Tanking Stat Weights

[Kitru]

Let me know. I need to update stuff for new gear anyway.

 

Assuming you're talking to me, I edited the math to be correct right after Will brought it up. Interval on EB is 10.3 seconds, average contribution of it is 14.6%, and average contribution of Power Screen is 1.73%.

Edited May 27, 2013 by Kitru

[WillLongstick]

what I meant with "least you can get" is, assuming you are in a scenario where you are not shielding any attacks and not getting lucky with ion cylinder procs, so the only thing reducing the cd is the rocket punch and rail shot. In other words what I meant was the higher cd timer possible, sorry for misleading

 

Triggering your cylinder with any ranged attack (15% chance from Cylinder, I don't see any talents to increase that) also procs a Heat Screen stack reducing the CD of Heat Blast. If you're not getting attacked at all, you can just spam Rapid Shots as it is 5 seperate attacks, giving you a .94*(1-.85^5) = .52 chance of proccing your cyliner (where .94 is your chance to hit). That may be an even higher chance as I feel like I'm overlooking a talent that brings the 15% chance to trigger even higher.

[Darth_Dreselus]

Are the weights for VGs and Guards calculated with the def rating available from the relic (just the weights not the AMRs). Since I didn't get that relic on either I am wondering the effect it has on the best dsa value.

[dipstik]

they are calced before the relics

[tXHereticXt]

Been looking through stuff since I decided to move onto ditching huge health pools and start gearing for optimal builds.

 

One of the things that makes this easier is using the appropriate mods and enhancements in order to make stat balancing with augments easier.

 

This is my primary issue. Just not know what mods to set up with what enhancements and augs in order to pull the numbers off. I'm terrible with math, obviously, but I can get the basic stuff.

 

However my concern of course is the lack of consistency to also get a second gear so by gearing my Sin and future Shadow for HM TFB/S&V is if certain bosses will negate all that hard work to set my characters up.

 

____

shadow tank 2.0 http://swtor.askmrrobot.com/character/65b987d7-c58f-4cde-bcb3-a4087603dcb5

 

Is this still a viable setup. I thought Kitru said stacking abs in augments was a bad idea, let alone the defense is 18. But this looks more less what I'd be looking for, a comprehensive gear guide. Granted I've barely touched into 72s yet so it's likely to be some weeks if not months -_- before I can get a set up like that.

Edited May 29, 2013 by tXHereticXt

[Kitru]

Is this still a viable setup.

 

Yes, it is. You'll find some minor variation between BiS loadouts depending on how said person calculates things like relic contribution (I take into account the average contribution of the proc relics as part of the itemization budget; dipstik doesn't; this causes me to take more Shield and less Defense than he does).

 

I thought Kitru said stacking abs in augments was a bad idea, let alone the defense is 18.

 

I don't think I've ever said that. The only augs a tank should *never* touch are Fortitude augs. Absorb, Defense, and Shield augs should all be game for perfecting your proper itemization distribution: my BiS loadout requires 3 Shield augs and 11 Absorb augs, but some of those Abs augs would need to be swapped out if I replaced some of my Bastion enhs with Bulwarks (they're both 32 Abs/Def).

 

As to the Defense value as it's listed, you can't really trust those numbers as written by AMR. For some reason, it misses some of the talent buffs for some classes and they're inconsistent in missing out buff contributions to boo: in my profile, it lists the contribution of skills/buffs as 5% when it really should be 6% (2% from Shadowsight; 4% from DBSD) whereas, for dipstik's, it lists just 1%. They're *both* missing the contributions, but, for some reason, those contributions are both different and both wrong.

 

It's for this exact reason that we keep telling people to stop looking at listed values and just look at rating. Ratings will never lie to you. Listed values will do so *often*.

[DogEyedBoy]

Nano-Infused Absorb Adrenals will be going from 375 shield and absorb to 500 with tomorrow's patch, does this now make them better than Exotech Absorb Adrenals? Edited May 30, 2013 by DogEyedBoy

[dipstik]

i doubt they will be better than the armor upgrade.

(for shadow)

starting form 563/868/1018 with post armor squish of 0.317674183

1575 armor for 15 seconds every 60*3=180 gives

0.267290848*(15/180)+(165/180)*0.317674183 = 0.3134755717

 

500 shield and 500 armor gives squish of 0.283944551, which is higher than the 0.267290848. so without even finding the weighted term we know its worse.

Edited May 30, 2013 by dipstik

[dipstik]

so i made a simulator for kinetic bulwark and for 563 def and 868 shield i got an average of 5.52% absorb (30 iterations, stdev of 0.5).

 

this gives new BiS d s a of 537 868 1044. very minor change.

 

total d s a mit slope/intercept

2000 378 790 832 0.3310 -0.000039

2100 431 790 879 0.3268 0.40867

2200 484 790 926 0.3227

2300 537 790 973 0.3187 3.9598E-05

2400 590 790 1020 0.3149 3.83053E-05

2500 643 790 1067 0.3112 3.70595E-05

2600 697 790 1113 0.3076 3.58588E-05

Edited May 30, 2013 by dipstik

[_REDRUM_]

How much difference does fixing shield rating at 790 affect the outcome? I can put on shield augment but I am still confused.

[dipstik]

How much difference does fixing shield rating at 790 affect the outcome? I can put on shield augment but I am still confused.

 

for 2200 budget i got

5.584980237 with standard deviation of 0.506233696

 

 

the average should be lower than the higher shield version, but for 30 iterations, it is still within the standard deviation. it doesnt really affect it too much based on 30 iterations.

 

oh, and shadows and juggs should never use shield augments.

Edited May 30, 2013 by dipstik

[Kitru]

oh, and shadows and juggs should never use shield augments.

 

You set Shield rating at a static 790 rather than having it operate as a minimum like it should. When you start getting into stat budgets above 2200, Shield augments start having their place as the optimal distribution without setting shield as a static value begins requiring shield augments to reach. At an itemization budget of 2500, a Shadow requires 3 Shield augs to arrive at the optimal distribution. The same is true for all of the tanks, in fact. The only reason your optimal values don't show this is because you assume Shield sits at a flat value rather than having it float with a given minimum.

[dipstik]

the optimal for juggs is below 790 at 2600. for shadows, optimum shield is above 790 around 2600 (this is new due to kinetic bulwark study). vanguards like lots of shield starting at 1600 or so...

 

due to the crafted ear and implants though, juggs and shadows dont need shield augments... not until next armor set maybe

 

when kell dragon gear is out i need to rework the numbers. i dont think juggs need much shield rating at all, but i will set it to a minimum that increases the total stat pool. in fact, juggs got better mitigation by decreasing stat budget by 14 points overall, to drop shield and gain defense... for the crafted enhancements... but sacraficing 2k hp to mitigate 4 dps wasnt worth it i dont htink...

Edited May 31, 2013 by dipstik

[dipstik]

info for 75 gearing: kell dragon ear/implants better than veracity for shadows (assuming 4500 dps). other tanks sacrafice 216 endurance for 48 mitigation... 48 mitigation is around 5 dps for guardians and vanguards. there are also arkanian crafted ear/implants, specifically redoubt ones that seem like BiS canidates: 181 enduance, 72 shield, 153 defense. this is more mitigation than the kell dragon and seems like BiS for shadows (dividing endurance by 2.5 and summing mitigation gives 297.4 which is greater than 273.8 that kell dragon gives. for juggs and guardians you sacrafice 108 endurance for 114 mitigation, which makes it seem like it is BiS for all tanks...

 

here is a link for shadows: http://swtor.askmrrobot.com/character/f86459f4-c432-4ad4-b89c-c5fb537676fb

 

not much changed with the relics:

 

weighted	relic	base	diff	29.95652174def proc	0.3037	0.2900	0.3085	0.0060	4981.79141abs proc	0.3049	0.2911	0.3097	0.0048	6191.718888shd use	0.3063	0.2895	0.3097	0.0034	8914.613561pvp		0.3054	0.3097	0.0043	6921.637423use def	0.3072	0.2946	0.3097	0.0025	11914.06878

 

 

i have to revise these numbers for the new armor rating. i will update the OP when i have done all the classes

Edited May 31, 2013 by dipstik

[tXHereticXt]

If I ever get there haha. 72s are loads more plausible for me given how quickly you can collect ulties in a week if you're dedicated enough during SMs.

 

What's with the random blank slot, dye? Had me thinking for a sec but then I figured it's just the dye slot.

[_REDRUM_]

You can craft Redoubt Arkanian in 2.2? I thought Arkanian is already purple and can't be improved further.

[dipstik]

i suggest people reverse engineer them to see if they learn variants of the original

[CharlieBrown]

Fun with combat log parsing, or how I learned to stop worrying and love the log.

 

In a (literal) feverish haze, I stumbled through the terror HM 8 man on Monday when I probably should have been resting and eating soup. Needless to say, was not performing anywhere near 100% (30% would be more accurate)

 

I noticed I stupidly had two proc relic equipped when I was trying to quantify the fail. Learn from my mistake!

 

Here's a link to the log.

 

The buffs I was looking for in the parser I wrote to look for these things were "Focused Defense" and "Shield Overcharge". When you look for just those two, the pattern becomes obvious.

 

Here are my procs in the first pull:

 

#	              Fight	Line                                                                               From Start	              In English
4	              1	00:43:55.191 Fiera gains Shield Overcharge.	              00:05.75	              6
27	              1	00:44:15.428 Fiera gains Shield Overcharge.	              00:25.99	              26
43	              1	00:44:35.556 Fiera gains Shield Overcharge.	              00:46.12	              46
62	              1	00:44:57.162 Fiera gains Shield Overcharge.	              01:07.72	              68
83	              1	00:45:33.072 Fiera gains Focused Defense.	              01:43.63	              104
103	              1	00:45:54.460 Fiera gains Shield Overcharge.	              02:05.02	              125
121	              1	00:46:14.652 Fiera gains Shield Overcharge.	              02:25.21	              145
140	              1	00:46:34.840 Fiera gains Focused Defense.	              02:45.40	              165
159	              1	00:47:06.673 Fiera gains Shield Overcharge.	              03:17.23	              197
180	              1	00:47:28.240 Fiera gains Focused Defense.	              03:38.80	              219
198	              1	00:47:48.484 Fiera gains Focused Defense.	              03:59.05	              239
212	              1	00:48:09.154 Fiera gains Focused Defense.	              04:19.72	              260
226	              1	00:48:29.396 Fiera gains Shield Overcharge.	              04:39.96	              280
232	              1	00:48:57.008 Fiera gains Shield Overcharge.	              05:07.57	              308

 

Like most of us, I have a bag full of relics so it's not a big deal to switch it around based on the fight, just do yourselves a favor and double check if you're neck deep in NSAIDs.

[JowyBlight]

So some one please boil down for me in black and white terms. I am a full 72 Vanguard Tank. What should my numbers be for the best possible mitigation?

 

My stats break down like this ATM when fully buffed:

 

 

DR is 50.7%

DC is 23.33% (1031)

SC is 38.79% (869)

SA is 34.68% (425)

 

How close is that to the ideal stats.

 

(I miss the old days, when it was just 600/600 shield/absorb van.)

Edited June 11, 2013 by JowyBlight

[Kitru]

DC is 23.33% (1031)

SC is 38.79% (869)

SA is 34.68% (425)

 

How close is that to the ideal stats.

 

Your itemization budget is ~2300, so your distribution should be 800/925/575. As such, you're about 150 light on Absorb, 50 light on Shield, and 200 heavy on Defense.

 

Your numbers are actually *very* close to optimal using KBN's assumptions (major difference being factoring in Riot Gas over time vs. treating it as a CD; KBN does, dipstik doesn't). For those, your optimal distro would be ~980/840/480, wherein you're *still* a little heavy on Defense, but you just need to transfer a bit of it (~50, so a single mod or 2 enhs) to Absorb.

 

Also, listing DR is largely redundant. DR is a largely static value since there aren't any customizable stats that affect it: armor rating is a static benefit attached to the armoring, regardless of what kind of armoring. Bringing it up in a discussion of how you should allot your stats means absolutely nothing because armor doesn't conflict with anything else.

[Ancaglon]

Also, listing DR is largely redundant. DR is a largely static value since there aren't any customizable stats that affect it: armor rating is a static benefit attached to the armoring, regardless of what kind of armoring. Bringing it up in a discussion of how you should allot your stats means absolutely nothing because armor doesn't conflict with anything else.

Going to have to quibble with this one, because "full 72 tank" doesn't autmoatlically equate "full Underworld 4-piece tank", and as I think you've already said, 4-piece bonus is worth a lot of armor/primary stat/endurance -- especially since now you can't do 2+2's between old and new. Edited June 11, 2013 by Ancaglon

[Kitru]

Going to have to quibble with this one, because "full 72 tank" doesn't autmoatlically equate "full Underworld 4-piece tank", and as I think you've already said, 4-piece bonus is worth a lot of armor/primary stat/endurance -- especially since now you can't do 2+2's between old and new.

 

And how does any of that relate to itemization budgets (beyond "as a VG, I don't have my 4 pc set bonus so I'm going to value Defense slightly less until I get it"; of course, none of the distributions calcs *don't* assume you don't have the set bonuses so, once again, null point: if you don't have 'em, get 'em)? The point is that listing DR when asking about how to properly distribute your tanking itemization is utterly pointless. DR has nothing to do with your stat itemization since it's only affected by one thing that doesn't affect *anything* else: armor rating. It's not like you can trade armor rating for Defense or Shield. If you *could*, it would be relevant to bring up but, until that happens, it's a null point when discussing itemization budgets and optimal distributions.

[JowyBlight]

Your itemization budget is ~2300, so your distribution should be 800/925/575. As such, you're about 150 light on Absorb, 50 light on Shield, and 200 heavy on Defense.

 

Your numbers are actually *very* close to optimal using KBN's assumptions (major difference being factoring in Riot Gas over time vs. treating it as a CD; KBN does, dipstik doesn't). For those, your optimal distro would be ~980/840/480, wherein you're *still* a little heavy on Defense, but you just need to transfer a bit of it (~50, so a single mod or 2 enhs) to Absorb..

 

Since i do pretty much use Riot Gas on off cooldown in boss fight, i suppose i'll KBN numbers for now. Maybe i'll try both out eventually and see which one works best for my playing style.

 

(also @Ancaglon when i said full 72, i didn't just mean the new set bonus, no i mean all my mods, ench, and armors are 72 :-)

[odawgg]

Thank you Dipstick and KBN I've read both your OPs but only had time to read through the discussion on this thread. Thanks also to all the others that contributed this has been a fun read! I have an assassin tank and learned a lot from this. I was getting conflicting advice between going balanced mitigation (def/absorb that is) vs limiting some of the spikiness that both of you suggest based on your ideal stat distribution, this helps clear that up.

 

I do however, currently have lettered mods, how important would you rate switching over to non-lettered mods in terms of survivability? 1-10, 10 being crucial. I have some that I can throw on but it would require some augment changes which will be a lil pain, if it's not that crucial, I may wait till Kell Dragon gear to optimize that. Thanks again!

[thasensei]

Thank you Dipstick and KBN I've read both your OPs but only had time to read through the discussion on this thread. Thanks also to all the others that contributed this has been a fun read! I have an assassin tank and learned a lot from this. I was getting conflicting advice between going balanced mitigation (def/absorb that is) vs limiting some of the spikiness that both of you suggest based on your ideal stat distribution, this helps clear that up.

 

I do however, currently have lettered mods, how important would you rate switching over to non-lettered mods in terms of survivability? 1-10, 10 being crucial. I have some that I can throw on but it would require some augment changes which will be a lil pain, if it's not that crucial, I may wait till Kell Dragon gear to optimize that. Thanks again!

 

I'm sure you're looking for a response from the numbers guys, but in the meantime I figured I'd drop my two cents.

 

Assuming you're using the B-letter mods (rather than the A-letter mods), I would rate the importance as a 2. If you're using the A-letter mods, I'd rate the importance as a 6 (those A-mods are painful).

 

The difference between unlettered and B-mods is roughly 1.5% absorb VS 2100 health and 15 HPS.

 

To put things in perspective using very basic numbers, lets assume 6000 damage per second, or 54,000 damage over 9 seconds (6 GCDs).

 

Unlettered mitigates about 383 extra damage over that 9 seconds.

B-lettered heals about 135 extra over that 9 seconds (and of course provides a larger health buffer)

 

In those 9 seconds, any single healing action will easily overcome that 250 extra damage that needs to be healed. Given that most (probably all, but I don't know for sure) healing actions have a difference of a couple hundred points between the min and max that they can do, I'd say this is really a non-factor in practice.