I'd like to also add that the idea of regemming for stamina instead of defense is not effective of a use for SG either. I realized that the exchange was just not enough. Let us look at it another way.
Now, what would your raid life have to be for equality with 4% parry? Consider a total net avoidance of about 70%. 1% avoidance is still worth about (1/30) of your total life.
4*(1/30)*total_life = 2700, or total_health = 20250 or 20.25k health.
I.E. if your buffed health is greater than this value, you still benefit more from swordshattering. You would have to be a brand new level 80 DK (or be completely unable to get 540 defense without SG) for SG to be worth it.
*Edit* I suppose I can see the value in regemming for hit/expertise assuming you don't need 4% avoidance. I have yet to see this problem though. Most of the times I have had threat problems was when frost presence was bugged or when I would try an inferior talent tank spec. *shrug*
Sigh, people still don't get it. It isn't about taking less damage or about maximizing your TTL(though your calculations only work on avoidable damage). Getting SG/Gemming stam is about maximizing you EH so you don't die to bad strings, so you survive Sartharion breaths. No matter how we gem we will have the highest TTL of any tank on an average fight, so why not cover our weakness? (ie our tendency to die when we don't avoid 3-5 times in a row)
Sometimes average survivability is worth sacrificing to improve your worst case survivability. If you were often taking in a situation with only 1 healer on you, taking overall less damage might be better, but generally tanks just get tons of overhealing.
Would you rather have 10k hp with 95% avoidance or 100k hp with 5% avoidance? I think the answer is pretty obvious. Finding the tipping point where gaining avoidance is better than gaining EH is tricky, but that tipping point in actuality is far more slanted to EH than a simple TTL calculation.
As an aside, I agree that there is no reason to be gemming for threat at the moment. Better off respecing to something higher threat, glyphing threat, or calling for extra misdirect/tricks if you are having problems(which you shouldn't be to be honest).
You might as well resign to the idea that this topic is going to be rediscussed frequently (as with any other tanking topic) in this thread because the OP doesn't update his post. There is no way someone has the time to dig through 100 pages and analyze varying view points (which often include faulty math that the poster later recognizes).
Digging through 100 pages is completely unreasonable. My gripe is that we get the same post on every page, and that people feel the need to re-do the math (frequently incorrectly as you point out) daily. Today's discussion has followed exactly the same path as yesterday's:
1) SG is only good for new tanks, otherwise it's patently inferior
2) No, SG is a tradeoff of 1% avoid for roughly item-points equivalent Sta, let's discuss which is more valuable!
3) No, SG is a tradeoff of at worst 1% avoid for ipe Sta because anyone taking SG is going to regem for something they value more than avoid rather than stay at 565, because if they valued avoid more than whatever else, then they would have stayed with SS in the first place.
Then everyone ignores (very important) point 3, discussion dies down, and the cycle repeats 24 hours later.
I'm not adverse to discussing tanking topics to death in the tanking thread, that's what it's for, but seeing discussion caught in a time loop drives me nuts.
I'll definitely agree with you that an up-to-date and accurate OP would help matters. I'm not sure how that's best achieved.
You might as well resign to the idea that this topic is going to be rediscussed frequently (as with any other tanking topic) in this thread because the OP doesn't update his post. There is no way someone has the time to dig through 100 pages and analyze varying view points (which often include faulty math that the poster later recognizes).
The only way to fix this problem is to redo the tank thread completely. Someone should strip out the best information here and post it and have this thread deleted. That's my $0.02 worth.
You can complain all you want about the topics being asked again and again but the means for readers to ascertain information is antiquated and inefficient. It's easier for them to reask the questions. As long as people are willing to post the answers then why complain? It isn't like this thread is aiming for efficiency anyways.
Bucknasty is working on remaking the thread so we have someone who can update the OP daily and luckily Buck tanks Sarth 3D and is an exceptional tank through my view which makes the fact that he is the one remaking the thread even better.
To comment on your SS vs SG post:
~1222 HP is the grand total net if you had a repelling charge.
~1550 HP is the grand total net if you had a seal of pantheon.
I highly doubt someone would deliberately use the essence of gossamer or something like that if they used the SS runeforge. I used the same layout you did but subtracted defense based on the trinket, but now that I think about it you don't have 3 trinket slots which means a lot of people would lose another 1200 stamina from dropping there Essence of Gossamer (if they were truly going for the most avoidance).
Digging through 100 pages is completely unreasonable. My gripe is that we get the same post on every page, and that people feel the need to re-do the math (frequently incorrectly as you point out) daily. Today's discussion has followed exactly the same path as yesterday's:
1) SG is only good for new tanks, otherwise it's patently inferior
2) No, SG is a tradeoff of 1% avoid for roughly item-points equivalent Sta, let's discuss which is more valuable!
3) No, SG is a tradeoff of at worst 1% avoid for ipe Sta because anyone taking SG is going to regem for something they value more than avoid rather than stay at 565, because if they valued avoid more than whatever else, then they would have stayed with SS in the first place.
Then everyone ignores (very important) point 3, discussion dies down, and the cycle repeats 24 hours later.
I'm not adverse to discussing tanking topics to death in the tanking thread, that's what it's for, but seeing discussion caught in a time loop drives me nuts.
I'll definitely agree with you that an up-to-date and accurate OP would help matters. I'm not sure how that's best achieved.
I completely agree. However maintaining and updating a post like that is time consuming and takes effort. It is EXACTLY what I'm trying to do with the FAQ, and each time we come up with something new I take note of it and try to add it into the FAQ somewhere. Currently, I'm working on just getting those repeat questions taken care of first. However, all that work is for naught when they come here and just ask and get answered anyway.
If you read thru the FAQ and find something to be wrong, please let me know either in thread or via PM and I will fix it straight away. New people coming in have told me how much of a help it is, and I want it to continue to be so. Any help I get is greatly appreciated.
I do hope Bucknasty does go thru with it, but like I mentioned before, its a pretty massive undertaking.
Sigh, people still don't get it. It isn't about taking less damage or about maximizing your TTL(though your calculations only work on avoidable damage). Getting SG/Gemming stam is about maximizing you EH so you don't die to bad strings, so you survive Sartharion breaths. No matter how we gem we will have the highest TTL of any tank on an average fight, so why not cover our weakness? (ie our tendency to die when we don't avoid 3-5 times in a row)
Sometimes average survivability is worth sacrificing to improve your worst case survivability. If you were often taking in a situation with only 1 healer on you, taking overall less damage might be better, but generally tanks just get tons of overhealing.
Would you rather have 10k hp with 95% avoidance or 100k hp with 5% avoidance? I think the answer is pretty obvious. Finding the tipping point where gaining avoidance is better than gaining EH is tricky, but that tipping point in actuality is far more slanted to EH than a simple TTL calculation.
As an aside, I agree that there is no reason to be gemming for threat at the moment. Better off respecing to something higher threat, glyphing threat, or calling for extra misdirect/tricks if you are having problems(which you shouldn't be to be honest).
I did say in a few posts ago that I can see it's value for a sarth 3D tank.
I am an advocate of gemming for EH if at all possible. But when SG was coming out, I just cannot fathom it being worth using (for me). At best I could drop my defense trinket and pick up the Gossamer, gaining me about 1.7k health. As nice as that is, if I was truly concerned for health I'd have a different talent build that included VotTW and/or SoD.
I am not recommending avoidance tanking. But as you pointed out, it's a balance. At 40k life I can't see the value of gaining another 500 health. If I were tanking sarth 3d and was using double stamina trinkets, I'd use SG in a heartbeat though.
It is spelled "through". If you are going to edit and update a FAQ, which I applaud you for doing, you should do so in proper English.
Originally Posted by FlamingCloud
Would you rather have 10k hp with 95% avoidance or 100k hp with 5% avoidance?
Pointless question is pointless. The mechanics of tanking in WoW require a balance of HP totals and mitigation to provide the most numerically consistent mitigation possible over a wide variety of fights and abilities. 100k HP and 5% avoidance would probably suck really bad against -% reduction abilities like Shear, but would be great against high damage, unavoidable attacks like Sarth3D breath.
The reason Stamina increases in value once a certain threshold of avoidance is reached is simple. Stamina scaling remains linear, and avoidance scaling diminishes rapidly over 50%. Add that to the FAQ and anyone who understands English should be able to figure how to gem for those two stats optimally (I'm kidding obviously, it needs to be explained in baby steps).
Pointless question is pointless. The mechanics of tanking in WoW require a balance of HP totals and mitigation to provide the most numerically consistent mitigation possible over a wide variety of fights and abilities. 100k HP and 5% avoidance would probably suck really bad against -% reduction abilities like Shear, but would be great against high damage, unavoidable attacks like Sarth3D breath.
The reason Stamina increases in value once a certain threshold of avoidance is reached is simple. Stamina scaling remains linear, and avoidance scaling diminishes rapidly over 50%. Add that to the FAQ and anyone who understands English should be able to figure how to gem for those two stats optimally (I'm kidding obviously, it needs to be explained in baby steps).
Actually diminishing returns on avoidance makes avoidance scale linearly, it was exponential before DR. Pointless question had a point, I was illustrating that things are not so cut and dry as what stat gives you the better TTL in a no heal vacuum. If you think you have a formula for calculating exactly when stamina is better than avoidance for a given fight, and at what health stamina is sufficient and you should switch back to avoidance please share it with the rest of us.
In my personal opinion as frost(Deep Frost) we have no major stamina buffs I wanted the 2%.
Being over defense lets my have some breathing room and I have only gotten 2 weeks of raiding to check where I stand on all my stats.
Also has anyone graphed the chart of IBF? Does it stop at 50%?
I have looked around and not found anything conclusive.
If IBF scales directly with Defense in a linear function would not having more defense to increase overall mitigation be worth it? Mostly for the Frost tanks especially with the T7 bonus.
If this is a repeat topic let me know and I will dig around the 100 pages more to find it.
The problem with that logic Okuno is that we don't need to increase our mitigation while IBF is up. As far as I know I have never died with IBF up, ever. What you want to work on is making survivability higher when cooldowns are not up.
Actually diminishing returns on avoidance makes avoidance scale linearly, it was exponential before DR. Pointless question had a point, I was illustrating that things are not so cut and dry as what stat gives you the better TTL in a no heal vacuum. If you think you have a formula for calculating exactly when stamina is better than avoidance for a given fight, and at what health stamina is sufficient and you should switch back to avoidance please share it with the rest of us.
I think we must be using two different definitions of linear then, or I am not using it precisely enough. Avoidance rating has a decay after a certain point, do we know that this decay makes it behave like stacking stamina? No, we know that it does not behave that way, whatever benefits stacking stamina may have, they are the same static values for each point of it. A point of dodge rating when you have 10% dodge is better than a point when you have 40%, thus dodge rating does not scale in a linear fashion.
I do not have a formula for calculating when stamina is better. But I can give you an educated guess of 55% avoidance from dodge and parry, based on current in game gear, and rating decay I am seeing on my own gearset. Maybe that jives with what you are saying, maybe it doesn't.
I think we must be using two different definitions of linear then, or I am not using it precisely enough. Avoidance rating has a decay after a certain point, do we know that this decay makes it behave like stacking stamina? No, we know that it does not behave that way, whatever benefits stacking stamina may have, they are the same static values for each point of it. A point of dodge rating when you have 10% dodge is better than a point when you have 40%, thus dodge rating does not scale in a linear fashion.
When you have lets say 50% avoidance the next 1% of avoidance decreases your total damage taken by 2%, if you had 98% the next 1% would reduce damage taken by 50%. DR essentially brings avoidance rating back to being linear(in theory). It is the same with armor, it suffers the same type of dr that avoidance does.
Avoidance does not become less effective the more you have, the amount you get from 1 rating decreases, but its effect essentially does not.
According to the findings in this thread, dodge and defense actually still scale positively even with the diminishing returns system. Thus, +25 dodge rating is worth slightly more when you have +500 dodge as when you have +100 dodge. Nowhere near to the same degree as in BC, of course.
Parry doesn't fair so well, presumably because the parry cap is so low compared to dodge.
In either case, it doesn't really help to answer the question of effective health vs avoidance. That question can only really be answered for individual encounters, and even then not completely (it would still depend on how willing you are to trust the RNG, your healers, the length of the fight, etc.)
Sartharion+3 has a pretty obvious answer; surviving breaths is the most critical part of the fight, so gearing for max health is the best solution.
In a hypothetical fight with 20 adds each attacking you once per second (and thus providing extremely steady incoming damage), gearing for avoidance would be the best solution.
Has anyone run any numbers on how defense point for point effects IBF?
Does it scale beyond 50%?
If so would you factor that into overall damage mitigation?
It make sense that if your defense put IBF up to 60% would that not lower your overall damage taken 30% of the time. It is only 21 seconds however that is an average of 3.3% less damage a minute over a 5 minute boss fight it adds up.
This idea applies mostly to a deep frost tank as their life tends to be lower than their blood/unholy brothers.
[EDIT : My math may be off but I hope the idea can still be interpreted]
I did say in a few posts ago that I can see it's value for a sarth 3D tank.
I am an advocate of gemming for EH if at all possible. But when SG was coming out, I just cannot fathom it being worth using (for me). At best I could drop my defense trinket and pick up the Gossamer, gaining me about 1.7k health. As nice as that is, if I was truly concerned for health I'd have a different talent build that included VotTW and/or SoD.
I am not recommending avoidance tanking. But as you pointed out, it's a balance. At 40k life I can't see the value of gaining another 500 health. If I were tanking sarth 3d and was using double stamina trinkets, I'd use SG in a heartbeat though.
I was not trying to say that SG is better or worse than SS. My point was that at my gear level, being able to gem for stam while using SG allows me to pick up gear with more avoidance stats themselves - effectively making SG vs SS neutral at best and possibly even making SS worse for my gear level (i.e. I could go with SS but I'd have to go with higher def. gear which has less dodge/parry/expertise/hit/stam) making that avoidance granted by SS moot. My issue was with your statement that only newly 80 DKs (lack of gear) or those tanking sarth3D (obvious reasons) should use SG. The enchant will have different values at varying gear levels. If you can get away with not using it and still be effective - I'm happy for you - for me, SG is a better choice. I'm 100% with you on the balance - if you need to use SG to achieve good balance, there is nothing wrong with that. RNG can suck at times.
Last edited by Lurne : 01/30/09 at 1:43 AM.
Reason: clarity
Has anyone run any numbers on how defense point for point effects IBF?
Does it scale beyond 50%?
If so would you factor that into overall damage mitigation?
It make sense that if your defense put IBF up to 60% would that not lower your overall damage taken 30% of the time. It is only 21 seconds however that is an average of 3.3% less damage a minute over a 5 minute boss fight it adds up.
This idea applies mostly to a deep frost tank as their life tends to be lower than their blood/unholy brothers.
[EDIT : My math may be off but I hope the idea can still be interpreted]
Erm, at 60% damage reduction from all sources for 21 seconds every 60 seconds, it's 21% less damage taken over a 5 minute fight (or a 10 minute one, or a 1 minute one). If you pop it before any predictable high incoming damage environment, it's even more. To be honest, we need to find someone who has well over 600 defense to be able to properly test how defense impacts IBF.
Anyway, napkin math seems to indicate that gemming/enchanting for defense > gemming/enchanting for dodge, even once you hit 540. Does a mob's chance to miss you get capped? If it does, I'll have to recalculate it. If there is no cap, then pile on the Defense!
Last edited by Raronikenshin : 01/30/09 at 3:50 AM.
Atm i can use IBF only when i need it, not every cd, so i dont think that stacking def for extra avoidance and better IBF have any value (at least for me). Also SG is way better than SS, simply because i could change 3-4 items with less/no defence but high avoidance (dodge mostly - druid items), and now with SG i am at same defence level as i was when using SS, but i have higher avoidance (much higher than from SS or SG itself).
I changed my back, ring, gems, and it took me to same avoidance level -4% lost from SS. By changing gems/items/enchants i gained ~7-8%, and now i'm at ~63% avoidance. Before that i was about 3% lower and additionally i have 2% more hp, more expertise and much more stamina from gems/enchants.
Recently i've noticed that it's better for my and my healers when im Effective Health tank, not Avoid tank, damage is more predictable, i don't have situations where i'm at 10-20% after spike dmg anymore.
So my opinion, from my personal experience is that HP tank with ~60% avoidance, stacking stamina and tps (exp, str, maybe hit later) stats is better for my group, because such MT can be healed easier, there are much much less "omg he almost died!" situations.
Erm, at 60% damage reduction from all sources for 21 seconds every 60 seconds, it's 21% less damage taken over a 5 minute fight (or a 10 minute one, or a 1 minute one). If you pop it before any predictable high incoming damage environment, it's even more. To be honest, we need to find someone who has well over 600 defense to be able to properly test how defense impacts IBF.
Anyway, napkin math seems to indicate that gemming/enchanting for defense > gemming/enchanting for dodge, even once you hit 540. Does a mob's chance to miss you get capped? If it does, I'll have to recalculate it. If there is no cap, then pile on the Defense!
Originally Posted by huntcaudata
Of course you can't ignore gem bonuses, etc. Consider, though, in my current gear a Dodge Gem gives me .306% avoidance, and a Defense Gem only .240%, or a difference of 3.5 dodge, or 4.4 parry rating worth. Your socket bonus would have to be better than that (not that difficult) to merit using the inferior gem. At the very least, it should make you use a dodge/defense gem (I think these exist...) rather than a pure defense gem. Or, it should convince you to put JC gems in Yellow slots instead of red slots. So, you do often get to choose between Dodge and Defense after all.
The point though, is to convince people to stop asserting things like, "I feel as though Defense must be better because it spreads out your avoidance," without having any theory to back it up.
In order to get to 60% damage reduction from IBF, you would need ~800def, or close to 2000 Def rating.
Each additional point of defense skill provides an additional ~0.11% absorption reduction, which works out to about 0.02% reduction per rating while IBF is up, or a total of an 0.008% of damage reduction over 1 minute, per point.
As for how additional defense affects (not effects, people: please learn the difference between these two words) IBF: the math has been done, it's linear.
Tastes like Awesome, because it's made of Awesome(TM)
In order to get to 60% damage reduction from IBF, you would need ~800def, or close to 2000 Def rating.
Each additional point of defense skill provides an additional ~0.11% absorption reduction, which works out to about 0.02% reduction per rating while IBF is up, or a total of an 0.008% of damage reduction over 1 minute, per point.
As for how additional defense affects (not effects, people: please learn the difference between these two words) IBF: the math has been done, it's linear.
I think you missed the post where IBF is not 35% at 540, but was tested to be 41%(or 0.15% per skill). Still, why improve IBF, you don't die with IBF up, and in general your healers aren't going to heal less. Unless we hit a fight where the fight has a 10-20 second window of danger once a minute you are better off with dodge or stam.
Originally Posted by Gurtogg
Test results I've seen measuring DoT damage show the figure is higher, that its linearly scaled at 0.15% per point of defense.
400 defense - 20% base IBF
540 defense - 21% increase to IBF
600 defense - 30% increase to IBF
I tested DoT damage at 400 (20%), 540 (41%) and 557 defense (43.55%). Results showed it does indeed scale at 0.15% per point.
Bucknasty is certainly going through with it, but the reality will be just short of my dreams of having a magic post with the best answer to every question to just drop on the community, then update from there. I will spend most of today editing/updating/reorganizing my op on this thread. Before I leave work I'll stick it in a new thread. From that point on it will receive regular attention from me during CST business hours. I play on the weekends, then theorycraft about my play once I'm getting paid.
Bucknasty is certainly going through with it, but the reality will be just short of my dreams of having a magic post with the best answer to every question to just drop on the community, then update from there. I will spend most of today editing/updating/reorganizing my op on this thread. Before I leave work I'll stick it in a new thread. From that point on it will receive regular attention from me during CST business hours. I play on the weekends, then theorycraft about my play once I'm getting paid.
Looking forward to it. I think my guild is going to push me to move to a Frost tanking build because we recently picked up some more melee and we do not have an enhancement shaman (IIT here I come). I haven't spec'd into Frost in a while so it will be interesting to see some numbers on it and different rotations used since 3.0.8 hit, as well as checking to see how much threat KM provides now.
When you move from Heroics to Naxx10 it is very difficult to use a lot of upgrades from blue to purple because of the requirement to be Def Soft capped. Using SG allows you to equipt those purples and get some extra health. So to say its for when you just hit 80 is very misleading.
At a certain point parry starts to seriously hit diminishing returns and since SS is seemingly not effected by DR that +4% parry is much better because it would require the equivalent of many more points of parry rating than the 4% represents.
If you assume DR is effecting parry by 25% then the +.33% parry from SG is only gaining .08%, the 4% from SS is gaining a full 1%, so it is clearly more of a benefit of 1% avoidance for 2% health tradeoff.
There is obviously an inflection point, but as a simple rule of thumb, when you can wear all your best gear and get all your gemming bonuses, only enchant your cloak and chest for defense and still be uncrittable, switch to SS
If you assume DR is effecting parry by 25% then the +.33% parry from SG is only gaining .08%, the 4% from SS is gaining a full 1%, so it is clearly more of a benefit of 1% avoidance for 2% health tradeoff.
There is obviously an inflection point, but as a simple rule of thumb, when you can wear all your best gear and get all your gemming bonuses, only enchant your cloak and chest for defense and still be uncrittable, switch to SS
Does someone remember where the testing that supported the "SG is not on DR" entry in the FAQ is located? I don't remember seeing it, but Shadai is generally not careless when adding things to the FAQ, so I'd like to think it's supported somewhere, and it'd be nice to be able to reference it when things like this come up.
As far as the second point goes, I guess I'll start the loop again:
SG is also preferable in any situation where any other stat is preferable to pure physical avoid in item points equivalent amounts. People have proposed gemming for threat/dps during farming, or gemming for EH for encounters where it's more useful (I can also see some value in having a deeper HP pool when trying to trim healers from farm raids or for DPS races).
I think it would help in general if we turned away from this sort of binary "use x until y" attitude, and instead looked at both runes (as well as spellshattering) as tools that we can flex between. The ability to re-enchant your weapon for the cost of a soul shard and about two minutes of raid time is an asset that should not be overlooked, especially on progression content where "good enough" might not do, and you might want to optimize from encounter to encounter.
Does someone remember where the testing that supported the "SG is not on DR" entry in the FAQ is located? I don't remember seeing it, but Shadai is generally not careless when adding things to the FAQ, so I'd like to think it's supported somewhere, and it'd be nice to be able to reference it when things like this come up.
As far as the second point goes, I guess I'll start the loop again:
SG is also preferable in any situation where any other stat is preferable to pure physical avoid in item points equivalent amounts. People have proposed gemming for threat/dps during farming, or gemming for EH for encounters where it's more useful (I can also see some value in having a deeper HP pool when trying to trim healers from farm raids or for DPS races).
I think it would help in general if we turned away from this sort of binary "use x until y" attitude, and instead looked at both runes (as well as spellshattering) as tools that we can flex between. The ability to re-enchant your weapon for the cost of a soul shard and about two minutes of raid time is an asset that should not be overlooked, especially on progression content where "good enough" might not do, and you might want to optimize from encounter to encounter.
This is testing that as with SS does not need to be determined. Take a look at the blue post monitor on mmo-champ and look at the GC post that states SG and SS are not affected by the DR for the answer.
I think it would help in general if we turned away from this sort of binary "use x until y" attitude, and instead looked at both runes (as well as spellshattering) as tools that we can flex between. The ability to re-enchant your weapon for the cost of a soul shard and about two minutes of raid time is an asset that should not be overlooked, especially on progression content where "good enough" might not do, and you might want to optimize from encounter to encounter.
Sorry if I was unclear. I was not saying it never pays to use SG once you are above a certain gear level, I was saying it doesn't pay to use SS until you can really afford to. That point is not when you first hit 80 or even close. And, that once you can, because of diminshing returns, the benefit of SS is better than the 1% avoidance difference being quoted.
There will always be situations when special gear choices need to be made, Sarth+3 is a prime example, needing extra threat is another, but for most general uses, once you are well geared, SS seems superior for avoidance.
It also seem likely that, when the new content is released, frost gear will be needed. SG will be of great use to stay uncrittable in that gear.
01/29/09, 5:33 PM
Similar Threads
