AC Parse, CS3 & ID5 & PE & DI5

[Nedrom]

Thanks goes to Kavhok for allowing me to run these tests.
These are the tests I did overnight. Every fight is 90 minutes long.

Level 65 Iksar Monk
No AA, no worn AC, 200 AGI

This is vs Level 60 mob from the Test NPC Summoner in the Arena on test server. I had myself backed into a corner so there were no "out of range" messages.

Raw AC AC Min % Max % Ave

0 853 4.08% 37.55% 360.19
100 1010 12.73 18.79 301.83
100 1010 12.46 18.20 302.44
100 1010 13.05 18.41 301.40
100 1010 13.15 18.41 301.61
200 1167 16.35 14.81 283.29
200 1167 18.33 12.34 271.20
200 1167 19.48 9.70 267.03
200 1167 19.76 10.85 264.73
200 1219 20.14 10.30 262.62 --- Level 70
200 1219 22.27 9.78 256.57 --- Level 70
300 1325 19.10 11.67 269.88
300 1325 20.12 11.49 265.62
300 1325 24.87 8.78 247.58
300 1325 22.27 8.06 250.75
300 1377 25.65 8.26 245.51 --- Level 70
300 1377 25.63 7.79 242.50 --- Level 70
400 1482 22.47 9.55 259.66
400 1482 23.81 8.92 254.06
400 1482 25.86 7.32 243.02
400 1482 25.08 8.09 244.75
400 1534 27.01 7.96 242.14 --- Level 70
400 1534 28.84 6.49 235.55 --- Level 70
500 1639 23.87 8.35 250.20
500 1639 26.35 7.64 242.55
500 1639 30.11 6.00 229.45
500 1639 29.35 5.72 234.23 --- ID5 #2
500 1639 30.78 7.44 233.65
500 1639 28.54 6.10 233.76 --- PE #2
500 1691 31.12 6.46 230.08 --- Level 70
500 1691 30.94 4.96 225.19 --- Level 70 ~1 hour parse
600 1798 25.71 7.28 242.67
600 1798 28.41 6.32 234.29
600 1798 31.47 4.68 224.32
600 1798 32.27 5.79 226.78 --- ID5 #2
600 1798 31.01 5.59 230.88
600 1798 32.54 5.72 226.86 --- PE #2
600 1798 33.29 5.63 227.40 --- PE #3
600 1850 31.90 5.18 226.22 --- Level 70
600 1850 33.99 4.74 223.42 --- Level 70
600 1850 34.06 4.83 221.75 --- Level 70 ~10 hour parse
1300 2951 46.48 1.33 191.51 --- Level 70

*all tests are 90 minutes unless stated otherwise

black = no defensive AA

blue = CS3

red = CS3 & ID5

purple = CS3 & ID5 & PE

yellow = CS3 & ID5 & PE & DI5

Are the parses all at 200 Agi?



and for the math wizzies, how do we tell the difference between an AC soft cap and the maximum benefit AC will give on said mob?



Without crunching numbers, the peak seems to have been met somwhere around the 300 worn AC mark, no?


http://www.geocities.com/evolvedbeyondyou/Amaru.txt</p>

That's interesting. It seems that AC is still a significant factor up to at least 1482 AC, based on these parses. I did notice that while the evasion in the first 4 or so parses stayed pretty much statistically the same (which I would expect), in the last parse it jumps 2-3%... was any part of the +400 ac added in that parse from an avoidance mod on an item or additional agi? If so, it would explain why the evasion percentage increased. It would also indicate that it took even less mitigation AC for the noted damage reduction than one would be initially led to believe by looking at the ac numbers in the previous parses as compared to this one, because the avoidance ac from the avoidance mod or agi would still show up as an increase in the shown AC number.



Euphrates


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=eup hrates>Euphrates</A> at: 5/13/05 10:09 am

[Ughbash]

Intersting.



What we see is a LARGE reduction in DPS when you add the first 100 AC, a smaller reduction when you add the second hundred, another small reduction when you add the third hundred, and then a larger reduction when you add the fourth.



First I would like to thank you for taking the 6 hours+ to do this parse.



Secondly I would like to note that this does NOT follow what I woudl expect with a softcap and then limited returns over that softcap.



If the mobs attack is high enough I would expect a linear, or close to linear, reduction until the softcap is reached, and then a slower yet still linear (from the point of the softcap) decrease in DPS.



If the mobs attack is NOT high enough, I would expect a sharp decrease in DPS as we add AC (similar to what you experienced adding the first 100 AC), but then a markedly less decrease in DPS and a flattening after AC got to a point sufficently beyond that AC.



The only reason I can think if why you had the big jump between 300 and 400 is if that wound up a sweet spot on the attack vs AC of the mob. Basically its attack was high enough that the movement between 100 and 200 and 300 was negligible but 400 put it over the top.



Is there any chance you could do two more test, the same mob with 500 and 600 worn AC to see if it creates a larger or smaller differnce in DPS.



Thank You,



Your Kinder Gentler Iksar,

Eagles


</p>

[Nedrom]

ill add the +500 and +600 ac later today



as for the tests, yes 200 agi for all



the only change from one test to the next was purely +100 ac per test only, nothing else changed at all.


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/13/05 10:16 am

Since % chance of a mob maxhitting on you is related to your AC and its attack, there will be a softcap for individual mobs beyond which you already mitigate the hits down to min damage and therefore no longer see a benefit from additional AC. To illustrate with fake numbers, say a mob has 50 atk and you have 8000 AC, you will mitigate nearly every hit down to min damage. Since this is the case, you will be unable to find the monk AC softcap when parsing against a mob with attack so low that the mob softcap is lower than the class softcap. However if you can find mobs that have their softcap at higher than the monk class softcap, you would then be able to see at which point AC stops having a noticeable benefit. Since return on AC over cap may be much more significant now, it may be necessary to do multiple parses against said mobs with different AC levels in order to quantify how much a certain amount of AC increases hit mitigation for that mob, then you would be able to find the AC level above which the return changes, which would be the monk AC softcap.



Euphrates


</p>

[nduma]

Very nice work Ned and thanks. This should provide good fodder for all the people that live by the numbers.



I know the last change definitely has had an impact on me.


</p>

[Trukoflanys]

the jump in dps from 300 to 400 is from the avoidance increase.



Nedrom can you post the Avg hit for just punches at each of the ac levels.


</p>

[Nedrom]

<blockquote>Quote:<hr>Nedrom can you post the Avg hit for just punches at each of the ac levels.<hr></blockquote>



Replaced Totals view with [punch] selected to show proper average, before it was showing [total] damage.



Sorry about that.


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Zonn]

Graph updated to current data and the color coding set to match the one Nedrom is using.



http://www.stasis-guild.com/MitigationParses/AverageDamage.jpg



P.S. Great stuff Nedrom. This is by far the most in depth mitigation parsing I've ever seen and it's really helpful showing the trends of our AA and gear choices on damage taken. Great job.



Oh, I know you've spent an incredible amount of time doing this, but if you continue to have the motivation I'd Love to see some Sheilding Parses with ID 5. I'd like to see how much further we can trend our damage down with it. Say at 10% and 20% if Kavhok could hook it up. Either way, I know I appreciate all the hard work.


<span style="color:red;font-family:helvectia;font-size:large;">Axiss Glox</span>

<span style="color:red;font-family:helvectia;font-size:small;">65 Transcendant of Stasis</span>

Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=10004)

Axiss@Stasis-Guild.com (Axiss@Stasis-Guild.com)</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=zon n>Zonn</A>* http://www.stasis-guild.com/stasis/images/stasis_icon.gif at: 5/28/05 1:18 pm

[samoa79]

Neato...Thanks Ned. Well...no worries about taking away our newfound better mitigation due to the fact that at my AC lvl 1450+ we are still hit for 9.55% max hits. I can get close to 1600ac w/ buff so Im eagerly awaiting the 1600 parse.


<div style="text-align:center">http://www.resilienceguild.org/files/Sigs/Kazmarsig.gif



My Stuff (http://www.magelo.com/eq_view_profile.html?num=243450)

Stone Fist</div></p>

[Redcloud WK]

It's a 22% reduction of max hits from 300 to 400 raw ac still. Nothing to sneeze at.



Thank you Nedrom.


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=red cloudwk>Redcloud WK</A> at: 5/13/05 11:56 am

[Evil Homer]

Soft cap at 65 appears to be ~193 raw AC.

Return over the cap appears to be 1/10 up from 1/40.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=evi lhomer@monklybusiness43508>Evil Homer</A>* http://www.urban-village.co.uk/thumbnails/Bruce%20Lee.jpg at: 5/13/05 12:18 pm

[Sarshish]

I would expect that the point of significant return would shift depending on mob attack rating. As the AC passes the mobs attack value then returns would level off. So, to properly see the curve you need to do the parses with an attack rating either much higher than your AC adjustment range or much lower. Otherwise the effect of passing the softcap will be impacted by the effect of passing the mobs attack score.


<a href=http://www.magelo.com/eq_view_profile.html?num=227968>Sarshish Silverstorm (Leezzard)</a></p>

[Nedrom]

added +500 ac parse



is this sufficient data or do you want me to re-parse from 0->600 AC with Combat Stability 3 added?


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/13/05 1:07 pm

[cebekis]

Very nice Ned! Many thanks for the effort on this. You've spent a lot of time on these parses and I appreciate it. Looks like the recent mitigation changes should significantly help.



Now stop testing and go raid! <img src=http://www.ezboard.com/images/emoticons/tongue.gif ALT=":b">










Baitstick (#)

Taxicab (#)

FA Dancer (#)

Medic! (#)

</p>

[Ughbash]

Would apprecaite it if you would post the 600 one. It looks jsut eyeballign it that the 400 one may be off slightly due to the increased avoidance. This is causing the 400 and the 500 AC tests to show the same DPS even thoguh there is a differnce in the AVG Hit.



And homer not really enough points to get a fair estimate of the softcap yet, and I would expect if the return was 1/10 rather then 1/40 the slope would have changed by a factor of 10. Looks to me like it might be closer to 1/3 or even dare I hope 1/2.



One last request if possible. Since you are obviously doing this on beta or test or in a controled environment where you can add exacting AC. Would it be possible to try this with two additional mobs. Specfically I would like to see how the ac affects a monk when he is fighting a Kyv from Kod Taz (notoriously high accuracy) and a Riftseekers trash mob.



Again thank you for the work you have put into this.



Your Kinder Gentler Iksar,

Eagles


</p>

[Zonn]

Any other data point you want to parse would be good, I'd especially like to see some more granular parses in the 100-200 region. I think it's clear your return diminishes just over 100 and is a completely linear reduction after that.



It would be nice to see where the cap moves with aa mitigation too.


<span style="color:red;font-family:helvectia;font-size:large;">Axiss Glox</span>

<span style="color:red;font-family:helvectia;font-size:small;">65 Transcendant of Stasis</span>

Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=10004)

Axiss@Stasis-Guild.com (Axiss@Stasis-Guild.com)</p>

[Soygen]

Good stuff, Ned.


-Soy

Stronger than never, ever before! SOYGEN is a drug against war!</p>

[Nedrom]

added +600 parse


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Evil Homer]

<blockquote>Quote:<hr>And homer not really enough points to get a fair estimate of the softcap yet, and I would expect if the return was 1/10 rather then 1/40 the slope would have changed by a factor of 10. Looks to me like it might be closer to 1/3 or even dare I hope 1/2.<hr></blockquote>



There are enough points...



Anyway, with the added point, the softcap went down to 187 raw AC? I'm missing something, the 600 Raw AC seems to have broken the linearity of 1/10 so it looks like 1/14ish over the cap now?



Math help. Simple coefficient and linearity, an excel spread sheet can do it. Double check me please.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Raptyr Tailfist]

/bows in reverence.


http://members.cox.net/jkay/images/ateamsig.gif



My Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=43561)

</p>

[Ugle]

First of all Ned, I want to thank you for all your efforts. You did what you could reasonably do.



However in my mind, all you've done is find the ATK cap....for one specific MOB....who is 10 levels lower then you are. I'd lay PP out that higher level/ATK MOB's are still going to hit for max at higher AC levels. At 1550'ish displayed AC, WoS nameds hit me for max more then min. However none of the trash MOB's in WoS/RCoD did. Who hits for max more often, Tormax or Vindi? It was the same problem I saw with Thepps parses. Complete as they were, it showed what you could expect....at different AC levels....On a specific type of MOB in a specific zone. Take a MOB who's max hit is the same from PoP and GoD, and the damage output/second is generally quite large. Add a third MOB whose max hit is just as large from OoW, and it will land somewhere in the middle.



My real rant is....why SOE, why? Make a Monk NPC. Give it super regen. Set it's AC to a certain point. Take it into MPG and let a L70 staticfist beat on it. Then an elite Kyv. Then take him to RSS. Let a Pyrilian beat on him. Take different parses based on different set AC levels. Hey I've been gaming for a long time, I've used Diablo Mods and screwed with my Bards Tale files, I'm sure your devs could get in and twink an NPC to have whatever stats you want it to have. I don't understand why it takes years of complaining about our damage taken, and numerous hours/thosands of log file lines for us to tell you something....that one bored Dev could have told you in a day of testing.


Natassi Dogym (http://www.magelo.com/eq_view_profile.html?num=198315&resize=true)

Dawulfie Duklin (http://www.magelo.com/eq_view_profile.html?num=406812&resize=true)

Lacesout Ferdafieldgole (http://www.magelo.com/eq_view_profile.html?num=1172520&resize=true)</p>

[Nedrom]

Ugle,



The monk in the parse is a dummy monk, level 65, 200 AGI, no AA, max defense skill for level 65 and the AC values I listed, not what is in my magelo, so the test NPC was 5 levels under.



The "only" variable is AC. Nothing else.


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/13/05 3:37 pm

[Ugle]

Ok, sorry I missed that point. 5 levels lower. That's the only point in dispute? Like I said I appreciate you taking the time to do this, I'm not trying to impugn you or your efforts. My question is how applicable the AC softcap you found on this particular MOB is going to be for me when I'm in RSS or the Nest and the tank drops and I have to grab it. Heck we don't even know how these numbers would change if you were L70. Like I said that's my real rant, SOE could parse 100 different level/AC/AA combos anytime they want, on any MOB they want. And yet due to damage output and the fact that EQ IS NOT OUR FULL TIME JOB it's not realistic for us to do the same. Why are we having to tell them what it's like to play their game?


Natassi Dogym (http://www.magelo.com/eq_view_profile.html?num=198315&resize=true)

Dawulfie Duklin (http://www.magelo.com/eq_view_profile.html?num=406812&resize=true)

Lacesout Ferdafieldgole (http://www.magelo.com/eq_view_profile.html?num=1172520&resize=true)</p>

[Nedrom]

These parses show that AC works, even with no Defensive AA and low end AC levels. What level the mob is, what level their attack is, and what zone they are in are irrelevant.



It's implied that high end zones like RSS are going to hit hard and fast, just like in Anguish or Txevu or Uqua or Tipt or Vxed. If every NPC had the same attack it would be a very boring and easy game.



Parsing against another NPC will prove nothing.





PS: From the graph that Axiss made, it appears that the softcap is a lot lower than everyone expected. There is a steep decline from 853 to 1010, and after that it evens out showing the "return over cap" working nicely.



Now I don't see how any monk can complain about the changes made and say it doesnt affect them because it does indeed affect all of us and makes all of us better monks and will make all of us take damage better.



Imo, great job Kavhok. Lets move on.


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/13/05 4:15 pm

<blockquote>Quote:<hr>Parsing against another NPC will prove nothing.<hr></blockquote>



I very much appreciate the posts and the logs you've done. However, I'd say that while we have a solid conclusion about worn AC at one given level of mob attack. The real test should be seeing on a very high attack mob if our AC has any meaning. My gut instinct is that our softcap is low enough that high attack mobs negates any gains we've recently made in AC. (As in, you need 1800AC+ before any high attack mob does anything but max hit you 90% of the time.)


















</p>

[Ugle]

AC has always worked. An 1800AC Monk gets hit for less then a 1K AC Monk. The question is at what point the extra AC starts showing limited/no returns. So since parsing is irrelevant according to you, then you can definitively state that the point at which your L65 Monk showed diminishing AC returns on that MOB is the same point that I will at L70 in RSS on a Pyrilian? It's a simple answer, yes or no. I don't have 4 hours of parses on one MOB, but from a fair number of fights, I can tell you that my hit distribution, leaving AC consistent, changes on higher level MOB's towards the top end.



I've got a Warrior friend who's stuck in the 80's. Well guess I should say 90's. His AC is <1600 unbuffed. He doesn't think he needs any more AC, because he's over the old Velious caps. He mitigates really well in places like the GoD sewers (specifically thinking of his parses in Sewers1 where they are 1K hitters, some unslowable). He gets hit really hard in MPG, even though max hit is pretty close in both places. According to you, his <span style="text-decoration:underline">average hit distribution</span> should be the same both places, why isn't it?


Natassi Dogym (http://www.magelo.com/eq_view_profile.html?num=198315&resize=true)

Dawulfie Duklin (http://www.magelo.com/eq_view_profile.html?num=406812&resize=true)

Lacesout Ferdafieldgole (http://www.magelo.com/eq_view_profile.html?num=1172520&resize=true)</p>

[Goldmund TargaRi]

Awesome job Nedrom, thanks for taking the time to do this.


http://www.goldmundtargari.com/morosig.jpg (http://www.magelo.com/eq_view_profile.html?num=65946)</p>

[Redcloud WK]

Great job, Nedrom and thanks a lot.


</p>

[Buhd]

Thanks greatly for takin the time and to Kavhok for setting you up.



Definately interesting information. Im lvl 65 with almost all the defensive AA available so I feel the information is relevant to me.



I do agree that it would be interesting to see the difference vs a mob with a drastically higher attck value though. Since this is all set up already, we shouldnt let that opportunity go by.



We can all make assumptions what will be the results..but until its proven, thats all they will be.






Sensei Brotor Tiller

unRetired

Rathe

League of Levity



</p>

[Shazzamx]

Well done Ned, good effort. If it's the holy grail we hoped for or not I don't know. But it's good enough for me and shows me what I needed to know, that the patch WORKED.



The talk about details is to me irrelevant (offcourse I have 1991ac unbuffed) because as already said it shows that our new ac returns works.



That a 1400-1500 monk will still get splattered in /rss and the nest should be a foregone conclusion. a lvl 70 Paladin friend of mine (dosen't have all his def aa's) have around 1700ac and can't tank in the Nest or Rss, so offcourse monks at that ac would also get pounded.



However monks with 1500 ac can now comfertably stand toe to toe and pull in zones where a tank of 1500ac is needed to tank and that is what we wanted wasen't it?


</p>

/applaud Nedrom.



I think we finally got what we wanted on mitigation/AC returns, I am satisfied on this issue.



I would like to see this community shift its focus to balancing dps, especially monk specials.



And upping the spawn rate of our epic 1.0 npc's. I hate Raster.



Time to give my alts a rest I guess and get back on the monk. With a few more AC points I should be able to solo a bit and duo effectively, albeit slowly (as opposed to glacially), which is what I've wanted all along.


Master Maark Goodsoul

58 Master of the Ninth AA

Permanoob Monk of Quellious

Karana





Maark the Monk (http://www.magelo.com/eq_view_profile.html?num=900571)</p>

[Daitan Niyan]

/bow Ned



Thanks man, for taking the time.



See ya in game =)



Regards,


Signature Graciously provided by the wonderful Immeni (http://www.rockin-originals.com)

http://thelionspride.net/images/daitan5.jpg (http://www.magelo.com/eq_view_profile.html?num=226108)

Remember, pulling is all about one goal... to bring mobs without wiping out your group or raid. The complete wipeout of a group or raid is the Greatest failure of any puller. It is better to die than wipe out your group or raid. Expect to die. Expect to die alot while learning. But with time, practice and experience, miracles are possible.

Do the impossible.

-Laotzu Qigong

</p>

[Shotan]

Good going Ned!



And its nice to see our AC doing something! Gives me hope.


http://knights-renown.com/signatures/shotan.jpg
<a href="http://www.magelo.com/eq_view_profile.html?num=1201869">Shotan - Monk of Qeynos
</a>
Following the teachings of Master Lu'Sun




Razeer - Druid of Surefall (http://www.magelo.com/eq_view_profile.html?num=1205005)

Follower of Karana's Will


Firiona Vie Server, Everquest

Token Monk of Wayfarer's Elite (http://p073.ezboard.com/bwayfarerselite)

My BLOG (http://knights-renown.com/b2)</p>

Excellent job, thanks for the effort!



/em claps and surreptitiously adds a parse with all Luclin and then all PoP defensive AAs to Ned's "to do when I am bored" list <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">


Scalia, 65 Monk (http://www.magelo.com/eq_view_profile.html?num=71792)




Scabon, 65 Shaman (http://www.magelo.com/eq_view_profile.html?num=576916)</p>

[Dythan Dabrave]

Thanks for the time Neo, great job.



Still really curious as to the new soft cap, i hit about 1550ac raid buffed, and trying to decide if i need oow migitation aa's or not.



been doing defence for a bit, would like to up my dps =P


Dithan Dabrrave

70th season monk

Forsaken Visage



The only difference between a Hero and a Fool is Success.</p>

Do the exact same thing with a warrior and a paladin. And every other class too... if that's not asking too much. =)


</p>

I've noticed the changes in mitigation, I am happy. I feel like I am back in the kunark era and have no mitigation nerf


<div style="text-align:center">http://www.sanctora.net/yada/heatedsig1.jpg (http://www.magelo.com/eq_view_profile.html?num=1183186)</div>

</p>

[Zonn]

I'm keeping the graph up to date as data is available. It's nice to see the average damage scales so nicely downward with more AC and we can clearly see the softcap making the plot turn. I'm not going to attempt to calculate exactly where that is, but the low 100's worn AC looks right since the slope between 100 and 200 is a good bit higher than from 300-600 but just barely more.



I'd be curious to see the same parse run with a level 70 to see where the softcap is at that level. I think it should be higher, correct? I could plot it out along side the 65 data to say for sure.


<span style="color:red;font-family:helvectia;font-size:large;">Axiss Glox</span>

<span style="color:red;font-family:helvectia;font-size:small;">65 Transcendant of Stasis</span>

Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=10004)

Axiss@Stasis-Guild.com (Axiss@Stasis-Guild.com)</p>

Would using a MOB with a known HIGH atk rating and then upping the ac rating on the monk until it it is obvious it has met the atk and then doing the division from where the ac is and where we know the atk is give any new information? I'm not entirely sure that is coherent and I was never big into the parses when I was playing anyway, but I was reading through and I wasn't sure if that would be a better way to do this.


<span style="color:navy;font-size:small;">Brother Desten: Retired</span></p>

[Kydaan]

If they're willing to copy a character to test and allow them to parse, why won't they just publish the AC softcap numbers?


</p>

<blockquote>Quote:<hr> If they're willing to copy a character to test and allow them to parse, why won't they just publish the AC softcap numbers?<hr></blockquote>



If I'm not mistaken, it looks like some people enjoyed crunching the numbers themselves.



It's a game, some people never do anything but bazaar, some only tradeskill, some crunch numbers, some get gear, some level, some AA...



EverQuest is one Hell of a good game with a ton, a frigging ton of content and things to do. I use to think they were lieing when they said, you don't have to, they weren't.



What they should have said is, "Have you tried one of the many other aspects of the game?"


http://i5.photobucket.com/albums/y193/tauzii/TauziiLFG.jpg (http://www.magelo.com/eq_view_profile.html?num=852244)

</p>

AC Ave hit Decrease AC Increase neg av dmg per AC

853 360.19 n/a n/a

1010 301.83 58.36 157 0.3717

1167 283.29 18.54 157 0.1180

1325 269.88 13.41 158 0.0848

1482 259.66 10.22 157 0.0650

1639 250.2 9.46 157 0.0602

1798 242.67 7.53 159 0.0473



ok so 100 "worn" ac gives a shown increase of roughly 157 ac.



The Softcap appears to be "around" ( not exactly ) the 1167 Ac mark as from then onwards the average hit decreases relatively steady ( although ever decreasing ) ammount per 100 AC.



As the Average hit decrease per ac is constantly dropping im not sure that they can say "you get a set return of real AC for shown exp over the softcap". If that was true the negative damage per AC gained should be a constant figure, but these parses show that the more AC you gain you constantly get a lower return for it, so the return for AC gained seems to be fairly fluid on these figures.



I know the RNG would create a variable in the results, but as each set of figures is consistantly lower I think it can be taken from this that the return for worn ac constantly decreases, rather than as advertised being set at a constant figure i.e 1/40 or 1/10 etc etc.



slaps


<div style="text-align:center">http://balsamo.9online.fr/slappah





Slappah Deazz (http://www.magelo.com/eq_view_profile.html?num=102344)

Officer of Wind of Tranquility (http://pub28.ezboard.com/bwindoftranquility)

Spiritslap Soulhunter (http://www.magelo.com/eq_view_profile.html?num=611853) forgotten son of the tribunal</div></p>

[azathool]

The Softcap appears to be "around" ( not exactly ) the 1167 Ac mark



We already knew this was the softcap, pre any recent mitigation changes.



Since then they were supposed to have increased the softcap and the return, then again last patch increased the return yet again.



So.. What exactly did they do 1st patch, since it's looking like the current softcap is the same as it ever was.


<hr>
<div align="left">
<table border="0" width="100%">
<tr>
<td width="15%"></td>
<td width="70%">
<p align="center">[Stone Fist] Azathool (http://www.magelo.com/eq_view_profile.html?num=597)
<Nightwatcher (http://www.nwguild.com/)></td>
<td width="15%">
<p align="center">
[ Shm (http://www.magelo.com/eq_view_profile.html?num=4174)
| Clr (http://www.magelo.com/eq_view_profile.html?num=1024677)
]</td>
</tr>
</table>
</div>

</p>

[SkyKungfu]

<blockquote>Quote:<hr>I think it can be taken from this that the return for worn ac constantly decreases, rather than as advertised being set at a constant figure i.e 1/40 or 1/10 etc etc.<hr></blockquote>



Maybe the return of AC is consistant but the dmg mitigation that you get for it in return is decreasing = the higher you get the more AC is needed for the same amount of mitigation increase. Just a tought.



Sky


</p>

[Kydaan]

<blockquote>Quote:<hr>If I'm not mistaken, it looks like some people enjoyed crunching the numbers themselves.<hr></blockquote>

And no caster does? They put all the information in SPDAT because anyone that plays a caster is not going to be interested in running parses? Obviously, some of the information in SPDAT needs to be there to implement client functionality (e.g. mana cost) but is there any reason to put something like fizzle adjustment in there other than so it doesn't have to be parsed?






</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=kyd aan>Kydaan</A> at: 5/14/05 6:08 pm

Based on Zonn's graph, the monk AC soft cap is somewhere between 100 and 200 raw AC. If you eliminate the lines (since they are not data points) and instead extrapolate a best fit curve, it looks to me that the AC soft cap is just under 150 raw AC. This is higher than the 118 raw AC soft cap pre-April 12 (there is still no indication that cap ever changes with level) but appears to be lower than the 163 raw AC cap pre-PoP.



The exact soft cap won't be found unless that test is repeated at 5 or 10AC intervals from 100 to 200 raw AC.



Sear

Five Rings


</p>

Cant forget this is the softcap for level 65 so 70 will be higher.


</p>

[Ughbash]

Ok, I still believe we are getting a better return then 10 to 1 for AC.



I looked a little closer at the data rather then "eyeballing" the

graph like I did on my first post.



In my opinion to tell the return over the cap we have to compare avg hit below cap to avg hit after cap. If 100 ac makes a 10 point change before cap, and a 5 point change after cap then we are getting a 1 for 2 return, not a 1 for 20. Other people may find other methodologies better.



The following lists the change in avg punch damage (I excluded kick and bash) as AC changes.



First 100 58.36

Second 100 18.54

Third 100 13.41

Fourth 100 10.22

Fifth 100 9.46

Sith 100 7.53 (not as much so don't join the dark side)



Ok the change is not totally linear but this is to be expected when above the attack value of the mob.



First change is 58.46 So we use that as bar for how much change we would LIKE to see. Second change is 18.54 this category part of it is likely pre softcap and part post softcap so not as useful for determing due to granularity. The average of the other 4 is about 10. This means AC for 100 is approximately 1/6th as effective after the softcap as before (58/10).



Conclusion we have approximately a 1 in 6 return on AC above the cap. More data will give better results.



And Nedrom, Ugle does have a point that differnt mobs will give differnt hit distributions. If [warrior troll 16786] parsed on orc pawns he could say we mitigated almost every hit down to minimum same as a warrior would. While it would be a pain I still think it would be informative to see how "testmonk" parsed on a RS trash mob, and a KYV from KT.



Thank You,

Eagles


</p>

[Zonn]

For a quick estimate of the effectiveness of AC over the softcap, you can compare the slope of the line from 0-100 against the slope from 200-600. Doing this quick calculation in excel I get the slope from 0-100 is 5.75 times larger than from 200-600. So, I agree the 1/6th number is a good estimate for effectiveness over the softcap.


<span style="color:red;font-family:helvectia;font-size:large;">Axiss Glox</span>

<span style="color:red;font-family:helvectia;font-size:small;">70 Transcendant of Stasis (http://www.stasis-guild.com)</span>

Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=10004)

Axiss@Stasis-Guild.com (Axiss@Stasis-Guild.com)</p>

[Evil Homer]

Well, I'm happy with the numbers. I'm seeing the softcap at about 187 RAW AC and the return (an average I like the other numbers people have posted better) over the AC is about 1/14 on average.



The 160ish number seems low. Please help me understand how you get that when the jump from 0 - 100 and the jump from 101 - 200 are closer than farther, when compared to the other slopes?


</p>

[Ancaglon]

I'd generally expect diminishing returns on increasing AC even without the cap. That is to say, I'd expect the average damage taken per hit to fall off a lot less with increasing AC - a 1/X form of curve, as it were, approaching the Damage Bonus asymptotically.



My best feel for the form of the equation is



AvDPS = DB + ( Power / MitAC ) * 20 * DI



where



MitAC = WornAC + OtherFactors (eg Defense bonus, Monk Bonus).



Power = 'hitting power' of the Mob (what we sometimes refer to as ATK, but that's mixing accuracy and damage). This may well be independent of the DB and DI of the mob. So, some mobs may hit for a large range of values (high DI) but once you have a decent amount of AC those hits will mostly be for minimum damage - hence low 'Power'. Another mob might have a small DB, but may need insane AC to offset its' high Power.



Doing simple straight line intercepts on the 0-100 raw AC range vs 200-300 gives a figure of 111 raw. I don't mean that this is the softcap, just that, FOR THIS MOB, increasing AC much beyond that point won't have as much effect (effective softcap around there). For a mob with higher 'Power', the equivalent figure may be much higher.



Anyone spot any obvious problems with this idea?




</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=anc aglon>Ancaglon</A>* http://www.ezboard.com/image/emoticons_classic/devil.gif at: 5/16/05 5:16 pm

[Nedrom]

ancaglon you make my eyes crossed, i just like to bash stuff <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Evil Homer]

No obvious problems that I can see.



What I did was take what had already been done here (http://p201.ezboard.com/fmonklybusiness43508frm31.showMessage?topicID=1.to pic) and compared to ned's new parses.



<blockquote>Quote:<hr>

CLS - Class

AC - AC with virtue

SHLD - Amount of shielding on items

AVD - Amount of avoidance on items

DB - Damage Bonus of the mob. This is modified by shielding.

DI - Damage interval of the mob. This is modified by warrior innate 5% reduction of the DI.

MIT% - 0% mitigation would be max hits every single time. 100% mitigation would be min hits every single time.

ATKS - Total number of punch attacks

HIT% - Hits that landed divided by total punch attacks

MISS% - Misses divided by hits that landed plus misses not including blocks, parries, dodges, or ripostes

BLK% - Blocks divided by total punch attacks

DDG% - Dodges divided by total punch attacks

PRY% - Parries divided by total punch attacks

RIP% - Ripostes divided by total punch attacks

DMG - Avg damage taken per punch attack, basically AVG * HIT%.

<hr></blockquote>



CLS AC SHLD AVD DB DI AVG MIT% ATKS HIT% MISS% BLK% DDG% PRY% RIP% | DMG

MNK 1854 2% 60 100 22.1 311.0 55.0 3476 41.1 47.2 12.0 5.4 NA 4.7 | 127.8

*MNK 1635 2% 60 100 22.1 313.2 54.5 2923 42.0 46.8 11.6 4.9 NA 4.7 | 131.5

MNK 1331 0% 0 102 22.1 317.4 54.0 2550 46.8 42.1 10.4 4.5 NA 4.4 | 148.4

MNK 1063 0% 0 102 22.1 344.4 52.5 2883 46.7 41.0 11.9 4.4 NA 4.6 | 160.8









<blockquote>Quote:<hr>

As you can see, at 1331 I mitigated nearly the same as at 1854 ac. The change was miniscule. Here is where we see a 'soft cap' on AC. I always thought the softcap would depend solely on the mob you were tanking. I thought you could keep piling on AC unitl you got to a point where the majority of hits were for min and thus you had your diminishing returns. It turns out that soft caps are on a per class basis. No matter how high my ac gets I do not think I could ever mitigate like a 1300ac warrior.<hr></blockquote>



Call me crazy or stupid or what ever but all I did was take the monk #s for AC and hit% and plot them against the new parses.



Edit: okay it looks good.



So I took:

AC and hit%



1854 41.1



1635 42.0



1331 46.8



1063 46.7



Slap that in a table and graph it with the new numbers as a different line.



853 47.96



1010 47.46



1167 46.69



1325 46.89



1462 49.05



1639 48.27



1798 48.98



Okay the old parses had shielding at 1854 and 1635. Shielding is good.



So we have 1331 and 1063 left for comparison and compared to 1167 and 1325 we look about right.



The mob Ned was testing was higher level and Ned didn't have Virtue which I think is calculated as an addtion to the AC softcap as are the Defensive AAs.



So what do I end up seeing? In short, somewhere just before 1325 we hit the cap.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=evi lhomer@monklybusiness43508>Evil Homer</A>* http://www.urban-village.co.uk/thumbnails/Bruce%20Lee.jpg at: 5/17/05 9:24 am

[Dythan Dabrave]

bleh, sitting at about 1550 raid buffed, was hoping i could put off oow mig aa's for a bit. guess not =P


Dithan Dabrrave

70th season monk

Forsaken Visage



The only difference between a Hero and a Fool is Success.</p>

[Ancaglon]

I'm really not sure about the methodology there -- it doesn't look at the mitigation part of it. How about graphing "MIT%"? For Nedrom's data that'd be (AvDMG-MinDMG)/(MaxDMG-MinDMG)*100%.

<blockquote>Quote:<hr>The mob Ned was testing was higher level<hr></blockquote>

The 'test dummy' was quoted at 60, Brodda Thep's parses were on a HoH Valorian Guard at 62.






http://www.xenaphon.demon.co.uk/anc/anc-new-sig.jpg (http://www.magelo.com/eq_view_profile.html?num=52287)</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=anc aglon>Ancaglon</A>* http://www.ezboard.com/image/emoticons_classic/devil.gif at: 5/17/05 10:09 am

[Evil Homer]

KK



<blockquote>Quote:<hr>(AvDMG-MinDMG)/(MaxDMG-MinDMG)*100%<hr></blockquote>



Using mitigation% would be: (Actual damage)/(total number of hits * maxdamage) * 100 = mitigation%



That's a question.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Ancaglon]

I actually got that backward, but from the Brodda Thep description:<blockquote>Quote:<hr>MIT% - 0% mitigation would be max hits every single time. 100% mitigation would be min hits every single time.<hr></blockquote>



That means mapping the average damage (of hits taken, thus removes effect of avoidance) into the min-max range. Since Shielding appears only to affect the DB (damage bonus) part, MIT% should be independent of Shielding as well as Avoidance.



That makes the formula



MIT% = (MaxDMG-AvDMG)/(MaxDMG-MinDMG) x 100%



This is pretty much directly comparable with Nedrom's parses, except that Brodda's parses only list displayed AC -- and Agility makes a significant difference to that, although it doesn't appear to make ANY difference to Mitigation.



This leads me to a question for Nedrom: what race was the lvl 65 monk used in this test? Given that your normal character is Iksar, I expect the test monk was also Iksar. I am also reasonably sure Brodda Thep's monk was Human. If the races weren't the same, then we do have the complicating factor of the Iksar AC bonus, which is believed to be a Mitigation bonus....



Does anyone recall for sure? eg do they have Brodda's Magelo number somewhere?


http://www.xenaphon.demon.co.uk/anc/anc-new-sig.jpg (http://www.magelo.com/eq_view_profile.html?num=52287)</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=anc aglon>Ancaglon</A>* http://www.ezboard.com/image/emoticons_classic/devil.gif at: 5/17/05 1:43 pm

[Evil Homer]

Alright then.



We shoudl take our paths and see where they lead.



I believe my calculation for mitigation is correct.



100% mitigation is avoidance to me.



Therefore I remove all zeros.



So I only count the hits = H

Total damage. = TD



Now how much could those total hits been for? # of hits * max damage= total max damage possible or TMDp



Therefore:

TD/(H*MD)*100 = %damage mitigated





For example:

853AC



986,976 total damage

max hit was 471

hits was 3005



986,976/(471*3005) * 100 = %



986,976/1,415,355 *100 = 69% of the maximum possible or 31% of maximum damage was mitigated.





amazingly ave hit 328.44/maxhit 471 * 100 = 69.9????


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=evi lhomer@monklybusiness43508>Evil Homer</A>* http://www.urban-village.co.uk/thumbnails/Bruce%20Lee.jpg at: 5/17/05 2:54 pm

The problem with that formula Homer is that it is not independent of the mobs minimum hit. You could test against two mobs with the exact same atk on a tank with the exact same AC and get different numbers for "mitigation %".



The effect of AC in fiddling with the distribution of hits is what should be quantified, and factors like mob minimum and maximum hits should be normalized out.


</p>

[Ganjin]

I have seen assumptions that the softcap increases just by leveling, but only saw one claim based on logged information from Thepp on TSW.



<blockquote>Quote:<hr> 10-31-2003 #6

Brodda Thep

Registered User





"Quote:

I wish I would have parsed the DPS my warrior takes in BoT. I leveled him in bot almost exclusively from 60-65 with very little change in his gear."







I did keep parses with no equipment changes at all.



Lvl 60 in ssra from a shissar taskmaster (entire lvl)

Defensive

40.6 hit %

176 avg hit



Offensive

35 average hit

Misses were turned off.

55.2 punch dps



lvl 61 in ssra from a shissar taskmaster (entire lvl)

Defensive

41.4 hit %

187 avg hit



Offensive

36 avg hit

57.5 punch dps (got 1 more to damage bonus at 61)



lvl 62 in ssra from a shissar taskmaster (somewhat small sample)

Defensive

39.9 hit %

191 avg hit



Offensive

37 avg hit

57.7 punch dps



Same buffs for all PoN parses.

lvl 62 in PoN tanking a horror hobgoblin (300,000 dmg taken)

Defensive

49.4 hit %

260 avg hit



Offensive

54 avg hit

90.4 punch dps



lvl 63 in PoN tanking a horror hobgoblin (700,000+ dmg taken)

Defensive

51.1 hit %

241 avg hit



Offensive

56 avg hit

94.2 punch dps (increase in damage table)



lvl 64 in PoN pulling a horror hobgoblin (500,000+ dmg taken)

Defensive

50.8 hit %

252 avg hit



Offensive

56 avg hit

96.6 punch dps (increase in damage bonus)



To me these indicate that, beyond damage table increases and damage bonus increases, I didn't do better damage or tank better. No where NEAR the difference you would see going from 55-60.



Of course my defensive avg hits on the goblins are pretty messy due to the fact that I wasn't tanking all the time and pulled with stone stance a lot.



If there are increases, then it is incredibly small. On the order of single percentages more than likely. ( (250 + 65)/(250 + 60) perhaps? ) At any rate, I highly doubt anyone that says they tanked noticeably better going from 60-65.<hr></blockquote>



I don't remember seeing any real parses of defense from 51 - 60 anywhere not involving AA's, so the question of the softcap increasing through these levels is also dependant on the removal of the skill cap at 50.



Then the question becomes; does the defense skill actuially increase the softcap or just change how raw ac is calculated for displayed ac?



I would think the best way to look at it is the softcap doesn't change past 60 unless you buy AA or use a shield.



Kavhok stated earlier that a level 51 monk was at the softcap with 163 raw item ac and 150 agi before the original nerf, which is where monks are supposed to be now.




</p>

[wolvurene]

OK here is a parse 87 minutes long in Riftseekers with me tanking. I had a cleric hitting me with elixir before the pull, and casting CH just before i started tanking each mob. Druid was providing spam heals over the top, and my shaman was spot healing in places where the druid/cleric heal delays overlapped. My AC was 1639, total buffed hps just under 10k. We could not find another real tank, so I stepped up to fill in. We had to choose the group according to that and so included the druid to help keep me up. Was it comfortable ? A quick extract from the log, it seems shaman regen is good for something. No heals hit me while i was unconscious, luckily the chimera didnt either for a couple seconds......



[Wed May 18 01:38:42 2005] a companion chimera was hit by non-melee for 12 points of damage.

[Wed May 18 01:38:42 2005] A companion chimera hits YOU for 1237 points of damage.

[Wed May 18 01:38:42 2005] a companion chimera was hit by non-melee for 12 points of damage.

[Wed May 18 01:38:42 2005] A companion chimera hits YOU for 725 points of damage.

[Wed May 18 01:38:42 2005] A companion chimera tries to hit YOU, but misses!

[Wed May 18 01:38:42 2005] You have been knocked unconscious!

[Wed May 18 01:38:43 2005] You cannot see your target.

[Wed May 18 01:38:43 2005] You are conscious again!

[Wed May 18 01:38:43 2005] Your target is too far away, get closer!

[Wed May 18 01:38:43 2005] You punch a companion chimera for 41 points of damage.

[Wed May 18 01:38:43 2005] Vtik bashes a companion chimera for 32 points of damage.

[Wed May 18 01:38:44 2005] cleric has healed you for 2861 points of damage.

[Wed May 18 01:38:44 2005] Your wounds fade.





http://i3.photobucket.com/albums/y66/wolvurene/RSS-1639AC.jpg


Wolvurene (http://www.magelo.com/eq_view_profile.html?num=921239)

Enlightened

Terror Australis (http://www.terror-australis.com)</p>

I had a few log files from December 2004 when I was dual-boxing my Monk/Shaman in Grieg's End courtyard. The average hit on my 65 Monk with CA1, CS1 and 1204-1240 AC (they dispell quite a bit) was 115 and peanuts. I had FA on, so all combat stats were maxed.



I did the same camp over the weekend. Level 52ish mobs, same place, same gear, same level, same AC, same AAs. The buffs were also the same give or take some dispelling. After almost exactly 1 AA worth of killing, I parsed my logs again using Yalp. The average hit was 116 and peanuts, so basically no change.



I should have checked the hit distribution, but it was late at night and Yalp was erroring out on me, so I stopped. I then flew back to the West Coast, so any further analysis will have to wait until I go back East. The only conclusion for now is that if your gear is as bad as mine (I had very little play time for a long time for RL reasons), don't go in expecting too much from the unnerf.


Scalia, 65 Monk (http://www.magelo.com/eq_view_profile.html?num=71792)




Scabon, 65 Shaman (http://www.magelo.com/eq_view_profile.html?num=576916)</p>

[sedusar]

Since you have so many dmg intervals, im guessing you had vie line on, which makes the parse more or less unuseable.



Nice job Ned, if I can find the time, ill check how the return is from 2000 to 2400, im guessing it flatlines somewhat after 1800 though.


</p>

[wolvurene]

vie for me when filling in as tank in RS is required, parsing is something that is just happening. If youu have 2k ac go parse there and post please.


Wolvurene (http://www.magelo.com/eq_view_profile.html?num=921239)

Enlightened

Terror Australis (http://www.terror-australis.com)</p>

[Evil Homer]

<blockquote>Quote:<hr>The effect of AC in fiddling with the distribution of hits is what should be quantified, and factors like mob minimum and maximum hits should be normalized out. <hr></blockquote>



In this case there is no need for normalization as the mob is the same and the max and min never change.



Distribution is the problem?



Then mitigation is the symptom.



Regardless I'm just looking for the AC softcap. All I want to do is make a line and see where the line breaks or bends over.



The line for me at least is bending over at ~ 187raw AC or 1200 -1250 AC.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Demandred]

<blockquote>Quote:<hr>Regardless I'm just looking for the AC softcap. All I want to do is make a line and see where the line breaks or bends over.

The line for me at least is bending over at ~ 187raw AC or 1200 -1250 AC.

<hr></blockquote>

Yes it bends over and if it didn't do this we would soon reach a 100% mitigation. I plotted Neds data to mitigation form and it's quite clear the softcap hasn't been reached at 600 raw ac, because if we continue the trend of this curve we would soon end up at warrior mitigation.



A cool thing to notice is that at around 1800 ac, 100 raw ac corresponds to 2% shielding. Of course I'm not saying ac = shielding because ac has another function aswell, it increases your survivability while shielding only lowers the average dagame taken.



http://users.du.se/~d98jja/mit.gif

The staples are aproximate 95% error estimations.












</p>

[samoa79]

Actually on your grid the softcap is obvious at 100ac...then we receive diminished returns after that. Note how much 0-100 is quite a boost but everything after that is a much smaller increase...that would be our softcap. Looks to be in the 100-150 range most likely.


<div style="text-align:center">http://www.resilienceguild.org/files/Sigs/Kazmarsig.gif



My Stuff (http://www.magelo.com/eq_view_profile.html?num=243450)

Stone Fist</div></p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=sam oa79>samoa79</A> at: 5/18/05 3:53 pm

[Evil Homer]

Oops! Thanks Demandred! (If you would can you make 200 on the one curve a zero and have that curve curve over the 0 - 200 curve please?)



Okay. see how his stick's bend evens out?



Well you can see and you even noted that that even bend isn't there from 100 - 200 raw AC.



Now all we do with math is look at the bendiness from 200 -600 and compare that to the bend from 100 - 200.



No much of a "noticable" bend from 100 - 200 as it is relly just two data points but!



We can guess where that bend starts by comparing the bends 200 - 600 and 100 -200.



When i do that in Excel I see the 200 - 600 bending start at 187 raw AC.



Make any sense?


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=evi lhomer@monklybusiness43508>Evil Homer</A>* http://www.urban-village.co.uk/thumbnails/Bruce%20Lee.jpg at: 5/18/05 4:14 pm

[samoa79]

Makes perfect sense.


</p>

[Demandred]

Ok well, I guess I wasn't clear enough. A nerfed monk wouldn't see any difference from 1350 to 1800 ac because he reached the so called softcap which in fact more felt like a hard cap because the returns were indeed diminishing. Now they changed the returns at 300 raw ac and because I'm used to the nerf, I guess I still expect a 'hard capped soft cap' so to speak, so I didn't consider a 2% improvement/100 AC, a soft cap :p



Question though, there are 3 defensive AA's:

Defensive instincts

Innate defense

Combat stability



What do these things do? Do they change the returns after reaching soft cap? Or do they raise the softcap?



In any case it would be very interesting to see the lvl 70 (with maxed aa's) version of this parse.


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=dem andred>Demandred</A> at: 5/18/05 5:23 pm

[santino]

<blockquote>Quote:<hr>What do these things do? Do they change the returns after reaching soft cap? Or do they raise the softcap?<hr></blockquote>



According to Kavhok, they raise the soft cap. (As does using a shield, apparently)



How much per AA (and if it varies depending on WHICH AA) is open to investigation. <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">


Fumika (http://www.magelo.com/eq_view_profile.html?num=983997)

Guild Leader of Court of the Crosswinds</p>

[Evil Homer]

Softcap is a return over the cap.



What we have is real AC value to X number followed by a reduced return on AC over that X.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Demandred]

pls go back to your excel bending


</p>

[Demandred]

<blockquote>Quote:<hr>Actually on your grid the softcap is obvious at 100ac...then we receive diminished returns after that. Note how much 0-100 is quite a boost but everything after that is a much smaller increase...that would be our softcap. Looks to be in the 100-150 range most likely.<hr></blockquote>

Actually you didn't read what I said and you see everything in lines while it very well could be a curve or more than two lines with different slopes. Once again, if you continue the trend, 2% / 100 raw AC, this would make a 3k ac monk equal to a knight with same AC (purely hypothetical since 3k AC monks doesnt exist). Now 100 AC for a monk would mean the same as it does for the knight. Since it's hard for a monk to reach more than 2200 ac without giving up other important stats, we have *DRUMROLL* BALANCED THE GAME!!! Do you honestly think SOE would do this intentionally? Me neither!


</p>

[Ancaglon]

I'd be astounded if the Softcap for a lvl 65 monk with NO defensive AA is as high as 600 raw, corresponding to 1798 displayed. (My optimistic expectation is that it's about 250 raw AC - about 1200 displayed - and that with the PoP defensive AA, this might hit 350 raw corresponding to about 1400 displayed).



However, I do agree with some of what Demandred is saying -- that the mitigation curve should indeed be a curve and not a straight line, EVEN IF THERE WERE NO CAP AT ALL. So, finding the actual cap is a tricky exercise in fitting curves.



Nedrom, any chance you can fill in a few more points in the 150-250 worn range?



PS: the number fiddling I've been doing with the formulae Magelo uses to calculate AC suggests that the Iksar AC bonus is equivalent to raw AC of (level*2/3): has anyone done any comparisons of this at various levels?




http://www.xenaphon.demon.co.uk/anc/anc-new-sig.jpg (http://www.magelo.com/eq_view_profile.html?num=52287)</p>

[Nedrom]

I won't be doing anymore parsing. What I have provided so far should be sufficient to show monks who have no defensive aa and no uber raid gear that AC really does work.


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/18/05 8:45 pm

It's not a curve at all over the cap. The data points are just too far apart and give the false impression of a curve.



Over the cap it's linear. It looks like a curve because the return is diminished but the return is linear, it never changes. All that is changing and giving the appearance of a curve is the large amount of AC used to draw the lines.



So, no. A Monk at 3,000AC would not, could not, ever have the same AC value as a Knight with 3,000AC. There is no curve leading to zero return.



You can put mommies statistical software away now.



<img src=http://www.ezboard.com/images/emoticons/wink.gif ALT=";)">



PS. I prefer the approach of someone attempting to make this easy to understand rather than someone attempting to Pontificate and then getting it completely wrong.


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=tau zii@monklybusiness43508>Tauzii</A>* http://i5.photobucket.com/albums/y193/tauzii/own.gif at: 5/19/05 7:27 am

[Ancaglon]

No, I think that's a mis-understanding of what the cap is understood to do: that is,



if AC =< softcap, Effective AC = actual AC

If AC > softcap, Effective AC = Cap + (actual AC - Cap)*return



so for example, if the Softcap is at 200 raw AC, and the return over the cap is 1 per 5,

250 raw ac would be worth 200 + (250-200)*(1/5) = 210 real.

300 raw ac would be worth 200 + (300-200)*(1/5) = 220 real. 350 raw ac would be worth 200 + (350-200)*(1/5) = 230 real.



-- that's a straight line, not a curve. I think the 'curves' come from looking at the ratio of attack power to effective AC (which is 'real' AC plus a factor based on defence skill, monk level, and as far as I can tell for Iksar, half the level as an extra bonus). I'll try and post some graphs today.


http://www.xenaphon.demon.co.uk/anc/anc-new-sig.jpg (http://www.magelo.com/eq_view_profile.html?num=52287)</p>

[Evil Homer]

<blockquote>Quote:<hr>pls go back to your excel bending <hr></blockquote>



Completely uncalled for.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Blazyn]

<blockquote>Quote:<hr>

When i do that in Excel I see the 200 - 600 bending start at 187 raw AC.

<hr></blockquote>

I'm stuck on the slope of the 0-100 line. I can see that the upper line passes through 187 - but I'm not sure it has intersected the first one yet. This still looks to me like a low 100's cap. I wish we had some additional data points between 100 and 200.


</p>

[Fisticuffs]

Doesnt even look like theres a (soft) cap to me.. looks like a curve that we overlayed straight lines onto every 100 ac (note that for the values >= 200.. the increase is less each time).


<div style="text-align:center">

http://rann2.home.mindspring.com/Final2.jpg

Strike me down and I shall become more powerful than you can possibly imagine.

Fisticuffs -- Stone Fist (http://www.magelo.com/eq_view_profile.html?num=395444)

Sladen -- Arcanist (http://www.magelo.com/eq_view_profile.html?num=475305)</div></p>

[Blazyn]

Sladen -



Imagine that for every 1 AC you added, you mitigated 1% better. At 100 AC, you would mitigate 100%. If you plotted this on a graph, you would see a straight line from 1AC and 1% mitigation on up to 100/100. If the line isn't straight, if instead it "bends" at some point, then the point where the bend occurs is the softcap. The existance of a softcap doesn't imply an upper limit where AC no longer does any good - we'd call that a hardcap. The softcap is the point where we begin to see diminishing returns.



In my fake example, let's say you got 1% mitigation up to 50 AC, and then after that each AC only gave you .5% mitigation. The softcap is 50, and now you'd only be at 75% mitigation at 100 AC.



Demandred's graph clearly shows a steep slope at first, then a more gradual one - implying a softcap somewhere between 100 and 200.


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=bla zyn>Blazyn</A> at: 5/19/05 4:39 pm

[Demandred]

I played around some more with Neds excellent data, because this line vs curve thing was annoying me. There is of course a softcap of some form but what meaning would it have if the mitigation plot turns out to be a curve? Basically none, so there has to be an explanation.



I started by looking at the frequency of max hits in Ned's parse:



http://users.du.se/~d98jja/max.gif



Clearly, this is not a straight line, but a curve of the form 1/x. So how can this lead to something that looks like a line in the mitigation plot? You got it, if you take the inverse of this curve you get:



http://users.du.se/~d98jja/inv_max.gif

Which is clearly a line. So what does this have to do mitigation?



The area under the graph in our first plot is the damage taken from max hits while mitigation basically is the damage we don't take, or mathematically, the inverse of the max hit distribution.



But what about the other hits that are non max hits? They also contribute to mitigation, but it turns out we can simplify this quite a bit by noting that all but the min/max hits are aproximately a uniform distribution. So we deal with the min hits first. I do the same thing as I did with max hits:



http://users.du.se/~d98jja/min.gif



and then I form the inverse of this plot:

http://users.du.se/~d98jja/inv_min.gif



In the next step I will subtract the inversed min plot from the inversed max plot. The explanation to this subtraction, is that min hits can't be mitigated. Then we have the the other hits that are non min/max hits. As mentioned earlier they also contribute to mitigation, but they will do this by an equal amount (remember I assumed a uniform distribution for these hits) at each ac point, so the only effect they will have is, some constant mitigation will be added to each data point. Since I already plotted the mitigation I will simply add a constant to each data point so that the estimated mitigation fits with the first data point in the previously posted plot.



http://users.du.se/~d98jja/est_mit.gif

Now we can compare this with the initial mitigation plot, in where I simply used the average, min and max hits.

http://users.du.se/~d98jja/mit.gif



Conclusion: The mitigation curve is basically determined from the max and min hit distributions. After reaching a certain AC the mitigation will be aproximately linear, but not completly, because the other hits, which I assumed were uniformly distributed, are not truly uniform.






















</p>

[Ancaglon]

I don't agree with the subtraction of 1/max from 1/min as being a valid estimator, because other hits in between can be mitigated to different extents (eg DI interval no. 2, if mitigated 5%, is going to end up as min damage, but anything more can't take it below DI1 either -- whereas DI 20 (=max hit) or DI 19 (next highest hit) will be effectively mitigated at all except the higest ACs).



However, your insight on plotting the inverse of the % of max hits is VERY interesting -- and looks amenable to fitting to determine the exact softcap. By eye though, it does seem to confirm that the softcap is between 100 and 200 raw AC, i.e. less than 1067 displayed.... if THIS is the INCREASED softcap at 65, I shudder to think what it was before -- and suggests to me that this data really isn't showing the same thing as Brodda Thep's parses vs the HoH mobs -- unless CS3+ID5 really amounted to increasing this to around 300 raw AC (1325 displayed).


</p>

Just plot without the 0 - 100 data.



The 0 raw AC data is obviously skewed by some penalty no one bothered to tell us about.



I'm pretty happy withe the notion that at 65 a person's softcap for AC is around 1200.



The return over the cap is what has "reversed" the nerf. The cap itself is stupidly low.




http://i5.photobucket.com/albums/y193/tauzii/TauziiLFG.jpg (http://www.magelo.com/eq_view_profile.html?num=852244)

</p>

[Evil Homer]

<blockquote>Quote:<hr>The area under the graph in our first plot is the damage taken from max hits while mitigation basically is the damage we don't take, or mathematically, the inverse of the max hit distribution.<hr></blockquote>



Not mathematically, but logically yes.



Just saying.



You are correct however.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Ancaglon]

I'd go along with that. It'd be nice to see exactly how the Mitigation AA's affect the curves, but this makes AC worth something for us again.



Curiously, this strikes me as a reflection of the situation for Tank classes at the time of the Nerf, in that at that point they'd been getting to the (then hard) cap and thereafter just loading up on HP, but after that AC was worth something again... our cap might not have been hard, but it was certainly not readily distinguishable from that until the current round of fixes.




http://www.xenaphon.demon.co.uk/anc/anc-new-sig.jpg (http://www.magelo.com/eq_view_profile.html?num=52287)</p>

[Demandred]

<blockquote>Quote:<hr>I don't agree with the subtraction of 1/max from 1/min as being a valid estimator, because other hits in between can be mitigated to different extents (eg DI interval no. 2, if mitigated 5%, is going to end up as min damage, but anything more can't take it below DI1 either -- whereas DI 20 (=max hit) or DI 19 (next highest hit) will be effectively mitigated at all except the higest ACs).<hr></blockquote>

Well if you look at the end result you see the estimation seems to work and the reason is that only the min and max hit's change in Ned's parse while the other DI's stay roughly equal throughout each parse. Anyway the intention was not to get the exact mitigation plot, since there are much easier ways to do this (only average, max and min hit values are needed).



I wanted to clarify for myself, the linear behaviour of the mitigation plot. I started to understand this when I saw that the inverse of max hit formed a line. I was still worried that the min hits would destroy the linear behaviour, but it turned out the min hit distribution was aproximately a line, which leads to a 1/x curve once we inverse it. So in the long run when we reach high ac values, the inversed min hit curve willl flatten out and the mitigation plot will behave linear after some certain point (soft cap).






</p>

[Ancaglon]

Hm. True. It appears to me that the 20-interval damage distribution is a kind of a "window" on a larger range of numbers - and that values that fall below the minimum get mapped to the minimum, and values above the maximum, to the maximum..




</p>

[Evil Homer]

<blockquote>Quote:<hr>values that fall below the minimum get mapped to the minimum, and values above the maximum, to the maximum<hr></blockquote>





You've hit on something here. There really isn't a true 21 point DI.



The DI is a creation, hence as you proved to me, the spikes at the top and bottom.



Could it be our mitigation change included a change to the conversion factors of the real numbers to the 21 DI table?



This also means that with a couple key strokes they could literally even out the distribution... Take 1/19 of hits over or under and put them here... This would open up a lot of game play and improve the combat system.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=evi lhomer@monklybusiness43508>Evil Homer</A>* http://www.urban-village.co.uk/thumbnails/Bruce%20Lee.jpg at: 5/20/05 11:11 am

[Ancaglon]

I had a thought or 2 about what an accumulated normal distribution looks like. Take a look at this:

http://www.xenaphon.demon.co.uk/anc/AC/small-fun-with-numbers-1.png (http://www.xenaphon.demon.co.uk/anc/AC/fun-with-numbers-1.png)

(The top and bottom 'buckets' are actually "this value and further out" totals).



Now increase the standard deviation further:

http://www.xenaphon.demon.co.uk/anc/AC/small-fun-with-numbers-2.png (http://www.xenaphon.demon.co.uk/anc/AC/fun-with-numbers-2.png)

Compare the bottom graph with Nedrom's graph for +200 AC ... I find it strikingly similar. Then, consider changing the Mean of the distribution:

http://www.xenaphon.demon.co.uk/anc/AC/small-fun-with-numbers-3.png (http://www.xenaphon.demon.co.uk/anc/AC/fun-with-numbers-3.png)



This is a bit of an aside with respect to AC parses, but IMO it gives a useful insight into what we actually see when we parse the data. It appears (to me) that the 'bell curve' we've been asking for has always been there, but as part of a larger range of values... perhaps someone with some formal statistical training can use this to take it further.



(If anyone wants to play with this themselves, the OpenOffice spreadsheet I used for these is available here (http://www.xenaphon.demon.co.uk/anc/AC/eqdmg.ods)).


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=anc aglon>Ancaglon</A>* http://www.ezboard.com/image/emoticons_classic/devil.gif at: 5/23/05 7:25 pm

[Ancaglon]

If the damage distribution is indeed as above, then the Average damage taken is a valid measure of mitigation - the min and max percentages are an artifact of the binning up process.



I then plotted (maxDmg-minDmg)/(avDmg-minDmg) against raw AC, and compared the gradients of the 2 sections. This shows that the gradient of the latter section is 2.8 times the gradient of the 1st segment -- which, to me, implies that our return on AC over the softcap is 1 per 2.8. The 'soft cap' implied this for an Iksar monk at level 65 is 118 worn AC -- for a Human monk this would be 150 worn AC.



http://www.xenaphon.demon.co.uk/anc/AC/invavdmg.gif


</p>

[Evil Homer]

Ancaglon,



Thanks.



For whatever reason I get the lines to cross at about 187 RAW AC using my formula...



I'm certain that RAW AC is capped at about 150-190AC.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Dythan Dabrave]

alrighty, been following this thread for awhile, and i don't mind telling you i don't understand at least half. You guys are doing awesome work, and i want to thank you for it.



Anyone think we can come to an agreement on when to buy our migitation aa's so as to get the full effect. can we post it like this.

(random guesses here)



1200 - Luclin Def

1350 - PoP

1500 - OoW



would be nice to have numbers, even if they are not 100% correct, and would kill alot of the new threads.


Dithan Dabrrave

70th season monk

Forsaken Visage



The only difference between a Hero and a Fool is Success.</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=dyt handabrave>Dythan Dabrave</A> at: 5/24/05 7:07 pm

[Nedrom]

im going to try to re-parse with CS3 over the next week


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Nedrom]

will have cs3 1798 ac parse in the morning for you

over the next few days ill include them with the other values of ac, i just have to let em run overnight



i had a peek at the log file after 50k damage, the results are very interesting <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/25/05 12:39 am

[azathool]

Is there any reason to believe that the cap at 70 is the same as at 65 ?



Reason I ask is, why do all the parses at lvl 65 if a large proportion of monks are lvl 70 ?


<hr>
<div align="left">
<table border="0" width="100%">
<tr>
<td width="15%"></td>
<td width="70%">
<p align="center">[Stone Fist] Azathool (http://www.magelo.com/eq_view_profile.html?num=597)
<Nightwatcher (http://www.nwguild.com/)></td>
<td width="15%">
<p align="center">
[ Shm (http://www.magelo.com/eq_view_profile.html?num=4174)
| Clr (http://www.magelo.com/eq_view_profile.html?num=1024677)
]</td>
</tr>
</table>
</div>

</p>

[samoa79]

Majority of monks on Emarr are lvl 66-68 from what Ive seen from doing /w all monk.


<div style="text-align:center">http://www.resilienceguild.org/files/Sigs/Kazmarsig.gif



My Stuff (http://www.magelo.com/eq_view_profile.html?num=243450)

Stone Fist</div></p>

[Demandred]

<blockquote>Quote:<hr>This is a bit of an aside with respect to AC parses, but IMO it gives a useful insight into what we actually see when we parse the data. It appears (to me) that the 'bell curve' we've been asking for has always been there, but as part of a larger range of values... perhaps someone with some formal statistical training can use this to take it further.<hr></blockquote>

Great work. I'm using MS office myself but I think I got your model right. I ran through the model and checked with the parse results and it's quite amazing how it predicts the hit distributions. Like look at this, using the parsed average hit (transformed to DI) in the model:



http://users.du.se/~d98jja/norm.gif



The staples are 95% confidence intervals for the parse results and the model scores 20/20! Coincidence? Probably not. It turns out the std=9 model works great for 300-600 rAC but is less accurate from 0-200 as expected, since the softcap does something to the distribution, which leads to a different deviation.




</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=dem andred>Demandred</A> at: 5/25/05 4:30 am

[Ancaglon]

Wow, that's pretty nice, considering that I just chose sd=9 'by eye' as an example from the graphs, increasing it by 2 each time, rather than trying to fit the distribution in any mathematical way. I suspect 9.5 or 10.0 might fit better (half of 19 or 20) but confidence intervals are a thing I don't remember how to use anymore <img src=http://www.ezboard.com/images/emoticons/wink.gif ALT=";)"> hence my plea to someone that does.



I would guess that level 65's being used because it's where Brodda Thep's parses were done. The form of the equation that's used by Magelo (and which is pretty much spot on for everyone except Monks and Iksar <img src=http://www.ezboard.com/images/emoticons/wink.gif ALT=";)"> ) is this for an unbuffed melee class:



DisplayedAC = trunc[(AvoidanceAC + MitigationAC) * (1000 / 847)]

AvoidanceAC = AGIfactor + trunc[DefenceSkill * 16/9]



(AvoidanceAC is set to 0 if the above total is less than zero).



(AGIfactor above level 40 and for AGI 126-199 is trunc[(AGI+374)/10 ] , and for AGI 200-305 (possibly higher) is trunc[(AGI+1060)/20 ] ).



and



MitigationAC = trunc[DefenseSkill/3] + trunc[raw AC * 4/3]



From what Kavhok wrote over in EQlive, the Monk bonus is equivalent to (Level+5) raw AC. Additionally, I estimate the Iksar bonus to be trunc[Level/2] worth of raw AC.



So, just levelling up from 65-70 will only increase the Mitigation AC by a very small amount, due to the increased Defence skill and the level bonus(es). Avoidance AC increases by more (due to Defence skill) and so Displayed AC also rises. (The 18 pt rise in Defense skill from 252 to 270 amounts to +6 mitigation AC, +32 avoidance AC, and hence 44-45 displayed AC).


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=anc aglon>Ancaglon</A>* http://www.ezboard.com/image/emoticons_classic/devil.gif at: 5/25/05 5:03 am

[Demandred]

9 is the magic number :p. I didn't find anything better.


</p>

[Nedrom]

<blockquote>Quote:<hr> Is there any reason to believe that the cap at 70 is the same as at 65 ?



Reason I ask is, why do all the parses at lvl 65 if a large proportion of monks are lvl 70 ?

<hr></blockquote>



because...a majority of the complaints on this forum are for non-raiding casual monks.



so i'm posting values for level 65.



page 1 now has 1789 AC CS3 parse, ill post more when i get it



AFIK...AC was not changed when OOW came out, so most likely caps were not changed from 65-70


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/25/05 7:15 am

[Ancaglon]

If I apply the gradient of the 'post-cap' line to Nedrom's CS3 plot, and calculate its' intercept with the 'pre-cap' line, I get a value of 178 AC for Iksar, which translates to 210 AC for a Human - an increase in the cap of 60. ANy other parses he does in the 150-350 range will help nail this down.



Speculating wildly, I would say CS1 increases cap by 10, CS2 by another 20, and CS3 by another 30.


</p>

[Evil Homer]

Ancaglon thanks. Ned thanks.



CS3 looks worth 2% more mitigation at 1700AC.



The line for mitigation as he decreases AC will be interesting to watch.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Ganjin]

<blockquote>Quote:<hr>Combat Stability: (3 skill levels) - Each skill level gives the character an increased ability to reduce the damage he takes in combat (damage mitigation). The first skill level increases the character's damage mitigation by 2%, the second skill level raises the damage mitigation increase to 5% and the third raises the increase to 10%.<hr></blockquote>



Add that to the stated softcap value of 163 rAC, plus any misc gains from level 51 - 65 and you come out between 190 - 200 ish rAC. ( assuming of course the 2, 5, 10% increase stacks instead of being an increase based on a fixed number. )



Interesting read, good luck tracking it down.


</p>

[Nedrom]

added 1639 + cs3 parse


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Ancaglon]

Graph of inverse average damage with CS3 points: although this is only 2 points, the gradient is steeper than that without Combat Stability by a factor of 1.12 (i.e 12% better) -- this may be what's meant by "mitigating 10% better", rather than simply increasing the soft cap.

http://www.xenaphon.demon.co.uk/anc/AC/netmit-500cs3.gif


</p>

[Demandred]

ya, once ac increase you'll get closer to a 10% mitigation improvement. So basically no need to parse CS3 anymore, just go on with those other aa's :)


</p>

[Ancaglon]

I'd like to see at least one more CS3 run at about 200-300 raw AC to see if it really IS a straight line also, and to pin down the actual effect...


http://www.xenaphon.demon.co.uk/anc/anc-new-sig.jpg (http://www.magelo.com/eq_view_profile.html?num=52287)</p>

[Nedrom]

added 1482 cs3 parse



i got a new webserver this month and it blows ass, so my bandwidth has exceeded...who woulda thought...



so anyway, i posted the numbers on the first thread at the top in the chart there



pictures should be fixed tonight if my friend fixes my account


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/25/05 8:53 pm

[Ancaglon]

Curiouser and Curiouser....

Updated to include CS3/400ac. It seems to converge on the AC cap line....

http://www.xenaphon.demon.co.uk/anc/AC/nedmit_400cs3.gif


</p>

[Nedrom]

ill have the 1300 parse tomorrow morning


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Demandred]

Watch them merge at 300 rAC :)


</p>

[Nedrom]

added 1325 AC CS3 parse


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Ancaglon]

Not quite... it still lies on a line that seems to lead back to where the pre-cap and post-cap lines cross.

http://www.xenaphon.demon.co.uk/anc/AC/nedmit_300cs3.gif


</p>

[Nedrom]

added 1167 parse


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Ancaglon]

Good grief. Now I have no clue what's going on.

http://www.xenaphon.demon.co.uk/anc/AC/nedmit_200cs3.gif


</p>

[Evil Homer]

It apears that CS3 only raises the soft cap and not the return on Ac over the cap, thus, the diminishing return normalizes then begins to appear (only apprears) to decrease.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Ancaglon]

I don't think that fits. That was my initial assumption and that's what I did to arrive at a softcap increase of +60.



If it did work that way, the CS3 line would parallel the post-cap noAA line on my plot.


http://www.xenaphon.demon.co.uk/anc/anc-new-sig.jpg (http://www.magelo.com/eq_view_profile.html?num=52287)</p>

[Evil Homer]

<blockquote>Quote:<hr>If it did work that way, the CS3 line would parallel the post-cap noAA line on my plot. <hr></blockquote>



It is isn't it? we're working backwards going from high AC to low. Is your line ploting the new points as we go or making a new line?



PS. please add a key for the lines.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=evi lhomer@monklybusiness43508>Evil Homer</A>* http://www.urban-village.co.uk/thumbnails/Bruce%20Lee.jpg at: 5/26/05 9:50 am

[Ancaglon]

While Nedrom is indeed working backwards, the lines are still being plotted as (maxdmg-mindmg)/(avgdmg-mindmg) against raw AC. If the only effect the CS3 had was to raise the soft cap, the CS3 line should be a straight line also, with a gradient identical to that of the noAA line, but have a higher intercept. As it stands, it is far from a straight line - so much so that I haven't even tried to draw a trend line to it.



I have added the line key to the last plot.



EDIT: Excel file used is here (http://www.xenaphon.demon.co.uk/anc/AC/ned_mit.xls).


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=anc aglon>Ancaglon</A>* http://www.ezboard.com/image/emoticons_classic/devil.gif at: 5/26/05 11:12 am

[Evil Homer]

Looks good to me. Thanks for the work.



I see the higher soft cap from CS3 and the diminished returns of the AC over the Cap due to CS3 then it proceeds as expected...



Old non-CS3 parse showing the diminished return of the AC sooner than the CS3 parse due to the rAC 100 points being so far appart.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Ancaglon]

Even if that 200 rAC / CS3 point is 'really' on the pre-cap line, the rest of the CS3 line doesn't make much sense; it actually appears to make the return on AC over that cap worse for a fair sized increase in raw AC, before picking up again. It 'feel's as if it's adding or subtracting a constant factor somewhe on the inverse of this plot.




http://www.xenaphon.demon.co.uk/anc/anc-new-sig.jpg (http://www.magelo.com/eq_view_profile.html?num=52287)</p>

[Nedrom]

added 1010 ac parse



from what i see there is no difference with or without cs3 for +100 rac



so...under the cap, cs3 doesn't help you, all it does is raise the cap


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/26/05 1:13 pm

[SkyKungfu]

Thanks for the work Ned.



Looking at the 1482 ac tests shows perfectly what I tought, CS 3 (and mostlikely my ID 5 as well) does nothing for me and every other monk below 1500 ac. <img src=http://www.ezboard.com/images/emoticons/embarassed.gif ALT=":o">



Sky


</p>

[Demandred]

old news


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=dem andred>Demandred</A> at: 5/27/05 2:04 am

[Trukoflanys]

<blockquote>Quote:<hr>Looking at the 1482 ac tests shows perfectly what I tought, CS 3 (and mostlikely my ID 5 as well) does nothing for me and every other monk below 1500 ac.<hr></blockquote>



i am confused how you came to this conclusion.



400 1482

Min % = 22.47

Max % = 9.55

Avg = 259.66



400CS3 1482

Min % = 23.81

Max % = 8.92

Avg = 254.06



how is that not an improvement from CS3?


[70 Stone Fist (Monk)]
Trukx (http://www.magelo.com/eq_view_profile.html?num=474627) (Iksar)<span class="guild">
<Dark Horizon (http://darkhorizon.angrybunni.org/)></p>

[Evil Homer]

Thanks Demandred.



I do not want to compete, please try the Homer formula of mitigation (Total#hits*max hit/actual damage * 100) and see what that does to both test cases?



I see rAC of 187 non-CS3 and rAC 212 CS3.



Thanks.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Nedrom]

sky the only place where there is zero improvement is at 100 worn ac, displayed 1010


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Evil Homer]

We should be able to see no improvement at 150 worn AC.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Ancaglon]

The CS3 data still looks a bit iffy to me - it "feels" as if either the mitigation at 200 rAC is too good, or the mitigation at 300 rAC is too poor.



Personally, I rather suspect the former, because then the lines would continue to converge on the previous intersection that I identified as the cap (118 iksar/150 human by my calculations), and that it is the SLOPE of the line that is the significant factor. This is interesting because it is open-ended, and scales up, more so than a fixed boost in the cap. It could then indeed be '10% better mitigation' than CS0 as it "says on the tin".



For another reason, it's been previously stated by Developers that, if you are below the cap, the AAs won't help. In the past, we've assumed that the AAs raise the cap, but that seems contradictory.



IF my ideas are correct, then practically every (main) monk 65+ will be at or over the Softcap already (just 118 rAC for an Iksar at 65) -- so the AA's will actually have value for every one of those monks.



The only way to be sure of that line is to re-do those 200 and 300 rAC CS3 parses - which only Nedrom can currently do... I appreciate this is eating into his play-time, but this could certainly help sort this one out once and for all.



(Even a single 250 rAC parse would help a lot, if the gear set permits...).


</p>

[Demandred]

<blockquote>Quote:<hr>please try the Homer formula of mitigation (Total#hits*max hit/actual damage * 100) and see what that does to both test cases?<hr></blockquote>

It's quite possible you get 187 rAC etc with this formula. But since min hits are not in there, this would be a different definition of mitigation which assumes 0 as min hit.


</p>

[Evil Homer]

The AC cap without CS3 is ~1200AC at 65 and it is ~1260AC with CS3 at 65.



Is this true? If so how is it we actually have a lower CS3 cap now than we had when thepp did his parses or were thepps parses *that* bad? I'm betting on the latter for CS3 data.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Ancaglon]

Brodda Thep had ID5 as well, I believe -- "all PoP AAs" has to include that.


http://www.xenaphon.demon.co.uk/anc/anc-new-sig.jpg (http://www.magelo.com/eq_view_profile.html?num=52287)</p>

[Demandred]

<blockquote>Quote:<hr>The CS3 data still looks a bit iffy to me - it "feels" as if either the mitigation at 200 rAC is too good, or the mitigation at 300 rAC is too poor.<hr></blockquote> Well it's there and you can't really ignore it. Don't forget that even with a 2500 sample size the mean error is roughly 1% (in the mitigation plot) with a 95% CI of +-2%.


</p>

[Nedrom]

<blockquote>Quote:<hr>The only way to be sure of that line is to re-do those 200 and 300 rAC CS3 parses - which only Nedrom can currently do... I appreciate this is eating into his play-time, but this could certainly help sort this one out once and for all.<hr></blockquote>



I did 90 minute parses of all tests, that is long enough to even out the parse. Even Frodlin and Emil recommended 90 minute parse.



Unfortuantely I cannot redo these parses, i'm allready working on ID5 parses to post as well.



if you want the raw log im more than happy to send it to you but i doubt that will help





edit: added 1798 ID5 parse


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/26/05 2:57 pm

[SkyKungfu]

<blockquote>Quote:<hr>Quote:

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

Looking at the 1482 ac tests shows perfectly what I tought, CS 3 (and mostlikely my ID 5 as well) does nothing for me and every other monk below 1500 ac.

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







i am confused how you came to this conclusion.



400 1482

Min % = 22.47

Max % = 9.55

Avg = 259.66



400CS3 1482

Min % = 23.81

Max % = 8.92

Avg = 254.06



how is that not an improvement from CS3?<hr></blockquote>





At 1482 ac the taken dps is:



123,69 without CS 3

124,17 with CS 3



So where does this show that one mitigates 10% better with CS 3?



Even with the avg hit of 259 / 254 I do not see a 10% better mitigation.



Sky








</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=sky kungfu>SkyKungfu</A> at: 5/26/05 3:03 pm

[Nedrom]

sky has a point...but i suck at math so <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Ancaglon]

<blockquote>Quote:<hr>Well it's there and you can't really ignore it. Don't forget that even with a 2500 sample size the mean error is roughly 1% (in the mitigation plot) with a 95% CI of +-2%.<hr></blockquote>

Be that as it may, fitting a straight line to include the rAC 200 point is clearly not reasonable - look at the r-squared value for that fit vs the fit for rAC 300 and upwards - so extrapolating a fit as you have there makes as little sense.



Nedrom - Okay. It's a pity, but ID5 is a lot more relevent to compare with Brodda Thep's stuff than just CS3.



Oh, and does anyone recall - is PE also supposed to help with CS?


</p>

[Nedrom]

PE does yes


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[SkyKungfu]

Was just thinking about it from a different point of view.



Lets say without any mitigation he would have hit us for 310 every time. Now, at 1482 ac, we mitigate (on average) 50 every hit. So we end up hit for 260 (actually 259, just for easier numbers) every time.



Now with CS 3 we get 10% better mitigation, 10% of 50 = 5. So we mitigate 5 points more on every hit (average). This way we end up at 255 (254).



I don't like the idea, but it might work that way ...



Or would we get hit for max every time without mitigating?



470 - 260 = 210

10% = 21

Avg hit with CS 3 should be 239 (238 ) than. The difference might be because ac is such a weird number containing agi, avoidance and god knows what else? Sigh.



Sky


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=sky kungfu>SkyKungfu</A> at: 5/26/05 3:38 pm

[Demandred]

<blockquote>Quote:<hr>Be that as it may, fitting a straight line to include the rAC 200 point is clearly not reasonable - look at the r-squared value for that fit vs the fit for rAC 300 and upwards - so extrapolating a fit as you have there makes as little sense.<hr></blockquote>



Well I used least square method and the result looks reasonable to me. Did you check the mitigation plot?


</p>

[Nedrom]

added 1639 ID5 parse


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Zonn]

I'm going to keep my graph on the first page updated as new data points come in. I tried adding some best fit lines to the data, but it got a little bit too busy. For now I'm satisfied with a raw data plot with lines. You can still see the damage trending down with just that.


<span style="color:red;font-family:helvectia;font-size:large;">Axiss Glox</span>

<span style="color:red;font-family:helvectia;font-size:small;">70 Transcendant of Stasis (http://www.stasis-guild.com)</span>

Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=10004)

Axiss@Stasis-Guild.com (Axiss@Stasis-Guild.com)</p>

[Nedrom]

added 1482 id5 parse


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Nedrom]

added 1325 ID5 parse


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[wolvurene]

Ned, its time to change your sig now mate. lazy doesnt cut it, thanks for all the work <img src=http://www.ezboard.com/images/emoticons/eek.gif ALT=":eek">


Wolvurene (http://www.magelo.com/eq_view_profile.html?num=921239)

Enlightened

Terror Australis (http://www.terror-australis.com)</p>

[Nedrom]

<blockquote>Quote:<hr> Ned, its time to change your sig now mate. lazy doesnt cut it, thanks for all the work <img src=http://www.ezboard.com/images/emoticons/eek.gif ALT=":eek"> <hr></blockquote>



i put lazy in there cuz i didnt like having to FD between Dragon Fang clicks before they removed the aggro =)






--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Nedrom]

added 1167 id5 parse



if i change the order, its interesting to note the following:





Raw AC AC Min % Max % Ave

0 853 4.08% 37.55% 360.19

<span style="color:blue;">100 1010 12.46 18.20 302.44</span>

100 1010 12.73 18.79 301.83

<span style="color:red;">100 1010 13.05 18.41 301.40</span>

200 1167 16.35 14.81 283.29

<span style="color:blue;">200 1167 18.33 12.34 271.20</span>

300 1325 19.10 11.67 269.88

<span style="color:red;">200 1167 19.48 9.70 267.03</span>

<span style="color:blue;">300 1325 20.12 11.49 265.62</span>

400 1482 22.47 9.55 259.66

<span style="color:blue;">400 1482 23.81 8.92 254.06</span>

500 1639 23.87 8.35 250.20

<span style="color:red;">300 1325 24.87 8.78 247.58</span>

<span style="color:red;">400 1482 25.86 7.32 243.02</span>

600 1798 25.71 7.28 242.67

<span style="color:blue;">500 1639 26.35 7.64 242.55</span>

<span style="color:blue;">600 1798 28.41 6.32 234.29</span>

<span style="color:red;">500 1639 30.11 6.00 229.45</span>

<span style="color:red;">600 1798 31.47 4.68 224.32</span>



if you dont have high AC, getting the AA's allows you to mitigate like someone with higher AC with no AA's



obviously having both helps in the long run...


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/27/05 7:15 am

[Demandred]

--removed graph--


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=dem andred>Demandred</A> at: 5/28/05 7:14 pm

[Ancaglon]

To me, it seems that CS3 "just" makes your AC worth more after the cap (gradient is 21% steeper with CS3 over noAA .... maybe coincidentally, 10% squared = 1.1**2 = 1.21) , while ID5 raises the cap and may make your AC worth more (ID5 line is 8% steeper than CS3 line). However, both lines are quite 'noisy' (r-squared values are not very tight for the CS3 and ID5 lines compared to the post-cap noAA line).

http://www.xenaphon.demon.co.uk/anc/AC/nedmit_cs3id5.gif



The upshot of all this is however as Nedrom stated. For example, having 300 raw AC and ID5 lets you take damage about as well as someone with 550 raw AC and no AA's.


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=anc aglon>Ancaglon</A>* http://www.ezboard.com/image/emoticons_classic/devil.gif at: 5/27/05 3:53 am

[Nedrom]

defensive aa's only raise the cap, that is all, they dont change the return of ac



added 1010 ac id5 parse...as you see, once again 1010 is the same from no aa to cs3 to id5



the only thing i dont understand is why is the damage jumpy....and not solid and consistant like when i had no aa used at all...maybe that is the reason why we see inconsistancies with the damage we take even after the changes?


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/27/05 7:13 am

[Evil Homer]

It's jumpy because components of the sum work like the "guard" spells do. There's only a chance your mitigation will be better. That's my opinion of course.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

[Ancaglon]

<blockquote>Quote:<hr> defensive aa's only raise the cap, that is all, they dont change the return of ac<hr></blockquote>

The fact that the gradients of the lines differ significantly argues otherwise.



Any plans for a parse with the GoD defensive AAs? <img src=http://www.ezboard.com/images/emoticons/happy.gif ALT=":D">


</p>

[Nedrom]

as soon as i get some more aa's from Kavhok I'll post the DI OOW Mitigation aa's...but i'm working on PE right now



and i asked, Kavhok told me that all the aa's do is raise the cap


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/27/05 12:37 pm

[Zonn]

Just wondering, to you plan any parses of Shielding once all AA deffensive abilities are capped? I'd be interested in say 10% and 20% parses to get an idea what type of impact it has.


<span style="color:red;font-family:helvectia;font-size:large;">Axiss Glox</span>

<span style="color:red;font-family:helvectia;font-size:small;">70 Transcendant of Stasis (http://www.stasis-guild.com)</span>

Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=10004)

Axiss@Stasis-Guild.com (Axiss@Stasis-Guild.com)</p>

[Nedrom]

ill ask if i can do that too


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Demandred]

<blockquote>Quote:<hr>Just wondering, to you plan any parses of Shielding once all AA deffensive abilities are capped? I'd be interested in say 10% and 20% parses to get an idea what type of impact it has.<hr></blockquote> But we already know what shielding does.


</p>

[Zonn]

Knowing what it does and seeing all of the data compiled in a single test case, then graphed gives a clearer picture of its effectiveness.


<span style="color:red;font-family:helvectia;font-size:large;">Axiss Glox</span>

<span style="color:red;font-family:helvectia;font-size:small;">70 Transcendant of Stasis (http://www.stasis-guild.com)</span>

Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=10004)

Axiss@Stasis-Guild.com (Axiss@Stasis-Guild.com)</p>

[azathool]

I dont see how shielding parses are any use.



It's a directly known amount. N% off the DB of the mob. So tell me the DB of the mob your parsing on, remove the N% of DB and hey presto, no need to parse at all. And also of course, since it's removing the DB portion that depends on the mob your fighting.



I think of more value would be to do the same above parses but for the avoidance AA's, and for avoidance mods, since there are unknown.


<hr>
<div align="left">
<table border="0" width="100%">
<tr>
<td width="15%"></td>
<td width="70%">
<p align="center">[Stone Fist] Azathool (http://www.magelo.com/eq_view_profile.html?num=597)
<Nightwatcher (http://www.nwguild.com/)></td>
<td width="15%">
<p align="center">
[ Shm (http://www.magelo.com/eq_view_profile.html?num=4174)
| Clr (http://www.magelo.com/eq_view_profile.html?num=1024677)
]</td>
</tr>
</table>
</div>

</p>

[Nedrom]

*grumbles*



someone casted buffs on me this morning so one of my PE parses is bad, gonna have to redo it, not happy :P



also when your attacking a summoned NPC your buff timers dont count down and disappear so thats how i know the parse was broken =(


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/27/05 4:16 pm

[santino]

One interesting point (to me at least) is that our vastly improved rate of return post-cap (compared to how it used to be), combined with our low soft cap, make the mitigation aas seem less useful.



ie. If Demandred's graph is correct, and cs3 raises the soft cap by a grand total of 19 raw ac... well, to me it then becomes questionable whether the 12 aas are worth it. Before, when we were getting 1/40 post-cap, raising the soft cap a few points was huge because doing so equated to a few hundred raw ac after the cap, (ie. raising my soft cap by 20 is equivalent to getting 800 more raw ac, post-cap).



With the post-cap returns so much more reasonable, the mitigation aas seem a lot less relevant... something to eventually get, perhaps, but nowhere near as important as they might once have been.



Incidentally, raw ac = the actual ac value of each item added up = the 'AC' component in Magelo's Mod tab, right? If so, then it seems to me even moderately geared monks will still exceed the cap... I'm at 303 myself right now, and am by no means uber.


Fumika (http://www.magelo.com/eq_view_profile.html?num=983997)

Guild Leader of Court of the Crosswinds</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=san tino>santino</A> at: 5/27/05 4:17 pm

[Zonn]

Then guestimate graph Shielding at 10% and 20% onto the existing graphs. The point is to show comparative effectiveness to the Mitigation AAs. If you show it graphically, with data it's much more quickly quantified in relative effectiveness. It can then help people decide how much to pursue Shielding along with AC in their gear choices.



Until you actually see the data integrated with the parsed information it's much more difficult to judge the quality. But if you plot out 20% shielding and see an very large dip in average damage, it sure would drive home the point.



Either way, adding the data to the mitigation graph is a useful exercise and parses to further backup guestimates would be beneficial.


<span style="color:red;font-family:helvectia;font-size:large;">Axiss Glox</span>

<span style="color:red;font-family:helvectia;font-size:small;">70 Transcendant of Stasis (http://www.stasis-guild.com)</span>

Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=10004)

Axiss@Stasis-Guild.com (Axiss@Stasis-Guild.com)</p>

[Nedrom]

<blockquote>Quote:<hr>even moderately geared monks will still exceed the cap... I'm at 303 myself right now, and am by no means uber.<hr></blockquote>



thats the whole point of this parse...to show you that the cap is allready exceeded by a lot more monks than you realize


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/27/05 4:20 pm

[Nedrom]

...hard to say if this is a bug...but i just added PE parse at 1798 AC



600 1798 25.71 7.28 242.67

600 1798 28.41 6.32 234.29

600 1798 31.47 4.68 224.32

600 1798 31.01 5.59 230.88



this includes all defensive aa, cs3, id5 and PE



im going to run the 1639 test now and see if we get the same results...



max hit is 17% worse and average hit is 3% worse after buying PE...



sounds odd but lets wait til i get the next test.


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/27/05 5:59 pm

[Demandred]

bug for sure


</p>

[Nedrom]

hmm, test server just went down, i guess no more parses today, too soon to start up another before raid =


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[santino]

<blockquote>Quote:<hr>thats the whole point of this parse...to show you that the cap is allready exceeded by a lot more monks than you realize<hr></blockquote>



Nod, which is what led to my post on the 'is mitigation fixed' post saying how these parses agree with my own experiences, and explained why the initial repeal and soft cap raising) felt far less significant than the drastic change to post-cap gains we most recently received.



As far as I know, there's no disagreement currently on that subject.



That conclusion was really tangential to the point of my post, however, which was that the mitigation fixes we've recently received actually make the mitigation aas less meaningful than they were previously. <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">




Fumika (http://www.magelo.com/eq_view_profile.html?num=983997)

Guild Leader of Court of the Crosswinds</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=san tino>santino</A> at: 5/27/05 6:59 pm

[Demandred]

<blockquote>Quote:<hr>That conclusion was really tangential to the point of my post, however, which was that the mitigation fixes we've recently received actually make the mitigation aas less meaningful than they were previously.<hr></blockquote>I'm not sure what you are smoking, but it's probably not good for your brain.


</p>

[Nedrom]

santino speaks the truth



lets make this hypothetical



before april may patch:



monk_a has 1000 AC

softcap is 900 AC

monk_a has CS3 which makes the softcap: 1000 AC



after patch:



monk_a has 1000 AC

new softcap is 1100 AC

monk_a has CS3 which makes the new softcap: 1200 AC



monk_a now gets less benefit from CS3 and has to actually search for more AC on gear to take advantage of it...hence why he posted the above statement



i do see where he is coming from...



but as you see from my parses, even ID5 has a huge benefit at low level AC, Ex: 1167 parse


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/27/05 7:39 pm

[Demandred]

wtf... let's say we have two monks with 1300 ac, way above the caps. Mitigation monk has cs3 and id5, while lazy monk has has no mitigation aa's. To make up for this, on this particular test mob, the lazy one would need 20% shielding to get the same average hit as the mitigation monk. It's a false statement to say these aa's become less meaningful after our fix. Back in our nerfed days the lazy one would be laughing his ass off because the mitigation monk had spent tons of aa, only to get a mitigation improvement of 2-3% shielding.


</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=dem andred>Demandred</A> at: 5/27/05 8:10 pm

[Demandred]

<blockquote>Quote:<hr>hmm, test server just went down, i guess no more parses today, too soon to start up another before raid =<hr></blockquote> PE fix incoming :)


</p>

[Nedrom]

there is no way i screwed up two 90 minute parses



i made no changes, i had no buffs, i had the exact same stats



i went from no aa, to cs3 to id5 and then added PE



now PE is making my defense worse...why?



500 1639 23.87 8.35 250.20

<span style="color:blue;">500 1639 26.35 7.64 242.55</span>

<span style="color:red;">500 1639 30.11 6.00 229.45</span>

<span style="color:purple;">500 1639 30.78 7.44 233.65</span>



600 1798 25.71 7.28 242.67

<span style="color:blue;">600 1798 28.41 6.32 234.29</span>

<span style="color:red;">600 1798 31.47 4.68 224.32</span>

<span style="color:purple;">600 1798 31.01 5.59 230.88</span>





I'm bringing this to Kavhok's attention



1639 is almost 20% worse for max % hits and average hit is 2% worse



if this is on live, who knows how long it has been bugged...


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/28/05 1:34 am

[Nedrom]

400 1482 22.47 9.55 259.66

<span style="color:blue;">400 1482 23.81 8.92 254.06</span>

<span style="color:red;">400 1482 25.86 7.32 243.02</span>

<span style="color:purple;">400 1482 25.08 8.09 244.75</span>



PE strikes again!


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[santino]

<blockquote>Quote:<hr>I'm not sure what you are smoking, but it's probably not good for your brain.<hr></blockquote>



I don't smoke, but I'm happy to start if it will make you smile. <img src=http://www.ezboard.com/images/emoticons/wink.gif ALT=";)">



The relative value of the mitigation aas per aa cost is less now than pre-fix.



Pre-fix, it was more important to get aas that raise the soft cap than it is now. Why? Because the ac gains post-cap were so much worse, (1:40 vs. 1: 2-7ish)



The cost of the aas remains static, the value increase per mitigation aa also is presumably static... yet the soft cap increases gained are less significant precisely because our post-cap gains are accelerated. Obviously, increasing the soft cap nets you 1 ac per each ac point the soft cap increases, then and now... but, whereas before that 1:1 is replacing a 1:40 ratio, now it's replacing a 1:7 (or whatever).



That's what I mean when I say the relative value now to pre-fix is worse.



Anyway, it's a minor point... just something that struck me while I was waiting for my project to compile.


Fumika (http://www.magelo.com/eq_view_profile.html?num=983997)

Guild Leader of Court of the Crosswinds</p>

[Ancaglon]

In absolute terms, you are better off now than before, just that the AAs aren't as critical to making use of the armour you already have. No loss there... it all helps <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">


</p>

[santino]

<blockquote>Quote:<hr>In absolute terms, you are better off now than before, just that the AAs aren't as critical to making use of the armour you already have. No loss there... it all helps <hr></blockquote>



Exactly. <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">


Fumika (http://www.magelo.com/eq_view_profile.html?num=983997)

Guild Leader of Court of the Crosswinds</p>

[Nedrom]

added 1325 PE parse, numbers are still off


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Nedrom]

added 1010 and 1167 PE parses



still odd numbers with PE...so i'm thinking there is something seriously wrong with it.



i did almost 20 tests with no weird results until i bought the PE aa



i was going to try to do the OOW DI aa parse this weekend but i dont think ill bother until we figure out what is going on with the PE aa


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 5/28/05 1:28 pm

[Ancaglon]

Ask Kavhok to check it out, this is too tricky for us to figure out just by parsing, I think.


</p>

[Nedrom]

can you provide an updated graph?


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Zonn]

My graph on the first page includes all the data points. It doesn't have best fit lines, but you can clearly see that PE actually made the mitigation worse in all but 1 data point.


<span style="color:red;font-family:helvectia;font-size:large;">Axiss Glox</span>

<span style="color:red;font-family:helvectia;font-size:small;">70 Transcendant of Stasis (http://www.stasis-guild.com)</span>

Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=10004)

Axiss@Stasis-Guild.com (Axiss@Stasis-Guild.com)</p>

[Nedrom]

Thanks Zonn, I linked your graph in the very first post so all the data is easy to find. Hope you don't mind <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Zonn]

Not at all, I'm trying to keep that single image updated as you post more data. So, it will stay as current as I keep it basically.


<span style="color:red;font-family:helvectia;font-size:large;">Axiss Glox</span>

<span style="color:red;font-family:helvectia;font-size:small;">70 Transcendant of Stasis (http://www.stasis-guild.com)</span>

Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=10004)

Axiss@Stasis-Guild.com (Axiss@Stasis-Guild.com)</p>

[Nedrom]

Zonn can you modify your graph legend to show:





*black = no defensive AA

*blue = CS3

*red = CS3 & ID5

*purple = CS3 & ID5 & PE



to avoid any confusion about what the graph lines mean




--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Sarkaukar]

<blockquote>Quote:<hr>still odd numbers with PE...so i'm thinking there is something seriously wrong with it.<hr></blockquote>



It wouldnt be the first time SOE programmers transposed numbers or forgot a sign. And if the problem is with the code, then all classes with ID/PE are likely affected by it.



PE should give an additional 2%.



From the parses so far, it appears that PE is reducing the ID portion of the avoidance AAs.


<div style="text-align:center">__________________________________________________ ____





Hunter Sarkaukar Half`Ear (http://www.magelo.com/eq_view_profile.html?num=27263)</div></p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=sar kaukar>Sarkaukar</A> at: 5/29/05 3:16 am

[Demandred]

Modified mit plot a bit. It doesn't explain why, but it's fun to look at.



http://users.du.se/~d98jja/mit_PE.gif






</p>

[Trinity Firestrike]

Ned, can you please ask Khav to double check the avoidance code portion of PE too?


http://img15.exs.cx/img15/555/betasig1.png</p>

[wolvurene]

be interesting to see how the OOW defensives parse out as well. But I think Ned has done enough for us :P


Wolvurene (http://www.magelo.com/eq_view_profile.html?num=921239)

Enlightened

Terror Australis (http://www.terror-australis.com)</p>

[sunol]

Great work everyone especially nedrom, don't know what it all means just yet <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">


http://65.254.93.38/images/avsigs/Sunol.jpg

Sunol 70 monk (http://www.magelo.com/eq_view_profile.html?num=195700)

Charlanne 66 Cleric (http://www.magelo.com/eq_view_profile.html?num=204375)

Besunol 65 Magician (http://www.magelo.com/eq_view_profile.html?num=568758)

Pleenk 65 Shaman

See What I helped kill (http://www.alteravita.net)

</p>

[Raptyr Tailfist]

I think it means that the nerf is revoked and we are actually a little better off than before. It also means that the PE aa is broken and people who bought it have been screwing themselves over.




http://members.cox.net/jkay/images/ateamsig.gif



My Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=43561)

</p>

[Evil Homer]

Told ya defensive AAs weren't worth it. To bad it took Ned to discover it was the PE making them worthless.



Jesus, how many years has this been in the game?





Soon as you finish defense Ned, best start parsing offensive AAs it's obvious no one has, yet.


Me. (http://www.thesoundarchive.com/somethingaboutmary/maryretards.wav)

Grabbit Allworth, Rogue Class Correspondent Who knew what was best for Monks. (http://bessiebee.com/Trivia/980927.wav)

Frodlin, Warrior Class Correspondent Who knew what was best for Monks. (http://www.reelwavs.com/movies/sounds/napoleon_dynamite/frigginidiot.wav)

The Moltenfists 15 minutes. (http://www.geocities.com/fozzy42/SoundClips/Movies/AnimalHouse/worthless.wav)</p>

wow, total weakness on PE AA, it really seems to hurt us by small amount that increases as AC gets higher #=X


#=D

Niha!

My gear is better then yours (http://www.magelo.com/eq_view_profile.html?num=680883)

http://www.creamedgeezer.com/quote/shakequote.php </p>

Wow ... PE hurting instead of helping .... what a feking joke. WTB a design/dev team with a clue Thanks for the parse work.



Fryarfej of Tarew Marr

Awakened

Drinal


</p>

[Raptyr Tailfist]

It's probably actually a programming error, not a Dev issue. One symbol in a line of code ...


http://members.cox.net/jkay/images/ateamsig.gif



My Magelo Profile (http://www.magelo.com/eq_view_profile.html?num=43561)

</p>

[Nedrom]

I still don't have any updates, I spoke to him last night but we're not sure what is going on.


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

we need to take this to the boards with our data and present it on the monk foums i think this is definatly a top 10 that should get fixed




</p>

[Nedrom]

PE is not just for Monks, every class that uses it could be affected.



there is allready a thread there as well:



eqforums.station.sony.com...ge.id=9262 (http://eqforums.station.sony.com/eq/board/message?board.id=monkbalance&message.id=9262)



plus the devs know about it, no need to pester them, im sure we'll figure it out, they cant have multiple melee classes wanting a refund it will just make everyone unhappy


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

Seem to remember when they changed the way the AA's where handled in code a while back the HP part of PE was broken . Prehaps the rest was affected as well but no one thought to check and as this is the first real set of controlled parses done it never came to light...


</p>

[Nedrom]

added some more 500 and 600 rac id5 and pe tests



/shrug



going to try to get DI tests done now



considering PE is only 2% difference on cap increase its possible that either i dont have enough data or my parse is not long enough so the numbers are too close to see a huge difference, plus 2% on the cap is not much



i can speed up the npc to include more data in the 90 min parse but i doubt this is really a bug just not enough parsing for 2% increase



ps: the code has been checked and the aa's are stacking fine and the server reports me as getting increased mitigation for every aa, just as an FYI


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 6/2/05 1:12 pm

[Nedrom]

added some level 70 tests, when those done im going to do DI5 (oow defensive)


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Demandred]

<blockquote>Quote:<hr>considering PE is only 2% difference on cap increase its possible that either i dont have enough data or my parse is not long enough so the numbers are too close to see a huge difference, plus 2% on the cap is not much<hr></blockquote>

Like I said before you can expect the average hit to vary +-3% from the real mean if you use a sample size of 2500 hits. To decrease this variation to 1.5% you would need 10,000 hits because the error decrease as 1/sqrt(n). The uncertainty of the mean hit may seem high when you consider the huge sample size of 2500. This is because of the huge deviation in the hit distribution (std dev=120 typically or roughly (Max-Min)/2). If our hitdistribution instead had the shape a of a bell curve, with a low deviation, we would only need a sample size of a few hundred hits to get a very accurate estimation of the mean value.



Anyway, 2500 hits should be ok for what you are trying to do. It was enough to show the difference between CS0 CS3 and ID5. That 2% increase in softcap (ID5->PE), should give a 0,5% increase in mitigation (0.00159*160*0.02=0.5%). The sample size you would need, to 'prove' there's a difference between ID5 and PE, is roughly 200,000 hits. Easy for SOE to simulate, but time consuming for us.


</p>

[Nedrom]

Demandred,



now that you put it that way, it makes sense.



i had ~3000 hits for all my parses.



all my tests now use a hasted NPC so that i get more data within the 90 minutes so that should help, but I'm sure if i ran a 12 hour test we would probably be able to notice that 2% better



i started a test for di5 when i left this morning so hopefully by the end of the day i'll have at least 2 DI parses for +600 and +500 rac up for you



as you can see with the level 70 and only cs3/id5/pe there is not much of a gain from simply raising the defense cap


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Scaleybutt]

I am curious at what point you guys think we will stop getting noticable returns on AC over the cap ?



Many think it's dumb, but I have sacraficed a lot of hp ( while still maintain over 11k ) to get my AC / end as high as i can with the augs in my possession. This has gotten me up to a buffed AC of 2632 ( conviction, guard of earth, tribute ac, 115 avoidance, 15 shielding, max AA ).



Before I farmed the 4 25ac augs and upgraded my earring, I buffed up to roughly 2350 AC. This allowed me to tank RS trash efficiently tossing a CH at roughly 60-65%. Now, 280ac later, I can toss that CH around 40-45% with no worries of a bad round flattening me at all. Very noticable difference in my last 5-6 hour sessions.



None of this is parsed numbers like you have here / are hunting for. But it's many hours of first hand mbox experience. And what I really want to know is, at what point you think the returns will dimish? For what I do, I've adjusted my standards aug wise, and it's paid off greatly; but I'm wondering if there is a point at which i'll find myself maxed out defensive wise.



Thank you for your time and hopefully non-flamatory input <img src=http://www.ezboard.com/images/emoticons/wink.gif ALT=";)">


<div style="text-align:center">"Revenge is a dish best served cold......."</div>

<div style="text-align:center">Sensei Sveir Castigo - Europa - Druzzil Ro (http://www.magelo.com/eq_view_profile.html?num=532893)</div>

<div style="text-align:center">Oracle Sveri O'Shamus (http://www.magelo.com/eq_view_profile.html?num=613401)</div>

<div style="text-align:center">Exarch Salanan (http://www.magelo.com/eq_view_profile.html?num=1156568)</div>

</p>

[Nedrom]

afik there is no cap on the softcap so as long as you keep adding ac, the return over cap still works



i can't tell you how to do it but from the numbers i have on the graph you should be able to determine the next one in the line easily if your a math wiz, but i am not



when done the di5 parse i can always run a test at 3000 ac and see what happens <img src=http://www.ezboard.com/images/emoticons/wink.gif ALT=";)">


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 6/3/05 11:01 am

[Nedrom]

600 1798 25.71 7.28 242.67 --- no AA

600 1798 28.41 6.32 234.29 --- CS3

600 1798 31.47 4.68 224.32 --- ID5

600 1798 32.27 5.79 226.78 --- ID5 #2

600 1798 31.01 5.59 230.88 --- PE

600 1798 32.54 5.72 226.86 --- PE #2

600 1798 33.29 5.63 227.40 --- PE #3

600 1850 31.90 5.18 226.22 --- Level 70

600 1850 33.99 4.74 223.42 --- DI5 Level 70 90 min parse

600 1850 34.06 4.83 221.75 --- DI5 Level 70 ~10 hour parse






--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Sarkaukar]

Demandred, I understand that and it may be true but going from no AA to CS1 is 2% increase and from CS1 to CS2 is going from a 2% to 5%.



Could Ned had been lucky on his previous pre-PE parses? Maybe. Or SOE is not looking at the code properly (wouldnt be the first time). As previously mentioned initially PE was screwed up, noticeable mostly from the sta-hp conversion.



Regardless, the overall parsing data have provided some interesting results. Thank ya Ned.


<div style="text-align:center">__________________________________________________ ____





Hunter Sarkaukar Half`Ear (http://www.magelo.com/eq_view_profile.html?num=27263)</div></p>

[Demandred]

hehe 10 hour parse, that's nice, 20k hits?



<blockquote>Quote:<hr>Could Ned had been lucky on his previous pre-PE parses? Maybe. Or SOE is not looking at the code properly (wouldnt be the first time). As previously mentioned initially PE was screwed up, noticeable mostly from the sta-hp conversion.<hr></blockquote>I'd say with the number of parses he has made so far he has hit a few good ones and a few bad ones (like PE). But atleast it's clear id5>cs3>cs0. Also with Ned's new parses at level 70 there's a significant difference between PE and DI5 at 600 rAC; roughly +2% in mitigation and 5% increase of softcap.


</p>

[Nedrom]

well it turns out in order for me to get an accurate measure of PE, considering it is only 2% difference, my parses will need to be much larger. I only have about ~3000 swings in the PE parses and i guess i'll need about 100,000+ in order to see the true changes.



i dont have time to redo all those tests with a larger scale of numbers. statistics shows that the above information can obviously not be as accurate as i had hoped considering a 2% change well...its the difference between cs0 and cs1...and while its not huge, may not always show up in every parse considering the other variables.



Kavhok checked with me on test server and watched my character as i bought cs3, then id5 then PE and saw that in each occurance my mitigation was reported as being higher by the server.



i think perhaps we judged the results too early over a 2% mitigation change, so i don't believe it is broken, i just believe that i dont have enough data.



its just like doing surveys...if you dont have enough data the results may be misleading.



either way, i should have the DI5 parses done by the end of the weekend.


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

<blockquote>Quote:<hr>I am curious at what point you guys think we will stop getting noticable returns on AC over the cap ?<hr></blockquote>



Without question, AC produces diminishing returns for every AC over 1. A min hit can't go any lower: it's still a min hit. Therefore, the number of hits that further mitigation works on will be lower and lower as you get more and more AC.



At level 65, 0 AA, 0 raw AC, the test monk was hit for min 4.08%. Therefore, the non-min hits made up 95.92% of the hits. Fast forward to level 70, 600 raw AC, CS, ID, DI and PE AAs. That monk had 65.94% of the hits land above the min hit.



In the first case, 95.92% of the hits will benefit from further mitigation. In the later case, only 65.94% of the hits will benefit from further mitigation.



Yes, at some point, when the monk is taking, say, 90+% min hits, AC will cease to matter since almost everything is a min hit already. However, it's safe to say that monks can't get that much AC with current gear fighting blue, experience giving mobs. In fact, a level 70 monk with 1800+ AC fighting a level 20ish mob (Guard Collin and Vallon in WC) only gets hit for min 44.03% of the time (before the monk mitigation changes though). It can't get much greener than that, and yet mitigation was nowhere near the 90+% level.



Sear


</p>

[Nedrom]

btw sorry for only 1 hour parse on +500 ac...test server went down and raids are starting so i just posted that, will do a longer one when i can...but it looks like min hit can only get so low on this level 60 mob


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[SkyKungfu]

<blockquote>Quote:<hr>600 1798 31.47 4.68 224.32 --- ID5

600 1798 32.27 5.79 226.78 --- ID5 #2

600 1798 31.01 5.59 230.88 --- PE

600 1798 32.54 5.72 226.86 --- PE #2

600 1798 33.29 5.63 227.40 --- PE #3<hr></blockquote>



<blockquote>Quote:<hr>Also with Ned's new parses at level 70 there's a significant difference between PE and DI5 at 600 rAC; roughly +2% in mitigation and 5% increase of softcap.<hr></blockquote>



Hmm somehow I fail to see this increase ...



Ah sorry, you meant the increase is there with DI 5 being better than PE? I tought you saw somewhere that PE worked =)



Sky


<span style="color:blue;"> SkyKungfu </span> lvl 66 Grandmaster of the Celestial Fists Guardians of Doom Bertoxxolous
SkyBasher lvl 60 Warlord / SkyTree lvl 62 Druid / Crakman lvl 66 Enchanter
Akabriel lvl 65 Arch Mage / Gospeed lvl 45 Shaman / Skyscat lvl 45 Beastlord


Mageloprofile SkyKungfu Lee (http://www.magelo.com/eq_view_profile.html?num=257891)


<span style="color:blue;"> World of Warcraft </span>
Miracculix, Tauren Druid, 27 - SkyFighter, Orc Warrior, 60 Bleeding Hollow / Horde

Braddy, Nightelf Hunter, 11 - Lothar / Alliance
</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=sky kungfu>SkyKungfu</A> at: 6/3/05 7:54 pm

[Nedrom]

finished DI5 tests



doing one last test at 2951 AC


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Nedrom]

added 1300 raw ac (2951 displayed) parse



this shows that even near 3000 AC we reap the benefits


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[SkyKungfu]

Guess AC has really a meaning again for monks. Thanks for the great job and the insane amount of your time investment.



/bow



Sky


</p>

[Nedrom]

was no sweat, i had season 1 and season 2 of Alias to watch <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)"> lol



it is nice though, to know that no matter how much AC we pile on, all of it will be worth it.






--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Dythan Dabrave]

So now the debate on....



hp vs ac on gear



1-2?

1-4?



and and what raw ac values do we suggest to buy each tier



/bow to Ned, you have done us all a great service.




Dithan

70th season monk

Officer

Forsaken Visage



The only difference between a Hero and a Fool is Success.</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=dyt handabrave>Dythan Dabrave</A> at: 6/4/05 7:36 pm

[Nedrom]

HP always better than AC if your in high end because ac wont save you from ae's, dots etc..



but if your casual and tank often ac is worth it


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

[Scaleybutt]

After one grouping Qvic goats last night, with me as the tank and one cleric ch'n with ease, I'm going to stick with my modified value in augs: 25AC aug > than 50HP aug, but NOT > than 50HP/END aug.



My real time experience with large AC numbers has left me with a smile on the whole change. I only hope our next goal in line brings us as noticable a change. <img src=http://www.ezboard.com/images/emoticons/wink.gif ALT=";)">


<div style="text-align:center">"Revenge is a dish best served cold......."</div>

<div style="text-align:center">Sensei Sveir Castigo - Europa - Druzzil Ro (http://www.magelo.com/eq_view_profile.html?num=532893)</div>

<div style="text-align:center">Oracle Sveri O'Shamus (http://www.magelo.com/eq_view_profile.html?num=613401)</div>

<div style="text-align:center">Exarch Salanan (http://www.magelo.com/eq_view_profile.html?num=1156568)</div>

</p>

[Shazzamx]

Really great work Ned. Any chance to get a Moderator to make your post into a sticky somewhere so it dosen't get lost?


</p>

From my perspective AC is not worth thinking about and HPs are.



I don't give my AC a second thought, especially now. While many see this data as meaning AC matters, the way I see it, AC is nice but like everything else in the game it is tiny increments of improvement in exchange for massive additions. /shrug AC will take care of itself.


http://i5.photobucket.com/albums/y193/tauzii/TauziiLFG.jpg (http://www.magelo.com/eq_view_profile.html?num=852244)

</p>

[Nedrom]

Zonn can you please update your graph to include DI5?

www.stasis-guild.com/Miti...Damage.jpg (http://www.stasis-guild.com/MitigationParses/AverageDamage.jpg)


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>Edited by: <A HREF=http://p201.ezboard.com/bmonklybusiness43508.showUserPublicProfile?gid=ned rom2004>Nedrom2004</A> at: 6/6/05 3:50 pm

[Zonn]

It's updated


<div style="text-align:center"><hr />Axiss Glox (http://www.magelo.com/eq_view_profile.html?num=10004)Level 70 Stone Fist [url="http://www.stasis-guild.com"]http://www.stasis-guild.com/Stasis/Images/Stas</div></p>

[Buhd]

/Bow



thanks Ned for all that..wow..



Its kinda strange to read good news..hehe



cheers


Sensei Brotor Tiller

unRetired

Rathe

League of Levity



</p>

[Demandred]

Q: You said the test mob was hasted. What kind of haste was that?


</p>

[Nedrom]

/say increase melee speed

kept doing this until he hit faster



/shrug



doesnt affect anything but the rate at which i took damage


--

---

Nedrom (http://www.magelo.com/eq_view_profile.html?num=1018594&resize=true)

The Laziest Level 70 Iksar Monk

</p>

Seeing this data is a real treat. However, I'd give a whole pile of platinum to see a few of some of those same tests redone with a warrior, and a paladin.


</p>

[Nedrom]

wont happen <img src=http://www.ezboard.com/images/emoticons/smile.gif ALT=":)">


</p>

[Tzaj]

I think I read on the EQlive boards that the aa's affected soft caps values, but only when you were at or above that value for your level. (softcaps changes by level so a 70 monk needed to reach a higher AC value than a 50 monk, the part I wasn't sure on was were there different AC softcaps for each batch of aa's to start working)



The question I would like to know the answer to is;



taking a level 70 monk as a starting point at what AC value will,

1) CA/CS

2) ID/LR

3) RM/DI

actually start to work?


</p>

[Bolehcim]

I gotta say... I'm surviving raid content more frequently... Used to be the boss would turn and just eat me with 4 max hits.



Lately, I get 1-2 max hits, and a couple lower end ones, letting me survive and FD @ 5% instead of just loading into the Guild Lobby.


</p>

/kneel ned



great work bro!


http://img321.imageshack.us/img321/2987/fvsig8es.gif



Brouce Leeroy

" Kiss my converse sucka! "</p>

[polkgaiimahn]

Gonna parse using a shield anytime soon?


</p>

[ssyxz]

amazingly ave hit 328.44/maxhit 471 * 100 = 69.9????

Not amazing since it's the exact same calculation you did in the first part of your post, just shorter.

[Nedrom]

Gonna parse using a shield anytime soon?



</p>

all a shield does is raise your ac softcap by whatever number the AC is on the shield.

so if shield is 100 AC, your softcap goes up 100

[wycca]

Bump.

Anyone save some of the graphs and mind re-hosting them?

[Myrilith]

Has anyone done parsing across a level spread of different mobs?

vs level 60, 65, 70, 75, 80 mobs?

And any parsing now with level 75 and the new AAs?

Hopefully we can get level 80 done soon also.


And I for one do think that a parse with full AAs and max shielding would be nice to have.