Tanker Primary "Tough Test"

[ninja surprise]

2 hours ago, Gulbasaur said:

Really interesting about Bio, though - I knew it was a bit OP, but that's really OP.

 

Not sure I agree with the "really" part of that... Bio is only really good against the things it's really good against, but then super weak where it has no protection or when slows or debuffs come in.

 

I don't know how much free time @Galaxy Brainhas, but it'd be super interesting to re-run the damage distribution to mimic some of the major enemy groups: Devouring Earth, Circle of Thorns, Arachnos, Crey, or Malta.

For instance, people love to cry about Malta Sappers, but big spawns will have lots of Gunslinger Lieut/Bosses with their freeze-ray guns stacking Stun and Hold magnitude.  On Crimson's arc last night my Rad/Fire Tank got held a couple times.

 

1 hour ago, Tsuko said:

Radiation less tanky than ice armor ?

I believe by default Defense outlasts Resist because it gets hit less.  Resist armors you have to max out and still be active with healing whereas Defense just stands there clicking glowies while enemy attacks whiff all around.

Plus, Icicles defeats the minions vs Rad armor doesn't have a damage aura so the entire spawn keeps attacking till the tank's dead.

 

[Tsuko]

26 minutes ago, Werner said:

And +6% defense and +8% to +18% scaling resistance from the defense and resistance uniques, which I tend to slot even on lowbie builds. And while few people would slot all of these, you could go nuts with uniques to at least get some semblance of IO benefits without the complexity of an IO build, still keeping the slotting completely standard across the primaries.

 

Shield Wall
Reactive Defenses
Numina
Panacea
Preventive Medicine
Regenerative Tissue
Unbreakable Guard
Steadfast Protection
Gladiator's Armor
Impervious Skin
Power Transfer

 

Call that your mid level test, maybe, instead of what I was saying I use as a baseline. And then there's the Tanker ATO, but that's harder to fit into a test like this, so I guess set that aside. I think the results would be you'd need to bump the level of the test up to +2 to continue to see a meaningful spread of results, and that you'd see a lot of shifting around in those results, though I won't even try to predict the order.

and the tanker ATO ^^

 

 

[Werner]

11 minutes ago, Tsuko said:

and the tanker ATO ^^

I'd love to include a double stack of it, but since it requires attacking, I don't see how to include it within the confines of the way the test is set up. You could run the defense amplifier. The extra 7.5% resistance is similar to a single stack of the ATO, and most people are going to get more defense from bonuses, so another 5% defense isn't unreasonable.

[Hyperstrike]

22 hours ago, Gobbledegook said:

In my experience of Tankers i would put Invulnerability or Shield defence as number 2-3. A SD Tanker was one of the better primaries for doing a MoITF buffed mobs no insps/deaths.

 

I certainly wouldn't put Rad Armour so low down.

 

I would put bio offensive below Rad. It is weak to none S/L damage and defences can be broken easily.

 

I think there are weakness to all sets whether defence debuffs, end drains etc that need to be counted for also.

Remember these are bare SO builds.  With no optimization whatsoever.
Also, because they're not attacking with their secondary, they get no power synergies.

All by itself, bare Rad armor is pretty squishy.

[Sovera]

32 minutes ago, Werner said:

I'd love to include a double stack of it, but since it requires attacking, I don't see how to include it within the confines of the way the test is set up. You could run the defense amplifier. The extra 7.5% resistance is similar to a single stack of the ATO, and most people are going to get more defense from bonuses, so another 5% defense isn't unreasonable.

Aura? Kinda works.

[Tsuko]

23 minutes ago, Hyperstrike said:

Remember these are bare SO builds.  With no optimization whatsoever.
Also, because they're not attacking with their secondary, they get no power synergies.

All by itself, bare Rad armor is pretty squishy.

That's my point about this tests.

Staying inactive in the middle of mobs is not a good indication of a tanker efficiency.

 

For instance, Radiation Armor needs Radiation Therapy / particle shielding with good recharge change ALL. It's even stronger with Dark Armor huge heal.

Don't get me wrong, i m very impress by @Galaxy Brain tests : tons of hard work.

But there about tankers, mitigation is not just about numbers : movement, position, synergy of armor / attack and scaling of gear and effect of power creep in the game hit directly Tankers.

 

Any tanker just using the P2W backpack to fly over 17 enemys will stay alive for so much more time.

Dam it's there i hate not being english to be more accurate 😕

 

 

Edited November 23, 2020 by Tsuko

[Saikochoro]

2 hours ago, ninja surprise said:

Not sure I agree with the "really" part of that... Bio is only really good against the things it's really good against, but then super weak where it has no protection or when slows or debuffs come in.

 

I don't know how much free time @Galaxy Brainhas, but it'd be super interesting to re-run the damage distribution to mimic some of the major enemy groups: Devouring Earth, Circle of Thorns, Arachnos, Crey, or Malta.

For instance, people love to cry about Malta Sappers, but big spawns will have lots of Gunslinger Lieut/Bosses with their freeze-ray guns stacking Stun and Hold magnitude.  On Crimson's arc last night my Rad/Fire Tank got held a couple times.

 

I believe by default Defense outlasts Resist because it gets hit less.  Resist armors you have to max out and still be active with healing whereas Defense just stands there clicking glowies while enemy attacks whiff all around.

Plus, Icicles defeats the minions vs Rad armor doesn't have a damage aura so the entire spawn keeps attacking till the tank's dead.

 

I agree that it is hard to draw the “really op” conclusion on bio armor. I am not surprised at all that it performs well at a base level. The layered mitigation really helps cover a lot of holes that other sets don’t have covered at their base level. 
 

However, fully built out and optimized there are several sets that are much tougher than bio armor in my opinion. Especially, if you are trying to maintain its offensive focus (ie rech bonuses and procs). This is even more true on non-tankers that don’t get the higher multipliers.  
 

Not saying bio won’t also be immortal, pretty much any tanker should be when built right. Just saying, an SO only build is quite different than a full set build.  Still I think base level information is good to have, we just need to be careful extrapolating conclusions that aren’t based on the test parameters. 

Edited November 23, 2020 by Saikochoro

[Hyperstrike]

1 minute ago, Tsuko said:

That's my point about this tests.

Staying inactive in the middle of mobs is not a good indication of a tanker efficiency.

The objective of this test is not "Tanker efficiency".  Tanker efficiency is a melange of characteristics including toughness and damage output.

What's being measured here is just raw toughness at a very basic, barely enhanced level.

"Stand there and take it, until you can't."

As was noted, for most the very HEAVILY optimized toughness Tanks, this test won't actually measure anything.
As, outside of certain mobs, these setups will set them well above the immortality line (where the enemy can't manage to do enough damage to kill you before you recover)..

[Bill Z Bubba]

This is a solid test of the armor as it stands alone while following the "this game is based on SOs" concept so heavily pushed by some. It is valuable in that it shows us a solid comparison between the armors. Nothing more, nothing less, and I think @Galaxy Braindid that quite intentionally.

 

Now he's adding in some pools which I think is a solid plan. @Werner's addition of the unique IOs is also a great place to get some good balance notes from. Damnation knows I use the hell out of all of them as soon as I'm able to slot 'em.

 

Yes, adding in a secondary's mitigation will at some point be useful as all hell as well. We've already seen how well claws does when you slap that KB--KD proc in shockwave and go to town with it. Knockdown is AMAZEBALLS for extra mitigation while not screwing up damage output. But we can get to that later. For now, I'm fully on board diggin in the dirt to take a good look at just the primary Tank armors and see what's what.

 

Total side note: It's threads like this that keep me coming back to the forums. There's no BS, no dev toadie yesmen, no crap goin on, just folks who like the numbers and the study. It's all good shit.

Edited November 23, 2020 by Bill Z Bubba

[Werner]

20 minutes ago, Bill Z Bubba said:

It's threads like this that keep me coming back to the forums. There's no BS, no dev toadie yesmen, no crap goin on, just folks who like the numbers and the study. It's all good shit.

Yeah, and I hope nothing I say suggesting changes or cautions comes across as impugning what @Galaxy Brain has done here. I think it's awesome, as long as we understand that it's totally unreasonable for me to conclude that we should...

 

Buff Shield Defense and Dark Armor!!! 😉 

 

(Though if you want to give Dark Armor a little DDR, I won't complain. Shhhh. It'll be our secret.)

Edited November 23, 2020 by Werner

[Marshal_General]

I wonder how much of a difference it would make with basic IOs.

[Werner]

9 minutes ago, Marshal_General said:

I wonder how much of a difference it would make with basic IOs.

Very little, I imagine. 3 SOs in an armor give 56% enhancement, 3 IOs in an armor give 58.48% enhancement, at least if you don't start +ing them.

[Hyperstrike]

21 minutes ago, Marshal_General said:

I wonder how much of a difference it would make with basic IOs.

All it would do is shift the entire scale sideways a bit.  As a whole. As everything would simply bump a miniscule 2.5%.

Actual differential would hardly be noticeable and there would be no rank changes.

[Marshal_General]

3 minutes ago, Hyperstrike said:

All it would do is shift the entire scale sideways a bit.  As a whole. As everything would simply bump a miniscule 2.5%.

Actual differential would hardly be noticeable and there would be no rank changes.

I would think SR would fare a bit better.

[Werner]

7 minutes ago, Marshal_General said:

I would think SR would fare a bit better.

A bit. 40.56% defense with 3 SOs, 41.20% with 3 IOs. That's about a 7% reduction in the number of hits taken.

[BZRKR]

1 hour ago, Werner said:

Yeah, and I hope nothing I say suggesting changes or cautions comes across as impugning what @Galaxy Brain has done here. I think it's awesome, as long as we understand that it's totally unreasonable for me to conclude that we should...

 

Buff Shield Defense and Dark Armor!!! 😉 

 

(Though if you want to give Dark Armor a little DDR, I won't complain. Shhhh. It'll be our secret.)

And Radiation too! 🙂

But seriously, thank you Galaxy Brain for taking the time to do this testing and present the findings in a digestible format. If you plan on doing further permutations of this test, may I suggest one where the enemy group is:

2 level 54 Rikti Magi

2 level 54 Rikti Priests

7 level 54 Rikti Headmen

… just off the top of my head, for no specific reason *whistle*

[Hew]

So @Werner, the value of procs (all the heal/end procs, the res procs, the def procs, the tank proc) are so huge, that they in many ways are worth more than 2 SOs slotted somewhere in some/many cases. When you consider that, comparing even a hybrid so/proc build to a straight so build is far far far from apples to apples. The performance boost of a handful of expensive procs is SO high, that in some cases you really can get by with just those and generic IOs, and totally deal with fairly beefy difficulties, like 1/4, 1/6, and up.

 

In fact, both def procs nets you 6%, which is close to weave at 3 power picks with no slotting, and _requires no end to run_.  2 res procs nets you somewhere between 6.5 - much higher (to all), which is ~ half of unslotted tough, _and requires no end to run_. When your tank (there are some!) has such heavy endurance costs that functioning is difficult, stripping out those two toggles for 4 procs can dramatically change how the tank (anything really) plays, and gives you 3 powers to pick that might otherwise be unpicked, like ohshit button armor buffs. On a generics only, or SOs only, that is a huge difference in function.

[Werner]

12 minutes ago, Hew said:

@Werner,... comparing even a hybrid so/proc build to a straight so build is far far far from apples to apples. The performance boost of a handful of expensive procs is SO high, that in some cases you really can get by with just those and generic IOs, and totally deal with fairly beefy difficulties, like 1/4, 1/6, and up.

I agree. I'm interested to see hybrid SO/proc builds compared to each other, not to straight SO builds. And I was guessing that +2x8 would be a good difficulty for comparing them to each other, though that's just a guess.

[Hew]

To minimize "overslotting", id recommend slotting both res procs on res builds (elec, rad), or both def procs on def (super reflexes, ice) builds; On hybrids (bio, shield), slot all 4 (its harder to meet survival competency when you are drawn across having to boost resist AND defense). 

 

General use procs should include the 7.5% hp boost, tank res proc, +absorb heal proc. 

 

The idea here is to maximize effect on the _primary_ function of a primary powerset (res vrs def vrs hybrid), without throwing a lot of extra juice in the mix.

 

Thoughts?

[Werner]

I don't think of it as overslotting. Yes, people are unlikely to slot all of them in a finished build, but they'll also be slotting sets for set bonuses, so we're still rather underslotted in that sense compared to a true IO build. And giving some primaries 2 uniques and some 4 sounds like a strange way to compare them. To me it seems simplest and fairest to give everyone everything.

[Hyperstrike]

4 hours ago, Marshal_General said:

I would think SR would fare a bit better.

Remember, it's not 2.5% better defense.  (Going from 25.0% to 27.5% defense.)
It's +2.5% base enhancement.  If unenhanced is 10% Defense, and SOs bump you to 15.6, 2.5% is going to bump to to 15.85%.

Improvement?
Yes!
Significant?
Not really.

[Galaxy Brain]

Alrighty, so for take 2 I added the 4 pool powers: Tough, Weave, Combat Jumping, and Maneuvers (except for Stone which doesn't allow CJ) for a total of +23.4% added SL res and +15.8% added defense.

 

Starting with Invuln, I noticed I was not getting dented at all... so I upped the difficulty to 2/8... nada

 

So I decided, why not: lets go right to +4/x8 

 

 

 

First thing to note, all sets except Fire actually lasted longer on average vs +4/x8 with pool powers than they did at 0/8 with no pool powers. 

 

Second, Granite is no surprise but Invuln holding up to the immortal line at the technical max difficulty is given the bare minimum investment!

 

Bio holds strong still, collectively taking the number 2 spot even in offensive mode (Again, not counting Granite). Invuln I put above Bio though as I only ever needed to use Dull Pain once, while Bio Defensive I needed to cycle the clicks often to survive despite not going down.

 

Ice is the same deal as before, though it never killed anyone. Willpower received a HUGE boost with added res and def moving above SR, which held up relatively the same with beyond soft-cap, but still getting the occasional hit in hurt badly. Speaking of, I can confirm scaling starts at 60%, but I would still love it to start at least at 90%.

 

Shield and Rad likewise jumped up, Shield much more so. The kicker here was like with WP, the added Res and Def stacks very nicely with the existing Res and Def in the set, and allowed for an additional Shield Charge to buy time per run on average. Rad bumped up two spots just on the note of being a bit tougher, and by nature of a certain other set plummeting.

 

Stone is about the same as before, as are Dark and Fire (though Fire actually fared worse), but the massive drop is with Elec Armor. The added res helps, as does a bit of defense, but at +4 the end drain potency has been stripped which allowed about half the runs on 0/8 to be immortal. Oh well. 

 

 

 

 

[Eclipse.]

1 hour ago, Hyperstrike said:

Remember, it's not 2.5% better defense.  (Going from 25.0% to 27.5% defense.)
.
.
.
Improvement?
Yes!
Significant?
Not really.

You're not wrong...but I'd argue when you put it like that it makes it sound a lot less significant than it is.

That +2.5% better defense = 10% more damage mitigation in that equation.

baseline 0% you take 1/2 (20/40) hits.
2.5% defense you take 19/40 hits
25% defense you take 1/4 (10/40) hits.
27.5% defense you take 9/40 hits.
30% defense you take 1/5th (8/40) hits

Every 2.5% defense = 2.5% (1/40) less hits taken as a static amount, but as your defense goes higher thats a larger percentage of increased damage mitigation. (inverse is also true...at 42.5% defense you receive 50% more damage than you do at 45%)

 

[Galaxy Brain]

12 minutes ago, Eclipse. said:

You're not wrong...but I'd argue when you put it like that it makes it sound a lot less significant than it is.

That +2.5% better defense = 10% more damage mitigation in that equation.

baseline 0% you take 1/2 (20/40) hits.
2.5% defense you take 19/40 hits
25% defense you take 1/4 (10/40) hits.
27.5% defense you take 9/40 hits.
30% defense you take 1/5th (8/40) hits

Every 2.5% defense = 2.5% (1/40) less hits taken as a static amount, but as your defense goes higher thats a larger percentage of increased damage mitigation. (inverse is also true...at 42.5% defense you receive 50% more damage than you do at 45%)

 

While true, it really would not make that significant a difference unless I ran SR like 25 times

[Eclipse.]

4 minutes ago, Galaxy Brain said:

While true, it really would not make that significant a difference unless I ran SR like 25 times

Wasn't targetting the comparative  SR performance based on the quoted, a few levels above mine. 

Just the diminishing of the significance of that 2.5% defense by Hyperstrike I specifically quoted.

Edited November 24, 2020 by Eclipse.