd caution any shaman right now against either trashing their T6 or avoiding upgrades to them contingent on how things end up in the coming weeks. I suspect the changes in store for us will be having constant changes as well once the PTR gets some more feedback.
Going off on a tangent slightly, but itemisation and therefore DKP, seems to be a funny issue.
Items for Ele seem to be totally polarised between a)items only an Ele would want (e.g Belts and Bracers fro Razuvious 25) and b) items half the raid will want e.g. Tier pieces, Dying curse).
So far for me DKP has not really been an issue, because most of the raid have a lot more 25-man items to take than me. Due to the seemingly poor Ele itemisation I know I won't get as much raid loot but I hope to get the pick of what I do want (e.g. Torch of Holy Fire - hubba hubba).
Even with my own unfavourable calculations sticking with T6 is not awful, and if you are in a very competitive DKP environment I imagine it could seem a good choice. For me though WotLK just feels like free epics land, where the easyness of the content + the quality of Heroic and Badge loot + the relative lack of Ele items... mean that it shouldn't be too hard to start thinking in terms of "best in slot" rather than "best for the DKP I have".
Hi there, I know that the "black Haste ranges" have already been discussed and I thought that I knew where to look to find that graph and numbers, I seem to be unable to find them again. I also saw that they might get added to the WoTLK TTT for Ele shamen but as of yet have not been. I was wondering if someone cold please repost those, as I have just started to work on my haste rating, with nice results I might add, and would like to know what ranges of haste to avoid.
You want to add haste until 11-12% (however much rating that is). At about 16% total (including Wrath of Air) you'll be able to use a very tight and efficient rotation. Mileage will vary, of course, depending on your latency. Personally I haven't settled on a haste value to compensate for the 1300+ ping on Mal'ganis Tuesday nights :P
Haste scales poorly but is unavoidable on our gear. You'll end up getting a fair amount (300+) just by taking upgrades. So don't sacrifice crit, hit or spellpower for haste rating and don't gem for it.
Hi there, I know that the "black Haste ranges" have already been discussed and I thought that I knew where to look to find that graph and numbers, I seem to be unable to find them again. I also saw that they might get added to the WoTLK TTT for Ele shamen but as of yet have not been. I was wondering if someone cold please repost those, as I have just started to work on my haste rating, with nice results I might add, and would like to know what ranges of haste to avoid.
The haste black spots were based around a fixed rotation. I haven't taken the "logic" that I put into SEIC to plot this out again.
All calculations show that haste is far better for us that crit (which is close to worthless).
Point for point haste is more than twice as useful as crit.
Once hit-capped, I would suggest you gear for spellpower and haste.
Actually, almost everything Ghando said is correct (haste doesn't scale that badly with the upcoming changes). Of course, we should probably add that you don't want to gem for crit either.
My findings on the new patch were different.
I suggest that the forthcoming changes boost the scaling of spellpower, thereby slightly reducing the relative weighting of haste.
My findings on the new patch were different.
I suggest that the forthcoming changes boost the scaling of spellpower, thereby slightly reducing the relative weighting of haste.
Because one stat increases in value, it doesn't mean that another decreases. There is no "solve for X" equation.
It seems to me that there are not enough elemental shaman providing independent analysis of our optimal dps.
I see people here promoting the view that we scale poorly with haste and indirectly implying that crit is a good stat for us.
These are directly counter to my findings.
While I appreciate that the mathematics of analysing a priority based cast system modified by various talents is too complex to surmise into a single useful post, I do feel that the more information we can gather the better.
I am quite happy for you to prove otherwise... providing of course that you can actually prove otherwise.
Yes, and I meant that, relatively, haste is more valuable than it used to be. But hey, it's not like I've put together a spreadsheet to look at this sort of stuff.
Oh, wait.
It's all very well and good to say stuff like that, but no body else is actually doing any work on the subject (or publishing the work, not just the results). Anytime damage goes up, it will increase the worth of haste, and vice versa. However, more spellpower does not change spellpower worth, while more haste will devalue haste slightly. In fact, some of the rotation changes may be to prevent haste rating being more valuable than spellpower, as was seen in T6/Sunwell gear.
Anyway, we have a Rawr plugin in the works, and I'm exploring a couple of other TC options as well.
12/11/08, 3:35 PM
Similar Threads
