Survivability Analysis - How Tough is Your Character?


Biospark

 

Posted

There was a similar Arcanaville analysis once [considerably pre- IOs]. I think she decided the "immortality line" gave misleading results- as you dial the damage up Regen is immortal, immortal, immortal, dead-in-10-seconds. I was always arguing that the immortality line for Invuln was disproportionately low because it's a slow-death set. Regen actually hits its lowest HP in the start of the fight and, as enemies get killed, it gains HP back. Invuln doesn't work like that. So maybe the dead-in-30-seconds line is worth considering. Of course, as long as you don't run out of endurance, anything that doesn't kill a Dark before Dark Regen comes back... won't.


Mini-guides: Force Field Defenders, Blasters, Market Self-Defense, Frankenslotting.

So you think you're a hero, huh.
@Boltcutter in game.

 

Posted

Hm. I've wondered about how these things compare when using differing kinds of mitigation. How, for instance, would you compare a dominator with permanent domination and mass control overlaps to a softcapped character?


 

Posted

@Fulmens,

Well then I guess that is the use of my "Fatal Burst" column. If that kind of damage ever hits you, you're dead.

But you have given me a great idea for another section to add to my tool. I'm going to figure out how to best do some kind of "Time to Fatality" analysis here.

Perhaps come at it as a function of the Sustainable DPS and max HP. At 25% over your DPS, you have 200 seconds to live. At 50% over, it's 60 seconds. Or whatever the math works out to be.


Where to now?
Check out all my guides and fiction pieces on my blog.
The MFing Warshade | The Last Rule of Tanking | The Got Dam Mastermind
Everything Dark Armor | The Softcap
don'T attempt to read tHis mEssaGe, And believe Me, it is not a codE.

 

Posted

Quote:
Originally Posted by Seldom View Post
Hm. I've wondered about how these things compare when using differing kinds of mitigation. How, for instance, would you compare a dominator with permanent domination and mass control overlaps to a softcapped character?
That's a good question, but I wouldn't consider him softcapped. In fact, I would put that in resistance. If you feel your Dominator can lock down 95% of the spawn all the time, then I would fill out 95% resistance across the board.

Or let's say he can definitely hold 50% of the group, but 100% will be immobilized. Then you can have resistance at 50% and average the ranged/AoE defense together.

Any resistances the character has from tough/patron powers should be multiplied by the "locked down" resistance.


Where to now?
Check out all my guides and fiction pieces on my blog.
The MFing Warshade | The Last Rule of Tanking | The Got Dam Mastermind
Everything Dark Armor | The Softcap
don'T attempt to read tHis mEssaGe, And believe Me, it is not a codE.

 

Posted

how is that resistance when the foes are not attacking?

I'd say thats even better than soft cap, as there is (as long as the dom has them secure) 0-chance of being hit back (as opposed to a 5%)


 

Posted

Quote:
Originally Posted by Player99 View Post
how is that resistance when the foes are not attacking?

I'd say thats even better than soft cap, as there is (as long as the dom has them secure) 0-chance of being hit back (as opposed to a 5%)
It's resistance in the sense that 50% or 75% or 100% of the mob is locked down and not attacking. I would say to never go above 95% because the chance to miss is always there.


Where to now?
Check out all my guides and fiction pieces on my blog.
The MFing Warshade | The Last Rule of Tanking | The Got Dam Mastermind
Everything Dark Armor | The Softcap
don'T attempt to read tHis mEssaGe, And believe Me, it is not a codE.

 

Posted

Heya Dechs,

Thats a great analysis tool. Good work !

I would like to copy some of the concepts over to my own Defender analysis that I have been working on, if you are OK with that.


BIOSPARK :: DARKTHORN :: SKYGUARD :: WILDMAGE
HEATSINK :: FASTHAND :: POWERCELL :: RUNESTAFF

 

Posted

Quote:
Originally Posted by Biospark View Post
Heya Dechs,

Thats a great analysis tool. Good work !

I would like to copy some of the concepts over to my own Defender analysis that I have been working on, if you are OK with that.
Thanks!

I don't mind if you copy some of it. I'd request to have a bit of a shout out, at least, but nothing more.


Where to now?
Check out all my guides and fiction pieces on my blog.
The MFing Warshade | The Last Rule of Tanking | The Got Dam Mastermind
Everything Dark Armor | The Softcap
don'T attempt to read tHis mEssaGe, And believe Me, it is not a codE.

 

Posted

Quote:
Originally Posted by Dechs Kaison View Post
It's resistance in the sense that 50% or 75% or 100% of the mob is locked down and not attacking. I would say to never go above 95% because the chance to miss is always there.
But if you miss with your AoE control, you can smack the ones you missed with single-target controls. You don't necessarily need to, but a mezz-based character can achieve >95% damage mitigation versus single spawns that way. On the other hand, they also have much bigger problems with things like adds and scattered mobs than a character with straight-up resistance would. And then you get into things like -recharge in /Ice Assault and my brain starts hurting.


Having Vengeance and Fallout slotted for recharge means never having to say you're sorry.

 

Posted

Creepy, I've been making on one of these in open office for the last few weeks.

GL all and have fun


Murphys Military Law

#23. Teamwork is essential; it gives the enemy other people to shoot at.

#46. If you can't remember, the Claymore is pointed towards you.

#54. Killing for peace is like screwing for virginity.

 

Posted

Quote:
Originally Posted by Dechs Kaison View Post
It's resistance in the sense that 50% or 75% or 100% of the mob is locked down and not attacking. I would say to never go above 95% because the chance to miss is always there.
Hm. The trick is that with a 95% chance to hit, you will probably hit 100% of a mob. (Assuming you're not at the target limit-16.) Add in control layers, and you'll be at 100%.

Sorry, I don't wish to derail here. It's just that whilst a lot of people like looking at survival directly, (res/def/regen/heal) the game has some that approach through other means. (Mez, -rech, -dam, -tohit) These go off of the enemy, so while not directly making a character 'tough,' functionally it's the same.


 

Posted

Ohh Shiny! I'll take a look at home!

Thanks Dechs!


 

Posted

Quote:
Originally Posted by Seldom View Post
Hm. The trick is that with a 95% chance to hit, you will probably hit 100% of a mob. (Assuming you're not at the target limit-16.) Add in control layers, and you'll be at 100%.
Alright, I'll agree to that. 100% mitigation. At that point, though, you don't need a tool. 100% mitigation means you won't die.

Quote:
Originally Posted by Seldom View Post
Sorry, I don't wish to derail here. It's just that whilst a lot of people like looking at survival directly, (res/def/regen/heal) the game has some that approach through other means. (Mez, -rech, -dam, -tohit) These go off of the enemy, so while not directly making a character 'tough,' functionally it's the same.
The derailment is fine! I would like to develop this into an all purpose tool that can take those things into account.

Mez we've discussed. -tohit could be done by adding to defense, since the two work the same way. You may have to use several sheets to account for the resistances of higher level enemies. -damage can likewise be added to resistance.

-recharge is tough to handle, but it doesn't exactly increase the DPS you can sustain. What it does is decrease the incoming DPS, so you'll have to bear in mind that a -recharging character will be more comfortable at higher levels of DPS.


Where to now?
Check out all my guides and fiction pieces on my blog.
The MFing Warshade | The Last Rule of Tanking | The Got Dam Mastermind
Everything Dark Armor | The Softcap
don'T attempt to read tHis mEssaGe, And believe Me, it is not a codE.

 

Posted

Quote:
Originally Posted by Seldom View Post
Hm. The trick is that with a 95% chance to hit, you will probably hit 100% of a mob. (Assuming you're not at the target limit-16.) Add in control layers, and you'll be at 100%.

Sorry, I don't wish to derail here. It's just that whilst a lot of people like looking at survival directly, (res/def/regen/heal) the game has some that approach through other means. (Mez, -rech, -dam, -tohit) These go off of the enemy, so while not directly making a character 'tough,' functionally it's the same.
Mez is fairly ignorable, because if it works (Fear aside) it's 100% mitigation and if it doesn't it is 0% mitigation so you only have to compare senarios where mez does or doesn't work.

-rech,-dam,-tohit can be added to an equation fairly easily, but you just have to compensate for any debuff resist the enemy might have.

Dechs, The numbers on the sheet look off. If a Tanker Has 1874.1 health at 50, how can it take 4459 smashing damage per second when 4459 * .45 * (1-.692)= 618 damage per second recieved on average. 1874.1/618= dead in 3.03 seconds. Am I reading it wrong?

GL all and have fun


Murphys Military Law

#23. Teamwork is essential; it gives the enemy other people to shoot at.

#46. If you can't remember, the Claymore is pointed towards you.

#54. Killing for peace is like screwing for virginity.

 

Posted

Quote:
Originally Posted by William_Valence View Post
Dechs, The numbers on the sheet look off. If a Tanker Has 1874.1 health at 50, how can it take 4459 smashing damage per second when 4459 * .45 * (1-.692)= 618 damage per second recieved on average. 1874.1/618= dead in 3.03 seconds. Am I reading it wrong?
First thing I notice is that my tank has 2508 hit points, not 1874.

Second thing I notice is that you're multiplying by .45 for defense, not .05. That'll make a huge difference.

I get 4459*(0.50-0.45)*(1-.692) = 68.7 damage per second received. Then 2508/68.7 = 36 seconds, which is enough time to fire off two dark regens.


Where to now?
Check out all my guides and fiction pieces on my blog.
The MFing Warshade | The Last Rule of Tanking | The Got Dam Mastermind
Everything Dark Armor | The Softcap
don'T attempt to read tHis mEssaGe, And believe Me, it is not a codE.

 

Posted

@Dechs...Gotcha.

Res/defense, then, are easy enough to calculate. They're based off of crunching the numbers of how much incoming damage there is, and it's accuracy, and comparing that to how much an armor will reduce/deflect that. While +res/+def powers skew the recieving scale, the debuffs -dam/-tohit change the incoming scale. The only trick is perhaps taking into account how much the foes resist these effects. All that seems easy enough.

+regen is basically the opposite of damage over a time period. Heals can be averaged into a regeneration-like value, based off of their recharge.

Wouldn't mez, then, be somewhat similar to how heals work? Disregarding immobilizes, all other mez stops an enemy from attacking you for a certain time. As such, the calculation would be what % of the time is an enemy incapacitated? If the power recharges in 30 seconds, and lasts 15, that's 50% damage survivability, assuming it hits. With full accuracy you're right about the 5% hit chance, but that doesn't apply vs. the second mez laid down. I'm betting the chances or two powers @95% accuracy missing the same target is a bit low.

The perma-dom, especially a perma-Mind dom sporting 3 AoE controls on fast enough recharges, is an extreme example. But since you brought in a dark armor character, I wonder how oppressive gloom and cloak of fear stack up? Or a dark miasma character that has an AoE fear/-tohit alongside +res and -dam/-tohit.

Hm. Fear is an odd one, as it's basically a -rech in function...for a certain duration, the enemy can only cower. Every 10 second, or when damaged, they have an action chance before they are returned to the cowering status. This continues until the fear duration is up.

Yikes, that gets complicated in a hurry.


 

Posted

Quote:
Originally Posted by Seldom View Post
Wouldn't mez, then, be somewhat similar to how heals work? Disregarding immobilizes, all other mez stops an enemy from attacking you for a certain time. As such, the calculation would be what % of the time is an enemy incapacitated? If the power recharges in 30 seconds, and lasts 15, that's 50% damage survivability, assuming it hits.
That's an interesting way to approach the issue. Non-obvious, but it could certainly be made to work.

I just thought about it though, and perhaps a good way to use the tool for a mezzing character is "This is how much damage I can survive from the stuff I miss with my mezzes."

Quote:
Originally Posted by Seldom View Post
Yikes, that gets complicated in a hurry.
Yes. Yes it does.


Where to now?
Check out all my guides and fiction pieces on my blog.
The MFing Warshade | The Last Rule of Tanking | The Got Dam Mastermind
Everything Dark Armor | The Softcap
don'T attempt to read tHis mEssaGe, And believe Me, it is not a codE.

 

Posted

Quote:
Originally Posted by Dechs Kaison View Post
I just thought about it though, and perhaps a good way to use the tool for a mezzing character is "This is how much damage I can survive from the stuff I miss with my mezzes."
It's the approach I'd take. There are just entirely too many variables to consider with debuffs and mezzes to really get meaningful numbers out of it, apart from highly-controlled situations like fighting a single target... and more esoteric powers get even worse. How much damage resistance is taking the alpha strike with Seeker Drones equivalent to? The answer is a resounding 'lol, I dunno.'


Having Vengeance and Fallout slotted for recharge means never having to say you're sorry.

 

Posted

Quote:
Originally Posted by Kelenar View Post
It's the approach I'd take. There are just entirely too many variables to consider with debuffs and mezzes to really get meaningful numbers out of it, apart from highly-controlled situations like fighting a single target... and more esoteric powers get even worse. How much damage resistance is taking the alpha strike with Seeker Drones equivalent to? The answer is a resounding 'lol, I dunno.'
In this case, the drones survivability would as a distraction would be used to subtract from the oncoming dps...so what %of the time would they be refocusing said DPS? This is based off of how long it takes to kill them, and how fast they recharge. The fact that they have a stun chance/-dam/-tohit in their radius would mean that one would have to see how many that radius would hit in a mob on average, how long/strong the -tohit/-dam would be, and how many, on average, would be mezzed for how long.


 

Posted

Quote:
Originally Posted by Dechs Kaison View Post
First thing I notice is that my tank has 2508 hit points, not 1874.

Second thing I notice is that you're multiplying by .45 for defense, not .05. That'll make a huge difference.

I get 4459*(0.50-0.45)*(1-.692) = 68.7 damage per second received. Then 2508/68.7 = 36 seconds, which is enough time to fire off two dark regens.
Yup, I was reading it wrong. When I saw 45% defense I was thinking 50% tohit - 5% def not 50% - softcap 45% defense. As for the health, I've been comparing scrapper/tanker/brute values lately and I completely ignored the health valued you had there and just put down the default. Sorry, working on muscle memory.

So I wanted to compare your sheet to what I have. I used Arcana's method of comparing percent health per second taken, with an equation of
DPS(in percent of health) = (.05 / (11.905/(1+(regen rate/100)))) / (.5-def) / (1-damdebuff)/ (1-res) / (1-(1/(1+hpbuff%)))

When something has different values at different times I can use weighted averages, but I Don't see a Dark/Fire having temporary hp or regen buffs, just set bonuses.

So when I put your stuff in my sheet, It said that you can survive 705.067296% of tanker Base health each second, Or 13212.405 DPS Thrown out, Or (13212.405*(0.5-0.45))*(1-0.692)*(1-(1/1.338242))= 51.427508 dps taken on average vs (((0.05/(11.905/2))+(0.351742/(17.3+1.17))))*100 = 2.74438% health restored per second, Or 1874.1 * .0274438 = 51.432426.

Increase both by the hpbuff% and you get 68.82245Damage vs 68.829032 restoration. Darn close to what you got above, but 13000+ is a good deal more than 4459.

I have no idea where I'm going wrong.


Murphys Military Law

#23. Teamwork is essential; it gives the enemy other people to shoot at.

#46. If you can't remember, the Claymore is pointed towards you.

#54. Killing for peace is like screwing for virginity.

 

Posted

Going back to more traditional mitigation calculations, one thing that might be good to add is a way to handle non-permanent max HP buffs. If you've got something like high pain tolerance or a +max hp set bonus, you can just put it into the HP box and be done. But if you've got something temporary like dull pain, it gets rather more complicated.

First off, your regeneration rates will be different with DP up and DP down. The overall regeneration rate will be an average weighted by the uptime of the power, which would be relatively simple to display pre-calculated on the sheet. Alternatively, you could simply calculate how much extra you regen over one cycle of the power compared to your base regen, and include that as a heal.

Second, the temporary +max HP effect will need to be quantified. The best way to deal with this, I find, is to treat it as a sort of pseudo-resistance. If you think in terms of 'fractions of your health bar', then a +max HP buff makes each incoming chunk of damage represent a smaller piece of your health and thus functions somewhat as a second layer of resistance that stacks multiplicatively with the first. The formula is:

pseudo resistance = 1 - (normal max HP/boosted HP)

This is then added multiplicatively with normal resistance as 1 - [(1 - resist)*(1 - pseudo resist)]. It could then be averaged in the same way as the tier 9 resistance values are.

I actually did this manually for my shield/mace build that's currently part of the document. One with the shield isn't providing the 49.6% S/L resistance indicated - it's providing 45%, and the other ~5% comes from the +max HP in the power. Similarly, my regen isn't actually 17.98 all the time - it's higher when OWtS is up, and lower when it's not, and that is just the weighted average. It would be nice to have this calculated automatically, though.

Finally, I second the suggestion for a 'X second survivability' column. I added a 60 second survivability column to my build's page, it's pretty trivial to calculate.


@MuonNeutrino
Student, Gamer, Altaholic, and future Astronomer.

This is what it means to be a tank!

 

Posted

I used to do this on excel except with the secondaries damage levels and control levels. It's alright if you enjoy getting into it as it takes alot to really get into and keep doing as the game keeps changing. At no point should anyone I feel be influenced by these things in deciding what tankers work best. They have to be played differently. Most survivability models assume a brain dead player in the middle of mobs where as I have no doubt that the Devs like to lean nearer assuming a skilled player amongst some brain dead mobs.


He will honor his words; he will definitely carry out his actions. What he promises he will fulfill. He does not care about his bodily self, putting his life and death aside to come forward for another's troubled besiegement. He does not boast about his ability, or shamelessly extol his own virtues. - Sima Qian.

 

Posted

Quote:
Originally Posted by New Dawn View Post
I used to do this on excel except with the secondaries damage levels and control levels. It's alright if you enjoy getting into it as it takes alot to really get into and keep doing as the game keeps changing. At no point should anyone I feel be influenced by these things in deciding what tankers work best. They have to be played differently. Most survivability models assume a brain dead player in the middle of mobs where as I have no doubt that the Devs like to lean nearer assuming a skilled player amongst some brain dead mobs.
You are absolutely correct New Dawn

Although trying to put scenarios down to encompass all likely possibilities would require a neural network or database to really get down to the heart of gameplay.

The fact that every villain type is different and then you have minions, lts and bosses all with different values really complicates things in a hurry. Then there is mob behavior. Does it stay back and shoot, run in to melee? How long before it decides to switch tactics? Does the player force this switch? etc, etc, etc

What I am attempting on my models is to just look at three minions as the baseline, come up with relevant models, then I can spawn a model with three Lts and another with 3 Bosses. Do this for every damage type, and come up with a summation based on what percentage of the game is represented by each type and villain. This will get me an average performance throughout the game, and also individual summations for each damage type and villain type.

So...... 2012 .... maybe


BIOSPARK :: DARKTHORN :: SKYGUARD :: WILDMAGE
HEATSINK :: FASTHAND :: POWERCELL :: RUNESTAFF

 

Posted

Is there anywhere to download a copy of the spreadsheet please? I can't seem to do that from Googledocs, nor can I seem to work on a copy of the template on googledocs.



Nice work. Thanks.





SAVE CoX info:
Titan Network efforts
Saving CoX events/FB info