Register

4.2 Kitty Gear Spreadsheet

Face-rippin fun.

Moderator: Forum Administrators

Honored
Posts: 116
Joined: Thu May 19, 2011 7:34 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Floofles » Mon May 30, 2011 3:15 am

I'll be watching this thread closely, I hate to be a dick but currently, not being able to see your working is making your work here quite difficult to follow.

Upon seeing this thread I have made my own sheet and have very different results to you, to the point that one or both of us have made an error would it be possible to upload the three mew profiles to somewhere and make your sheet accessible please?

In addition to this, i'd suggest completely ignoring the RSVs for now, make your own sets of gear based on crit heavy, haste heavy, hit/expertise focused, mastery heavy. This is really the only reliable way of testing.

Exalted
User avatar
Posts: 1147
Joined: Fri Jun 04, 2010 4:28 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Leafkiller » Mon May 30, 2011 3:27 am

Konungr wrote:
Leafkiller wrote:100 is default...you already did that. Try a larger number...


Oops. It had 0 as default.


"0" means use default - which is 100...I asked Yawning about using 1000 - he said that would be a little slow...

But I echo what Floofles said (which echoes what I have blogged about). Ignore RSVs - just build different profiles and test them. 99% of the time I disable RSV generation while using Mew (although it is not as bad now as it used to be when it was slow - Yawning improved RSV performance a lot in the 4.2 code).

Exalted
User avatar
Posts: 568
Joined: Sat May 28, 2011 3:43 pm

Re: 4.2 Kitty Gear Spreadsheet

Postby Konungr » Mon May 30, 2011 6:54 am

@ Floofleas

Send me a message with your e-mail and I will send you the 3 Spreadsheets and the 3 Mew profiles.

I'm not too sure where the problem lies, on the WoW Post it lists all pieces of gear and enchants used. It also lists the exact amount of stats inputted into Mew for each test based on the reforges.
I will go back and create more profiles with different reforges using the same gear.
Haste > Crit > Mastery
Haste > Mastery > Crit
Exp/Hit > Haste > Crit
Exp/Hit > Mastery > Haste
Crit > Haste > Mastery
Crit > Mastery > Haste

@Leafkiller. Not sure what you mean by disabling the RSVs, don't see an option to stop generating them.

Exalted
User avatar
Posts: 1147
Joined: Fri Jun 04, 2010 4:28 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Leafkiller » Mon May 30, 2011 1:35 pm

In the top section of the Model Parameters just above where you set the "Relative stat value scale" there is a checkbox for whether or not to generate RSVs.

Since RSV generation is much faster now, it may not be a big deal to leave RSV generation on - although at the risk of sounding like a broken record, RSVs don't provide much value and overall dps is a much better indicator of a decent setup (bearing in mind a Pathwork fight will undervalue hit and expertise).

Exalted
User avatar
Posts: 568
Joined: Sat May 28, 2011 3:43 pm

Re: 4.2 Kitty Gear Spreadsheet

Postby Konungr » Mon May 30, 2011 2:32 pm

Leafkiller wrote:In the top section of the Model Parameters just above where you set the "Relative stat value scale" there is a checkbox for whether or not to generate RSVs.

Since RSV generation is much faster now, it may not be a big deal to leave RSV generation on - although at the risk of sounding like a broken record, RSVs don't provide much value and overall dps is a much better indicator of a decent setup (bearing in mind a Pathwork fight will undervalue hit and expertise).


Wow. I ran 13 Profiles this morning. With 100k Iterations and it took a few hours. I just disabled RSVs and it took less than 2 minutes.

http://i449.photobucket.com/albums/qq21 ... 1306780052

Title in Blue is the Name of the Profile, shortened. 1st Number is a calculation based on stats and their stat values. 2nd number is the DPS spit out by Mew. Blue Numbers are the stats used in Mew. Orange Numbers are the Stat Values. Green numbers are the calculations. If you look at S-V, 10-22 It gives a table of the 13 Profiles and their DPS output, Sorted by best to last. I am going to run trinket tests using the top 3 profiles here in a bit.

100k Iterations. RNG Seed, HiRes Timer. Full Raid Buffs, Minus Shattering Throw. Full DPS Spec w/ 2/3 Furor. Pro-Pot, Winds, Seafood. Ancient Seed & Hungerer used for all tests.

Exalted
User avatar
Posts: 1147
Joined: Fri Jun 04, 2010 4:28 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Leafkiller » Mon May 30, 2011 3:25 pm

You included Tricks, Unholy Frenzy and DI in the buffs?

I generally leave those off as most of the time DI goes on spriests or boomkins, rogues like to trade their tricks and unholy dks (if you still have any who have not gone 2H frost) usually use their UF as a personal cooldown.

Those three together will change dps output by around 10% - which could change which of your profiles comes out on top.

If indeed you did include those, you should consider some runs with those disabled to see if the order still holds true. Just do the runs with RSVs disabled...

Next up would be to repeat the tests with a different fight scenario. I will try to find some time to create a 4.2 version of an Atramedes sim script. That will provide some contrast and a chance to see if hit/exp jump up on a fight with interruptions to combat.

Exalted
User avatar
Posts: 568
Joined: Sat May 28, 2011 3:43 pm

Re: 4.2 Kitty Gear Spreadsheet

Postby Konungr » Mon May 30, 2011 3:31 pm

Leafkiller wrote:You included Tricks, Unholy Frenzy and DI in the buffs?

I generally leave those off as most of the time DI goes on spriests or boomkins, rogues like to trade their tricks and unholy dks (if you still have any who have not gone 2H frost) usually use their UF as a personal cooldown.

Those three together will change dps output by around 10% - which could change which of your profiles comes out on top.

If indeed you did include those, you should consider some runs with those disabled to see if the order still holds true. Just do the runs with RSVs disabled...

Next up would be to repeat the tests with a different fight scenario. I will try to find some time to create a 4.2 version of an Atramedes sim script. That will provide some contrast and a chance to see if hit/exp jump up on a fight with interruptions to combat.


No, I didn't include those. I consider those Personal Buffs, not Raid Buffs. Should have been specific. I am currently in the Process of testing the 378 Trinkets and Unheeded on the Top 4 profiles: Haste>Crit>Mastery, Haste>Hit/Exp, Hit/Exp>Mastery, and 378 Reforge None.
If you would like a copy of the 13 Excel Spreadsheets and the corresponding 13 mew profiles for the stats, send me a PM here with your email and I'll forward you the message I send Floofles today.

Exalted
User avatar
Posts: 568
Joined: Sat May 28, 2011 3:43 pm

Re: 4.2 Kitty Gear Spreadsheet

Postby Konungr » Mon May 30, 2011 4:13 pm

Information on Druid Forum Post has been updated to reflect current test results using Mew Sims over Toskk Formulation. This includes Overall DPS numbers given 13 different Reforge Profiles and 6 Different Trinket Combos using the Top 3 DPS Reforge Profiles. Slight discrepency between Top 2 Profiles, depending on the combo of trinkets, Haste > Hit/Exp is quite often higher DPS than Haste > Crit > Mastery > Hit/Exp. Highest Profile/Trinket combo is Matrix/Unheeded w/ Haste > Hit/Exp Profile.

Exalted
User avatar
Posts: 1147
Joined: Fri Jun 04, 2010 4:28 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Leafkiller » Mon May 30, 2011 4:31 pm

I just uploaded a version of the cat sim script that simulates Atramedes. If you resync to top of tree, you will get it. To use this, you need to go to the "Custom Strategy" label on the Model Parameters tab and hit "Load". When you want to go back to the default script, just hit "Clear".

It lives in the dist/Mew/samples directory. The name is Atramedes4-2 - so it is hard to miss :P

Exalted
User avatar
Posts: 568
Joined: Sat May 28, 2011 3:43 pm

Re: 4.2 Kitty Gear Spreadsheet

Postby Konungr » Mon May 30, 2011 5:10 pm

Leafkiller wrote:I just uploaded a version of the cat sim script that simulates Atramedes. If you resync to top of tree, you will get it. To use this, you need to go to the "Custom Strategy" label on the Model Parameters tab and hit "Load". When you want to go back to the default script, just hit "Clear".

It lives in the dist/Mew/samples directory. The name is Atramedes4-2 - so it is hard to miss :P



I see it on the site. I just don't see a way to download it.

Exalted
User avatar
Posts: 1147
Joined: Fri Jun 04, 2010 4:28 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Leafkiller » Mon May 30, 2011 5:27 pm

How did you download the existing Mew? It should be part of the SVN download now.

Exalted
User avatar
Posts: 568
Joined: Sat May 28, 2011 3:43 pm

Re: 4.2 Kitty Gear Spreadsheet

Postby Konungr » Mon May 30, 2011 6:43 pm

Leafkiller wrote:How did you download the existing Mew? It should be part of the SVN download now.


Nevermind. Instead of looking for a download, I just copy and pasted it into Notepad and saved it as Atramedes.script.

Honored
Posts: 116
Joined: Thu May 19, 2011 7:34 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Floofles » Mon May 30, 2011 9:32 pm

Not yet had time to have a look in detail at your stuff yet and it's 2:30 am so bed time. Just one more question before I do so we're both looking at it from the same thingy. Which version of mew are you using?

Exalted
User avatar
Posts: 568
Joined: Sat May 28, 2011 3:43 pm

Re: 4.2 Kitty Gear Spreadsheet

Postby Konungr » Mon May 30, 2011 9:36 pm

Floofles wrote:Not yet had time to have a look in detail at your stuff yet and it's 2:30 am so bed time. Just one more question before I do so we're both looking at it from the same thingy. Which version of mew are you using?


The 4.2 PTR version uploaded/updated 3 days ago. ( Mew-20110526 )

Honored
Posts: 116
Joined: Thu May 19, 2011 7:34 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Floofles » Tue May 31, 2011 3:21 am

First thing I notice is that your calculation for AP, str and agi are wrong. This is partly Yawning's fault as I assume you ripped the base stats from failcat.

The strength and agility of failcat are correct for a worgen (you're using a night elf) but the AP is not correct for any race.

Night elf druids have 255 base AP, 92 base strength (only 82 of this gives AP) and 93 agility.

That means that a night elf druid who is naked and unspecced will have 337 AP. When you spec them feral this is multiplied by aggression which is a 25% increase. AP display in game seems to always round up.

Eventually when you add a 20 stats enchant and 190 attack power on legs a night elf druid will have 112 strength and 684 AP. I did calculate for other races but as i'm not anything other than a night elf and i'm incredibly lazy and selfish I can't be bothered to go and find where I wrote them down.

I'm having difficulty seeing the formulae in the cells as I do not own a copy of excel and the web excel viewer doesn't allow you to view formulae. Uploading them to google docs so I can see the working has failed a couple of times. However i've seen enough just from the final numbers.

As I can't see the working right now, this is how reforging works;

Item A has 321 hit on it. Reforging first takes 40% away from that 321 hit which leaves us at 192.6. It will then round this up and give the item 193 of the original stat. That 193 is then subtracted from the original number and the remainder is the new stat. In this case it's 128.

Item B has 257 crit on it. Reforging first takes 40% away from that, which gives us 154.2 crit remaining on the item. Surely we'd round down? Not so young jimmy my lad! It always rounds up. This means that we are left with 155 crit and 102 of the new stat.

EDIT: I can now see the formula in your reforging box thing and can see that the reforging is done correctly. Will leave the information there in case someone is interested.

As for the results and what can be drawn from them, well, I can say this; the conclusions you are drawing regarding trinkets are rather far fetched i'm afraid. Claiming Unheeded Warning to "remain best in slot" is not really entirely correct. How can I say that? Well having made a profile with mastery > expertise > hit reforging using 378 items (+ sinestra belt and fluid death) i'm within 60 dps of your highest results. It's quite difficult to evaluate the 378/391 trinkets properly as all of them seem to be horribly under budget (look at the averaged out stat points of Essence of the Cyclone vs, for example, Hungerer) and when mastery procs on the restabilizer, Mew doesn't attempt to maximise DPS by reapplying bleeds so that it gets two full rips modified by the mastery. It's unlikely that Yawning will add support for that at any point so that trinket will most likely be undervalued if you end up proccing mastery with it.

TL;DR

There are however some conclusions which can safely be drawn from your work and i'm afraid they aren't very conclusive:

Pick any secondary stat you like, it makes no difference what so ever. Hit and expertise will always be superior for swiping as swipe does not return energy on dodges or misses.

Mastery is NOT devalued as everyone is claiming, it's still valuable when you play around with simulators. In real world situations you'll find the value varies massively as mastery will most likely be valued higher if you renew dots prior to target switching and of course mastery doesn't work optimally with the restabiliser in mew currently.

Trinkets are pretty inconclusive too, the 359/372 trinkets have higher proc budgets than the new trinkets, meaning that even fluid death is broadly similar DPS to the new ones when simulated in mew.
Last edited by Floofles on Tue May 31, 2011 3:25 am, edited 1 time in total.

Honored
Posts: 116
Joined: Thu May 19, 2011 7:34 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Floofles » Tue May 31, 2011 3:24 am

I've been making some profiles using T11 2 + T12 2 set using T11 helm/gloves and T12 legs/shoulders. As you've got far more samples, would you be willing to try switching to double two set in a couple of profiles?

The DPS seems to be at worst neutral and in some cases better than using the T12 4 set, when the T12 items are 378.

Honored
Posts: 116
Joined: Thu May 19, 2011 7:34 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Floofles » Tue May 31, 2011 4:57 am

T11/T12 double 2 set mew profile using the same settings as your profiles.

https://docs.google.com/leaf?id=0B1j_y5 ... k&hl=en_GB

Gear's basically 372 helmet/gloves, t12 legs/shoulders, sinestra belt, otherwise just 378 crap.

Reforged to emphasise mastery > expertise > hit.

Here's the spreadsheet I made in about 5 minutes to make the profile. It's not very detailed, advanced, pretty or usable but it shows what i've done, more or less. I'm too lazy to make it more usable or advanced.

https://spreadsheets0.google.com/spread ... 470L#gid=0

Honored
Posts: 116
Joined: Thu May 19, 2011 7:34 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Floofles » Tue May 31, 2011 5:02 am

I'm good at thinking of extra things to say and then replying to myself aren't I?

Later on today if I can be arsed i'll try and make a couple of 391 profiles that make sense and test them later since no one really cares about 378.

Yawning added really basic support for reapplying bleeds during the matrix proc if it procs mastery, it wasn't an increase in DPS but the support was pretty basic. He's reverted the changes as it seemed to be a loss. However, if you can make the matrix thing proc mastery due to the long duration of the proc you'll most likely get two rips and two rakes out of it anyway.

I can almost guarantee these trinkets will get buffed in the future though as they are all pretty under budget.

EDIT:

28814 DPS profile with half assed reforging (it's basically the same thing as my last post minus the FD stats and plus the seed.)

https://docs.google.com/leaf?id=0B1j_y5 ... 5&hl=en_GB

Exalted
User avatar
Posts: 568
Joined: Sat May 28, 2011 3:43 pm

Re: 4.2 Kitty Gear Spreadsheet

Postby Konungr » Tue May 31, 2011 7:55 am

It's going to take me a minute to get through the majority of the posts. I agree with Strength and AP being wrong, I knew they would. Agility is not wrong however. The stats I use are directly off of my spreadsheets which only take the Agility of the gear,enchants, and gems into account with the additional 93 Base Agility we get as Night Elf Druids. I know, I took off all my gear just to check and added it in manually to the Agi table on the Spreadsheet.

Item A has 321 hit on it. Reforging first takes 40% away from that 321 hit which leaves us at 192.6. It will then round this up and give the item 193 of the original stat. That 193 is then subtracted from the original number and the remainder is the new stat. In this case it's 128.


No I tested this from in-game. Any stat, when divided by 40% will always round down for the new stat and up for the old stat. I grabbed my calculator and looked through all the different reforging options. On one piece the 40% would have given me 71.8 When I looked in-game, It said I had 71 points to reforge. Erm, nevermind. After reading the second line I realize you were talking about the New Stat rounding up. I agree with the assessment that the Existing Stat is always rounded up.


----------------------------------------------------------------

I will take the current top 3 Reforging Profiles: Haste > Crit > Mastery, Haste > Hit/Exp, and Hit/Exp > Mastery and change up the use of the T11 Helm/Gloves, and see where that gets us.

Posts: 11
Joined: Sat Jun 04, 2011 2:23 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Odas » Sat Jun 04, 2011 2:57 am

I ran some tests with the 391 profile given with mew. I ran three different tests with 1million iterations each to try to get a decent confidence interval on the RSV. All of the tests used all debuffs and all buffs except Tricks, Unholy Frenzy, and Dark Intent. I also included external bleeds, faerie fire, mangle, and a 5% encounter duration randomization. The original profile measured at 33571.25637 +/- 1.57892 DPS and stat ordering of haste>crit>mastery>hit=exp. I then tried moving 600 hit to haste to see what changed. The second test measured 33633.61321 +/- 1.58482 DPS, however the stat values changed. While haste was still in the lead, hit and expertise jumped ahead of crit, becoming the second most valuable stats. Finally, I tried moving points from mastery back into hit and expertise and I got 33663.52666 +/- 1.58307 DPS. The final RSV that I got are as follows:

Relative Stat Values:

Weapon DPS: 6.968851273871333 +/- 0.022656854881093084
Agility: 3.726726327173965 +/- 0.022490710599647124
Haste Rating: 1.4956719715695364 +/- 0.022407085760273532
Hit Rating: 1.3887411671859444 +/- 0.022379132991703642
Strength: 1.3873242667535524 +/- 0.022429998055242702
Expertise Rating: 1.3759111250008573 +/- 0.022346412745575934
Crit Rating: 1.3628134343370766 +/- 0.02240641286657295
Mastery Rating: 1.3471437861096638 +/- 0.022405095839958665
Attack Power: 1.3212612064319547 +/- 0.022429998055242702


My guess is that hit rating was undervalued on the original profile because it was so close to the hit cap. I do find it odd however that both hit and expertise were closely rated for that profile. What can be drawn from this is that haste will definitely become our best secondary stat with full firelands gear. The other stats however are very close in value and it would likely be advantageous at this point to become hit and expertise capped to smooth the rotation.

I did a small check for the possibility of using other trinkets (namely Unheeded Warning and Ancient Petrified Seed). Both trinkets proved around 200 DPS losses when compared to 391 Hungerer and Matrix Restabilizer so I did not bother to investigate further. I did not check the effectiveness of 2t11/372 2t12/391.


Here is a link to all three of the reports generated by mew: https://docs.google.com/document/d/15ec ... t?hl=en_US

Exalted
User avatar
Posts: 1147
Joined: Fri Jun 04, 2010 4:28 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Leafkiller » Sat Jun 04, 2011 4:23 am

Odas - you could run a billion iterations and your conclusion would still be flawed.

RSVs only test values in a 100 stat point range.

The 391 profile has 6044 stat points in the secondary stats - at least 40% of which can be moved around.

We just changed Mew so that RSV generation is turned off by default when you make new profiles because people are misinterpreting it so much.

Haste may very well be the best stat - but we will only determine that by reforging to different schemes and testing overall dps.

Honored
Posts: 76
Joined: Sat Jun 04, 2011 4:21 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Yawning » Sat Jun 04, 2011 5:53 am

TLDR: Don't use RSVs for anything beyond "Should I gem anything other than Agility ever?" and *possibly* "How should I enchant?". Use lots of profiles, and compare DPS.

Sigh, hopefully people will gain some insight from this.

This is how the RSV generation code works.
1) Calculate a baseline performance value (in the case of the cat model, this would be baseline DPS).
2) For each stat that the user is interested in:
2a) For hit and expertise, if hitRating + X (or expertiseRating + X) would put you over the melee hit (or dodge 'soft' cap), it calculates a new baseline exactly at the cap, and a sample point X rating below the cap. And saves the difference as deltaDPS.
2b) For other stats, add X, and saves the difference from the baseline calculated in step 1 as deltaDPS.
3) Each stat's RSV is deltaDPS / X.

There are a few things that astute readers should figure out from this.
a) Comparisons are done in a vacuum. What you are essentially figuring out is "how much will my DPS increase if I magically received X amount of free stats".
b) It will always attempt to give "sane" values for hit/expertise. The way Mew is doing this makes looking at X beyond ~1k rating somewhat problematic (the values of Hit/Expertise will be even more worthless).
c) The values are highly dependent on X, especially if the spec being modeled has a complicated interrelationship between stats. (Hint: Feral Druids do.)

Under the present climate, the only time RSVs end up being useful are for deciding gemming/enchanting since that's the only time you end up increasing a single isolated stat with no opportunity cost. Now, you may ask "Well, if they're useful only in such a niche case, why include them at all? And how am I supposed to look into more complex questions such as scaling or reforging?".

RSV generation is included in Mew when I wrote it for a few reasons:
a) When I wrote Mew, it was my opinion that stat inflection would be relatively rare. ArPen was gone. Mastery was totally terrible during much of Cataclysm Beta, and when it Feral as a whole was buffed, Mastery became so strong that it was extremely unlikely that any solution other than "Spam Agi, Spam Mastery Everywhere" would be correct.
b) The correct solution to solving gearing is extremely computationally intensive. There is a problem in mathematics called the knapsack problem, that reads as thus "Given a set of items, each with a weight and a value, determine the number of each item to include in a collection so that the total weight is less than or equal to a given limit and the total value is as large as possible." If you are thinking "Hey, that sounds a lot like gearing", you would be correct, since the problems are identical. The knapsack problem is NP-complete, which boils down to "believed[0] to be extremely time consuming to solve exactly". There are a few ways to approximate a "good enough" solution that don't require solving the full problem set but with how fast the Simulator runs, they would still take too long.
c) I am a terrible UI programmer, and implementing a gear based UI was not something I was comfortable doing.
d) Other specs have less stat swing. For example, the bear mitigation model's stat valuations are significantly more consistent.
e) As a extremely rough back of the envelope estimate, they can be used to give vague hints when attempting to solve this the correct way, much like how gutting a goat and playing with the entrails will give you insight into how the harvest will go this year.

The correct way to look into any sort of complex question (like scaling, reforging, rotational analysis) would be:
1) Make a profile with all the gear you want to look at on chardev.org
2) Make another profile with a different gear set.
3) Configure the Simulator.
4) Run the Simulator for each profile.
5) Compare DPS. If satisfied stop, otherwise repeat.

This should be relatively easy give that Mew will happily let you save 3 zillion different data files all alike, and once you've come up with a base profile, making incremental adjustments is relatively easy given that the only real variable is reforging. Note the use of the Simulator. The Formulation model is only really useful for getting a ballpark estimate of DPS, since it currently makes a few rather liberal assumptions.

Hope this helps clarify things.

[0]: P=NP is still open to debate. If you feel like getting a guaranteed teaching position at your university of choice, a Fields Medal, and a $1 million cash prize, feel free to attempt to find a proof either way.
"Understanding is a three-edged sword."

Mew: http://code.google.com/p/mew-wow-druid-model/

Exalted
User avatar
Posts: 1147
Joined: Fri Jun 04, 2010 4:28 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Leafkiller » Sat Jun 04, 2011 2:30 pm

Good afternoon Yawning! (and welcome to the posting side of these forums). I hope you did not stay up too much longer beyond when I went to bed last night ;)

Posts: 11
Joined: Sat Jun 04, 2011 2:23 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Odas » Sat Jun 04, 2011 7:12 pm

I understand the way RSVs are calculated and how they can be misleading. The point of doing those tests was to get a feel for the way that stats changed with the 4.2 shift of our damage from bleeds to direct attacks. Seeing how haste is always significantly higher than mastery for example, I can guide which reforging strategies to focus on.

Do you have a list of the items used to produce the 391 ptr profile that comes with mew? When I tried to make a BiS list I ended up with about 200 less secondary stats. https://spreadsheets.google.com/spreads ... c&hl=en_US

Honored
Posts: 76
Joined: Sat Jun 04, 2011 4:21 am

Re: 4.2 Kitty Gear Spreadsheet

Postby Yawning » Sat Jun 04, 2011 8:47 pm

It's in the change logs:
http://code.google.com/p/mew-wow-druid- ... tail?r=640

As a general rule, I do show my work. It was something I threw together quickly so there may be mistakes in it, but my rationale for making one was to get something that was approximately correct so I can eventually start working on re-tuning the script.
"Understanding is a three-edged sword."

Mew: http://code.google.com/p/mew-wow-druid-model/

PreviousNext

Return to Kitty DPS

Who is online

Users browsing this forum: eronn, Yahoo [Bot] and 8 guests