Jump to content

Tanking Stat Weights


dipstik

Recommended Posts

ive started a thread like this in the pts forums. i looked at 90% for k/e coming from w/r and the other 10% from f/t, but looking at keyboardninja's boss dps numbers, the average is 70% of damage from w/r and 30% from f/t, in terms of w/e damage, so defense is almost useless. i will most likey end up posting the numbers for 70/30 split, but it is really just weighting absorb to shield...

 

i think defense should increase resistance to give defense the same number of rolls as shield.

 

right now for eevry 10 w/r attacks we get 7 defense rolls and 10 shield rolls (well... except the part when if defense roll wins we dont roll shield... but u know what i mean)

Link to comment
Share on other sites

  • Replies 328
  • Created
  • Last Reply

Top Posters In This Topic

Kitru i have a question regarding these stat weights for NiM Kephass. Would you keep these same stat weights on a shadow for that fight or stack a bit for defense for i minor loss in absorb? Edited by Daelars
Link to comment
Share on other sites

Kitru i have a question regarding these stat weights for NiM Kephass. Would you keep these same stat weights on a shadow for that fight or stack a bit for defense for i minor loss in absorb?

 

Kephess is all about high amounts of incoming damage, and the problem that makes Shadows squishy is the general spikiness of incoming damage and high number of attacks per second. As such, I would honestly focus primarily on maximizing total mitigation rather than trying to preserve KW stacks since those stacks are going to vanish almost instantly no matter what you do, not to mention that focusing on Defense just makes you incoming damage even more spiky.

Link to comment
Share on other sites

  • 2 weeks later...

Other than "I'm doing it wrong", which I know has to be in there somewhere, I'm falling short on getting the correct calculation results relative to Defense %.

I get different stats when putting on Mr. Robot than I do in game. AND using the same defense rating (678 from IG stat sheet, not Mr. Robot) I get wildly different values from LagunaD's sheet and the formula provided in this thread.

 

Excel tells me with a 678 def rating =1/(100)*(5 +12+10+5+ 30 * ( 1 - ( 1 - ( 0.01 / 0.3 ) )^( ( (678+46) / 50 ) / 0.55 ) )) results in a def rating of .497115. ***? 49% def? using Mr. Robot's 804 I get an even greater result.

 

Using a 678 def rating in LagunaD's sheet I get a Def rating % of 17.8% and a total of 39.8%.

 

I know that all of the items/mods/enhancements in Mr. Robot are what I have IG, but they aren't matching up from character stat sheet to the resulting %s in Mr. Robot.

 

In short, I'm trying to optimize here but I'm getting all sorts of wonky results that are not helping in any way. Help?

 

http://swtor.askmrrobot.com/character/4bebb4eb-71a7-4720-9cea-c6e16dc4a65b

Link to comment
Share on other sites

As long as the ratings themselves are what you are comparing. The % results of the spreadsheet includes accuracy debuff and blade barrier, ask mr robot and in game character sheet don't. Ask mr robot may differ, make sure you have the exo stim on/off depending on if you have one active in game. Edited by Marb
Link to comment
Share on other sites

Other than "I'm doing it wrong", which I know has to be in there somewhere, I'm falling short on getting the correct calculation results relative to Defense %.

I get different stats when putting on Mr. Robot than I do in game. AND using the same defense rating (678 from IG stat sheet, not Mr. Robot) I get wildly different values from LagunaD's sheet and the formula provided in this thread.

 

Excel tells me with a 678 def rating =1/(100)*(5 +12+10+5+ 30 * ( 1 - ( 1 - ( 0.01 / 0.3 ) )^( ( (678+46) / 50 ) / 0.55 ) )) results in a def rating of .497115. ***? 49% def? using Mr. Robot's 804 I get an even greater result.

 

Using a 678 def rating in LagunaD's sheet I get a Def rating % of 17.8% and a total of 39.8%.

 

I know that all of the items/mods/enhancements in Mr. Robot are what I have IG, but they aren't matching up from character stat sheet to the resulting %s in Mr. Robot.

 

In short, I'm trying to optimize here but I'm getting all sorts of wonky results that are not helping in any way. Help?

 

http://swtor.askmrrobot.com/character/4bebb4eb-71a7-4720-9cea-c6e16dc4a65b

 

Askmrrobot for Shadows/Sins Tank Defense calculations have been inaccurate by I think 4-5% for quite a few patches now :( Sadly they claim to be updating it but their main focus is to WoW and LoL, with SWTOR getting table scraps whenever they decide to drop them. Its falling on its face way behind, its DB is still 1.4 and missing/inaccurate on several items so don't use it as 100% measure.

Link to comment
Share on other sites

This is what's become the "standard" hybrid tank spec. There are some variations upon it (some don't take Overhead Slash and instead take a second point in Swelling Winds; some take Stasis Mastery or Courage instead of Swelling Winds), but it's a difference of 1-2 points of allocation at most and there haven't been appreciable differences in performance discovered between them.

 

This is a great thread. Thanks a lot for crunching the numbers for us.

 

For what its worth, on my hybrid guardian tank, I actually take 1 point in force rush instead of swelling wings or courage or stasis mastery. If you get in the habit of using OS every cooldown, it means you will crit on most (60% or so.... I'm not a mathy type) of your BS and dispatches.

Link to comment
Share on other sites

  • 4 weeks later...

POWERTECH/VANGAURD

 

total defense shield absorb

1400 314 586 500

1500 335 634 531

1600 358 681 561

1700 382 727 590

1800 409 772 619

1900 435 817 647

2000 463 862 675

 

> d90(dr):=1/(100)*(5 +2+10+ 30 * ( 1 - ( 1 - ( 0.01 / 0.3 ) )^( ( (dr+46) / 50 ) / 0.55 ) )): ;

> sh(shr):=1/(100)*(5+23+ 50 * ( 1 - ( 1 - ( 0.01 / 0.5 ) )^( ( shr / 50 ) / 0.32 ) )):

> ab(abr):=1/(100)*(20+4+8+50 * ( 1 - ( 1 - ( 0.01 / 0.5 ) )^( ( abr / 50 ) / 0.18 ) )):

> d100(dr):=1/(100)*(5 +2+ 30 * ( 1 - ( 1 - ( 0.01 / 0.3 ) )^( ( (dr+46) / 50 ) / 0.55 ) )): ;

> ac90 := .5;

> ac100 := .5;

LagrangeMultipliers(ac90*(1-d90(dr))*(1-sh(shr)*ab(abr))+ac100*(1-d100(dr))*(1-sh(shr)*ab(abr)), [abr+dr+shr-N], [dr, shr, abr])

 

ASSASSIN/SHADOW

1400 453 430 518

1500 471 479 550

1600 492 527 581

1700 514 575 611

1800 538 621 641

1900 563 667 670

2000 590 712 698

 

> d90(dr):=1/(100)*(10 +6+15+ 30 * ( 1 - ( 1 - ( 0.01 / 0.3 ) )^( ( (dr+46) / 50 ) / 0.55 ) )):

> d100(dr):=1/(100)*(10 +6+5+ 30 * ( 1 - ( 1 - ( 0.01 / 0.3 ) )^( ( (dr+46) / 50 ) / 0.55 ) )):

> sh(shr):=1/(100)*(5+15+20+ 50 * ( 1 - ( 1 - ( 0.01 / 0.5 ) )^( ( shr / 50 ) / 0.32 ) )):

> ab(abr):=1/(100)*(20+4+50 * ( 1 - ( 1 - ( 0.01 / 0.5 ) )^( ( abr / 50 ) / 0.18 ) )):

 

 

JUGGERNAUT/GUARDIAN (HYBRID)

 

pool defense shield absorb

1400 838 268 295

1500 837 327 337 (counterintuitive)

1600 843 381 375

1700 854 434 412

1800 870 484 446

1900 888 533 479

2000 909 581 511

 

 

> d90(dr):=1/(100)*(5 +12+10+5+ 30 * ( 1 - ( 1 - ( 0.01 / 0.3 ) )^( ( (dr+46) / 50 ) / 0.55 ) )): ;

> d100(dr):=1/(100)*(5 +12+5+ 30 * ( 1 - ( 1 - ( 0.01 / 0.3 ) )^( ( (dr+46) / 50 ) / 0.55 ) )):

> sh(shr):=1/(100)*(5+15+ 50 * ( 1 - ( 1 - ( 0.01 / 0.5 ) )^( ( shr / 50 ) / 0.32 ) )):

> ab(abr):=1/(100)*(20+50 * ( 1 - ( 1 - ( 0.01 / 0.5 ) )^( ( abr / 50 ) / 0.18 ) )):

 

 

FULL JUGG

pool defense shield absorb

1400 769 290 342

1500 775 345 381

1600 786 392 417

1700 800 448 452

1800 818 498 485

1900 858 546 516

2000 860 593 547

 

 

 

 

> d90(dr):=1/(100)*(5 +12+10+5+ 30 * ( 1 - ( 1 - ( 0.01 / 0.3 ) )^( ( (dr+46) / 50 ) / 0.55 ) )): ;

> d100(dr):=1/(100)*(5 +12+5+ 30 * ( 1 - ( 1 - ( 0.01 / 0.3 ) )^( ( (dr+46) / 50 ) / 0.55 ) )):

> sh(shr):=1/(100)*(5+19+ 50 * ( 1 - ( 1 - ( 0.01 / 0.5 ) )^( ( shr / 50 ) / 0.32 ) )):

> ab(abr):=1/(100)*(20+50 * ( 1 - ( 1 - ( 0.01 / 0.5 ) )^( ( abr / 50 ) / 0.18 ) )):

 

Is this updated for 2.0?

Link to comment
Share on other sites

this is old info. thanks for all kind words. i hope to be doing more theorycrafting in the end of may. super busy til then.

 

im still throwing my 2 cents to key board ninja when i can though. its much harder to do all this work without askmrrobot.

Link to comment
Share on other sites

i took the proc cool dowsn as 23 seconds. the mitigation with the relics effect is just relic. so i would take base as 2200 abalnced build wwith underworld gear and add to that the relic effect.

 

so for procs youhave (6/23)*relic+(17/23)*base= time averaged value. i used 30/120 and 20/120 for the on use relics.

 

i included the static defense for the proc defense relic. for pvp i took the slope of the post armor mitigation as a function of mitigation pool and multiplied that by 120 for the elite war hero.

 

jugg:

 

relic analysis. lower number is better: i took mitigation with applied stats and did a time average using uptime as wieghts to the mitigation. the base and applied(relic) stats are given, and tyhe time weighted result.

 

 

weighted relic base

def proc 0.44142418 0.432999241 0.444397688

abs proc 0.441603447 0.433686429 0.444397688

shd use 0.442049775 0.435006034 0.444397688

pvp 0.441760428 0.444397688

use def 0.442955845 0.435746626 0.444397688

 

proc defense and elite war hero. next best is proc absorb if you dont have proc defense

 

 

relics for shadows:

 

healing does 634/23 s... need 7200 dps+ to make any relic better than healing relic

 

weighted relic base

def proc 0.436610018 0.425744183 0.440445019

abs proc 0.436841222 0.426630466 0.440445019

shd use 0.437086044 0.427009121 0.440445019

pvp 0.437047992 0.440445019

use def 0.43858544 0.429287549 0.440445019

 

heal relic and proc defense

 

PT:

 

weighted relic base

def proc 0.416283504 0.409265271 0.418760528

abs proc 0.416159806 0.408791095 0.418760528

shd use 0.416181263 0.408443469 0.418760528

pvp 0.416397707

use def 0.417499447 0.411194043 0.418760528

 

absorb proc and on use shield/absorb

Edited by dipstik
Link to comment
Share on other sites

suggested 2.0 gearing:

 

suggested 2.0 gearing:

 

you can trade 17endurance for 20 mitigation in the enhancements. for shadows the trade isnt good enough. at 3k dps you self heal for 9 per second, versus mitigating 12 per second with the 20*7 mitigation being traded.

 

shadow: http://swtor.askmrrobot.com/character/65b987d7-c58f-4cde-bcb3-a4087603dcb5

 

guardian: http://swtor.askmrrobot.com/character/f4f059f3-a83f-4d82-8bb1-ff52c0cf84e6

 

the past few points in the skill trees are up for grabs in both specs. note that double strike does about 60% of the damage that project does in fights. thats why i went with extra project damage.

 

there gearings are REALLY close to optimal stat weights (plus minus 10), where i have assumed shield to be at 790.

 

will get a PT one up soon

Link to comment
Share on other sites

As far as I can tell, the Immunity 31 and Sturdiness 31 enhancements don't exist in gear. Unless they turn out to be a schematic drop somehwere, I don't think we'll actually be able to get our hands on these enhancements. I think we'll be stuck using the Bastion and Bulwark enhancements instead. Edited by thasensei
Link to comment
Share on other sites

As far as I can tell, the Immunity 31 and Sturdiness 31 enhancements don't exist in gear. Unless they turn out to be a schematic drop somehwere, I don't think we'll actually be able to get our hands on these enhancements. I think we'll be stuck using the Bastion and Bulwark enhancements instead.

 

They don't drop anywhere. There is no such thing as 72 crafting. PT and juggs want high mitigation mods while sin tanks want high endurance b mods. It is insanely easy to gear up a sin tank as it is.

Link to comment
Share on other sites

They don't drop anywhere. There is no such thing as 72 crafting. PT and juggs want high mitigation mods while sin tanks want high endurance b mods. It is insanely easy to gear up a sin tank as it is.

 

My post was in response to the posted AMR builds. I was just pointing out that the builds are using enhancements which are currently unavailable, so it would be worth adjusting those accordingly. But thanks for the info!

Link to comment
Share on other sites

A very sincere thank you to everyone who participated, particularly dipstik, for all of the incredibly informative information and hard work.

 

Thank you very much. your thanks is appreciated. you are very welcome.

 

That being said. I updated the OP with relic info and links to BiS Builds.

Link to comment
Share on other sites

×
×
  • Create New...