[WOTLK 4.0] Cats - Page 20

JaymzHendo · 10/12/10, 1:06 PM

Again sound logic - but with immediate increases to stam (today) and the fact that I'm smart enough to move out of dumb stuff and being faster (as a kitty to do so) I personally am not worried about it - if I do see myself taking that much more damage than normal I still have Barkskin to use and I also could move 1 point from Stampede into NI, and as I saw others suggest Primal Madness isn't so great considering haste regen so a point could come out of it as well. Going forward to 85 there is no reason not to pick up NI, but my initial reactions are for today and the next 2 months.

Shonassir · 10/12/10, 1:37 PM

I was looking more or less at this build:
Talent Calculator - World of Warcraft

Furor I don't see as a necessary talent, given the fact we've been using 3/5 (of the original) points for the longest time, and given the new energy regeneration rate, it shouldn't be that big of a deal. Using the 2 points from that to pick up NI as well as maintaining 2pts into Stampede.

I think the talents like Brutal Impact and Primal Madness are where the points could vary and be moved around. Brutal Impact is nice utility, but if you run with multiple rogues/fury warriors/frost DKs, the likelihood of you having to interrupt something is very low. You could rotate those 2 points back into Furor, or 1 pt Furor, 1 pt Survival Instincts for even greater survivability.

Aradiel · 10/12/10, 2:20 PM

I made this chart to help bear tanking. It shows how different stats affect mitigation, survivability and threat. I'm thinking of adding rage generation in there as well and I'll probably make it easier to read if I have the time. Anyways if you see any mistakes there let me know.

http://i91.photobucket.com/albums/k312/mdk2k/chart.jpg

Odas · 10/12/10, 2:24 PM

Originally Posted by psuman99

Agility (2.42), Mastery (1.95), Haste (1.72), and Hit/Exp (1.47)

Thats surprising that hit and expertise are rating so low. Are you able to gem agility and still be below the crit soft cap? Do you remember what your value for strength was? I'm wondering if it is still beneficial to use a Nightmare Tear for blue sockets instead of an Agi/Hit gem.

Edit: How about crit rating? I'm assuming we're going to be reforging crit->mastery but I could be wrong.

Koril · 10/12/10, 3:13 PM

I have seen some really interesting feedback on the new thorns (on french community forum) : since the new thorn has his damage incresaed by both spell power and attack power, it seems quite efficient for a feral druid. They says that with decent gear, thorns ticks around 4K on each hit (more than a bonkin's one, by the way), making it necessary in our rotation (well it depends on the boss' attack speed but I guess it will worth it to get in caster form, loosing 2-3 melee swing to get your tank buffed for 20 sec). It is also a funny way to help your tank with threat, and I guess it would be VERY efficient in aoe tanking. I still wonder if thorns can now stack with severql druids : since it looks like it is a signifiant part of our dps, that would be a problem if only one druid can use it in the raid. Remember that with the (major) glyph, thorns last 20 sec for a 25 sec cooldown, making it impossible for several druids to "share" the uptime on their tank.

For this reason, I would value furor much more than we used to : if we have to get in caster form in 25 sec, I guess than 2/3 furor will be necessary, if not 3/3. Even if we hit thorns while at 0 energy, we have about 3 sec (2GCD+latence) to get back in cat form and make your energy cap back to 100, and with good haste, you would get about 50 senergy. I guess I go for 2/3 furor at level 80.

That would bring me to this template :
Wowhead - Talent calculator
Since NI, SI and Brutal impact (only 1/2

) doesn't seems optional for me. Well it is a personal point of view but I think it is more usefull, in the next few weeks, to get familiar with our utilitary talents than to have our dps passively increased by a very few amount. And as mentionned sooner, the first point in stampede is important, but the second is way less important. I used the glyph of FB here, but I will first give a try to the new FB without glyph to see if it makes me miss some energy. If not, then I guess we shouldn't glyph FB since it is better to use extra energy in FB than in shreds.

psuman99 · 10/12/10, 4:35 PM

Originally Posted by Odas

Thats surprising that hit and expertise are rating so low. Are you able to gem agility and still be below the crit soft cap? Do you remember what your value for strength was? I'm wondering if it is still beneficial to use a Nightmare Tear for blue sockets instead of an Agi/Hit gem.

Edit: How about crit rating? I'm assuming we're going to be reforging crit->mastery but I could be wrong.

I don't remember strength exactly but I think it was roughly the value of haste (~1.8). so it appears that a nightmare tear would be at least equal to agi/hit. Plus the extra stam and int do have some value. As far as crit, I reforged all my gear to reduce crit in exchange for mastery so I was no where near the cap. The value of crit was below the others I listed above, some where close to 1.4 I think.

kalbear · 10/12/10, 6:22 PM

Assuming you're using tiger's fury on cooldown, the glyph would result in ~2.2% more uptime or 0.33% damage from the buff itself. You would also gain (80/27)-(80/30) =~ 0.296 energy per second. Thats equivalent to
(0.296 energy)*(1% haste/(0.1 energy/sec))*(32.79 haste rating/haste%) =~ 97 haste rating. I haven't heard of anyone generating any specific numbers for the relative values of stats but if its anything close to live values in BiS 97 haste rating is between 1.5% and 2% damage.

Yeah, you're likely right. It helps more with getting more rips and rakes in the sweet spot, but the actual overall damage boost isn't comparable to savage roar, at least not currently.

Odas · 10/12/10, 8:05 PM

Originally Posted by kalbear

Yeah, you're likely right. It helps more with getting more rips and rakes in the sweet spot, but the actual overall damage boost isn't comparable to savage roar, at least not currently.

What I meant to show was that the main damage gain from the glyph is from the increased energy, not the extra uptime on the 15% damage. However, when I came up with that 1.5 to 2% damage number it was based on the total value of haste. The glyph only gives you extra energy not extra white swings, which is the majority of the value of haste (I think). So that would lower the total value of the tiger's fury glyph down to probably lower than the savage roar glyph.

This is all assuming a 100% uptime on a boss. During transitions you can have tiger's fury cooling down which would greatly diminish the value of the tiger's fury glyph while it would only slightly lower the value of the savage roar glyph.

a civilian · 10/12/10, 8:37 PM

Originally Posted by psuman99

Your logic is flawed here.

The amount of combo points consumed doesn't really matter since our rotation is based on only using FB when you don't need the combo points to apply Rip or refresh SR. There is an opportunity cost in NOT using those combo points for a FB in that situation. Since FB does more damage per energy than shred or mangle, it is worth using for the initial 35 energy. However, wasting the additional 35 energy to not build combo points can lead to Rip falling off the target or SR dropping off, each of which is a large dps loss compared to the extra damage that additional 35 energy did vs. using it for a shred and 1-2 combo points. Now obviously, if you KNOW you can use that extra 35 energy and still maintain your bleeds and SR, it will be a dps increase but to say that across the board it is always worth it to use that extra 35 energy is an inaccurate statement.

Your argument has merit, but let us look a little more closely.

Suppose you are short on combo points such that using FB at 70 energy means delaying your next rip by 35 energy, perhaps 2.7 seconds with haste. In this case you lose 2.7/22(rip damage) and gain roughly (FB damage)+(-35+2.7/22*30)(shred damage)/40. (Gain a ferocious bite, lose 35 energy worth of shreds, and gain 2.7/22 of rip's energy worth of shreds.)

I'm not sure of the precise comparison between rip, FB, and shred damage in 4.0.1. But suppose that rip does 3.5 and shred does 0.7 times as much damage as ferocious bite. The former is based on the figures a page back, including mastery (base plus about 4 points) and glyph; the latter is based on 3.3.5 values. (If anyone has better values let me know - I'm not able to download 4.0.1 at present.) In this case the net gain is about 0.02 (FB damage) - essentially break even. So even now we're not losing damage.

Note, however, that this assumes it is certain that spending an additional 35 energy will delay rip by the equivalent amount of time. In reality, the time to generate 5 CP follows a fairly broad probability distribution. In most cases, it is far from certain that rip will be delayed by the full 2.7 seconds; and if you did set up a case where this was near certain, you would also expect to lose a lot of rip uptime just due to the base FB cost.

Let me make this more precise. As a rough estimate, the time to generate 5 CP looks something like

$t \approx \frac{5}{(1+h)(10/38+\overline{l})(1+\overline{c})}$

where h is the haste, l is the clearcast rate, and c is the crit rate. I assumed for simplicity that each CP generator costs 38 energy. I've also ignored tiger's fury and berserk - let's say they are on cooldown. Let me assume 40% crit, as kalbear suggested, and 30% haste (as above). If it hasn't changed in 4.0.1, we expect a clearcast rate of 3.5/60. Note the standard deviations in the mean clearcast and crit rates are

$\Delta\overline{c} = \sqrt{\frac{c(1-c)}{n}}$
$\Delta\overline{l} = \sqrt{\frac{l(1-l)}{m}}$

where n is the number of CP generators used and m is the number of autoattacks. Strictly speaking m and n depend on the values of c and l, but since this is just a rough approximation I will assume they are constant. It takes approximately n = 5/1.4 = 3.6 CP generators to generate 5 CP and approximately m = 5/(1.4*(10/38+3.5/60)) = 11.1 autoattacks are made (I will ignore fury swipes for simplicity). Hence $\Delta\overline{c} = 0.26$ and $\Delta\overline{l} = 0.07$ .

Now t = 8.54 seconds from above and its standard deviation is

$\Delta t \approx t \sqrt{\left(\frac{\Delta\overline{l}}{10/38+\overline{l}}\right)^{2}+\left(\frac{\Delta\overline{c}}{1+\overline{c}}\right)^{2 }} \approx 2.44 s$

[Remark: If someone else, perhaps someone trained in statistics, could come up with a better estimate for the probability distribution of the time to generate 5 CP, this is certainly something that would be of great use in general to non-simulation feral theorycraft.]

What does this mean? Suppose we have 35 energy and rip has 8.54 seconds remaining, so we expect to generate 5 CP in rip's remaining duration.

First, let us FB now. The expected amount of lost rip uptime is approximately
$\int_{0}^{\infty}{dx \frac{x}{\sqrt{2\pi (2.44s)^2}} e^{-\frac{x^{2}}{2(2.44s)^2}}} \approx 0.97s$

Next, let us try using FB at 70 energy. In this case the expected loss of rip uptime is approximately
$\int_{0}^{\infty}{dx \frac{x}{\sqrt{2\pi (2.44s)^2}} e^{-\frac{(x-2.7s)^{2}}{2(2.44s)^2}}} \approx 2.87s$

Here, using FB at 70 energy is only costing us an additional 1.9 seconds of rip uptime; this is a moderate damage gain (about 13% of ferocious bite's damage) according to the initial analysis.

Sadirin · 10/12/10, 8:57 PM

Thorns damage reduced by 66%, now deals [ 16.8% of Spell Power + 178.58 ] or [ 16.8% of AP + 178.58 ] depending on the highest.

No more 4k Thorns for us. More around 1-2k now.

dudutank50 · 10/13/10, 1:29 AM

goodmorning all, any one have some ideas on how reforge item in 4.0.x for bear?

Schuakia · 10/13/10, 1:56 AM

I did not see it posted here since this seems to be mostly targeted at feral cats, but what is the general consensus for conversion to mastery as bear tanks? I am assuming haste. I missed the previous post since I did not refresh my page prior to posting but looks to be the same cunundrum as I have.

Mihir · 10/13/10, 3:06 AM

Originally Posted by Odas

Thats surprising that hit and expertise are rating so low. Are you able to gem agility and still be below the crit soft cap? Do you remember what your value for strength was? I'm wondering if it is still beneficial to use a Nightmare Tear for blue sockets instead of an Agi/Hit gem.

Edit: How about crit rating? I'm assuming we're going to be reforging crit->mastery but I could be wrong.

Crit is slightly lower than hit/exp, while str is around the same value as mastery. I've regemmed according to mew (so full 20agi/34agi, 10agi+10haste in +4 or better bonus per yellow sockets and 1 nightmare tear) and with reforging I still end up below the crit cap.

DPS: 12123.61939
Chance Miss/Dodge (%): 8.87486
Crit % for White: 64.99985
Crit Cap % for White: 67.12514

One potential problem I see is that agi and mastery scale too well with each other; with 2427 agi and 537 mastery, Mew gives the relative stat values as Agi: 3.16, Mastery: 2.20, Str, 2.01, Hit/Exp: 1.72, Crit: 1.68, Haste: 1.55. Also the lack of mastery gems forces the gemming of (almost too much) haste.

Koril · 10/13/10, 4:11 AM

About bear reforging (I haven't seen anything about that, so I could be completely out the way), I wonder if these options are possible/efficient :
- Reforge offensive stats on leather into mastery : if lacerate is buffed by mastery, you could keep enough threat, if you don't, you could get some hunters! This is probably better for your survivability, but haste/crit also makes savage defense proc more often, I guess some math are needed here.
- Reforge offensive stat on leather into dodge : is it simply possible? If it is, you definitely loose some threat, but still, hunters/rogue are our friends. For the same reason, I wonder if it is really a survivability increase, since diminishing return on dodge is huge, and for the next few weeks, ICC debuff is still here.

psuman99 · 10/13/10, 4:23 AM

When I finally got into game tonight to work on gemming and reforging, I got slightly different results than Mihir.

At first, I reforged all my crit into mastery and gemmed agi/haste in yellow sockets with +4 bonuses and agi/hit for blue sockets with +4 bonuses. I then used Mew and it showed that Crit was more valuable than haste. I then reforged all my haste into mastery (instead of crit) while keeping my gems the same. Mew then showed that haste was better than crit. Clearly, I was running into crit cap problems. I ended up reforging haste into mastery on all of my gear except for three pieces, which I reforged the crit into mastery. I kept the same gem set up as before, however the model indicates it is not worth going for socket bonuses unless they are +6 or better (I ran out of gems so my armory still shows me gemming for +4 sockets).

Doing this and rerunning Mew showed that Agility and Mastery are significantly better than any other stat. Haste, Crit, Hit/Exp are all almost equal in value for me now (ignoring Str since we won't be gemming or gearing for Str) with hit/exp having a very slight edge.