Archived

This topic is now archived and is closed to further replies.

Esoth

[Cataclysm] - Survival Hunter

627 posts in this topic

1. Addressing serpent vs arcane if calculations are correct and 50 dps more from arcane glyph on perfect 1 target fight in most of fight you will keep serpent on more targets then one (or aoe with serpent spread).

Nefarian 2 targets

Cho'gall 2 targets

Ascendant Council 2 targets

Valiona and Theralion 2 targets on some part of fight

Halfus Wyrmbreaker 3-4 targets in same time depends of tactics you use

Maloriak green phase aoe

Magmaw adds if you chose to kill them (we just kite them to end of fight by frost dk)

Omnotron green adds aoe

Conclave of Wind adds aoe

So in only 3 fights arcane could prove better then serpent making serpent winner overall.

2. Thing i wont to discus is benefit of improved stady shot talent as survival.

For me with 4% haste with me with around 350 ilvl i never use arcane shots (my guild pop lust on start so even then i spam cobra 60% more crit and 1.1s cast time) so i think getting 15% haste is big dps incise with uptime i could have it would be around 11% effective haste.

I can stick 2 stedis and one cobra/ba/arcane in be-twine every explosive shot making solid uptime i would lost cobra shot damage because stedy hitting less and 3 talent points (1 from serpent spread and 2 of sic'em) but gaining atleast 11% haste.

i plan to have dual spec one with normal spec and one for single target dps with ISS.

3. survival tactics or hunter vs wild should be dropped for entrapment bellicose it is big help on kiting on Magmaw Maloriak and green adds on Omnotron.

Share this post


Link to post
Share on other sites
Yup but 1% crit is not the same DPS increase as 1% mastery. while mastery increasing your magic damage, let´s say, about 80% of your total damage, every point of mastery adds the same amount of damage to that portion. On the other hand, the more crit you have the less valuable it is, as each point of crit added has a lesser impact on your DPS. There might be a shifting moment where mastery surpass crit in absolute value.

no, there is no diminishing return on crit (besides the DR on GFTT). Go try it out in femaledwarf.

Ignoring procs and chaotic meta for the sake of simplicity, going from 0% crit to 1% crit is exactly a 1% dps gain. More realistically going from 20% crit to 21% is a 0.833% dps increase. This isn't diminishing returns however. Even agility is better at low levels of gear if you look at it as a percent increase.

Your first percentage of mastery is not a 1% increase to your elemental damage. It's a (1.09/1.08)= 0.925% increase. As you get more mastery rating, it will suffer from the same fake diminishing returns.

Share this post


Link to post
Share on other sites

Also, regarding the crit vs mastery debate, note that most players aren't able to use the 3% crit damage metagem which, once fixed, will somewhat increase the value of crit. Fiddling with FD suggests that crit will stay slightly superior to mastery all the way until pet crit cap, I think.

Share this post


Link to post
Share on other sites
I personally think, that KC is a good focus dump in situations with dangeriously high focus, because when youre attempting to min-max too much with AS or SS, a TotH-procc can pretty much mess you up.

Mess you up? If Thrill procs from Arcane, you have just spent 22 focus and gained 9. If that caps you out, something is wrong. And that is most likely not even a full waste. It will only ever be a full waste if you start at 100 and it procs. That's near impossible.

Thrill proccing is always better than not (I haven't read anything to indicate it has an internal cooldown). It's a bonus. Trying to make sure it doesn't proc seesm to be rather counterproductive. You should hope it procs on every ability possible. Besides it can't proc from Serpent Sting anyway. As I said, KC needs to be 60% more damage than the Improved Serpent Sting proc. If you are at 100 when you cast it, then within 1 second you need to regen 25 focus for it to cap again, this is the only time KC needs to do less than 60% more damage. Obviously you would use Arcane Shot after the second LnL Explosive.

Share this post


Link to post
Share on other sites
Unrelated question: Has anyone tested if our mastery affects non-hunter procs and abilities? For instance, the goblin rocket racial, [item]Darkmoon Card: Hurricane[/item], etc.

Enhancement shaman's mastery increases the damage of the darkmoon card(they have pretty much the same one as us), so I don't see why SV's mastery wouldn't increase the damage of the same things.

Share this post


Link to post
Share on other sites
You need 179.28 haste rating to get 1% haste.

I was under the impression that 128.07 haste rating is needed to get 1% haste. Using Zeherah's Haste calculator, 128 reflects 1% Haste.

Share this post


Link to post
Share on other sites
I was under the impression that 128.07 haste rating is needed to get 1% haste. Using Zeherah's Haste calculator, 128 reflects 1% Haste.

You are correct, I must have looked at the value for Crit or Mastery on accident since those are both 179.28. Edited the OP to reflect this.

Share this post


Link to post
Share on other sites

I now tried to include KC into the rotation without great dmg increase. Vice versa I experienced a slight loss of ~400dps. I am aware of saying that everything went like hitting the dummy in Stormwind. It was testified within the Baradin Hold. Using AS seems worth it. I will now focus on the MS thing.

Share this post


Link to post
Share on other sites

The KC would only ever be worth looking into when you're LnL procced with ~full focus AND everything else already on CD anyway, it's so situational that you really wouldnt get many come up in a fight unless you were very unlucky. As such, if you noticed a 400dps loss, I would say you were actually using KC too much. I would be surprised if you had to hit it more than twice in a 5 minute fight.

You can probably look at your raw numbers and work out whether KC or MS will work better for you in this specific situation, though there are a few things to consider; as soon as more than one mob is involved, MS would beat out KC. Likewise if any pet shenanigans are going on, KC might not even be an option.

The only other factor to consider is whether its a dps loss to burn that focus on a sub par ability and waste the GCD, or accept you're focus capped, hit another cobra instead of standing around waiting, and dumping focus like mad on AS the second LnL dissapears.

Share this post


Link to post
Share on other sites

-A question to make it clear; on a single target fight 5min+, assuming I have ganerated enough focus, is it better to re-cast SS (imp SS) or always refresh SS through CS?

Share this post


Link to post
Share on other sites

-Finally a question to make it clear; on a single target fight 5min+, assuming I have ganerated enough focus, is it better to re-cast SS (imp SS) or always refresh SS through CS?

It is better to refresh your sting with CoS. CoS for one is nature damage that goes along with being boosted by your mastery anyway. Also there is a bug right now that cuts down the regen of focus by 10% instead of boosting it 10%. This results in very little focus to dump on things like arcane shot.

Share this post


Link to post
Share on other sites

Based on my personal testing, the focus regen being reduced by Hunting Party seems to be fixed.

To test, I'm using the following macro:

/script DEFAULT_CHAT_FRAME:AddMessage("Focus: " .. UnitPower("player") .. " at " .. GetTime())

What I do, is take all my gear off so I have 0% haste on my character sheet, but still have the 10% from hunting party.

I spam trap launcher to get my focus down to <10.

I hit the macro.

I wait until my focus is >90.

I hit the macro again.

Take the total amount of focus gained and divide by the number of seconds between macro activations.

It keeps coming out to 4 focus/sec, give or take a couple % which is due to the focus value being limited to integers.

Unless I'm misunderstanding the issue, my testing is showing that it's fixed.

Share this post


Link to post
Share on other sites

Are you able to repeat the experiment with consistent results, and also with a talentless spec as a form of control? I would do it but am at work posting from my phone.

Share this post


Link to post
Share on other sites

Unrelated question: Has anyone tested if our mastery affects non-hunter procs and abilities? For instance, the goblin rocket racial, [item]Darkmoon Card: Hurricane[/item], etc.

It does. Same goes for the engineering glove-enchant and other various elemental procs

Share this post


Link to post
Share on other sites

no, there is no diminishing return on crit (besides the DR on GFTT). Go try it out in femaledwarf.

Ignoring procs and chaotic meta for the sake of simplicity, going from 0% crit to 1% crit is exactly a 1% dps gain. More realistically going from 20% crit to 21% is a 0.833% dps increase. This isn't diminishing returns however. Even agility is better at low levels of gear if you look at it as a percent increase.

Your first percentage of mastery is not a 1% increase to your elemental damage. It's a (1.09/1.08)= 0.925% increase. As you get more mastery rating, it will suffer from the same fake diminishing returns.

Pichua did not say anything about diminishing returns. He is referring to exactly what you posted, gradual decreases in the dps percent increase value of crit. What needs to be accounted for is that Mastery is a percentage effect. It's value to our dps increases the more damage we are doing with our spell damage. As our crit gets higher providing more total damage and becoming less valuable percentage-wise, mastery may be increasing in value percentage-wise because of the extra crits and what not.

Let's look at an extreme case as an example:

100 shots @ 1000 damage per shot @ 1 shot per second @ 99% crit @ 0 mastery

Total Damage: 199000

100 shots @ 1000 damage per shot @ 1 shot per second @ 100% crit @ 0 mastery

Total Damage: 200000

100 shots @ 1000 damage per shot @ 1 shot per second @ 99% crit @ 1 mastery

Total Damage: ((100*1000) + (99*1000))*(1+.8*.01)

Total Damage: 200592

Clearly, this is a fictitious example. Before you go down the path of thought "well, it's only 592 over", remember that this is a fictitious example that is only proving that mastery can eventually overtake crit. Mastery is more valuable the more damage we are causing. Each point of mastery is, more or less, equal to .8% increase in dps. When the percent dps increase of crit drops below .8%, mastery becomes more valuable.

I guess, as a note, the only thing not completely solid in this regard is that I'm assuming mastery applies to 80% of the damage dealt. Clearly, this figure is not completely accurate and can fluctuate from battle to battle due to LnL procs and what not.

Someone could back me up on this, but off hand I am pretty sure that as our crit % goes up, the percentage of damage effected by mastery goes up, since all of our magical junk hits harder than our auto shots. This would mean that as crit% goes up, so does the swapping percentage when mastery overtakes crit.

Share this post


Link to post
Share on other sites

Someone could back me up on this, but off hand I am pretty sure that as our crit % goes up, the percentage of damage effected by mastery goes up, since all of our magical junk hits harder than our auto shots. This would mean that as crit% goes up, so does the swapping percentage when mastery overtakes crit.

You're right, the equation for average damage (ignoring things like ability-specific crit modifiers and the 3% metagem, for the moment) is:

(1+crit%)*(1+mastery%)*(average elemental damage) + (1+crit%)*(average physical damage)

This means that the elemental portion of our damage will be maximized when crit% = mastery%, assuming that you're converting crit rating directly into mastery rating and vice versa, by reforging, regemming, switching gear, etc.

Assuming that our autoshots/killshots are roughly 20% of our total damage, then in reality you don't get a better return from mastery until your crit% is three or four times greater than your mastery%. Take into account crit-based procs and crit rating affecting our pets, and crit becomes better pretty much all of the time.

Share this post


Link to post
Share on other sites
You're right, the equation for average damage (ignoring things like ability-specific crit modifiers and the 3% metagem, for the moment) is:

(1+crit%)*(1+mastery%)*(average elemental damage) + (1+crit%)*(average physical damage)

This means that the elemental portion of our damage will be maximized when crit% = mastery%, assuming that you're converting crit rating directly into mastery rating and vice versa, by reforging, regemming, switching gear, etc.

Assuming that our autoshots/killshots are roughly 20% of our total damage, then in reality you don't get a better return from mastery until your crit% is three or four times greater than your mastery%. Take into account crit-based procs and crit rating affecting our pets, and crit becomes better pretty much all of the time.

Why is the elemental portion maximized when Master% = crit %? In the equation above, the elemental portion of the damage will always increase as mastery increases, so it's maximized when mastery is infinitely large.

Why do you not get a better return until crit% is 3 to 4 times greater than mastery? I believe you are saying 20%/80% for 4 times greater? I believe you are over simplifying the problem, but I would love to see some math to the contrary if you can provide it.

Now, other considerations you posted have some relevance which lead to the fuzzy area of dps maximization. Mastery will definitely not help our crit based procs, etc. Also my math did not include the possibility of a 3% crit meta (although I doubt anyone in top tear will use such a meta until they change the red/blue requirement). We can make attempts to factor in some value for these considerations, but they do not make my previous post any less accurate. Mastery will overtake crit at some point.

For example, at 26% crit the dps increase for 1% crit is already down to .8%. Here's some fun numbers:

0-1% = 1% increase

25-26% = .8% increase

50-51% = .67% increase

75-76% = .57% increase

Keep in mind that the percent increase of dps for 1% in mastery will always be:

(percentage of total damage effected by mastery)*(.01)

Also keep in mind that as crit % increases, the (percentage of total damage effected by mastery) most likely also increases.

Does this make mroe sense to you? The % increase value of mastery only every increases while the % increase value of crit only decreases. When mastery surpasses crit, mastery becomes more valuable. Based on the 80% assumption, this will occur around 25% crit.

Share this post


Link to post
Share on other sites

Sorry, I see what you were saying about maximization. Brain spasm.

Also, my assumption that mastery will have the same % increase value is incorrect, it diminishes for the same reasons (and in a similar fashion) that crit does.

Also, I verified in a model that you are correct, mastery is only more beneficial when crit is 3-4 times greater than mastery when 80% of the damage dealt is elemental. However, that modifier (3-4 times greater) drops quickly as the percentage of damage dealt by elemental sources increases above 80.

For instance, say you could theoretically deal 90% of your damage in elemental attacks (maybe using a cunning pet? Does mastery work that way?). The benchmark that was "if crit is 3 times greater" has been reduced to "if crit is 1.33 times greater" -- that's less than half of what it was.

I think that to maximize damage, we will really have to monitor that percentage of our damage. With an 8% mastery base, it will take awhile for our crit to get high enough for us to worry about mastery with a benchmark of 3-4 since we really can't avoid having mastery on our gear. However, if we begin doing elemental damage more than 80% of the time, that benchmark may collapse to a more pertinent number.

Share this post


Link to post
Share on other sites

Keep in mind that the percent increase of dps for 1% in mastery will always be:

(percentage of total damage effected by mastery)*(.01)

No, that was the point of the original point where I brought that up. Mastery suffers from the same fake diminishing returns as crit. Going from 8 mastery(base) to 9 is a 0.925% increase to our elemental damage. Going from 20 mastery to 21 is a 0.833% increase to our elemental damage. Going from 40 mastery to 41 mastery is a 0.714% increase to our elemental damage. The situations where mastery overtakes crit are not realistic. Mastery overtaking haste at certain soft caps is likely to happen, though.

Share this post


Link to post
Share on other sites
No, that was the point of the original point where I brought that up. Mastery suffers from the same fake diminishing returns as crit. Going from 8 mastery(base) to 9 is a 0.925% increase to our elemental damage. Going from 20 mastery to 21 is a 0.833% increase to our elemental damage. Going from 40 mastery to 41 mastery is a 0.714% increase to our elemental damage. The situations where mastery overtakes crit are not realistic. Mastery overtaking haste at certain soft caps is likely to happen, though.

Thanks Tragos, I see that now. Poor assumption on my part and edited the above post while you were typing this one I think. :)

Footloop has the right equation to model from and the %damage dealt by elemental attacks will be the key to determining when mastery will be better than crit.

Share this post


Link to post
Share on other sites
Why is the elemental portion maximized when Master% = crit %? In the equation above, the elemental portion of the damage will always increase as mastery increases, so it's maximized when mastery is infinitely large.

Sorry, I should have elaborated. Since we can transform mastery rating to crit rating and since the same number of points are required for 1% of mastery and 1% of crit, we can determine a relationship between them. Take 't' to be your total crit%+mastery%, so t = c + m. We can transform crit <-> mastery at a 1:1 rate (not quite limitlessly, but freely enough for our purposes), so t = (c+1) + (m-1) => t = c+m and we can see that the equation holds as we vary crit and mastery rating. Describing crit as a function of mastery, we have c = t - m.

Plugging this back into our formula for elemental damage, we end up with (1+c)*(1+m)*(damage) = (1+t-m)*(1+m)*(damage) = (-(m^2) + t*m + 1+t)*(damage). Average base damage doesn't change with respect to mastery, so our damage is going to be maximized when the equation in the first set of brackets is maximized. Take the first derivative to find a maximum, and you end up with -2m + t = 0. Solving for m we find m = t/2, so plugging back into our first equation we find that this coincides with c = t/2, and hence c = m.

Now that we know our maximum damage is at c% = m%, we want to figure out how this changes when we throw physical damage into the picture. We can do this by describing our total damage as a weighted average of physical and elemental damage, so that the equation in my previous post becomes:

(1+c)*(1+m)*(elemental) + (1+c)*(physical) = (1+c)*(1+m)*(elemental% * total damage) + (1+c)*(physical% * total damage)

Our earlier assumption of 20% physical damage, which is a fairly rough estimate but matches up decently to what I've seen from raid logs, gives us: (1+c)*(1+m)*(.8 * d) + (1+c)*(.2 * d). Making the same t = c+m substitution as before:

.8*(-(m^2)+t*m+1+t)*(d) + .2*(1+t-m)*(d) = d * (.8*(-(m^2)+t*m+t)+.2*(1+t-m))

again, damage shouldn't vary so we need to find a maximum of the value in the brackets, which has a maximum at:

-1.6m + .8*t -.2 = 0, or t = 2m + .25. This means that c = m + 0.25.

So it turns out that my initial statement of 3-4x was incorrect; you really want your crit% to be 25% higher than your mastery%. Keep in mind that this is ignoring your pet damage, which will cause crit to be worth a more, and also ignoring +crit% talents, which will cause crit to be worth a little less (if you're only looking at your character sheet numbers).

e: oops, had to go away for a bit and left this tab open without refreshing. I guess the first half you already figured out, but I think the last calculation is still pretty interesting. Assuming my math is right, at least.

double e: generalizing further, you want c% = m% + (%physical)/(%elemental). Again keep in mind this is without the pet, so crit is going to be worth even more than this indicates.

Share this post


Link to post
Share on other sites

Edit** Seems footloop beat me to it by a few minutes but I'll leave this here because my explination may be a bit easier to follow for some.

No, that was the point of the original point where I brought that up. Mastery suffers from the same fake diminishing returns as crit. Going from 8 mastery(base) to 9 is a 0.925% increase to our elemental damage. Going from 20 mastery to 21 is a 0.833% increase to our elemental damage. Going from 40 mastery to 41 mastery is a 0.714% increase to our elemental damage. The situations where mastery overtakes crit are not realistic. Mastery overtaking haste at certain soft caps is likely to happen, though.

To toss in some simple math, if you expand on the equation shown earlier:

D = D_0*(1+0.01*Crit)*(1+ 0.01*PercentElementalDamage*Mastery)

D_0 is just the base damage without any Crit or Mastery. You can solve, using partial derivatives ( dD/dCrit = dD/DMastery) for the point where these two overlap and for this particular case you get Mastery is equivalent to Crit when

Master + 100(1-PercentElementalDamge)/(PercentElementalDamage) = Crit

If we use the case mentioned above of 80% elemental damage (P=0.8) then we get the trade off occurs at:

Crit = Mastery + 25

In other words if your Crit chance is 30% and your mastery is 10% it's still better to stack Crit over mastery but if instead it were 40% Crit and 10% Mastery you'd be better off stacking mastery. Keep in mind though, this is an over simplification of this trade off only to demonstrate that a trade off does exist at some point. Where that point is precisely is probably best left up to those interested in doing numerical simulations.

Share this post


Link to post
Share on other sites

The 20/80 model is not correct, since a Ferocity/Cunning pet (yeah they are quite close) contributes about 15-17% of the Hunter's damage. Auto alone is worth more than 10%, and then Kill Shot. It becomes a lot more like 30/70. At which point any discussion about Mastery vs Crit is moot, the unavoidable values of both stats will make sure that we will not reach such a situation where Mastery should beat Crit.

This, of course, is based on the previous equations.

The only worthy 'goal' is the battle between Mastery and Haste. Now that's a more equal fight.

Share this post


Link to post
Share on other sites
Based on my personal testing, the focus regen being reduced by Hunting Party seems to be fixed.

To test, I'm using the following macro:

/script DEFAULT_CHAT_FRAME:AddMessage("Focus: " .. UnitPower("player") .. " at " .. GetTime())

Thank you for this macro, it's been helpful for my testing. My tests don't show it as being fixed, although without completely respeccing it's hard to be totally sure. A big limitation of our testing is the inability to tell exactly when the focus ticks happen (since focus does not seem to flow smoothly). I've been testing by only hitting the macro when I see the tick appear on my focus bar, so that hopefully it's always after the additional focus registers.

I also haven't been testing by using shots, but rather by swapping specs, which reduces your focus most of the way, then I just measure the time it takes to get to almost maxed focus and use the original and final focus in the math.

I have 1% haste from talents in my BM spec, and 3% in my SV spec, plus hunting party.

In BM I'm measuring between 4-4.03 fps. This is roughly in line with the expected 4.04fps that should be expected from 1% haste.

In SV I'm measuring about 3.74 fps. If hunting party was working properly, I'd expect to get 4.532 fps. Reducing focus regen by 10% instead of increasing it, then applying 3% haste comes out to 3.74 fps.

Share this post


Link to post
Share on other sites

The new engineering tinker [Tazik's Shocker] does seem to scale with mastery, as the tooltip will reflect changes in equipping gear. Using Zeherah's simulator with best raid buffs and debuffs enabled, the wrath glove tinker [Hyperspeed Accelerators] is slightly better with my current gear. However, when I reduced 240 haste in the custom gear stat options, the new tinker took a small lead.

Sadly, the new tinkers are weak, and engineering is not even approaching an optimal raiding profession for the time being.

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.

  • Recently Browsing   0 members

    No registered users viewing this page.