I'm sorry, but I can't do that. As much as I'd love to discuss it, I somewhat respect the NDA and these forums are way too public to discus it right now. I couldn't care less about official forums, as the thread will be deleted sooner or later, but EJ should keep up its reputation, imo.
Well, there is already posts about Mage, Warlock, Rogue, Balance Druid and now recently also Enhancement Shamans. Read somewhere a comment that past alpha and beta discussions have taken place here on EJ and it is not frowned upon. No idea if it is an "official" EJ-stance though.
Personally I respect an NDA that I've signed, or possibly, an NDA that I sympathize with. I'm not really inclined to discuss leaked talents on the wow-forum, due to I wager that any discussion will be disrupted by other classes whining about percieved injustices, whimsical shamans that keep suggesting totally overpowered ideas on what the Shaman needs or by Shamans screaming "nerf" when a leaked change is reverted. It is from this point of view I see the alpha-NDA. Blizzard don't want to have their forums flooded by people whining and crying about things that are not in the game.
Well, no point to discuss this further here. I think I'll create that thread that I pushed for and see what happens. I respect Binkenstein's wish to keep the One Stop Elemental Shop on topic (and I appoligize for this little wotlk outburst).
Edit:
Oh, good. No need for me to make a wotlk-thread.
Ok, so looking at mana use, specifically Elemental Focus.
I graphed out my expected figures, and got where Y = mana cost on average after EF, and X = crit rate.
Generating some random values, I replicated roughly 15,000 LB casts and ended up with
Everything seems to be set up correctly, and the 10 values I'm generating from the 15k cast list all average out to the trend line formula quite neatly, so if anything I think I'm currently under-estimating mana use. Except I'm overestimating it based on observations from in-game.
First I calculated the chance for the last cast to not crit and generate an EF (including the possible LO crit)
.2*[(1-crit)*(1-crit)] + .8*(1-crit)
so the chance for 2 casts not to crit is
is simply [.2*[(1-crit)*(1-crit)] + .8*(1-crit) ]^2
so the chance for EF to be up on a cast is 1 - [.2*[(1-crit)*(1-crit)] + .8*(1-crit) ]^2
First thing I checked :p A few runs I was waaaaaaaaaaaaaaaaay under (and not just 10%, I think I had a total of 12% crit one try. Not kidding). However, most runs were fine. That's why it doesn't make sense, everything except the damage checked out. No idea, perhaps I was just unlucky and only hit / crit CL-4LB rotation in the lower end of spell's innate damage. That could have brought about quite a damage difference too...
Removed since Daidalos covered my thoughts with a small typo in the last two formulas. They should be 1 - [.2*[(1-crit) * (1-crit)] + .8*(1-crit) ]^2
The result should be y = 1 - 0.96x + 0.736x² - 0.192x^3 + 0.016x^4.
I looked at my post for 5 mins before I even found the typo. I was looking at it saying "but thats what I have" ...ohhh.. oops. OK typos fixed now thanks for pointing that out.
The only way to check partial resists that I'm aware of is to compare average damage per bolt to spreadsheet one. Yes, the numbers checked out and I've only had higher resist rate twice on Chain Lightning (since we wiped early and the relative resist rate could not go down before that).
As I said, it has to have something to do with delay, but I have no idea how, if numbers are almost perfect on Brutallus and if LB spam produces almost perfect results. Anyway, whenever we're on Muru, I'll continue observing the results and do a few screenies. I'm determined to get to the bottom of it :p
Creativity requires the courage to let go of certainties.
The only way to check partial resists that I'm aware of is to compare average damage per bolt to spreadsheet one. Yes, the numbers checked out and I've only had higher resist rate twice on Chain Lightning (since we wiped early and the relative resist rate could not go down before that).
As I said, it has to have something to do with delay, but I have no idea how, if numbers are almost perfect on Brutallus and if LB spam produces almost perfect results. Anyway, whenever we're on Muru, I'll continue observing the results and do a few screenies. I'm determined to get to the bottom of it :p
If you look at the full spell breakdown on the WWS reports, it should list partials on there, I don't remember offhand what it's called, I can check later if need be. Either way, if that's not your problem, then it's negligible.
Please do let us know though if you figure it out!
I'm sorry if this doesn't contribute to the discussion really, but I was playing around with the haste cap and got a pic of it in case anyone interested of the Dr. Dmg summary/information. http://static.pici.se/pictures/XNhSlaIzA.jpg
I know I should have switched totems and all that, but it was mainly "for the lulz" so to speak.
I'm a shaman, a conduit of the ancient forces of nature. A master of the elements, except fire, water or earth really... nor do I have any REAL control of nature. But I do hurl bolts of lightning!
This was discussed quite a few pages back, but I don't remember if it was said; can LO proc of a LO? I've seen examples of that on CL, but with TLC there's usually just too much LBs to notice something like that...
This was discussed quite a few pages back, but I don't remember if it was said; can LO proc of a LO? I've seen examples of that on CL, but with TLC there's usually just too much LBs to notice something like that...
I think someone tested this for hours and never got a LO to proc a LO. I don't think CL LOs proc CL LOs, however each jump of the CL can proc an LO if thats what you are thinking of.
I think someone tested this for hours and never got a LO to proc a LO. I don't think CL LOs proc CL LOs, however each jump of the CL can proc an LO if thats what you are thinking of.
I tested it quite extensively, and I'm more than willing to say it can't happen, at least with the most recent version of LO.
I think someone tested this for hours and never got a LO to proc a LO. I don't think CL LOs proc CL LOs, however each jump of the CL can proc an LO if thats what you are thinking of.
If you want to do some real testing, try 5 min + tests, replicated 10 or more times. That way the numbers will be more accurate (ie: less skewing of results if you got a few lucky crit strings)
So a couple weeks ago I started designing my gear around much of what is written here as well as a few other random sources. Much of this makes sense for a T4/T5 Ele Shammy. Now I have a little bit of my own theory that I've been working on. I understand the DPS standpoint but how about the DPBF? (Damage per boss fight)
I'll reiterate a little bit of theory I am basing this off of. First and foremost to effectively make your mana pool last in a boss fight, crit is essential due to Elemental Focus. Outside of talents and your Totem of Wrath the number to hit seems to be 27% before the point of diminishing returns takes affect where a stat that would normally net you 1% crit will only net .7%. Once you've reached that point spell damage seems to be the best stat to start stacking as long as your spell hit is capped which is effortless as an elemental shaman. (3% outside of talents, racial and Totem of Wrath.)
Then Haste starts to become a question of effectiveness over spell damage or crit. The 3lb/1cl rotation makes a lot of sense and I can easily make it through a long boss fight without depleting my mana pool as long as I don't miss the cool down of my mana pots. Next comes the 4lb/1cl rotation which makes sense for maximum DPS but in order to cut that lb down to 1.5 sec cast requires 25% faster cast or 394 haste ratung. After looking over all the options for gear, that number seems impossible to reach and still maintain the crit% that is needed to retain low mana usage. So now I'm casting faster which uses more mana and criting less which saves less mana? Doesn't sound like your DPS is gonna make up for running out of mana and standing around.
So here is my thoughts on a good haste gear setup for the Ele Shammy:
A target haste rating of 197 which equates to a 1.75 sec lb cast and using a solid lb rotation only. To figure out the effectivness of not only mana usage but overall damage I had to figure out how many lb casts at 1.75 fit into the 7.5 sec cast rotation of 4lb/1cl. To simplify it I eliminated using fractions of casts in the equation:
(2lb x 2sec) +(7cl x 1.5sec) = 52.5 sec cast rotation
30lb x 1.75sec = 52.5 sec cast rotation
Now the first figure I want to show is the mana usage over that duration.
(21lb x 270m) + (7cl x 684m) = 10,458 mana used over 54.5sec
30lb x 270m = 8,100 mana used over 52.5sec
This saves 2,358 mana every 52.5sec
Now to figure out the total DPS for each rotation. For this I have figured only the base damage done by the lb and cl without factoring crits to keep it simple which is 684 for lb and 880 for cl.
(21lb x 684) + (7cl x 880) = 20,524sd over 52.5 sec or 342.06 DPS
30lb x 684 = 20,520sd over 52.5 sec or 342 DPS
That seems to make alot more sense than running the perfect 4lb/1cl cast rotation and depleting mana by the end of the fight. The only problem is the DPS doesn't really change. So is there even an advantage to stacking haste on an elemental shaman at all? Is it worth the 2,358 mana saved per 52.5sec? I think so. With that you could even use the 4lb/1cl rotation and not worry about casting the cl as soon as the cooldown is up. Now the DPS will increase and you're still saving mana. Of course there are other factors that come into play here like crit percentages and crit bonuses which I really have no idea how to even begin trying to calculate. That is why I posted this here. Hopefully some of you math geniuses can play with this a little more and determine if it makes sense at all.
I have to disagree with the wording of statements made that a Shaman should aim for X% of spell critical rating. Spell haste and spell damage are always a better dps choice, point for point, than spell critical rating is (assuming some modicum of gearing, even as low as full T4). Saying "40% spell critical should be the goal to aim for" might sound good on paper, but the reality is that a raiding Shaman should always take a point of spell damage over a point of spell critical whenever he has the opportunity.
Your response consists merely of opinion and shows no reason why the spell damage is more important. Simply put, you need to validate your response with some sort of proof. If you had actually read what I wrote, you would have noticed that the reason for crit over sd is simple; the more a shaman crits the less said shaman uses mana due to the talent Elemental Focus. Maybe this will help:
Elemental Focus
Instant
After landing a critical strike with a Fire, Frost, or Nature damage spell, you enter a Clearcasting state. The Clearcasting state reduces the mana cost of your next 2 damage spells by 40%.
Your response also states that for max DPS spell damage is more important. My post clearly states that it is not or max DPS but for longevity in a boss fight which in the end will net more damage than the other builds. Hopefully the next person to reply to this has a more intelligent input whether it be in agreement or disagreement. I do, however, wish to add one point to my theory which, until your response, I overlooked. I assumed that any Shaman reading this and trying to achieve it would be in the T4/T5 range and therefore have a good amount of sd already. So, let me clarify that. A shaman should see his/her sd in the range of 900 - 1000 bonus damage before trying to achieve this 40% crit rating.
On a side not, it appears that my post should not have been in this thread. Is it possible to have a moderator move this to class mechanics?
Removed since Daidalos covered my thoughts with a small typo in the last two formulas. They should be 1 - [.2*[(1-crit) * (1-crit)] + .8*(1-crit) ]^2
The result should be y = 1 - 0.96x + 0.736x² - 0.192x^3 + 0.016x^4.
Plugged that into my test sheet (finally) and it matches the test data I'd produced, varying by 0.25% at most.
Putting a 2nd power polynomial trend line gives y = 1 - 0.9416x + 0.6126x^2
This appears to be, at most, 0.08 off the fourth power equation, so it looks like we're all correct.
Your response consists merely of opinion and shows no reason why the spell damage is more important. Simply put, you need to validate your response with some sort of proof. If you had actually read what I wrote, you would have noticed that the reason for crit over sd is simple; the more a shaman crits the less said shaman uses mana due to the talent Elemental Focus. Maybe this will help:
Elemental Focus
Instant
After landing a critical strike with a Fire, Frost, or Nature damage spell, you enter a Clearcasting state. The Clearcasting state reduces the mana cost of your next 2 damage spells by 40%.
Your response also states that for max DPS spell damage is more important. My post clearly states that it is not or max DPS but for longevity in a boss fight which in the end will net more damage than the other builds. Hopefully the next person to reply to this has a more intelligent input whether it be in agreement or disagreement. I do, however, wish to add one point to my theory which, until your response, I overlooked. I assumed that any Shaman reading this and trying to achieve it would be in the T4/T5 range and therefore have a good amount of sd already. So, let me clarify that. A shaman should see his/her sd in the range of 900 - 1000 bonus damage before trying to achieve this 40% crit rating.
On a side not, it appears that my post should not have been in this thread. Is it possible to have a moderator move this to class mechanics?
Most raiding elemental shaman work on the assumption that a shadow priest is in their group. With this assumption, your arguement is rather moot and highest dps = highest encounter damage.
I've been trying to use SEIC 3 (which I hope is the latest version available) to do some dps calculations for my shaman. More importantly, I'm interested in the trade off between dps and sustainability (basically, I'm trying to figure out how much crit I can give up for more spell haste, given the current itemization available to my char). I know most people argue that we should have a SP in group always as elemental, but the sad truth is my raid usually puts them with the healers and there ARE some occassions when I have to do without.
So I'm concerned about how long can I sustain my dps without a shadow priest, just relying on chain potting, consumables, buffs and my crit ofc. In order to check this, I went to the spreadsheet and filled out all the relevant stuff then looked at the TTOOM (time til out of mana, I guess) and the number looked way too low given my experience. I'm currently raiding late T4, early T5 and most fights are in the 4 - 7 min range. Usually I can handle easily up to 5 and start having a bit of trouble with longer fights (I spam LB without a SP). Problem is, the spreadsheet says I should be running OOM in less than 3 min!
Thus, I checked the calculations and found 2 things I think are quite relevant:
First, the procs of water shield are not being modelled if I'm correct. In several fights, water shield procs provide some nice extra mana simply because everyone in raid takes some damage. Is there any way to include this? maybe an option to just include our own estimate of the PPM (procs per minute) of water shield?
Second, there's something fishy with the way the TTOM calculation is done. If I'm right, the idea is to do something like: (Total mana)/(Mana used per second - Mana regain per second). But I don't get the "Mana used per second" formula used. The one I have in mind (taking LB for example) would be (LB mana cost * EF)/((2-LM)/(1+haste) + delay), using the terms from the spreadsheet. But what I find in the calculations is (LB mana cost * EF)/(((2-LM)/1.3)//(1haste) + delay). I don't understand what that 1.3 is doing there at all. My best guess is that it's the modifier for heroism/bloodlust, but obviously that is not up during the whole fight so the formula needs to be changed?
I'm really sorry for the long post and hope this hasn't been already addressed/answered elsewhere.
Thanks for a great tool, btw!
Last edited by Berenonehand : 06/29/08 at 10:13 AM.
1) This is not the thread to discuss my spreadsheet
2) The mana calculations were indeed off, and have been fixed (will be in the b4 release), but you would know this if you had read the thread.
3) I have not yet attempted to model water shield charges, partially as it's a case by case basis, and partially because it's a PITA to work out. Suffice to say, this is a general guideline tool, so I'm not going to attempt to get everything in.
4) the /1.3 is part of the heroism calculations for DP values
06/13/08, 7:45 AM
Similar Threads
