Please upgrade your browser for the best possible experience.

Chrome Firefox Internet Explorer
×

Tweaking End Game Tanking Gear - Full Campaign Modding

STAR WARS: The Old Republic > English > Classes
Tweaking End Game Tanking Gear - Full Campaign Modding

rob_y's Avatar


rob_y
10.12.2012 , 10:07 AM | #11
Quote: Originally Posted by KeyboardNinja View Post
In augmented Campaign/Dread Guard gear, my defense chance is 28.85% with stim. According to my (extensive) calculations, this is within 0.0001% of ideal, modulo the fact that shield chance is inflated due to poor mod itemization. The higher defense chance comes from the fact that I'm using the proc absorb relic, which slightly deflates the value of the static absorb stat.

So, I agree that 30% defense chance is too much, but it's not far wrong.
My math says 29.21% is ideal, not 28.85%. Please back up your assertion with some spreadsheets, a readable abstract, and some solid conclusions.
Yan (55 - Scoundrel) Silch (50 - Guardian) Hoeven (50 - Commando)
Thith'yan (50 - Sorceror) D'lamia (50 - Marauder) Brazo (55 - Powertech)
RIVAL
The Fatman -> Prophecy of the Five

adagioing's Avatar


adagioing
10.12.2012 , 11:17 AM | #12
Stacking endurance is only good if your self-heals make up for the lost mitigation, i.e. once boss dps is high enough, mitigation is better (assuming you can avoid the damage).

My personal strategy has been to take 2 endurance heavier pieces with defense, but take two war hero defense relics, with all other pieces mitigation heavy. I consider this a decent trade-off: I'm stacking mitigation and defense is my weakest stat at this point, however as mentioned there is *some* unavoidable damage so a little extra health isn't the worst idea.

Buffed and stimmed:
25.2k hp
30.95% defense
66.2% shield (dark ward)
61.4% absorb
These are optimized for my budget and shield rating. With wither/discharge on target, this gives 76.6% average damage reduction (for avoidable damage), which surprisingly is almost as good as the other tank classes.

Also my calculations show that the campaign proc absorb relic is weaker than the war hero defense relics. That's purely statistical, not even counting that you can't choose when the proc happens.

grallmate's Avatar


grallmate
10.12.2012 , 06:50 PM | #13
Quote: Originally Posted by adagioing View Post
Also my calculations show that the campaign proc absorb relic is weaker than the war hero defense relics. That's purely statistical, not even counting that you can't choose when the proc happens.
My calculations (and logic) agree that the proc relic is a sub-optimal option. It provides a very nice boost to mean mitigation but only improves the mitigation of attacks that are already being mitigated. It doesn't increase the chance to mitigate damage. A further argument against it, is that it is proc driven and not controllable mitigation.

What it comes down to is that taking 10% less damage on shielded attacks looks really good on paper. But the shielded attacks aren't the ones that kill you. I'd much rather increase my change to shield or defend an attack. Or use a controllable clicky relic to boost my mitigation when needed.

Once you mean mitigation is enough to reduce incoming damage to a healable level there are a couple of schools of thought:
- Reduce it more to increase TTL under average RNG
- Increase HP to increase TTL under bad RNG
- Increase utility
- Increase threat generation to limit threat loss due to bad RNG
I take a bit out of all of them. On my gear I reach about 24-25k HP and then stack mitigation, fairly balanced between defense/shield/absorb, I use one WH relic and one clicky relic which I normally pair with an armor adrenal for an extra (pseudo) CD.

My main is a Guardian tank, so the absorb proc relic is worth less to me than a Shadow or VG, but I feel even for them it is sub-optimal.

Marb's Avatar


Marb
10.13.2012 , 10:32 PM | #14
Quote: Originally Posted by grallmate View Post
My calculations (and logic) agree that the proc relic is a sub-optimal option. It provides a very nice boost to mean mitigation but only improves the mitigation of attacks that are already being mitigated. It doesn't increase the chance to mitigate damage. A further argument against it, is that it is proc driven and not controllable mitigation.

What it comes down to is that taking 10% less damage on shielded attacks looks really good on paper. But the shielded attacks aren't the ones that kill you. I'd much rather increase my change to shield or defend an attack. Or use a controllable clicky relic to boost my mitigation when needed.
I can understand using a click relic, but I don't get your reasoning for not using the absorb proc relic at all. It makes the biggest contribution to mean mitigation out of ALL the relics.

You say that the proc does not increase your chance to mitigate damage, but in a way it actually does. Your shield is only ever as good as the damage it absorbs. Shield and Absorb are intimately linked to each other. If your thinking about increasing your chance to mitigate damage, you really need to think of shield and absorb as the same stat when comparing to defence. Defence is the chance to ignore an attack, shield is the chance to absorb an attack.

Now with the way most have their stats set up, 113 shield/def is not going to provide that much real mitigation because of diminishing returns, especially shield rating. Using 2 passive war hero relics is a bit silly. The proc flat out gives you a whopping 405 absorb rating, and it does actually have a huge amount of uptime.

Most tank relic set-ups should look like this:

Absorb Proc
Passive War Hero or Defence Activation. <This is what you change depending on the encounter.
Harbinger

KeyboardNinja's Avatar


KeyboardNinja
10.13.2012 , 11:24 PM | #15
Quote: Originally Posted by rob_y View Post
My math says 29.21% is ideal, not 28.85%. Please back up your assertion with some spreadsheets, a readable abstract, and some solid conclusions.
"A readable abstract"?! You're getting a spreadsheet, and you're getting an explanation.

Note that I'm not using the War Hero defense relic. Instead, I'm using the proc heal as my second relic. If I were using the defense relic, it would give me a larger mitigation budget to work with, which would push up the ideal defense chance a bit closer to your numbers.

Also note that I've gone with the 26/27B mods in a couple of places. This is because the stat budget is higher on these mods (albeit skewed toward endurance). The specific ratio of endurance to <mitigation> is 21:9, which is high enough that the extra endurance actually works out to *increase* survivability through the self-heal, despite the loss of mitigation. See my spreadsheet for how the self-heal value is calculated in terms of mitigation points.

Also worth noting that my numbers were specifically referencing a shadow tank. Ideal defense chance is much higher for a guardian (in the 33% range) and much lower for a vanguard (around 16-17%).

Quote: Originally Posted by grallmate View Post
My calculations (and logic) agree that the proc relic is a sub-optimal option. It provides a very nice boost to mean mitigation but only improves the mitigation of attacks that are already being mitigated. It doesn't increase the chance to mitigate damage. A further argument against it, is that it is proc driven and not controllable mitigation.
From a purely statistical standpoint, the war hero defense relic *does* provide slightly higher weighted survivability benefits than the campaign proc absorb...by 0.0013%. Not enough that I'll lose sleep over it. It would be slightly higher if I readjusted my stat budget to compensate for the fact that I have less statistical absorb and more defense. Note that at lower stat budgets (non-augmented Rakata), the proc absorb relic is statistically much *more* valuable than the War Hero defense relic, which is interesting. I suspect this distinction is due to the fairly generous DR curve on defense rating when compared to the curve on absorb.

A bit of quick algebra indicates that there are real values of defense, shield and absorb where the proc absorb relic is superior. Computing whether or not these n-tuples are ideal loci is a much more complicated problem, and one that would probably cause Wolfram|Alpha to just time out.

Random side note: on high damage fights, the War Hero defense relic edges out the proc heal relic in terms of mean survivability contribution, but not by a lot. On Nightmare Pilgrim, the War Hero defense relic provides a net 28 DtPS benefit (with my current itemization), while the proc heal relic provides 20 HPS. On HM Kephess, it provides 21 DtPS, which is almost equivalent to the proc heal. On fights like HM Toth & Zorn, the damage is sufficiently low that the proc heal relic becomes dramatically better in terms of survivability contribution.
Computer Programmer. Theory Crafter. Streaming Dilettante on The Ebon Hawk.
Tam (shadow tank) Tov-ren (commando healer) Aveo (retired sentinel) Nimri (ruffian scoundrel)
Averith (marksman sniper) Alish (lightning sorcerer) Aresham (vengeance jugg) Effek (pyro pt)

grallmate's Avatar


grallmate
10.14.2012 , 03:18 PM | #16
Quote: Originally Posted by Marb View Post
I can understand using a click relic, but I don't get your reasoning for not using the absorb proc relic at all. It makes the biggest contribution to mean mitigation out of ALL the relics.

You say that the proc does not increase your chance to mitigate damage, but in a way it actually does. Your shield is only ever as good as the damage it absorbs. Shield and Absorb are intimately linked to each other. If your thinking about increasing your chance to mitigate damage, you really need to think of shield and absorb as the same stat when comparing to defence. Defence is the chance to ignore an attack, shield is the chance to absorb an attack.

Now with the way most have their stats set up, 113 shield/def is not going to provide that much real mitigation because of diminishing returns, especially shield rating. Using 2 passive war hero relics is a bit silly. The proc flat out gives you a whopping 405 absorb rating, and it does actually have a huge amount of uptime.

Most tank relic set-ups should look like this:

Absorb Proc
Passive War Hero or Defence Activation. <This is what you change depending on the encounter.
Again I disagree with your blanket statement about relic set up.

The absorb proc relic is awesome for the 6 seconds it is active. For the 14 seconds it isn't, it does almost nothing.

Pros:
- Mean mitigation numbers are very good
- "set and forget"
Cons:
- Uptime is only 30%
- Uptime is not controllable
- Those hits would normally be mitigated, just to a lesser extent

Sure, in the 6 seconds it is active you aren't likely to die. In the 14 seconds it isn't you might as well not have anything equipped (passive power and endurance boost aside). Now think about 6/20 vs 30/120 seconds in most fights. Is that 6 seconds of higher mitigation going to save your life? Is that 30 seconds of higher mitigation going to save your life?

Now think about Zorn/Toth or Firebrand/Stormcaller. 30 seconds of controllable uptime is MUCH better than 6 seconds of uncontrollable uptime. If you use tank swaps you can have your relic active for almost an entire Toth/Firebrand timeslice, every other tank swap. This would push the value of those relics up even higher relative to the proc one, which has it's usefulness halved.

Now we move onto diminishing returns and this bit varies for everyone due to different stat budgets and distributions. 113 Defence is not going to push you into serious DR unless you already stack defence heavily. Add to this that unless you've been ignoring absorb, 415 probably will likely push you into high DR territory. This is compunded by the DR curve on Absorb being very steep (worth a lot until it suddenly becomes worth nearly nothing) compared to Defence being very shallow (not worth a lot to start with and progressively loses value). On top of that, you need to consider that for all 3 tanks these have different weights (my Guardian having less value from the proc relic than the other 2 due to lower shield chance).

Now at no point did I recommend having 2 passive relics. I agree this is not an ideal set up (but better than 2 clicky relics). The reason the dual passive set up is spouted all over the place is for DPS whose BiS set up (for sustained DPS) is usually 2 passive power relics.

I look at it this way: tanks must adapt as the fight changes and benefit heavily from increased utility (ie an extra CD). Since using 1 clicky relic triggers a CD on the other, dual clickies is not an optimal combination, so lets trade one out. Now we look at the other options: proc absorb relic, passive WH relics or (for Shadows) proc heal relic. Depending on class and other stats, grab what is most useful. My set up is a WH Defence relic and a Campaign Shield/Absorb clicky relic. I use the Shield/Absorb clicky because it is less spiky than the clicky Defence one despite it giving less mean mitigation.

I like to compare the proc Absorb relic to the Sorc/Sage AoE heal in a WZ. It produces some awesome numbers, but it doesn't really keep anyone alive.

Marb's Avatar


Marb
10.15.2012 , 04:56 AM | #17
I can see the reasoning behind using a click relic instead of the proc on SOME fights. But again, if you want the best overall contribution to your mitigation, absorb proc relic is the way to go.

And I retract what I said about the relic setup, you are correct that its not always applicable for all classes in that way.

I still disagree with opting for not using the proc at all however. If the fight does not call for the click relic, you should be using the proc.
Harbinger

grallmate's Avatar


grallmate
10.16.2012 , 08:23 PM | #18
Quote: Originally Posted by Marb View Post
I can see the reasoning behind using a click relic instead of the proc on SOME fights. But again, if you want the best overall contribution to your mitigation, absorb proc relic is the way to go.

And I retract what I said about the relic setup, you are correct that its not always applicable for all classes in that way.

I still disagree with opting for not using the proc at all however. If the fight does not call for the click relic, you should be using the proc.
As I stated above, for Guardian's the proc relic is far from ideal due to lower shield chance. As KeyboardNinja stated above the WH Defence relic is (negligibly) better than the Absorb proc relic for his Shadow tank. These will always vary for different characters with different gear set ups and for some it may provide a higher contribution to mean mitigation.

Yes, the proc relic can greatly increase mean mitigation. I still feel that it is far from ideal for the reasons I highlighted above.