Leafkiller

Jun 262011
 

The last few days before a patch are always very painful for anyone who maintains an addon, or in my case a script for an addon. When Cata first came out, I picked up support for the feral Ovale script that we maintain on the Fluid Druid Forums. In keeping with the tradition established by Nightcrowler (the author of FeralbyNight, an addon we used back in BC) I have always based the rotation in my Ovale script on an optimized Mew simulation script. While that sounds simple enough, it means I spend time optimizing the sim script in Mew, then updating my Ovale script to match, and finally debugging the Ovale script on the PTR. And, I pretty much have to wait until just before the patch is released due to the ongoing changes to the feral class that we have seen in 4.2 and the accompanying changes that Yawning has been making to Mew.

The good news is, barring any unforeseen additional changes to 4.2 and/or Mew, I just completed my work on the Mew sim script and a first version of a 4.2 compatible Ovale script, which means the changes to the rotation for 4.2 are fresh in my mind.

Starting from the top… in case anyone has not heard, the optimum third glyph to use is now Glyph of Berserk. So be ready to grab some Dust of Disappearance, and swap out Glyph of TF (if you are still using it). This means that the base duration of Berserk will be 25 seconds, and the duration of TF will be 30 seconds. With the 4 piece tier 12 set bonus, we will sometimes see Berserks that are over 40 seconds in length (during Heroism) and will typically see 4T12 Berserks last just over 30 seconds. The result is that the optimum time to cast Berserk is during TF (preferably at the same time). As a matter of fact, the only time you would ever want to cast Berserk without TF is if you are in a situation where the Berserk is coming off of cooldown and is going to be clipped if you wait for TF (like the end of a fight) – and even then, if TF is coming off of cooldown soon (6 seconds seems to be a good rule of thumb) you should wait for it. So make sure you have TF macroed into your Berserk, and hit both together.

Next up, casting Ferocious Bite during Berserk. Clearly, this is important to do if you have the 4 piece T12 set bonus. As it happens, it is also a small dps up to cast FB during Berserk even if you do not have the 4 piece set bonus. No more complaining about endless Shred spamming during Berserk, FB is back. There are, of course, guard times. Make sure there is at least 5 seconds left on Rip and 3 seconds left on Savage Roar when you cast your 5 combo point FBs.

In case people did not get the message in 4.1, trying to refresh Savage Roar early in order to avoid having it come off of cooldown around the same time as Rip is no longer a dps gain. However, Savage Roar has been buffed in 4.2, and the rules around when to refresh it have changed. In the 4.1 rotation, there was a 6 second Rip guard time on Savage Roar. Basically an optimized rotation would wait to refresh Savage Roar if Rip was going to expire within 6 seconds, even if this meant an extended period of downtime on Savage Roar (as much as 8 seconds). The scales have tipped, if Savage Roar is coming off of cooldown, you don’t check for when Rip is going to expire anymore, you just refresh it. Keep in mind, Rip is still higher in the priority list than Savage Roar, and we do try to refresh it about 2 seconds before it expires, so this is really about the cases where Rip has more than 2 seconds left and/or we have less than 5 combo points. Nonetheless, it is trading off uptime on Rip for uptime on Savage Roar.

Did anyone else feel a cool “Wrath-like” breeze blow by just now? Not only is it a dps up to cast Ferocious Bite during Berserk, but it is also a dps increase to cast it other times. The guard times are very restrictive (can you say “level 80 Wrath”?), 14 seconds for Rip and 10 seconds for Savage Roar, but it does happen. I spent time on the PTR training dummy earlier tonight and it popped up in my Ovale script occasionally. Just be fast, because the window closes quickly.

The final change worth mentioning is in the area of energy pooling. Keep doing it, only more than before. When you are not in the middle of TF or Berserk, refreshing a DoT or debuff or hustling to get a few extra combo points because Rip and/or Savage Roar are about to expire, get your energy up to at least 85. This is a good 10-15 energy higher than what was needed in 4.1.

Now, having gone through all of the changes, it is important to remind all of you to keep these changes in context. Specifically we have a complex priority-based rotation, and all of these changes are still subject to a set of well defined priorities. To understand those, you should spend some time reading either my new Ovale script and/or the latest, greatest Mew sim script (which should be released in the next day or two and is available via SVN). If you really want to get a good feel for the rotation, download Ovale, grab my 4.2 script from the Fluid Druid forums, and play with it on the PTR. The 4.2 Ovale script is incredibly close to the sim script, closer than any other script I have done to date. Just get on a training dummy, setup both Ovale and a cooldown monitor of some sort, and follow the Ovale suggestions. You will quickly get a feel for how the rotation has changed, especially if you keep in mind what I have covered here.

In closing, I like the rotation changes. It has brought some life back into feral play. The endless Shred spamming during Berserk was mind-numbing. I think most of you will appreciate the changes. It does feel a lot more like the Wrath rotation did. Also, the best thing about testing with a premade T12 feral was not having to keep SoTP up and running. That is probably my least favorite set bonus ever. I can’t wait to get 2 piece T12 so I can lose the 4 piece T11 set bonus (hopefully forever).

Best of luck to everyone in Firelands. Soon we will be having Fire Kitty processions and dances! I am looking forward to seeing some cool Fire Kitty videos on Youtube ;)

P.S. there will no longer be zero range Feral Charges. We have estimated that the 4.2 changes are on the order of a 9% dps increase over 4.1 , but on fights where we used to be able to Feral Charge on cooldown while standing next to the boss, expect the improvement to be half of that.

Apr 222011
 

I feel like RSVs (Relative Stat values) are the theoretician’s equivalent to Archeology. No matter how much time you spend doing it, the return you get for your time spent is far less than doing just about anything else.

In his recent post about Mew and RSVs, Tangedyn wrote “Increasing the number of iterations by 100 times to 1million will reduce the Error by 10 times, to ~0.015 requiring 0.0212 difference in RSV, which is far more acceptable.” That is a lot of iterations. In this post, Tangedyn elaborated on the time involved in calculating RSVs: “Takes about 2 hours to test a profile with 1M iterations :/.”

In case you have never done it, download Mew, grab a profile, make sure that the model selected is “Simulator”, and then do two runs, one with RSVs enabled and one without RSVs. You will quickly get an appreciation for how time consuming it is to calculate RSVs. While most people would agree that anything that is extremely time consuming should have a decent return on investment, I hesitate to say that about WoW players who will spend hours upon hours farming for a rare pet. However, RSV calculations do not reward us with a pet. Do they reward us with increased understanding that will help us gear better and conquer the content faster?

Let’s do a few tests and see. First of all, I am not going to do tests with 1 million iterations. I will use 40k iterations (which is still pretty slow). My goal is not to calculate a set of RSVs that can used for my gearing. Rather it is to try several tests to see how how RSVs change as I move my secondary stats around and to draw some conclusions as to whether or not the RSVs can help increase my dps. In my previous post about the homogenization of the secondary stats in 4.06, I already took the position that the difference in the secondary stats is not that much since the 4.06 changes. That would argue against spending time worrying about RSV values. Lets see if that holds true or not.

The setup I am using is starting with a balanced stats profile – Weapon DPS: 703.5, Strength: 142, Agility: 5118, Attack Power: 887, Haste Rating: 936, Hit rating: 604, Crit Rating: 954, Expertise Rating: 188, and Mastery rating: 1987 (this is an ilvl 362 setup). The testing is based on the 4.1 Berserk along with the Berserk changes I covered in my last post. I am not going to make an effort to reforge to different setups, rather I will simply subtract and add to my secondary stats in multiples of 100. For example, I might lower haste rating by 500 and increase crit rating by 500 if I want to test a profile with more crit. The first and most important thing you need to understand about RSVs (other than it can take 2 hours to get numbers that have a high confidence level) is that RSVs are calculated in a small range – about 100 stat points (I move around 91 stat points when I reforge my ilvl 369 chest). If you reforge more than a single piece of gear, it is likely you are changing your stats around more than the RSV code has tested. Enough discussion  – let’s see how different things affect my RSV.

Does the type of fight affect my RSVs:

Fight Description DPS Str Agi AP Haste Crit Mastery Hit Exp Error
5:00 Patchwork 24090.3 2.31 3.04 1.10 .81 1.02 1.19 .99 .98 .08
5:00 Mid-Combat Ravage 25133.7 2.40 3.16 1.14 .86 1.05 1.19 1.03 1.02 .08
5:00 Atramedes 19900.9 1.89 2.50 .90 .61 .84 .97 .89 .89 .07

Across all thee fights, taking into account the error (.08 x sqrt(2) = .11 and .07 x sqrt(2) = .10) we can safely say that haste is the worst stat for my profile on all three fights. It looks like mastery is the best on all three fights, but on the Atramedes fight we cannot say so with a 95% confidence without doing more iterations. It also looks like hit and expertise improve relative to the other stats on Atramedes, but again, we need to do more iterations to have a 95% confidence interval. So fight type does seem to have some effect on the ordering of the RSVs – but not a lot (obviously a fight like Atramedes has a statistically significant impact on all of the RSVs due to lost combat time).

How about the length of the fight?

Fight Description DPS Str Agi AP Haste Crit Mastery Hit Exp Error
5:00 Patchwork 24090.3 2.31 3.04 1.10 .81 1.02 1.19 .99 .98 .08
3:00 Patchwork 24673.0 2.34 3.10 1.11 1.21 1.05 1.19 1.10 1.09 .11
7:00 Patchwork 23977.9 2.31 3.05 1.10 1.03 1.03 1.18 .98 .98 .07

Look at the variation in haste. Even with only 40k iterations per sim run, that is significant. Haste is clearly sensitive to fight length – and in a way that is not obvious without doing more tests. It looks like hit and expertise are more valuable on a short fight but we cannot say that with a 95% confidence – so more iterations are needed to validate that.

What happens if I switch to Glyph of Berserk in place of Glyph of Tiger’s Fury? The previous tests were run with GoTF. Lets try the three fights with GoB and see what happens:

Fight Description DPS Str Agi AP Haste Crit Mastery Hit Exp Error
5:00 Patchwork 24000.7 2.29 3.04 1.09 .90 1.03 1.16 1.04 1.03 .08
5:00 Mid-Combat Ravage 24878.5 2.37 3.12 1.13 .85 1.04 1.16 1.04 1.02 .08
5:00 Atramedes 19948.2 1.89 2.51 .90 .59 .85 .97 .81 .81 .07

Comparing the GoB numbers to the GoTF above, it looks like the difference is not that great – at least for these three fight scenarios.

What if I move around stat points, taking from the lowest and putting them in the highest? I am going to focus on the 5:00 Patchwork fight where Crit is clearly better than Haste – so I am going to shift points from Haste to Crit and see what happens. Technically, Mastery is the better than Crit, but in my actual reforging I am already maxed on on Mastery.

Fight Description DPS Str Agi AP Haste Crit Mastery Hit Exp Error
5:00 Patchwork 24090.3 2.31 3.04 1.10 .81 1.02 1.19 .99 .98 .08
5:00 Patchwork, move 100 24097.7 2.31 3.06 1.10 .95 1.03 1.19 .96 .95 .08
5:00 Patchwork, move 200 24096.1 2.31 3.06 1.10 1.04 1.02 1.20 .97 .96 .08
5:00 Patchwork, move 500 24083.8 2.31 3.04 1.10 .97 1.01 1.21 .92 .91 .08

Once again Haste seems to be more volatile than the other stats, with some movement from Hit and Expertise, although more iterations will be needed to say that at a 95% confidence. The main thing to look at is the DPS line. I shifted 500 stat points from Haste to Crit (ending up with 436 Haste Rating and 1454 Crit Rating) and along the way, the range of the DPS varies only 14 dps from the high to the low. So even if, after a two hour run we can safely say that Crit rating has a higher RSV than Haste, does the potential 7 dps gain from moving 100 stat points from haste to crit justify the time spent?

TLDR: To get valid RSV values takes hours of simulation work. RSV values are only meaningful in a range of 100 stat points. Haste seems to be the most volatile stat (quoting Tangedyn from this post: “I’ve seen Haste’s RSV fluctuate enough that there is probably more than meet the eye, but at the moment there is no real proof pointing to anything as a cause of the the fluctuation or a Haste breakpoint.“). Even large swings (500 stat points) in how you allocate your secondary stats will have very little impact on your overall dps.  Quoting myself from my Homogenization post “The secondary stats are so close in value now that you you can pretty much choose any combination of crit/haste/exp/hit without significantly impacting your dps.” Bottom line – RSV calculations may be valuable to theoreticians, but do not use them as a guide for how to reforge your gear. As always, your time will be much better spent focusing on the fight mechanics and your rotation.

Apr 172011
 

In my last post I discussed rotation modifications for dealing with Berserk in 4.1. In so doing, I also touched on the relative performance of the Glyph of Tiger’s Fury (GoTF) and the Glyph of Berserk(GoB) on different fights. As often happens, a fair amount of the commentary by people on that post has been on the relative value of the glyphs rather than on the changes to the rotation with Berserk no longer being on the GCD. Rather than continue to discuss the glyphs on that post, I decided to toss up this post as a place to begin a more in-depth discussion of GoTF and GoB. The tricky part about comparing GoB with GoTF is that either can be better depending on the fight mechanics and fight length. GoB is particularly sensitive to fight length due to the three minute cooldown and GoTF is particularly impacted on a fight like Atramedes where the repeating air phases are 31.5 seconds with no chance to cast TF while the GoB cooldown continues to tick down.

My feeling is that the choice of whether to use GoB or GoTF is dependent on both fight mechanics and raid needs. GoTF will provide more overall dps on fights where there is no interruption on attacks (Chimaeron is a good example) but GoB will provide more burst dps (which is also applicable to Chimaeron where burst is important in phase 3). It really depends on the role one is playing in the fight.

My goal of this post is not to say which glyph is better, but rather to help people gain some insight into how fight mechanics and fight length affect each of the glyphs. Atremedes is a really interesting fight to show that on due to the periodic air phases. Per DBM the timing in Atramedes is:

  1. Initial 90 second ground phase
  2. 31.5 second air phase
  3. 85 second ground phase
  4. 31.5 second air phase
  5. repeat the 85/31.5 second pattern until either Atremedes is dead or the enrage timer is hit

A couple of things to keep in mind. The dps script has not been optimized for this fight. It will try to cast Berserk as early and often as possible with the constraint that it will try not to delay the next TF. To optimize the script for this fight, there are likely times where it would be better to cast Berserk at a point where it will delay TF in order to get as much Berserk time as possible before the next air phase. Additionally, it is difficult to know what to set the Blood in the Water timing to so I am leaving it at 25% – which really impacts dps when Atramedes dies during or right after an air phase. I am going to test a series of fights that use the same profile, for both the GoB and the GoTF with the fight length increasing in 30 second intervals from 5:00 to 8:30. The setup I am using is a balanced stats profile – Weapon DPS: 703.5, Strength: 142, Agility: 5118, AttackPower: 887, Haste Rating: 936, Hit rating: 604, Crit Rating: 954, Expertise Rating: 188, and Mastery rating: 1987 (this is an ilvl 362 setup). The testing is based on the 4.1 Berserk along with the Berserk changes I covered in my last post.

So what does the data below tell us? First off, you should use GoB on Atramedes. At 6:30 and 7:00 you can see the impact of the Berserk logic waiting for TF to come  off of cooldown resulting in Berserk being clipped when using the GoTF. Putting in some additional fight specific logic to minimize clipping of Berserk right before an air phase,  I just got 19347 dps, 12.3 TFs, and 2.9 Berserks for the 6:30 GoTF fight and 19606 dps, 13.5 TFs and 2.9 Berserks for the 7:00 minute GoTF fight – so pushing off a TF in order to avoid clipping a Berserk can be a small dps up – but it is not that significant. Notice that the dps difference between GoTF and GoB goes from a low of 23dps on the 5:00 fights to a high of 292dps on the 7:00 fights which is a pretty big swing. You can see how much fight length and mechanics affect dps – which has a high of 20273 on the 5:00 GoB fight to a low of 18126 on the 8:00 GoTF fight.

Sometime soon, I will do a followup post comparing the two glyphs on a patchwork style fight at different fight lengths. That will favor GoTF and  will further illustrate how fight length and mechanics affect TF, Berserk, overall dps and the choice of which glyph to use, GoTF or GoB.

 

Glyph of Tiger’s Fury Glyph of Berserk
Fight Length DPS # of TFs # of Berserks DPS # of TFs # of Berserks
5:00 20250 10.2 2 20273 8 2
5:30 20068 10.6 2 20217 9 2
6:00 18261 11.5 2 18418 9.9 2
6:30 19348 12.5 2.7 19573 10.2 3
7:00 19561 13.6 2.8 19853 11.7 3
7:30 19319 13.9 3 19518 12 3
8:00 18126 14.8 3 18362 13 3
8:30 18841 15.9 3 19010 14 3
Apr 172011
 

April 12th, Yawning posts a new version of Mew with the known 4.1 changes included and a few hours later Blizzard posts new 4.1 feral changes that make this new version of Mew out of date. Personally I think it is a conspiracy.

Fast forward to April 15th. I chatted with Tangedyn last night and he told me that the change to Berserk was already checked into the Mew source tree. He also walked me through setting up my Eclipse environment so that I can compile Mew locally, which gives me access to the latest changes without having to bug Yawning to post a special version for me (he is retired from the game after all, although apparently he is bored and continues to make enhancements to Mew). So there I was, with a new Mew in hand (my version of a new toy) and the burning question, “what does the change to Berserk mean to the feral rotation?” More specifically, what does it look like when I model it in my Mew script and what changes will be needed in my feral Ovale script?

Taking Berserk off of the global cooldown makes it castable at the same time as Tiger’s Fury and the next GCD ability. The default 4.0.6 sim script constrains Berserk by ensuring there is an upper limit to how much energy you can have when casting it. This is to prevent energy loss during the GCD that Berserk currently triggers. In my 4.0.6 sim script, I also set a low bound – a minimum energy level for when to cast Berserk. This makes a very small difference with the Glyph of Tiger’s Fury (GoTF), and a larger difference with Glyph of Berserk (GoB). With the 4.1 changes, I removed the upper energy bound from Berserk and also synchronized it to not be cast during a GCD. The same rule applies to TF – you should wait for the previous GCD to be finished before casting TF to maximize the number of GCD abilities that can be cast during the TF. For the purposes of this post, I tested my rotation with three versions of Berserk code, the default 4.06 code, my 4.0.6 code and my new 4.1 code (which I revised while writing this post). I posted an updated version of my sim script which includes all three code snippets with two of them commented out – so anyone who wants to can easily compare all three.

For the testing, I decided to do a comparison test of all three methods using both GoB and GoTF with a single profile that has balanced stats – Weapon DPS: 703.5, Strength: 142, Agility: 5118, AttackPower: 887, Haste Rating: 936, Hit rating: 604, Crit Rating: 954, Expertise Rating: 188, and Mastery rating: 1987 (this is an ilvl 362 setup). I used the following five combat scenarios:

  1. A 5 minute patchwork style fight (“Patch”)
  2. A 5 minute patchwork style fight with in-combat Ravages with no travel time (“5 min. Ravage 0″)
  3. A 7 minute fight that emulates Atremedes (7min. Atramedes)
  4. A 5 minute fight that emulates Atremedes (5min. Atramedes)
  5. A 7 minute patchwork style fight with in-combat Ravages with no travel time (” 7min. Ravage 0″)

During the course of the testing I noticed that my 4.0.6 Berserk code worked slightly better than the default 4.06 code on the patchwork style fights, but significantly worse on the Atramedes fights. This led me to revise the new 4.1 Berserk code taking elements from both of the 4.0.6 Berserk codes. Now, if TF is up, the 4.1 Berserk code does not have a minimum energy level, but there is a minimum energy level if TF is not up. This code performs well for both GoB and GoTF. Since the Berserk code has no energy constraints during TF, it opens up the possibility of using a macro to cast both together when both are off of cooldown, energy is low enough to cast TF and it is at a time you want to cast both TF and Berserk. The new 4.1 Berserk code performs better than both of the 4.0.6 Berserk codes in all test scenarios. A simplified explanation of when to optimally cast Berserk in 4.1 is:

  1. If the Tiger’s Fury buff is up and you can cast Berserk, do so.
  2. If the Tiger’s Fury buff is not up and you have at least 65 energy and the next cast of Tiger’s Fury will not be delayed by Berserk, cast Berserk.
  3. If the Tiger’s Fury buff is not up but the fight (mob) will end before Berserk does if you wait for TF to be up, cast Berserk.

Obviously, these rules should be applied at a time when you want to use Berserk – which is fight dependent. For example you might find yourself casting Berserk on a key add phase knowing that it will delay TF.

Comparing GoTF to GoB, GoTF is consistently better on the patchwork tests, especially those that allow for melee range Feral Charge (think Chimaeron) and GoB is consistently better on the Atramedes tests. I did not do enough tests to make a thorough comparison of the two glyphs, but some patterns can be seen in the data. Unfortunately there are still a lot of combat scenarios that are not possible to model in Mew, especially add fights. I did have a chance to chat with Yawning today, and asked him to add a couple of new toggles to Mew that will make it possible to test some additional scenarios in the future.

GoTF GoB
4.06 Patch 24437 24336
4.06 5min. Ravage 0 25498 25205
4.06 7min. Ravage 0 25221 24942
4.06 7min. Atramedes 19532 19735
4.06 5min. Atramedes 20224 20201
4.06 Leaf Patch 24451 24362
4.06 Leaf 5min. Ravage 0 25514 25234
4.06 Leaf 7min. Ravage 0 25184 24913
4.06 Leaf 7min. Atramedes 19372 19594
4.06 Leaf 5min. Atramedes 20227 20198
4.1 Patch 24480 24398
4.1 5min. Ravage 0 25531 25279
4.1 7min. Ravage 0 25277 25014
4.1 7min. Atramedes 19561 19853
4.1 5min. Atramedes 20250 20273
Mar 292011
 

The cat threads across the various forums are littered with people asking questions about what gems and enchants should be used, and how to reforge. The advice is varied and in many cases out of date as people have not fully embraced the changes that 4.06 made. The following three rules are all you need. If you spend more time thinking about gemming, enchanting and reforging than the amount of time it takes to read this post, you are spending your time on the wrong subject. You would be better suiting thinking about your rotation and the various boss encounters your guild is working on if you want to make significant improvements to your dps.

1. Always gem for agility. Agility is generally worth about three times the stat value of any of the secondary stats. Unless a socket bonus is worth 14+ agility for a single gem socket, ignore it. In the current content there are only a few items in the game where it is worthwhile to get the socket bonus (helms, tier chest, Sinestra belt). Everything else should be gemmed with agility (delicate). As it happens, in all three cases where it is worth getting the socket bonus, the socket color is blue – so you will be using agility/hit gems for the blue sockets in those items (glinting).

2. Use the known BiS enchants. The BiS enchants are well understood. Agility/haste for the helm; agility/mastery for the shoulders; agility for the back*, bracers, weapon and boots; stats for the chest; Attack power/crit for the legs; strength for the gloves. Don’t use any other enchants. * 65 crit is a viable alternative for the back.

3. Get mastery on every item. Mastery continues to be the best of the secondary stats and is generally worth about 20% more than the other secondary stats. If an item does not have mastery on it, reforge the highest available secondary stat to mastery. Per my post 4.06 – the Homogenization of Secondary Stats the difference in relative values of all of the secondary stats other than mastery are pretty close and varying them will have minimal impact on your dps. Once you have mastery on every item, then you can reforge selectively as needed (hit until 4.1 for interrupts) or desired (for example some people like to reforge to expertise on Al’akir). In general having haste and crit balanced provides good results, although some people prefer to reforge to hit/expertise in order to have a more stable rotation and consistent results on fights with target switching.

Mar 232011
 

In response to some discussions we have been having on the Fluid Druid forums, I decided to revisit the relative values of the trinkets, Unheeded Warning, Fluid Death, and the 372 Essence of Cyclone. As I have all three of these, the stat values I am modeling with are actual values taken from my toon. In preparation for this, I reforged to a balanced haste/crit setup. I did the testing against the same four scenarios I used in my previous post about the homogenization of secondary stats. As a reminder, here are the four combat scenarios:

  1. A 5 minute patchwork style fight (“Patch”)
  2. A 5 minute patchwork style fight with in combat Ravages with no travel time (“Ravage 0″)
  3. A 5 minute patchwork style fight with in combat Ravages with 2 second travel times (“Ravage 2″)
  4. A 7 minute fight that alternates 90 seconds of combat with 30 seconds of downtime which is similar to the air phases on Atramedes (“90/30″)

All tests used the Glyphs of Rip and Shred, and I did full runs with the third glyph being either TF or Berserk.

My test results were not surprising, but it is useful to see consistency across a set of combat scenarios rather than just looking at the trinkets in a patchwork style fight. The combination of Unheeded Warning and Essence of Cyclone continue to test out better than combining either of them with Fluid Death. If you look at the test data you will see a 50-100dps difference depending on the combat scenario. One thing worth taking away from this post (other than the best combination of trinkets) is that there is no single BiS trinket. If you combine either Unheeded Warning or Essence of Cyclone with Fluid Death in some of the tests the combination with Essence of Cyclone is better and in other tests, the combination with Unheeded Warning is better.

FD+UH FD+EoC UH+EoC
Weapon DPS 703.5 703.5 703.5
Str 142 142 142
Agi 4697 4739 5060
AP 887 887 887
Haste 1165 1165 1165
Hit 459 459 266
Crit 1102 1102 1102
Exp 188 188 188
Mastery 2010 2010 1882
TF Patch 23479 23509 23603
TF Ravage 0 24707 24665 24831
TF Ravage 2 24051 24047 24172
TF 90/30 21092 21109 21154
Berserk Patch 23443 23465 23538
Berserk Ravage 0 24497 24459 24607
Berserk Ravage 2 23903 23910 24004
Berserk 90/30 20953 20966 21007
Mar 182011
 

Greetings to the feral community. I am going to keep my intro to a bear minimum (pun intended). I started playing WoW about 3 weeks after the game was released. My first level 60 was a rogue with about 40 days played. My second level 60 was a feral druid with over 378 days played, otherwise known as “Leafkiller.” Let’s just say I liked playing the druid more. Currently I maintain an Ovale script that is posted in the cat forums of The Fluid Druid and can be found here. My script is based on the work of many ferals over the years dating back to the early move suggesters such as FBN. In keeping with the tradition that was set by nightcrowler, my script has been tuned by using a cat DPS simulator (I use Mew). If you want more background on Mew, read the Mew N’ You post by Qbear.

For better or worse, my posts will generally be focused on answering questions through simulation testing and will be a bit on the dry side. It is in my nature to prefer hard data to opinions. Many times on the forums I see people post answers that lack a numerical foundation and more often then not, the answers are incomplete, wrong or in some cases about something that is trivial and not worth discussing.

A question I see asked a lot is what secondary stats people should reforge to. Before 4.06 the question was pretty easy to answer – mastery was significantly ahead of the rest of the secondary stats and for most people crit was the clear second choice. 4.06 changed that by making our DoTs weaker and Shred/Mangle stronger. Mastery has come down in value while stats such as Hit and Expertise have come up in value. Back in October, when 4.0 first came out, I posted on EJB about dropping hit and expertise from my gearing and seeing a dps increase of over 500 (this was at level 80). The ensuing discussion was pretty lively, and (to me) culminated in a post from tangedyn which he concluded with the following statement:

It’s time we drop the notion that it is necessary to cap hit and exp. The game has changed, and we either adapt or fall back in the damage meters.

With the changes in 4.06 bringing the secondary stats closer in value to one another, I decided to retest exactly how much the dps cost was for reforging to the hit cap. 500 dps at level 80 was (to me at least) pretty substantial. It was 3% of my total damage. With 4.06 I knew the numbers had changed substantially and I wanted to know what the cost was to reforge to the hit cap. There are advantages to being hit capped which include not missing interrupts (changing in 4.01), not missing cowers, and a more stable dps rotation. There have also been many posts where people have claimed better dps on add fights with more hit. I did not try to get both hit and expertise capped as that would have resulted in gemming/enchant changes. One thing that has not changed with 4.06 is that the relative value of agility is still around three times that of any of the secondary stats and the relative value of strength is still more than twice that of the secondary stats and there are not enough secondary stats available to reforge to get both hit and expertise capped.

To test the impact of emphasizing crit, haste or the hit cap, I reforged my current gear the following four ways:

  1. First I emphasized crit rating after mastery.
  2. Then I emphasized haste rating after mastery.
  3. The third setup had haste rating and crit rating almost equal.
  4. For the fourth setup, I got as close to the hit cap as I could (960 hit rating for 7.99% hit) which did result in a slightly lower mastery rating.

I tested this against four combat scenarios as follows:

  1. A 5 minute patchwork style fight (“Patch”)
  2. A 5 minute patchwork style fight with in combat Ravages with no travel time (“Ravage 0″)
  3. A 5 minute patchwork style fight with in combat Ravages with 2 second travel times (“Ravage 2″)
  4. A 7 minute fight that alternates 90 seconds of combat with 30 seconds of downtime which is similar to the air phases on Atramedes (“90/30″)

All tests used the Glyphs of Rip and Shred, and I did full runs with the third glyph being either TF, Berserk, or SR (this is captured as row prefixes in the data below). The testing I did is not intended to comment on the relative merits of the TF, Berserk, and SR glyphs, but was done to make sure that the behavior of a hit capped profile was consistent regardless of which of those three glyphs was used. I will probably post a followup to discuss the relative merits of each of those glyphs. I have included all of the data I measured at the end of this post. I did the testing with my Mew script, which is posted here. The differences between my script and the default one that ships with Mew are pretty minor with my script typically producing slightly higher dps. To emulate the Atramedes air phases I added some code to the top of the script to return a null result after every 90 seconds for 30 seconds. I made the fight length 7 minutes to ensure that the fight ended during a combat phase.

Looking at the data below, the differences in value between haste and crit are low (the largest difference was with the glyph of berserk on the 90/30 fight where emphasizing haste comes out about 50dps ahead of stacking crit). Reforging to the hit cap is a dps loss in most scenarios – but it is not nearly as significant as it was before the 4.06 patch came out. Using it in conjunction with the glyph of TF generally showed about a 60dps loss while using it with the glyph of berserk was closer to a 100dps loss. It would be a mistake to make too many assumptions about the glyph of berserk results as it is particularly sensitive to fight length and mechanics (for example on the 90/30 fight, Mew is showing less than 3 full berserks which means the second berserk is being clipped). What was interesting was that with the the glyph of TF on the 90/30 fight the hitcap setup was on par with the other setups – which lends some credence to the notion that being hitcapped helps with dps on fights with target switching/adds.

The bottom line is that the dps loss for reforging to the hitcap is about .25% of the measured dps. The secondary stats are so close in value now that you you can pretty much choose any combination of crit/haste/exp/hit without significantly impacting your dps. Mastery is still preferred, but even giving up a small amount of mastery to get hit capped is not that significant (I had to give up 91 mastery with my gear).

-Leaf

*Updated the data with agility at 5060 not 5360.

Crit Haste Balanced Hitcap (7.99%)
Weapon DPS 703.5 703.5 703.5 703.5
Str 142 142 142 142
Agi 5060 5060 5060 5060
AP 887 887 887 887
Haste 663 1260 1134 782
Hit 266 266 266 960
Crit 1604 1007 1133 957
Exp 188 188 188 113
Mastery 1882 1882 1882 1791
TF Patch 23562 23588 23580 23529
TF Ravage 0 24797 24813 24807 24759
TF Ravage 2 24154 24142 24152 24048
TF 90/30 20127 20165 20147 20170
Berserk Patch 23492 23527 23514 23418
Berserk Ravage 0 24593 24598 24601 24526
Berserk Ravage 2 23994 24008 23995 23883
Berserk 90/30 19859 19909 19870 19827
SR Patch 23429 23462 23459 23372
SR Ravage 0 24534 24547 24536 24469
SR Ravage 2 23937 23925 23917 23840
SR 90/30 19797 19842 19811 19783