AC Mitigation

[Gorroth]

Haven't posted this in bugs because I know the devs are aware of this but I'd like to know whether it's been said if AC mitigation will be looked at once more in the following months.

Here's a short example of what's currently wrong with it, for those who don't know.

I was in the area of an old exping solo camp I used from 44 to 50. As a shaman, with close to 200 more AC than what I then had (from 660 to 860), I figured I'd give it a shot, see how easier it's become.

Well it hasn't, or barely. Mobs still hit me for their max most of the time or just miss; there simply isn't much AC mitigation going on, as in, mobs don't hit me often for lower amounts of damage, it's either all or nothing.

The mobs I ran this short test are still dark blue at level 50 (range of 38-42). On light blues and greens, it seems better, although I get a lot of missed attempts; mobs still hit for their max frequently when they do manage to land a hit. It seems that, as of now, there are hardly any benefits found in stacking AC which puts "tankier" classes even more at a disadvantage compared to pure casters (fierce pets, OP charm) when it comes to tackling content.

With Kunark slowly approaching (hopefully), the level cap being raised and mobs hitting even harder, will this get addressed at some point?


Edit: added parse on page 2.

[Derazor]

Well, I dont know if AC works or doesnt work really. But, you need data, and generalizing about the amount of damage you take from mobs isnt it. You need to parse their DPS over an extended period of time with 660 ac and again at 860ac. If the parse comes back nearly identical, you can assume AC is broken.

[Gorroth]

Oh please, it's well known AC is wonky, any tank will vouch for it. I just gave an example from my own experiences to illustrate what's going on.

What I want to know is if it's been said that it'll get looked at again in the near future.

[Kainzo]

I'm sure it will most certainly be looked into. While they're at it - hopefully the hit ratio will be given another glance too.

[Bumamgar]

Oh please, it's well known AC is wonky, any tank will vouch for it. I just gave an example from my own experiences to illustrate what's going on.

What I want to know is if it's been said that it'll get looked at again in the near future.

"well known" or not, without parses and empirical evidence, this sort of issue tends to sit at the bottom of the pile, since what you are really saying is: "I can't be arsed to do any research into the subject. Dev's, please spend a lot of time doing parses and verifying my anecdotal claims, thanks!"

If you provide parses, statistics and details, then it becomes:
"Dev's, here's a proven issue that I've thoroughly researched. Please take this data and use it to aid your investigation into the problem."

That tends to move issues much much higher in the priority list, because it means the dev can focus on checking the code and working on the problem, instead of parsing logs and verifying that the problem exists in the first place.

[Ripcord]

Im a ranger and stack ac whenever its possible, buffed I'm close to 800 at lvl 35.I feel it does make a noticable difference, but the max hit thing can be tricky when you are watching the numbers come in. Mobs are intended to hit hard ensuring tanking never becomes trivial unless you have like 20 lvls on the mob. I'm not sure about actual numbers tho and agree a parse is needed to really prove anything, but I've found in stat gear with low or no ac I'm much gimper at tanking for any substantial time.

[Gorroth]

The big survivability difference I've noticed when stacking "survival gear" comes, for the most part, from the increased HP pool, allowing you to withstand more attacks. Since mobs don't hit very fast (1 round = 6 seconds), even surviving one or two more rounds due to a larger HP pool is a noticeable increase in survivability.

To the naysayers, remember that the role of AC has already been addressed at the beginning of the server, when it was much worse. The current "solution" was found to be adequate for classic content, but when mobs start hitting harder and faster, the role of AC will have to be looked at once more, unless we just end up having all raid content tanked by charmed mobs.

Also, if anyone knows of a suitable parser for EQ, I'll gladly spend the time gathering data and post the results.

[Uaellaen]

they are working on it yeah, i think it was our nuclear physicist that does ... and he asked for parses parses parses to help him out ...

[Uaellaen]

http://www.project1999.org/forums/showthread.php?t=7775&page=22

here is the bugs ... thread / flame, whatever you wanna call it

[Lazortag]

How come no one ever complains when mobs are missing them constantly or doing very little damage?

[girth]

Well for me Lazortag, it's because when that happens, I am fighting green cons.

[guineapig]

Well, I dont know if AC works or doesnt work really. But, you need data, and generalizing about the amount of damage you take from mobs isnt it. You need to parse their DPS over an extended period of time with 660 ac and again at 860ac. If the parse comes back nearly identical, you can assume AC is broken.

There have been many threads about people complaining about AC mitigation and the like.

In every instance the thread ended up dying and getting buried when it came time for people to submit test logs or stats from parses.

Deranzor is absolutely right. It's not enough to start these threads. You need to provide this info to the devs if you have any. They don't get paid for this you know.

[km2783]

Seriously people. Provide evidence or be quiet ;) NOTHING will get done unless you offer up more than generalized statements.

That said, Haynor I think has stated that they/he will be looking at mitigation again.

How come no one ever complains when mobs are missing them constantly or doing very little damage?

Does this happen to you? parse and report it :D

[girth]

Shit guys, Haynar made a post that said he worked on mitigation tables for like 10 hours or so the other weekend and THAT WASN'T even his main project at the time.

He knows, probably better than 99.9% of us, that there is an issue and what it is. He also is putting a lot of effort into it he says, and I'm sure we will see the fruit of his labors soon enough.

Give him time.

[Uaellaen]

Give him time.

Or better, give him parses from all of your levels fighting mobs ...

[Gorroth]

http://www.project1999.org/forums/showthread.php?t=7775&page=22

here is the bugs ... thread / flame, whatever you wanna call it

The thread was about melee miss rates, from PCs to NPCs.

I'm asking about the reverse, that is, incoming damage from NPCs to PCs. More specifically, how AC determines not just your chance to get hit, but your chance to get hit for lower damage values.

[Uaellaen]

I think its related to the melee formulars, wich i think was discussed in that thread, i wont read 23 pages again thou >< especialy since its mostly gibberish ...

but as far as i remember haynar stated that he is working on melee damage in general ...


EDIT: yeah arround page 20 they take AC into the account ... maybe even earlier

[Gorroth]

Parse Results - 30.05.2010

Method and Parameters

Parser used: EQ Companion (http://www.eq-companion.com/)


Gorroth
Level 50 Troll Shaman (PC)

V.S.

Captain Boshinko
Level 40 Human Warrior (NPC)


Two separate logs of 5 combat sessions each have been recorded.

Combat log 1: AC 709 ¦ AGI 92

Combat log 2: AC 818 ¦ AGI 99


An earlier attempt to further reduce my AC while retaining part of my HP pool (needed because the mob hits hard) was discarded due to my AGI dropping too low. The thinking was that AGI might affect survivability by other means than simple AC values, i.e. increasing pure avoidance without taking AC into account, like a secondary avoidance check. Therefore, low AGI would lead to biased results.


Relevant Values:

DPS

Slash Hit Average (AvgH)

Hit Percentage (Hit%)

Number of Max Hits / Total Damage


Results

Click for raw data (http://www.sendspace.com/file/9he80z).

Combat log 1: AC 709 ¦ AGI 92
http://img210.imageshack.us/img210/2492/boshinkolow709ac92agi.jpg
Combat log 1: AC 709 ¦ AGI 92

------------------------------------

Combat log 2: AC 818 ¦ AGI 99
http://img171.imageshack.us/img171/9610/boshinkomed818ac99agi.jpg
Combat log 2: AC 818 ¦ AGI 99


Rough estimate of percentage of total damage due to max hits: 28-40%


Conclusions

I've never played a melee class on live. I did, however, play a shaman for many years and have endured a lot of hits and blows and have a fair idea how "tough" a class should be.

Shamans aren't tanks. Nevertheless, they tend to get hit a lot and thus, appropriately benefit from the "chain" mitigation table.

Conjecture: As I said, I never played a melee on live and thus and therefore do not know the proper values of how everything was when it comes to tanking. What I do know is that in the hierarchy of enduring damage, Plate > Chain > Leather > Cloth. This makes sense, as the classes higher up on the mitigation tables, that is, benefiting from a greater protection from harm, are those who tend to get hit the most (except enchanters, one of EQ's cruel jokes).

I am aware of special, intermediate mitigation tables for special classes, such as the monk, but I believe most classes were rigidly contained within this hierarchy.

Hence, the shaman is supposed to be able to take hits. Not as good as a tank (war/pal/sk) or a bard or cleric, who lack the HP of tanks, but on equal footing with rogues and rangers (they have taunt, so they can take hits, remember?).

As a shaman, back on live, noticing the tendency of most of my spells to draw aggro, I gave preference to survival equipment. Furthermore, due to the shaman's hp-to-mana conversion ability, lost HP meant lost mana. Thus, AC > HP > Wis > Sta > Mana became how I chose equipment. I was never interested in parses because I didn't need to tank mobs during raids, all I knew is that the tougher I was, the easier soloing became.

And it has served me well, very well in fact. Everyone has heard tales of shamans soloing tough encounters, such as Sebilis, Skyfire or Western Wastes mobs. I was one of them.

Back to P99, where I sought to apply the same "toughness über alles" mentality, with mixed results. I had the feeling that, although my HP helped me withstand more blows, my AC wasn't hindering those blows from hitting me as hard.


TLDR VERSION

As the four relevant statistics show (DPS, AvgH, Hit%, #MaxHit / TotDmg), and keeping in mind the relatively low number of events coupled with the presence of statistical outliers, there is virtually no difference between the two parses.

This implies that, as a shaman, going from 709 AC to 818 AC has no effect. A 100 difference in AC has virtually no effect.


This means two things:

- My AC is nearly capped at 709 (chain mitigation). This is ridiculous, as there'd be no point in pursuing better gear in the later two expansions. Kunark and especially Velious gear puts old-world equipment to shame.

- AC has a negligible effect. Instead, what explains the current "tankability" of a tank is his HP pool, innate mitigation table and possibly other parameters which influence avoidance (e.g. dodge, parry, agility, etc.). This should not happen, as I vividly remember tanks trying their best to hit high AC values by any means. The "AC is king" became somewhat of a mantra during the Velious era.

As I said in my first post, the current situation is acceptable for old world encounters. Once Kunark and Sky open, people might start to notice an overall lack of survivability.

So, once more, I'm asking anyone if a developer once said that AC mitigation will be looked at in the near future.


PS: I thank the devs for their hard work and respect the sacrifice of their free time and ask them to forgive me if what little effort I spent in producing these parses brings them to, once more, to spend time on such a complicated matter.

[Uaellaen]

dude ... right in the thread i linked ... aeolwind on page 20 ...

I'll double check some values, but I think when I found the way to tweak the melee values that I still left it actually short of what the rest of the emulator uses (Yep, I'm a dick =D). I'm on the process of recalculating AC's around the raid content and I can tell you that it is WAY off and most likely is the extra 2-3% over we're seeing AND the missing amount that I left for tweaking in the DB.

[Gorroth]

I've read Aeolwind's answers on page 20 of that shit fest of a thread and it seems to solely pertain to PC => NPC interactions, not NPC => PC.

I guess my question still stands.



Edit:
I'll double check some values, but I think when I found the way to tweak the melee values that I still left it actually short of what the rest of the emulator uses (Yep, I'm a dick =D). I'm on the process of recalculating AC's around the raid content* and I can tell you that it is WAY off and most likely is the extra 2-3% over we're seeing AND the missing amount that I left for tweaking in the DB.

* Players aren't "content", right? =p


His other posts clearly talk about PC to NPC interactions.

Or are the formulas used for mob AC and player AC the same?

[fabric9]

The AC softcap (beyond this point, AC from gear gives less return - the only exception being if it's a shield, as it goes beyond the softcap) was raised with every expansion. I don't have a good enough memory to recall what it was in Classic, but I'm fairly sure it wasn't over 1000 for plate classes.

I wouldn't expect to see a much better return for shamans past 700-800AC in classic, but I would expect that number to rise with Kunark. Do your tests at lower AC if possible to find the softcap.

[Aeolwind]

Yes, I'm making mobs harder. Everyone was so encensed over the OOC changes I slid in under the radar. <3. :D

But yes, Haynar has been looking at AC's. Combat in general as it were. I'm hitting AC from the otherside and making sure mobs have the right amount.

[Gorroth]

Thanks a lot for clearing that up! It's good to know it's being worked on. =)

I might do some more parses at lower AC values and re-do my original parses at a later date. The problem with going lower in AC is that I also lower my HP pool, which limits my safety net and cuts into the parsed time due to having more difficulties managing my total HP while getting hit (i.e. corpse runs suck!).

[YendorLootmonkey]

Gorroth -- kudos for correctly going about this and doing the parsing to provide quantitative data!!

[Audacious93c]

Every since I EXP'd in Oasis quite some time ago, I knew this. Hell, spells hit for a hell of a lot more here on P1999 as well. However, its kind of hard to prove something when the original isnt what it used to be by any means.

As for the raid content; itll be a thankful addition to the game in my opinion. Tanking gods and dragons is down right laughable. If it wasnt for Death Touch... ;)

So yes, please bring on the AC changes! I know Im going to appreciate the extra challenge :) Thanks again for your free work broseph.