Jump to content


Photo

Redcape's Ret DPS Spreadsheet 5.21 - 4.3 ready (Updated 01/31/12)


  • Please log in to reply
60 replies to this topic

#41 Redcape

Redcape

    King Hippo

  • Members
  • 543 posts

Posted 17 February 2011 - 02:26 PM

It is possible, certainly, but I doubt I will do it. My spreadsheet is much more focused around theoretical rotation testing and developing stat weights rather than testing gear combinations. I could make it do that but since at the moment I have absolutely no idea how to do so it would take a lot of time to learn and implement which I don't have at the moment.

It may happen in the future but don't hold your breath.
My Ret paladin spreadsheet: ->here<-

For questions regarding my spreadsheet or otherwise please contact me via PM on the EJ boards and not in game. Thanks.

Philosophy, Psychology and other fun stuff:

WOW and gaming in general:

#42 Redcape

Redcape

    King Hippo

  • Members
  • 543 posts

Posted 20 February 2011 - 01:09 AM

New version uploaded. 5.15

Has reverberating metagem, new gearset with new reforging (should be BIS 359 with t11 4p) and a few tiny bugfixes.
My Ret paladin spreadsheet: ->here<-

For questions regarding my spreadsheet or otherwise please contact me via PM on the EJ boards and not in game. Thanks.

Philosophy, Psychology and other fun stuff:

WOW and gaming in general:

#43 Redcape

Redcape

    King Hippo

  • Members
  • 543 posts

Posted 23 February 2011 - 06:17 PM

I have put up 5.16 version. Added 50 crit to cloak enchant and fixed a bug with calculating stat returns on hit/exp. This will not affect gearing decisions as it only mattered to those who were below hit cap, which is still a terrible idea.

Newest weights:

Weapon Speed 187923
Weapon DPS 862

Strength 227
Hit rating 172
Exp rating 125
Crit rating 95
Haste rating 81
Mastery rating 101
My Ret paladin spreadsheet: ->here<-

For questions regarding my spreadsheet or otherwise please contact me via PM on the EJ boards and not in game. Thanks.

Philosophy, Psychology and other fun stuff:

WOW and gaming in general:

#44 Vexryn

Vexryn

    Glass Joe

  • Members
  • 3 posts

Posted 27 February 2011 - 04:58 PM

Redcape,
I'm getting some odd bug behavior in the spreadsheet. Not even sure how to report it.

I entered in my gear stats and the associated enchants.

When I update the stat weightings with the "Make Rotation" button, I'm getting some surprising results:
Str 207
Hit 146
Exp 100
Crit 83
Haste 95
Mastery 71


The piece that jumps out at me is the Haste 95! Why is Haste showing up as a top secondary stat?

I was so incredulous, I tried verifying this in the gear section myself. I tested it easily by just adding 100 points of any particular stat, and looking at the dps impact.

As expected, 100 STR gave me 207 dps. 100 Crit gives me 83 dps. 100 Mastery gives me 71 dps.

But 100 haste, entered directly into the gear sheet, gives me only 36 dps, not 95. If I rerun the rotation in light of the higher haste, it turns out the increased haste only gives me 19 dps.

So all the other stat weightings appear to be accurate, but the stat weighting report for Haste is coming up very out-of-whack.

I entered a relatively "normal" set of gear and enchants. The only fight conditions I changed were to add 0.15sec latency, and to set the target swap duration equal to the fight duration.


Any thoughts about what might be going on here? :)

#45 Nätion

Nätion

    Piston Honda

  • Members
  • 109 posts

Posted 27 February 2011 - 10:15 PM

I'm not 100% sure how Redcape's spreadsheet calculates the numbers, but it is possible you are near a valley of haste value in that gear, where a tiny bit more would grant alot of additional gain, but subsequent points would not add much/anything. 100 may just be too much, try, 1, 10, 20, 50 or something. As always stated the stat weights are relative numbers, and your best bet to determine gearing choices is to run your gear against the upgrade and see which comes out on top.

#46 Vexryn

Vexryn

    Glass Joe

  • Members
  • 3 posts

Posted 27 February 2011 - 10:18 PM

I'm not 100% sure how Redcape's spreadsheet calculates the numbers, but it is possible you are near a valley of haste value in that gear, where a tiny bit more would grant alot of additional gain, but subsequent points would not add much/anything. 100 may just be too much, try, 1, 10, 20, 50 or something. As always stated the stat weights are relative numbers, and your best bet to determine gearing choices is to run your gear against the upgrade and see which comes out on top.


I thought of that and tried a few increments besides 100; results were similar (I just forgot to post here). 10 haste gave me 4 dps (with rounding); 30 haste gave me 11dps; 50 haste was 18, and 100 haste was 36. Even 1000 haste was 359 dps, so it appears to be calculating in a purely linear fashion.

#47 Exemplar

Exemplar

    The One-Eyed Man

  • • Guide Author
  • 1817 posts

Posted 28 February 2011 - 01:55 PM

I thought of that and tried a few increments besides 100; results were similar (I just forgot to post here). 10 haste gave me 4 dps (with rounding); 30 haste gave me 11dps; 50 haste was 18, and 100 haste was 36. Even 1000 haste was 359 dps, so it appears to be calculating in a purely linear fashion.


Are you placing this value in the "Testing" fields? This will provide a linear increase, because it is not re-running the simulation. It is simply adding the Haste directly to weaponswing, CS CD, and Spell GCD without recalculating how this would impact Exo procs or CS/Filler usage (collisions and priority decisions). "Testing" fields are perfectly legitimate for every other stat (since only Haste throws a wrench in the simulation).

If you want to see how DPS will truly change, go to the Gear tab. Find an item with no Haste. Add a value of Haste to this item. Press "Make Rotation." It will run the simulation and you will get an actual result. I suspect you will find that you rapidly climb out of your trough and further Haste will suddenly be less than pleasant.

Repeat this many times by small increments (a few times I've built a function in his sheet to increment it by 10s and list on a fresh tab to generate a graph, just to compare with my sheet) and you should see Haste value rise and fall - there may even be a point where enough extra Haste is less DPS than zero extra Haste. It's an unavoidable flaw inherent in how Haste is modeled via Simulation and nothing that Redcape has done incorrectly or can remove from the spreadsheet.
Rock: "We're sub-standard DPS. Nerf Paper, Scissors are fine."
Paper: "OMG, WTF, Scissors!"
Scissors: "Rock is OP and Paper are QQers. We need PvP buffs."

#48 Redcape

Redcape

    King Hippo

  • Members
  • 543 posts

Posted 28 February 2011 - 03:16 PM

Exemplar is quite right. There are a few more things I should add though:

First, you can pretty much entirely eliminate the haste valleys and hills through longer simulation times. You have to be willing to sit for 30 seconds while it crunches (or more on an old machine) but at 100,000 seconds the value for haste is actually very stable and can handle fairly small changes in haste pretty well. The old sheet had a limit of 30,000 seconds which you can change by altering data validation if you want to, the new limit is 100,000. Using short simulation times will always give strange values for haste. This is a limitation of the way the simulator works and is not fixable.

Note that adding latency is going to change the value for haste substantially so when you change that variable you will definitely see results that differ from the ones normally accepted as standard. I can't reasonably test all combinations of different settings myself to report interesting behaviour but if you find the sheet reporting values that differ from the accepted norms with particular settings (and have used a long sim time) please do report them as they might be useful.
My Ret paladin spreadsheet: ->here<-

For questions regarding my spreadsheet or otherwise please contact me via PM on the EJ boards and not in game. Thanks.

Philosophy, Psychology and other fun stuff:

WOW and gaming in general:

#49 slipey

slipey

    Glass Joe

  • Members
  • 23 posts

Posted 07 March 2011 - 10:17 PM

I just downloaded 5.16 and wanted to do some analysis of the T11 4p bonus however, I am encountering some strange behavior. If I disable the T11 4p bonus in the Set bonuses section of the Talents, Buffs, Enchants page and then run "Make Rotation" again, I get the same result as with the bonus enabled. Even the "Ability Uses" section on the Main page is identical after running "Make Rotation" with the option either enabled or disabled. I am just using the default gearing and options that were pre-configured with the spreadsheet. Is there a reason the DPS estimates and ability usages stats would be the same with 4p t11 both enabled and disabled or is this possibly a bug?

#50 Redcape

Redcape

    King Hippo

  • Members
  • 543 posts

Posted 08 March 2011 - 02:24 PM

This is a bug. It turns out I introduced it when I added the 5% CS from glove field. The t10 4P status was being taken from the t10 2P field. This only matters in the instance where you set the sheet up to have 4P "No" and 2P "Yes". Given that all my published weightings and such are still going to be correct, thankfully, since obviously everybody is using 4P and all relevant calculations were done with both set bonuses enabled.

The bug has been corrected and the new version will have that update. Thanks.
My Ret paladin spreadsheet: ->here<-

For questions regarding my spreadsheet or otherwise please contact me via PM on the EJ boards and not in game. Thanks.

Philosophy, Psychology and other fun stuff:

WOW and gaming in general:

#51 Redcape

Redcape

    King Hippo

  • Members
  • 543 posts

Posted 14 March 2011 - 05:03 PM

New version 5.17 now up.

This includes a few minor bugfixes that do not affect combat choices in current gear and the new

TV - DP
Inq - DP

choices in the priority list. This allows the user to set up different priorities for TV and Inq based on HP status and DP status independently. The newest best rotation seems to be:

Inq
CS (0-2 HP)
TV (DP)
Exo
HoW
TV (3 HP)
Judgement
HW
Consecration

clcinfo supports this rotation already so it should be easy to implement for those using that addon. The new ability to specify Inq as either HP or DP does not matter as placing both first in rotation is optimal. This may change with new mechanics.
My Ret paladin spreadsheet: ->here<-

For questions regarding my spreadsheet or otherwise please contact me via PM on the EJ boards and not in game. Thanks.

Philosophy, Psychology and other fun stuff:

WOW and gaming in general:

#52 Panzerlol

Panzerlol

    Glass Joe

  • Members
  • 9 posts

Posted 17 March 2011 - 12:10 PM

First time poster, so bear with me. :)

I noticed some minor issues with your spreadsheet.
1) The +50 crit back enchant is counting as +100 crit, since it is repeated twice in the excel "code" for back enchant selection.
2) +10 Mast gem bonus isn't implemented. The drop-down meny offers only +"4 Mast", while the code that's summing up mastery is only checking if .IF(B20:Q20; "10 Mast")*10, thus will any +10 Mast gem bonuses be ignored.

And as a request, would it be possible to implement DMC:H as a trinket option, or is it possible to approximate it as a STR bonus?

#53 Redcape

Redcape

    King Hippo

  • Members
  • 543 posts

Posted 17 March 2011 - 03:24 PM

Good points. I have posted a new version, 5.18, where both the back enchant and the mastery socket bonus are fixed. I also added in DMC:H support for good measure, although since we don't have a working model of it yet (to the best of my knowledge, anyhow) I just built a kludge that models it correctly for current gear levels. Should be fine for current content and DMC:H isn't relevant for the next tier of content anyhow.
My Ret paladin spreadsheet: ->here<-

For questions regarding my spreadsheet or otherwise please contact me via PM on the EJ boards and not in game. Thanks.

Philosophy, Psychology and other fun stuff:

WOW and gaming in general:

#54 Panzerlol

Panzerlol

    Glass Joe

  • Members
  • 9 posts

Posted 07 April 2011 - 10:32 PM

I have recently discovered wowreforge.com, a tool that allows for very efficient reforging for the respective caps, and then further stack the stat with the highest stat weight. I very much appreciate the fact that it fetches everyting from armory, as it can be quite tedious to update manually the spreadsheet every time i acquire new gear and change my reforgings, and this makes it very user-friendly.

The only issue that has bothered me to some extent, is the fact that the stat weights are static. You can of course change them manually, but part of the deal with retribution is that the stat weights change with your gear. The tool uses the same stat weights no matter what/how your gear is.

Do you think it would be possible to implement some of your dps spreadsheet functionality into the tool present there, or perhaps the most important aspect, namely the calculation of specific stat weights? Another possibility could be to implement it online, like the case is for hunters. I do not know much about how much work this might prove to be, but I might have a site able to host it.

#55 Redcape

Redcape

    King Hippo

  • Members
  • 543 posts

Posted 08 April 2011 - 02:23 PM

Recalculating stat weights dynamically as your change your reforgings/gemmings would be colossally time intensive. It is certainly possible for me to create a reforger that checks new stat weights after each change but to get decent numbers you would have to leave excel running for a month? year? lifetime? to work it all out. There is no way with current computing power to work out new stat weights with new rotations derived from new stats on gear to optimize reforging in a rational time period. The search space is just too large.

I have, at times, been asked to incorporate an armory grabber into my sheet and I may tackle that someday but not right at the moment.
My Ret paladin spreadsheet: ->here<-

For questions regarding my spreadsheet or otherwise please contact me via PM on the EJ boards and not in game. Thanks.

Philosophy, Psychology and other fun stuff:

WOW and gaming in general:

#56 Panzerlol

Panzerlol

    Glass Joe

  • Members
  • 9 posts

Posted 09 April 2011 - 08:23 AM

I'm sorry, I worded it badly.
What I had in mind was to do one round of calculations, based on your current setup, and then provide stat weights based on that.
I know this wont be perfect, but at least it will give some directions of what your best secondary stats would be from that point.

Anyway, great job on the spreadsheet! It has really helped my DPS since I started using it.

#57 Guest_AceRider_*

Guest_AceRider_*
  • Guests

Posted 09 April 2011 - 09:10 AM

What I had in mind was to do one round of calculations, based on your current setup, and then provide stat weights based on that.
I know this wont be perfect, but at least it will give some directions of what your best secondary stats would be from that point.


I think what you're looking for is already within the spreadsheet. If you look on the 'Main' sheet there is a heading Stat weightings for SoT. The stat table underneath this is your current stat weights for the gear used in the spreadsheet. Perhaps not perfect but Redcapes stat weights are usually one of the most accurate I've found.

#58 Redcape

Redcape

    King Hippo

  • Members
  • 543 posts

Posted 29 April 2011 - 03:09 PM

New version up. A few very minor bugfixes.
My Ret paladin spreadsheet: ->here<-

For questions regarding my spreadsheet or otherwise please contact me via PM on the EJ boards and not in game. Thanks.

Philosophy, Psychology and other fun stuff:

WOW and gaming in general:

#59 Redcape

Redcape

    King Hippo

  • Members
  • 543 posts

Posted 25 June 2011 - 09:17 PM

A new version is up, 5.20. It is current with the mechanics changes coming in 4.2 including new tier set bonuses.

I have retired from WOW so I will no longer be keeping current with upcoming changes in the game. I am happy to continue to support mechanics changes and new set bonuses as those take minimal time investment but there will no longer be substantial upgrades or rewrites to the sheet. I will no longer be checking this thread or forum so please PM me with any new mechanics that need to be added in.

Good luck in all your adventures in Azeroth. May you find many shiny piles of loot and may your numbers always get bigger.
My Ret paladin spreadsheet: ->here<-

For questions regarding my spreadsheet or otherwise please contact me via PM on the EJ boards and not in game. Thanks.

Philosophy, Psychology and other fun stuff:

WOW and gaming in general:

#60 Fordragon

Fordragon

    Von Kaiser

  • Members
  • 24 posts

Posted 25 June 2011 - 09:37 PM

A new version is up, 5.20. It is current with the mechanics changes coming in 4.2 including new tier set bonuses.

I have retired from WOW so I will no longer be keeping current with upcoming changes in the game. I am happy to continue to support mechanics changes and new set bonuses as those take minimal time investment but there will no longer be substantial upgrades or rewrites to the sheet. I will no longer be checking this thread or forum so please PM me with any new mechanics that need to be added in.

Good luck in all your adventures in Azeroth. May you find many shiny piles of loot and may your numbers always get bigger.




Really sad to see you leaving.Good luck with real life mate.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users