Yes it does have the same type of DR as avoidance does
Originally crit gets you white attack crits and special attack crits
at 76% you only gain yellow attack crits
Thus the value of your crit has diminished.
Now if you want to argue that you won't see 76% crit through current gear that is fine, but Crit does have a DR in terms of what the phrase stands for.
No, this is not what DR means. For avoidance, every point gives you less flat value than the previous point. It is a fluent curve based on a maximum value.
Crit does not have diminishing returns. Every point of crit has the same flat value as the previous point. It has "soft caps". Obviously, the hard cap is 100%. The first soft cap is 50% for FB (with RnT). The second soft cap is at ~76.6% for glancing blows on white attacks. Note, all of these values are actually 4.8% higher due to boss crit suppression (54.8%, 81.4%, 104.8%).
Diminishing returns, to me, means that at some point you get less value from the same amount than you did at one time.
Hit once you reach the hit cap has diminishing returns in a very literal sense. I see the distinction that the value you get from the same amount of hit is constant, but the value you get from it is still diminished in usage. Since I did not coin either soft cap or diminishing returns I can not speak to the validity, but in the very literal aspect, the value is still diminished to the player at some point. This point of inflection is called the "soft cap" because after the value of the stat is "diminished". <-- My view
But like I started the reply with, it is pretty much semantics, the important part here is:
-50% crit to cap FB with a bleed
- ~76.6% crit to cap white attack crits
-4.8% boss crit supression
Diminishing returns, to me, means that at some point you get less value from the same amount than you did at one time.
But this is technically true of EVERY SINGLE STAT. When you have enough AP to one-shot every boss in the game, AP suffers from your definition of "diminishing returns". The fact is that you get a roughly constant value per point of crit rating or agi at all levels of attainable gear. Even if you are technically correct, stating that crit suffers from DR is misleading.
Avoidance suffers from DR at every level of avoidance. Hit and expertise suffer from DR because soft capping is so easy to do with current levels of gear. Crit, for all intents and purposes, does not suffer from DR because no gear in the game will ever allow you to hit the point at which its value diminishes. I suppose it is a matter of semantics, but I still believe that your definition of DR -- decreasing in value at some theoretical, unobtainable value -- is not useful.
It is common sense that every stat has "diminishing returns" because when you get an additional point of attack power, the relative value of attack power will go down. And by contrast, every other stat, such as crit, hit, armor pen, and haste will rise in value in relative to attack power. This is entirely different from avoidance's artificial implanted diminishing return, where each additional point of dodge rating will give less dodge percentage in relative to the point before, past a certain point.
In terms of actual avoidance versus damage taken (this is talking about TBC, without DR in place), avoidance actually had increasing return because for each percentage you gain in avoidance, the loss in damage taken increases dramatically. I do believe the DR on avoidance brought scaling down to almost the exponential increase down to close to linear level, but I am not sure.
So, semantics regarding diminishing returns aside, come 3.1 will there be a realistic point at which AGI is no longer the preferred stat? 76% crit seems to me to be an unrealistic goal to reach, but I am new to the feral life (resto since birth). So, is there a point at which STR or some other stat benefits us more than stacking ever increasing amounts of AGI?
It is common sense that every stat has "diminishing returns" because when you get an additional point of attack power, the relative value of attack power will go down. And by contrast, every other stat, such as crit, hit, armor pen, and haste will rise in value in relative to attack power. This is entirely different from avoidance's artificial implanted diminishing return, where each additional point of dodge rating will give less dodge percentage in relative to the point before, past a certain point.
In terms of actual avoidance versus damage taken (this is talking about TBC, without DR in place), avoidance actually had increasing return because for each percentage you gain in avoidance, the loss in damage taken increases dramatically. I do believe the DR on avoidance brought scaling down to almost the exponential increase down to close to linear level, but I am not sure.
Now this is semantics. AP has DR inasmuch as it increases your DPS by a lower and lower *percentage* per point of AP. But it always increases your DPS by the same constant value. Specifically, suppose adding 10 AP increases your DPS by 5. If you do 1000 DPS or 2000, 10 AP will increase your DPS by 5 at both levels, but 5 is a smaller percentage of 2000 than 1000.
Also, *holding all else equal*, if you are at 1000 DPS with 0% hit or 1000 DPS with 7% hit, adding 1% hit will increase your DPS by approximately the same amount in both cases. This is why we argue that hit, etc. are not affected by DR except at the soft caps. It's all a matter of how you frame up the problem
Avoidance is a different monster altogether, and probably shouldn't be discussed here.
Edit: Treescendent, unless Blizzard decides for another round of drastic changes in 3.1, Agi will almost certainly be the preferred stat at every gear level.
Yes it does have the same type of DR as avoidance does
Originally Posted by Eilt
I suppose it is a semantics issue at this point:
Diminishing returns, to me, means that at some point you get less value from the same amount than you did at one time.
Hit once you reach the hit cap has diminishing returns in a very literal sense. I see the distinction that the value you get from the same amount of hit is constant, but the value you get from it is still diminished in usage. Since I did not coin either soft cap or diminishing returns I can not speak to the validity, but in the very literal aspect, the value is still diminished to the player at some point. This point of inflection is called the "soft cap" because after the value of the stat is "diminished". <-- My view
But like I started the reply with, it is pretty much semantics, the important part here is:
-50% crit to cap FB with a bleed
- ~76.6% crit to cap white attack crits
-4.8% boss crit supression
Regardless of your interpretation of what diminishing returns means to you, the fact is that crit does not have the same type of DR that avoidance does. It's not semantics. Hit rating does not have soft caps, only a hard cap. It has value before the cap and none after it. Expertise only has a soft cap if you are attacking from the front.
Perhaps you can argue that any reduction in value is "diminished" in terms of strict definition, but it really serves no purpose to refer to it that way in relation to WoW. Soft cap, hard cap, and diminishing returns all refer to different game mechanics. Getting these terms mixed up only serves to confuse and frustrate people who actually understand what they mean.
Crit has soft caps at 54.8% (if you FB) and ~81.4% (for white damage). Even with these soft caps, the amount of damage we get from these attacks pretty small in relation to our total damage so the reduced value is still high.
So I need some advice on my situation. I did this test on the heroic training dummy for 5 mins and got this result:
I used mangle, mangle for 2 combo points. SR, rake, shred to 5 cp and rip. From then on I shred on any cc procs, used mangle, rake and FFF whenever they were about to fall off, TF and berserk when their cooldown came off and shred any other time. After 2 cp I used SR then after 5 rip, 2 sr, 5 rip etc.
If anyone can suggest how I can pull more DPS, I have a feeling Im not doing as much as I should be.
Edit: Shit I covered the recount window, but it was basically at about 2700 for 5 mins.
You used Ferocious Bite only once.
You've got 48,55% Critchance. That should generate much more Combo Points, than you can spend in Rip and Savage Roar alone.
Try to fit in some more FB's.
Apart from this, your gear does not look like best in slot, although I can't identify all your items. 2700 DPS selfbuffed with your gear seems very satisfying imho.
You used Ferocious Bite only once.
You've got 48,55% Critchance. That should generate much more Combo Points, than you can spend in Rip and Savage Roar alone.
Try to fit in some more FB's.
I have to disagree, I can have trouble keeping rip up all the time on heroic dummy with 48% crit, and that's while not using ferocious bite. ;s
Crit on heroic dummy is almost always very low for some reason. Let's blame it on luck, or whatever, but the only time I'm able to fit in ferocious bite on dummy without sacrificing rip uptime is during berserk.
BiS really isn't that subjective, and there's quite a few upgrades for you to get, drop randomness being what it is however, my advice for the moment is primarily to increase your Rip uptime (89%) and especially Rake uptime (83%). It also looks like you did an extra mangle somewhere in there (18x18=324 seconds). Not that it would help you reach your Rawr maximum, but Icewalker and Precision enchants are superior DPS, and could help you smooth your cycle.
I used mangle, mangle for 2 combo points. SR, rake, shred to 5 cp and rip. From then on I shred on any cc procs, used mangle, rake and FFF whenever they were about to fall off, TF and berserk when their cooldown came off and shred any other time. After 2 cp I used SR then after 5 rip, 2 sr, 5 rip etc.
If anyone can suggest how I can pull more DPS, I have a feeling Im not doing as much as I should be.
Edit: Shit I covered the recount window, but it was basically at about 2700 for 5 mins.
I am assuming this is on live, and not on PTR.
I'd also like to add that you shouldn't clip rakes - you say "whenever they were about to fall off", implying you are "refreshing" it. Don't do that, wait for it to tick full duration and then apply a new one.
Also 2cps SR seems very conservative, I do at least 3cps testing self buffed, or even 5, though 5 is harder without raid buffs for me, resulting in lower uptime overall.
Your relative Shred damage seems too low (percentage of overall damage), granted I cant really say with 100% certainty how low is that for that gear level.
One more thing: make Snapper hit food, 205 hit rating is way to low even for testing purposes, while you have more expertise than needed (i am guessing those are Concealment shoulderpads and Trollwoven girdle?) - but this issue will be rectified by further upgrades you get.
One more thing: make Snapper hit food, 205 hit rating is way to low even for testing purposes, while you have more expertise than needed (i am guessing those are Concealment shoulderpads and Trollwoven girdle?) - but this issue will be rectified by further upgrades you get.
I disagree with this. In my understanding of feral mechanics (Edit: from on Rawr, Tossk's, and discussions here), there's really no magic number of hit a feral should reach. We don't scale with hit the way rogues or casters do. It's a strong stat, but strength (on live) or especially agility (on the PTR) is stronger in current gear. So, I'd gem for str/agi and eat str/agi food. Gearing for hit over the "best" DPS can be beneficial depending on your playstyle, but don't think there's some amount of hit you have to have to do good DPS. My preference is to use hit and expertise pieces over nearly-equivalent ones, but to still use the stats that give more theoretical DPS increase.
I disagree with this. In my understanding of feral mechanics (Edit: from on Rawr, Tossk's, and discussions here), there's really no magic number of hit a feral should reach. We don't scale with hit the way rogues or casters do. It's a strong stat, but strength (on live) or especially agility (on the PTR) is stronger in current gear. So, I'd gem for str/agi and eat str/agi food. Gearing for hit over the "best" DPS can be beneficial depending on your playstyle, but don't think there's some amount of hit you have to have to do good DPS. My preference is to use hit and expertise pieces over nearly-equivalent ones, but to still use the stats that give more theoretical DPS increase.
This is an old and, it seems, futile discussion. I personally go for the target of 263 hit rating, and i will gem/food around it if necessary - and I might slip a tiny bit below (say, 253) but never more than that.
The point you are making that there is no amount you must have is moot, since the results stemming from any randomly chosen number between 0 and the hard cap can not be statistically relevant, since they would either be derived from a purely personal/ subjective experience ("it didnt feel like I was missing much") or from pure theorycrafting, neither of which would produce irrefutable numbers.
Ie: 205 hit rating might in some individual situations produce very distinctive and mutually incomparable "overall damage/dps" results, very circumstantially dependent on many RNG factors during a relatively short time frames of any single boss fight currently in the game; add to that all the differences in raid buffs, gear, playstyle, etc and you just cannot get any stable "cut-off" number... Make numbers for lets say, 100x5 minutes dummy hitting tests (with full "rotation", not just auto attacks) with 205 and 263 hit rating each and maybe we can talk some more beyond personal beliefs. But because that would be a colossal waste of time, we turn to Rawr and other theorycrafting utilities.
Rawr puts 202 hit rating + 83 expertise (closest I could get with acceptable highly wanted gear) to a theoretical 1.839% chance to miss + 1.468% to dodge and 3334 dps with optimal rotation. With just putting +hit and +expertise gems instead of strength (easiest way to not change the gear pieces) i get 256 hit rating + 110 expertise (0.6453% to dodge and 0.1927% to miss) and 3331 dps optimal rotation - which does not seem like a cataclysmic dps loss to me.
You make out of it what you like; I prefer not to have that 1.839%.
Last edited by Wickedgirl : 03/10/09 at 7:18 PM.
Reason: clarification of points
I know kitties are getting buffed pretty good on the PTR, but up until last night I didn't actually know how significant of a DPS upgrade I'm actually getting.
For starters, I only have 2 BiS items, the rest of my stuff being good, but not great, so I'm not seeing the crazy high numbers that a lot of the posters here are. On most naxx bosses I'm sitting at around 3100-3400 dps. It's not great, but it's respectible. Anyway..
I wanted to finally see first hand how the changes were goign to effect me. My first sample was on the Live realm, buffed with only elixirs, food, and MotW.
In 5, 5 minute fights, I was getting 2400-2600 dps. Normally melee hits and shred were my top 2 damaging abilities, and rake and rip were the next two.
I copied my druid over to the PTR, without even moving him, when I got there, I spent all of my talent points, and the only change I made to the spec was taking 1 out of nurturing instinct and putting it in to primal gore.
I did another 5, 5 minute fights, and my dps was always up around 3800 with rip as my new most damaging attack, followed by melee, shred, and rake. (I don't think I let rake drop, I'm pretty OCD about keeping it up)
That all aside, I'm just confused as to how I'm really recieving a 1200 dps buff in the patch. That seems a bit extreme to me. This was also without the SR glyph, so the actual increase should be even more than what I saw.
So my question is: Are other ferals seeing the same kind of increase on the ptr? Is a 5 minute test really enough to draw conclusions? I thought about doing some 10 minute tests, but I figured that 5 minutes was closer to an actuall boss fight.
I am seeing a similar increase in DPS as you, going from 2600 DPS on Live to about 3600 DPS as of build 9658. I picked 5 minute fights for the same reason, this seemed closer to an actual (static) fight. No Savage Roar glyph used either. The only difference to your results is the abilities' percentage in total damage done: First Rip, then Rake, Melee and Shred. I'm using a SR3+, Rip5 priority list with Ferocious Bite thrown in whenever the situation allows it, and try to keep Rip and Rake uptime as high as possible (well, obviously). Could be just different gear, but I'm inclined to do more tests to see if these numbers even out over time or if there's some underlying problem in my cycle.
I am seeing a similar increase in DPS as you, going from 2600 DPS on Live to about 3600 DPS as of build 9658. I picked 5 minute fights for the same reason, this seemed closer to an actual (static) fight. No Savage Roar glyph used either. The only difference to your results is the abilities' percentage in total damage done: First Rip, then Rake, Melee and Shred. I'm using a SR3+, Rip5 priority list with Ferocious Bite thrown in whenever the situation allows it, and try to keep Rip and Rake uptime as high as possible (well, obviously). Could be just different gear, but I'm inclined to do more tests to see if these numbers even out over time or if there's some underlying problem in my cycle.
Rake being #4 on my damage done was really confusing to me too. I assumed it was my fault as I do tend to clip rake quite a bit out of habit. The "oh, 3 seconds left on rip, I should refresh that" doesn't so much hold true with Rake, but I did not think I clipped it enough to have it come in fourth.
Thanks for the reassurance though. It's good to hear that my numbers were not some random fluke RNG thing. I'll have to test this more when I get home to figure out my problem with rake.
If by refreshing you mean reapplying before it ends, then you should not refresh rip and rake. Unless i miss something.
My tests on ptr: rip 29%, melee 22%, shred 20%, rake 17%, fb 5%, mangle 4%
If you only have 2 or 3 seconds left on rip and have 5cp, I think it's better to refresh rip. Of course, it's best to let rip run out and then reapply it. My old main is a mage and ! am having trouble getting out of the "make sure debuffs dont fall off" mentality.
As for your percentages, that's about the same numbers I have, but I figured rake would make up for more of our damage.
I have been trying 5SR rotation, and I keep finding myself in the situation where i have 5 combo points, 20+ seconds on SR, and 10 second on rip with my energy ticking up from 60... what should I do in this situation? I have mainly just been using mangle or shred during this time to keep my energy below 100, but I am not sure if it is optimal. I am pulling 4k+ unbuffed ATM.
I have been trying 5SR rotation, and I keep finding myself in the situation where i have 5 combo points, 20+ seconds on SR, and 10 second on rip with my energy ticking up from 60... what should I do in this situation? I have mainly just been using mangle or shred during this time to keep my energy below 100, but I am not sure if it is optimal. I am pulling 4k+ unbuffed ATM.
In a case like this, particularly if I have TF available, I pop a FB if my energy is around 35-40. With 10 seconds left before Rip drops, that 10 seconds and a TF should be enough to get back another 5 CPs to continue with your cycle. However if I'm in that position and my energy is near max, I just pop a shred to lower it and then FB if there is still time enough waiting on Rip.
Item - Druid T8 Feral 2P Bonus (Class: Druid) -- The periodic damage dealt by your Rake, Rip, and Lacerate abilites has a chance to cause you to enter a Clearcasting state.
Item - Druid T8 Feral 4P Bonus (Class: Druid) -- Increases the duration of Savage Roar and Survival Instinct by 8 sec.
This is really going to make some interesting cycle variations. SR can last long enough to fully cover 2 Rips in some cases now (depending on glyphs). Gonna have a lot of research and testing to do on this. The extra CP from the 2t8 clearcasts might go help sustaining a cycle or even more FBs between Rips.
03/09/09, 3:42 PM
[/IMG]
[/IMG]
Similar Threads
