[RAID] Boss armor values - Page 3

Valerian · 10/05/07, 11:08 AM

Originally Posted by Sarlunas

Your claim is that "time to death" is the true metric and therefore DPS has diminishing returns while armour and armour pen are linear in value. I would make the opposite claim that it is in fact "time to death" that has diminishing returns and thus DPS is the better metric.

To me it seems fairly obvious that shaving 10 sec off a short bossfight is generally far more valuable than shaving 10 sec off a very long bossfight, this is assuming 100% time on target.

DPS is what really matters, which means armour does have diminishing returns and armour pen has "increasing returns".

Hmm I dont know about that. You want the boss killed in the shortest amount of time. Longer fights lead to more chances for things to go wrong whether due to player error, poor RNG strings or boss enrages. Looking at it from a direct time to death view gives you a direct feel of how long the fight will take. It also puts a lot of this theorycraft stuff into perspective when you see that a 100 DPS increase via optimization results in something like 2 seconds less of fighting. Whether thats a good thing or something kinda shows all this a waste of time, I dont know :P

In any case it doesnt change the numbers in any way. ArP IS better to add than straight DPS increasers (assuming proper item value which I havent looked at), Im not arguing it isnt.
Its simply a matter of perspective. As I said I posted it because the whole "diminishing returns on armor" thing is commonly known and could be brought up here without realizing that, if you look at it that way, you need to also notice that the straight up DPS increasers have diminishing returns (from this perspective).

• Aldriana · 10/05/07, 12:32 PM

Originally Posted by Valerian

I dont know about the rogue spreadsheet, but Toskk's druid dps calculator is already giving a weighting for ArP. I dont think there'd be an easy way to see the benefit of it vs. say hit/crit/ap without a spreadsheet/calculator to determine the exact values.

The Rogue Gear sheet has a model for armor penetration, although it's accuracy has always been a bit suspect due to not knowing the base armor of mobs. This should help quite a bit, so kudos to those who have done the research.

Originally Posted by Valerian

Hmm I dont know about that. You want the boss killed in the shortest amount of time. Longer fights lead to more chances for things to go wrong whether due to player error, poor RNG strings or boss enrages. Looking at it from a direct time to death view gives you a direct feel of how long the fight will take. It also puts a lot of this theorycraft stuff into perspective when you see that a 100 DPS increase via optimization results in something like 2 seconds less of fighting. Whether thats a good thing or something kinda shows all this a waste of time, I dont know :P

In any case it doesnt change the numbers in any way. ArP IS better to add than straight DPS increasers (assuming proper item value which I havent looked at), Im not arguing it isnt.
Its simply a matter of perspective. As I said I posted it because the whole "diminishing returns on armor" thing is commonly known and could be brought up here without realizing that, if you look at it that way, you need to also notice that the straight up DPS increasers have diminishing returns (from this perspective).

Sure, you want the boss to die as quickly as possible; however, I would argue that the true metric of benefit in terms of reducing the fight duration is percentage decrease of duration; that is, if you cut 6 seconds off a 10 min fight, you've reduced the fight duration by 1%, and thereby reduced the chance of random flukes killing you by 1%. However, if you reduce a 2 min fight by 6 seconds, that's now a 5% reduction in the chance of a random fluke killing you. Hence, I subscribe to the school of thought that DPS is the correct way of measuring the output of a DPSer, rather than "time to kill"

Additionally, it might be noted that "time to kill" assumes that you're the only one attacking, which you're not. When you factor in the fact that there are other people attacking the mob in question, in terms of "total time to kill", armor penetration effects have increasing returns.

Regarding the utility of theorycraft: it's true, theorycrafting one's own DPS output does little to reduce the overall fight duration. However, if all 15 DPS in the raid theorycraft and all squeeze out an extra 100 DPS, suddenly we're talking about a 10% reduction in fight duration, which can easily trim a minute or more off fights. So I wouldn't think of it as reducing the fight duration by a couple of seconds; I'd think of it as doing your part to cut a minute off the fight. I'd also think of it as doing your part to avoid 1% wipes.

PSGarak · 10/05/07, 5:43 PM

"Time to death" is generally calculated as a percent, not an amount. DPS having diminishing returns, but that makes the time-to-death ratio linear. If you want to, you can think of time-to-death as time-for-XXX-damage, or spd (1/dps) times some big number of seconds so it's not a stupidly small fraction. Seconds per 1k damage sounds like a reasonable metric.

• songster · 10/07/07, 3:16 AM

Hmm, I guess this seems like the most appropriate thread. This is a potentially very silly question about Expose Armor.

Expose doesn't stack with Sunder, and Sunder can't be applied while EA is present. How does this interact with Devastate? If you put up a stack of 5 sunders, and then EA - does this prevent Devastate being used?

Garithras · 10/07/07, 3:26 AM

No. I suppose I'm talking out of my ass because I haven't actually tested this, but following the normal game rules, EA should overwrite the sunders, which causes Devastate to hit for less damage, therefore causing less threat. There's no reason currently that EA would make Devastate unusable.

FAME · 10/07/07, 10:17 AM

While Devastate will hit for less, other threat abilities will hit for slightly higher due to the drop in boss' AC. If threat is not an issue it might even be worth trying something like that to see how it plays out. However from the wroding in the changes being done to Devastate in the comming patch, it would seem that sunders will greatly increase its threat.

schmurfy · 10/08/07, 9:29 AM

Great thread

Originally Posted by enno

uh? How exactly do you mean this?
As a rogue I already died to her punching my lights out after I took agro from our tank. (hey, if the raidleader says full nuke, that's exactly what I do) :P Now I just use vanish after blowing all cooldowns in the first 20 or so seconds.

Solarian does have aggro, she will randomly hit the first melee player in aggro list with a ~10k that will kill almost any melee dps, in the same way if any caster get in melee range when she decides to hit he will also die

Maybe it was changed with the recent changes but I doubt it.

Illundai · 10/08/07, 10:03 AM

That's how mobs work, Aran does it in a similar fashion. Afaik, it has something to do with how mobs queue autoswings - if they queue the swing during a cast they often hit right after the cast is done. Haven't really thoroughly tested or investigated that, but that's my experience.

Sleepy · 10/11/07, 4:20 PM

Originally Posted by Dayve

The answer to that isn't simple; all of those things are dependant on how much armor the boss has before debuffs, as well as what other debuffs are on the boss.

Armor reduction of a level 73 boss attacked by a level 70 player is 37% for 6200 armor, and 42.17% for 7700 armor. Sunder Armor x5 is -2600 armor, Faerie Fire is -610, and Curse of Recklessness is -800.

For a boss with 6200/7700 armor, SA x5 gives a damage increase of 18.36%/16.61%.

If we assume now that the boss has SA stacked x5 on it, adding just FF gives a further 4.5%/4.05% increase, CoR gives 5.99%/5.38%, and with both on they give 11.06%/9.9% increase over just SA.

Armor Penetration is a bit more complex because it comes it varied amounts. If the boss has just SA x5 on it, then the effect of ArP is as follows:
100 ArP: 0.71%/0.64%
200 ArP: 1.43%/1.29%
300 ArP: 2.16%/1.95%
400 ArP: 2.91%/2.62%
500 ArP: 3.66%/3.3%

For a fully debuffed boss (SA x5, FF, CoR) the increase from ArP is:
100 ArP: 0.79%/0.71%
200 ArP: 1.59%/1.42%
300 ArP: 2.41%/2.15%
400 ArP: 3.24%/2.89%
500 ArP: 4.08%/3.64%

Been playing around in excel with the numbers, but i cant seem to get the same results as you. Im wondering what im forgetting so can you put up a couple of formulas?

Currently i get
- With sunder
100 0,53 0,43
200 1,07 0,87
300 1,61 1,32
400 2,17 1,77
500 2,73 2,22

- With sunder + FF + CoR
100 0,65 0,52
200 1,32 1,05
300 1,99 1,59
400 2,68 2,14
500 3,38 2,69

If i want to hit the same values as you, i need to drop armor another 1250 (so 6200 starts as 4950)

Can anyone double check the numbers and maybe post some formulas if you locate the error

Spline · 10/11/07, 7:31 PM

I'm a Warlock, and with these armor values I've been able to exactly quantify casting Curse of Recklessness based on the raid makeup. To be more exact, I figured out how to rank the curses Shadow, Elements, and Recklessness according to damage contribution, by equating the curse to a percentage of a single person's DPS. Admittedly, this only begins to matter when you bring two or less Warlocks, but it's useful to know.

The value of Shadow is simply equal to 10 (or 13) times the number of Shadow/Arcane users in the raid. Likewise, Elements is 10 (or 13) times the elemental users. Recklessness is 5-6 times the number of physical damage dealers, considering Dayve's math:

If we assume now that the boss has SA stacked x5 on it, adding just FF gives a further 4.5%/4.05% increase, CoR gives 5.99%/5.38%, and with both on they give 11.06%/9.9% increase over just SA.

Take for example a raid of 4 Rogues, 2 DPS Warriors, 2 Hunters, 3 Warlocks, 1 Shadowpriest, 1 Arcane and 2 Fire mages, supported by 3 tanks and 7 healers. No warlock has Malediction.

Shadow is worth 40%. Elements is worth 20%. Recklessness is worth 40%. If one of the warlocks has to go, it's the Fire mages who are getting the shaft, curse-wise.

Sleepy · 10/12/07, 9:11 AM

Can anyone give some feedback on the numbers above (my last post)

With the new gear comming, armor pen seems to be even more widespread, so accurate numbers is needed.

intressepil · 10/12/07, 10:08 AM

Lactose/Dayve, was there ever any consensus reached regarding this? Which numbers are the correct ones?

ArkamisImm · 10/12/07, 12:44 PM

Originally Posted by Dayve

I've been doing a fair bit of testing on mob armor over the past month. I've been using pretty much the same method as Lactose, although I use imp. Thunder Clap (which has a fixed damage amount).

To make things easier, I wrote a simple addon that watches for TC hitting my target, calculates how much it should have hit for before armor reduction, and outputs into a chat window the range of armor values that could have produced the observed hit (taking into account any armor debuffs of course). My estimates were then taken from the intersection of all the ranges that I recorded. From the size of the ranges, the accuracy of the results should be about +/- 30.

So far, I've managed to get results for almost every trash mob and boss in SSC, TK, BT, and Hyjal. Some of the values are different from the ones Lactose obtained. I'll just list the boss armor values, because those are the useful ones

.

Serpentshrine Cavern:
Hydross the Unstable: 7700
The Lurker Below: 7700
Leotheras the Blind: 7700
Fathom-Lord Karathress 6200
Morogrim Tidewalker: 7700
Lady Vashj: 6200

Tempest Keep:
Void Reaver: 8800
High Astromancer Solarian: 6200
Al'ar: 7700
Kael'thas Sunstrider: 6200

Hyjal Summit:
Rage Winterchill: 6200
Anetheron: 6200
Kaz'rogal: 6200
Azgalor: 6200
Archimonde: 6200

Black Temple:
High Warlord Naj'entus: 7700
Supremus: 7700
Shade of Akama: 7700
Teron Gorefiend: 6200
Gurtogg Bloodboil: 7700
Reliquary of Souls:
- Essence of Suffering: 0
- Essence of Desire: 7700
- Essence of Anger: 7700
Mother Shahraz: 6200
Illidari Council:
- Gathios the Shatterer: 6200

Notice that all but 2 of the bosses have either 6200 or 7700 armor. A similar pattern occurs in the trash mobs: 5475/5700/5950 or 6800/7100/7400 for level 70/71/72 mobs. In terms of percentage reduction, these values result in either 34.15% or 39.15% armor reduction vs attackers of the same level as the mob.

If you can give me your raw data I can write a Householder Transformation that will give a much more accurate estimate.

It's a method that gives a more accurate estimate of observed values when you have an overdefined system.

Grailwatcher · 10/13/07, 2:43 PM

How would personal ArP affect personal dps % assuming that only sunderx5 and CoR were up on the boss. I'm not the most solid at math, but it looks like it would have a lower %increase in damage vs having all 3 of the debuffs up. Is this correct?

ArkamisImm · 10/13/07, 2:51 PM

Originally Posted by Grailwatcher

How would personal ArP affect personal dps % assuming that only sunderx5 and CoR were up on the boss. I'm not the most solid at math, but it looks like it would have a lower %increase in damage vs having all 3 of the debuffs up. Is this correct?

I'm running those numbers as we speak =)

dazed420 · 10/16/07, 8:51 AM

any word on the math you were working on?

Platina · 10/18/07, 8:13 AM

I put some work on Numbers and my conclusion is that the TimeToDeath or TimeToLife changes linear to +/- Armorclass.

My Blueprints are in an Excel Sheet: http://www.dragonfire.de/theory/AP.xls

My Opponent/Boss hast 7500 AC and 500.000hitpoints. I would like to kill him within 600seconds (TTL).

-7500 AC = 1425,33 NeedDPS
-7000 AC = 1385,86 NeedDPS
-6500 AC = 1346,40 NeedDPS
-6000 AC = 1306,93 NeedDPS
-5500 AC = 1267,46 NeedDPS
-5000 AC = 1228,00 NeedDPS
-4500 AC = 1188,53 NeedDPS
-4000 AC = 1149,06 NeedDPS
-3500 AC = 1109,60 NeedDPS
-3000 AC = 1070,13 NeedDPS
-2500 AC = 1030,67 NeedDPS
-2000 AC = 991,20 NeedDPS
-1500 AC = 951,73 NeedDPS
-1000 AC = 912,27 NeedDPS
-0500 AC = 872,80 NeedDPS
-0000 AC = 833,33 NeedDPS

500 AC compensates EVER 39,47 DPS. The First 500, and the last 500 until the Cap (0) is reached. This is NOT a general universal usable Term!

Conclusion: "The First -AC is as good as the Last -AC". No Stacking, No Boost, nothing special.

How good -AC is on you're current Gear isn't that easy. You must definte/calculate your own "TrueDPS w/o Armor" Value. Then you must decide. More -AC or more AP/Crit/Hit, whatever.

Easy Overview:

(AP + Crit + Hit + Miss + Dodge + Weapon) * (Armor - Armorpierce)

What do you think about this?

Sarlunas · 10/19/07, 9:49 AM

It has been established long ago that TTD/TTL have a linear relationship to armour, there is no need for further math on this as far as I can tell. The reason that people still regard armour as having diminishing returns is because we don't agree with the assumption that TTD/TTL are good metrics to use in the first place, at least not in the way you are using them.

There has been some discussion about this matter in this thread already but the gist is that while I assume that your math is correct I will never agree with your conclusion that armour has linear scaling because I don't agree with your basic assumptions. As far as I'm concerned DPS is, was and always will be the true metric. Which means that armour has diminishing returns.

From your perspective -AC is linear and all other dps stats have diminishing returns of various degrees, in any case claiming that the scaling of -AC is "nothing special" is clearly erroneous as it differs from all the other stats.

Takkara · 10/19/07, 10:15 AM

No, the reason the TTD metric is bad is that not all damage done to the boss is physical, and not all members of the raid are operating under the same amount of -armor. Also, TTD doesn't translate into any physical measure.

For example, even if armor penetration has a linear affect on TTD, it doesn't have a linear performance boost in a typical raid.

The only situation where TTD and linear scaling armor penetration has a realistic and consistent affect on raid performance is when all damage done in the raid is physical and all members operate on the same level of armor penetration.

This is already been hashed over by Aldrianna.

Indora · 10/23/07, 2:29 PM

this is exactly what we (in our german forum) are saying @platina

Caesar · 10/24/07, 7:25 AM

Originally Posted by Dayve

The answer to that isn't simple; all of those things are dependant on how much armor the boss has before debuffs, as well as what other debuffs are on the boss.

Armor reduction of a level 73 boss attacked by a level 70 player is 37% for 6200 armor, and 42.17% for 7700 armor. Sunder Armor x5 is -2600 armor, Faerie Fire is -610, and Curse of Recklessness is -800.

For a boss with 6200/7700 armor, SA x5 gives a damage increase of 18.36%/16.61%.

If we assume now that the boss has SA stacked x5 on it, adding just FF gives a further 4.5%/4.05% increase, CoR gives 5.99%/5.38%, and with both on they give 11.06%/9.9% increase over just SA.

Armor Penetration is a bit more complex because it comes it varied amounts. If the boss has just SA x5 on it, then the effect of ArP is as follows:
100 ArP: 0.71%/0.64%
200 ArP: 1.43%/1.29%
300 ArP: 2.16%/1.95%
400 ArP: 2.91%/2.62%
500 ArP: 3.66%/3.3%

For a fully debuffed boss (SA x5, FF, CoR) the increase from ArP is:
100 ArP: 0.79%/0.71%
200 ArP: 1.59%/1.42%
300 ArP: 2.41%/2.15%
400 ArP: 3.24%/2.89%
500 ArP: 4.08%/3.64%

very interesting!

Also anyone know if Owl Screech stacks with the above mentioned abilities?

◊ Lactose · 10/24/07, 8:01 AM

Owl screech has no effect, since it does not affect armor of target. Screech reduces the target's Melee Attack Power.

Caesar · 10/24/07, 8:54 AM

>.<

Need more coffee. Yes you are absolutely right, I was thinking use CoR and then use owl screech to remove added atp, and then I got brainfreeze. I ment to ask if screech stacks with demoshout. but I think I already found the answer by googling (and he answer was yes it does stack)

Zure · 10/24/07, 4:42 PM

These numbers seem to indicate that for any reasonable number of physical dps, imp expose armor is now very much better than sunder + rupture. 475 armor pen is about 3.5-5.0% more damage per physical dps (depending on equipped armor pen) and rupture seems to account for about 6% of the average rogue's damage output.

If expertise geared tanks can hold agro with weaker devastates (possibly with SS, Revenge, Shield Bash, Dev, repeat with 2x dev instead of shield bash), that's a noticeable jump in raid dps.