2.2 Tank Builds

[Kitru]

hope that helps

 

None of that would explain the differences you guys have for Shadows (2500 = 643 790 1067 v. 539 912 1049). The minimum Shield assumption you make is bypassed by the fact that the 912 is above said static minimum. There isn't any variable other than different attack ratios that would account for the difference since Shadows don't have a def/shield/abs mechanism that is accounted for differently in your systems.

[dipstik]

my absorb value is a bit different. i use .0552 for bulwark based on my simulations. he uses 0.036.

 

912/2500=0.3648, which is his minimum.

Edited July 15, 2013 by dipstik

[Kitru]

he uses 0.036.

 

912/2500=0.3648, which is his minimum.

 

That is *ludicrously* low for Kinetic Bulwark. Wow.

[wetslampigduex]

So what I'm getting from the last part of this thread is... Dip's numbers are slightly more accurate because he factors in a third type of dmg?

[Kitru]

So what I'm getting from the last part of this thread is... Dip's numbers are slightly more accurate because he factors in a third type of dmg?

 

The third damage type wouldn't affect the mitigation numbers because F/T I/E damage isn't affected by Def, Shield, or Abs.

 

From what I can tell, the differences, for Shadows, are caused by a different assumption of Kinetic Bulwark's contribution (KBN's is *really* low imo; even dipstik's is a bit low as I see it) and, for Vanguards, KBN averages out the contribution of Riot Gas whereas dipstik ignores it (KBN sees it as a static benefit used on CD; dipstik sees it as a tank CD in and of itself). The final difference from then is how they determine the existence of Shield rating (KBN assumes a percentage of total budget whereas dipstik assigns a given minimum equal to the Shield you would get from a full allotment of Bastion/Bulwark enhs for the given tier).

[dipstik]

what do u assume for bulwark?

 

my simulator is primitive. assuming the swing timer is 0.87 seconds, and you have 20 seconds before you have to refresh kinetic ward, you get 23 chances to proc bulwark. so i generate 23 random numbers between 0 and 1, and if the number is greater than (1-d)*s (which i have set to .2937 and 0.5643) then the trial succeeds (i tried using binomial at first but things like 23 choose 8,each with a different effective absorb value gets hairy). so if the i'th trial succeeds out of 23, then it gets to stay up for (23-i)/23 fraction of the time. then i set it so if 8 trials before i succeeded, then they were not allowed to succeed. then i take the sum of the (23-i)/23 to get the value for that iteration, and i did 30 iterations and took the average.

 

so if you get a proc after 0.87 seconds, it would contribute (23-.87)/23=0.96 absorb, while if the 8th proc happens on the 23, you get no contribution. i could have set it so the 1st swing happens right at the beginning of the kw buff, and that would gives a proc on the 23'rd chance a 1/23=.04 absorb contribution, but you really only get 22.9 swings and not 23 using .87 in 20, so i went with 0 contribution for 23.

[Kitru]

what do u assume for bulwark?

 

I assume a flat 6% for napkin math but 5.6% for more complicated stuff (I honestly don't feel the need to apply much specificity to the value since it a slides based upon the encounter in question; as long as the value is *close* to the real average value, it's good enough since you're going to naturally have some minor variance in reality).

 

My math operates under the assumption that all 15 stacks are going to be consumed at or before the 20 second duration comes up (i.e. the "intended" manner). For fights with low incoming attacks/sec, KB is naturally going to deviate from the given value, but those fights naturally diminish the value of KW and KB overall and aren't really ones that I ever particularly worry about. In ops, you're pretty much guaranteed to blow through your 15 in under 20.

 

As such, with 15 attacks, I look for the sum Absorb contribution from KB for each KW stack consumed (0 + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8*7) and average it out over the 15 attacks, for an average contribution of 5.6%. It's an admittedly simple method but, in my experience, I never really have my KW stacks timing out rather than running out while I'm tanking in an ops (FPs, on the other hand, most of the time, on bosses, I'll only have 3-4 stacks of KB when I have to refresh KW).

[wetslampigduex]

The third damage type wouldn't affect the mitigation numbers because F/T I/E damage isn't affected by Def, Shield, or Abs.

 

From what I can tell, the differences, for Shadows, are caused by a different assumption of Kinetic Bulwark's contribution (KBN's is *really* low imo; even dipstik's is a bit low as I see it) and, for Vanguards, KBN averages out the contribution of Riot Gas whereas dipstik ignores it (KBN sees it as a static benefit used on CD; dipstik sees it as a tank CD in and of itself). The final difference from then is how they determine the existence of Shield rating (KBN assumes a percentage of total budget whereas dipstik assigns a given minimum equal to the Shield you would get from a full allotment of Bastion/Bulwark enhs for the given tier).

 

Ok so that clears up Sin's and Pt's for me, I think.(sins different values for the amount of time KB/DW are up, VG/PT because KB uses RG/OS on CD(which is gen what I do) and Dip uses it as an O sh*t CD) So for what reason are the values so different for a JG/SW....

 

Dips

2000 777 860 363 0.3339 -3.12971E-05

2100 830 860 410 0.3305 0.396258281

2200 883 860 457 0.3273

2300 936 860 504 0.3241 3.176E-05

2400 989 860 551 0.3210 3.07639E-05

2500 1042 860 598 0.3180 2.98021E-05

2600 1094 860 646 0.3151 2.88735E-05

 

KB

{2000,{defense->977,shield->729,absorb->294}}

{2100,{defense->990,shield->766,absorb->344}}

{2200,{defense->1004,shield->802,absorb->393}}

{2300,{defense->1019,shield->839,absorb->442}}

{2400,{defense->1034,shield->875,absorb->491}}

{2500,{defense->1050,shield->912,absorb->539}}

{2600,{defense->1066,shield->948,absorb->586}}

 

I apologize for all the questions.. I took 2/3 months off and everything has changed im just trying to get a solid understanding on this... I play all 3 tanks(kinda shelved my Sin after reading the majority of you guys theorycrafting)

[dipstik]

because the shield value in kbn numbers are assumed to be a min of 36% of the budget. (divide his shield value by the total and you will get 36.x% for all of them)

 

i use the shield value of that tier of gear (or higher if optimal distribution is higher, like for VG) and min max around the shield which is as close to optimal as can be... namely 720, 790, 860 or the weird numbers for underwolrd and arkanian that use veracity ear/implant.

Edited July 17, 2013 by dipstik

[MillionsKNives]

Do we know yet if the shield proc and the defense proc relics can be used together? My jug has the Arkanian defense proc relic and I'm not sure which other relic I should replace my Battlemaster shield relic with. I was going to take the clicky shield/absorb, but I remembered reading that the shield proc is better, assuming that the two procs work.

[dipstik]

there has been so much back and forth about relics i dont want to say one way or the other.

 

im pretty sure you cant have two proc relics of the same tier (underworld-underworld). i have heard some say you can use the pvp proc relics with the pve proc, but i would test it out. u get 2 hours to resell them.

[MillionsKNives]

I think I will try that, shouldn't be too hard I think if I go let some mobs beat on me, to check for overlap.

 

I also figured out part of my confusion, I had remembered seeing the following patch note, but had forgotten that it was still only on the PTS:

Arkanian, Underworld and Kell Dragon Proc Relics will no longer prevent other Arkanian, Underworld and Kell Dragon Procs of a different type from activating. For example, characters can now have both a Power proc relic and a damage proc relic equipped and expect them both to activate correctly.

 

Hopefully they both work as expected after 2.3, if the patch note is to be believed.

[dipstik]

i still kind of doubt they will let the def and abs proc work at same time. if they did, those would be bis for jugg and vg, and maybe even shadows depending on nim incoming dps numbers.

[Yngow]

i still kind of doubt they will let the def and abs proc work at same time. if they did, those would be bis for jugg and vg, and maybe even shadows depending on nim incoming dps numbers.

 

I also doubt that and if that true we can also expect a fix to the pvp and pve relics stacking since they already did that with dps relics (apparently they were incapable of even testing the tank relics and seeing they had the same glitch).

 

And if they allow us to use 2 proc relics it would be really stupid since the clicky relics would have no reason at all to be in game

[Kitru]

And if they allow us to use 2 proc relics it would be really stupid since the clicky relics would have no reason at all to be in game

 

On fights with long swap tank cycles (i.e. on a target where Shield/Abs is useful for 30 seconds and then either no boss or on a boss where higher Shield/Abs is useless or unneeded for 30 seconds), the use relics are actually better than the proc relics. The math is pretty simple because you're getting functional 50% uptime (30 seconds tanking with buff up, 30 seconds tanking without buff, 60 seconds in a state where the proc relic isn't going to matter) where the proc relic is still just getting ~25% uptime. Of course, the only one that's really like that are Writhing Horror and TfB (you could *maybe* bring some of the S&V bosses in on that, but no one really tanks swaps those because it's not really necessary).

[Septarra]

Using Old TFB Data (not NiM) i get 63% weapon/kinetic, 29% tech/kinetic, and 8% tech/internal

 

this gives the following BiS estimates:

 

Shadow:

Defense 200

Shield 1056

Absorb 1319

 

Jugg:

Defense 947

Shield 860

Absorb 887

 

Powertech:

Defense 371

Shield 1356

Absorb 966

 

Am I reading this right? So going into say NiM/HM TfB, as a PT/Vanguard you want to strip defense down to a minimum?

[Yngow]

Pretty much for every class due to the dmg profile on the bosses, but being a pt/vang def is the least valuable defensive stat anyway

[Kitru]

Pretty much for every class due to the dmg profile on the bosses, but being a pt/vang def is the least valuable defensive stat anyway

 

Actually, for the HM incoming damage ratios, Defense was the most valuable tanking stat for VGs by a large margin. Shadows were the ones that didn't care all that much about Defense.

[Septarra]

Hmmmm... So another mixed bag of answers. Hearing from quite a few sources that Defense isn't worth stacking again, stick to Shield/Absorb. Never sure what to believe though.

[dipstik]

the numbers are not in for either NiM, so unless they have the data and have done the calcs, no one knows how to gear for nim.

 

assuming HM damage weights carry into nim, you want more defense in SnV than you do in TFB.

 

there are arguments that you should gear for the tank checks... problem is, i think each of the tank types have different checks. for a shdown, titan might be a tank check, whereas for a jugg, the terror might be the tank check... or even operator.

 

since 2.0 we have been using SnV hard mode damage weights to calculate best in slot. this seems reasonable since thrasher had the highest damage of all the bosses in 55 hard modes. if we find there is some new super high damage boss trhat beats thrasher, then we might want to gear for that operation (using damage weights between bosses etc). some people even go so far as to have different gear sets for different bosses.

 

long story short: defense seems much better for snv than tfb.

Edited August 1, 2013 by dipstik

[Yngow]

You will definitely get mixed answers when asking this kind of stuff because it involves a lot of personal play style, what this thread gives you is the build which would provide the most mitigation overall, or would make you take less dmg, it doesn't necessarily means it is the *better* choice. Also those number are all for S&V HM and on average, not for a specific encounter.

Every boss will have a slightly different dmg profile which would favor one type of gear or another, and also some people like to gear more towards a dmg profile rather than for less dmg taken, which means even though defense heavy build would provide you less dmg taken overall some people go with heavy shield/absorb to be as less spiky as possible, even if they take more dmg by doing that.

It is very inviable to have a different gear set for each boss, but I really like to have a selection of different relics and swap accordingly to each boss fight (in my case it's shield amplification and fortunate redoubt).

 

I also strongly recommend that you talk to your healers, their inputs are really good for gearing your tank, some healers have no problem with spikiness and prefer if you take less dmg overall, some healers are the opposite and prefer to keep a certain amount of healing on you all the time instead having to give heavy bursts from time to time

Edited August 2, 2013 by Yngow

[Berjiz]

It is very inviable to have a different gear set for each boss, but I really like to have a selection of different relics and swap accordingly to each boss fight (in my case it's shield amplification and fortunate redoubt).

I think a collection of all bosses with their dmg types and attacks togheter with tips and tricks would be very useful. Is there any collection like that somewhere?

[Yngow]

KN posted all the dmg profiles for HM S&V and pretty sure he did the same with HM TFB, the same data is also on dipstik's thread about BIS equipment on 2.2: http://www.swtor.com/community/showthread.php?t=645948

 

As for what to do with the data, I recommend one of the two: You can use http://www.tor-tank.com which is a tool designed to follow KN's numbers, but on advanced options you can change the dmg ratios and get an optimization for a specific situation, also you can use redclaw's spreadsheet for all tanks: http://www.swtor.com/community/showthread.php?t=660767

Never really tried this one, but it seems to be very good for that purpose.

[Kitru]

Never really tried this one, but it seems to be very good for that purpose.

 

It's still got a fair share of screwed up numbers. The added up values on Guardians are kind of correct, but the contributions are screwed up (it's got 5% from Soresu and 4% from Guardian Slash), and it's still missing the 5% acc debuff. The values for Shadows are inflated by 5% on Shield (it's saying 20% from stance instead of 15%) and it doesn't include the acc debuff either. It's good for VGs though.

[dipstik]

updated OP for double proc relics. on use shield is good for tank swap/ bursty fights.

 

and i want to plug tor-tank.com. advanced options for damage weights and shield ratio (a static number or am check bos to not change shield would be ncie).

Edited August 6, 2013 by dipstik