TheSafehouse



The test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function. -F.Scott Fitzgerald




Safehouse Forums



Safehouse Site Info



 

The end is nigh


 

 

 







FAQ Members List Calendar Mark Forums Read

Reply
 
Thread Tools
Imasu
Imasu is offline
Registered Member

Join Date: Dec 2007


 
Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #1

Hello my rogue friends. Last evening I spent an hour or so looking at the top ranked players on both the site eqleaderboards as well as magelo. What I'd initially expected to find was an itemization error mainly heroic strength or heroic dexterity. However it turns out that across berserkers, monks, rangers and rogues we're really not that far off from peak and average levels of these two stats.

Instead I ended up noticing a fairly significant gap in ATK between us rogues and those other 3 classes. Now granted both the eqleaderboards and magelo are not 100% accurate (really I dont trust either of them fully) however there is a trend forming and I would like to hear ya'lls thoughts. In particular im looking to draw a connection between classes ATK values and their dps outputs. Rangers being hybrids are going to be allowed much room for argument simply by virtue of not having all their dps done via melee/ranged. I would like to point out though im told they'd been given significant self ATK buff upgrades this xpac.. coincidence? Complete hearsay? Who knows for sure, its just food for thought. Anyhow here are the trends.

Alphabetically
Berserkers,
Custom Leaderboard filter all servers, berserker order by ATK
Top ATK = 3,263
51st ranked = 3,181
100th ranked = 3,135

Monks,
Custom Leaderboard filter all servers, monk order by ATK
Top ATK = 3,065
51st ranked = 2,974
100th ranked = 2,955

Rangers,
Custom Leaderboard filter all servers, ranger order by ATK
Top ATK = 3,180
51st ranked = 3,088
100th ranked = 3,064

Rogues,
Custom Leaderboard filter all servers, rogue order by ATK
Top ATK = 3,025
51st ranked = 2,908
100th ranked = 2,877

Berserkers,
Magelo filter all servers, L90 to L90 berserker
Top ATK = 3128
26th ranked = 3072
50th ranked = 3045
100th ranked = 2949

Monks,
Magelo filter all servers, L90 to L90 monk
Top ATK = 2923
26th ranked = 2860
50th ranked = 2842
100th ranked = 2807

Rangers,
Magelo filter all servers, L90 to L90 ranger
Top ATK = 3021
26th ranked = 2972
50th ranked = 2952
100th ranked = 2917

Rogues,
Magelo filter all servers, L90 to L90 rogue
Top ATK = 2877
26th ranked = 2835
50th ranked = 2815
100th ranked = 2780

So all that said.. my questions are what does ATK mean to actual dps? Are we talking every 10/25/50 ATK you get a flat dmg adjustment or what? How is displayed ATK calculated before buffs? Can some mechanic masters/experts weigh in on this to help further this observation please? I'm having a gut feeling this is were our passive dps issues are coming from after only a brief look in to these numbers. I just dont have the info/experience to truly make the argument.

Edit1:
Here's the formula to calculate displayed ATK per monkly business. Note this was posted 10-20-2007 and therefore may no longer be relevant? Looking for the relevant formula on magelo.. having some issues so ill edit that in later once figured out. Anyhow!

Weapon_Skill*2.7 + Offense_Skill*4/3 + Worn_Atk*4/3 + Str>75*9/10 + 10
Note: I believe the portion Worn_Atk*4/3 varies from class to class. Will look into this more.

So the weapon skills, offensives and worn atk maximums of our 4 classes are,

Berserker 470 2hb/2hs/piercing, throwing 370, 290 offense, 550 worn atk
Monk 390 2hb/h2h, throwing 370, 290 offense, 550 worn atk
Ranger 370 1hs/2hb/2hs/piercing, archery 390, 290 offense, 550 worn atk
Rogue 370 1hs/1hb/piercing, throwing 370, 290 offense, 550 worn atk

Any thoughts? More lawl?



Last edited by Imasu; 03-11-11 at 02:48 PM.
Pyro
Pyro is offline
Registered Member

Join Date: Mar 2010


Pyro's Avatar

 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #2

lol


skewerzjoo
skewerzjoo is offline
Veteran

Join Date: Oct 2003


skewerzjoo's Avatar

 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #3

I know a ranger in our guild buffs to over 7K attack during raids..I'm no where near that...I may be just a slacker, though...



Skewerz
Feochadan
Feochadan is offline
Registered Member

Join Date: Nov 2003


 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #4

ATK affects how you hit the damage you can do. More ATK will not give you more damage, just give you more chance to see the highest scores than the lowest. This is how it affects DPS.
And no, there is no relation between class DPS and ATK. Rangers have always had a much higher ATK than other classes, but the current nonsense DPS they have is related to some outrageously strong AAs they got in HoT.


Imasu
Imasu is offline
Registered Member

Join Date: Dec 2007


 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #5

Exactly. Moving our average damage per hit upwards through the DI is along the same fundamental change that all the discussions on innate crit/flurry/passivedpsboosting conversations are wanting to achieve.


Padgin
Padgin is offline
Registered Member

Join Date: Jun 2004
Location: The Zone....


Padgin's Avatar

 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #6

Imasu said:
Berserker 470 2hb/2hs/piercing, throwing 370, 290 offense, 550 worn atk
Monk 390 2hb/h2h, throwing 370, 290 offense, 550 worn atk
Ranger 370 1hs/2hb/2hs/piercing, archery 390, 290 offense, 550 worn atk
Rogue 370 1hs/1hb/piercing, throwing 370, 290 offense, 550 worn atk

Any thoughts? More lawl?
Imasu, you missed one very important skill most beloved to us rogues......
Backstab.... 280 with the backstab skill mod of 12% 313. Now that might be something to compare, rather than atk wich fluctuates depending on the rank of shammy and ranger buffs you get. Which for me most day's is 3850, although it can hit 3950 (no idea on that one cept mabe a larger over buff on rank 3 pred and tracker.

And I think 3850 is our cap atm, cause even if I add 70 more atk in augs or gear... no change in atk


Bargai_Stromm
Bargai_Stromm is offline
Registered Member

Join Date: Jul 2006


 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #7

Imasu said:
Exactly. Moving our average damage per hit upwards through the DI is along the same fundamental change that all the discussions on innate crit/flurry/passivedpsboosting conversations are wanting to achieve.
Except for one thing. Increasing attack does not ALWAYS move the average hit upwards though the DI tables.

Attack is used in conjunction with the targets Armor Class. The whole purpose of Attack is to defeat armor. Once Attack reaches a certain point, it can overwhelm a mob's AC. When this happens, DPS increases associated with Attack increases are so small as to be almost (not entirely) unparsable unless a massive amount of Attack is being gained. Essentially, the amount of Attack needed to boost DPS changes from mob to mob.

The Attack values you listed (even for the 100th ranked) are above what is required to overwhelm most mob's AC in game. This even holds true for many raid mobs.


Now don't get me wrong, more Attack is always better. It certainly won't hurt. But if rogues wage a two year campaign to get their attack increased by 500 over everyone elses, the results will be disappointing unless SOE starts increasing mob AC. (Note: except for rare instance mob AC has not significantly changed since around the OoW timeframe.) SOE is famous for giving the player what is asked for but not what is wanted. "You asked for more Attack and you got it. What's the problem." "It doesn't do much is the problem." "You got the Attack you asked for. Live with it."


Imasu
Imasu is offline
Registered Member

Join Date: Dec 2007


 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #8

Yeah im hearing what ya'll are saying. However I dont know that I agree with the statement that even the lowliest rogues atk is far exceeding all mobs AC. So far in this thread ive seen a wild range of atk's people are buffing up to. Which, btw, buffed atk supposedly affects your damage output differently than nonbuffed. I looked at is as per the difference between AC boosts. Of course I would have to cite material for this, so at the moment im just leaving it in as speculation. Cant find the source material so chalking this comment up to pure nonsense!

Anyhow I dont believe there IS a cap to atk right now via buffs. Some are quoting 3500 range, up to just shy of 4k. Last night though I was sitting at roughly 4500 before a burn started and getting all the atk buffs associated with clickies during that. Im positive I go well over 5k and my damage output shows it.

As to what it does, check out http://www.thesteelwarrior.org/forum...t=16017&page=2 in particular post #19 by Lluianae and the chart showing what happens to your average hit when your attack overwhelms a targets AC. Granted this was all done back in 01-28-2007 but just imagine if our atk was high enough to replace even say 50% of our minimal hits with something middle of the road.

Keep in mind however that a ton of base class skills like offense (which hasnt gone up since then) 1hs, piercing, 2handed w/e etc came in roughly.. march 2007, TSS era I believe. Since that time berserkers have been out pacing us significantly in skillups. Monks and rangers are still in line with us. In other words they're the same margin ahead of us that they've always been. Bererkers are not, and thats not even taking into account their aa's in 2hands no mercy or w/e its called which increases the skill values of their 2handed weapon use.

Id like to see what max atk's are being achieved by these 4 classes including buffs, clickies, kitchen sink. Tonight ill check and see how high I can get both irl and in game.

Btw special attacks like BS, frenzy etc are on a different calculation. The atk score used in them is different. Also archery is calculated differently using dex instead of str I believe.



Last edited by Imasu; 03-14-11 at 02:45 PM.
Imasu
Imasu is offline
Registered Member

Join Date: Dec 2007


 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #9

Okay to further illustrate this attack disparity. Pop open your gamparse, go to graphs clicks on hits, then pop in your name and fiddle around through there a bit. Notice the range of damage distribution for YOU. Now check out the rangers, zerkers, and monks.

Its highly possible its just that all my peers are uber in comparison to me and the other rogues ive had the luxury to parse. So im curious if your distribution of hits lines up better with your peers (your raid or group).

What I suspect you're going to see is that special attacks across zerkers, rogues and monks (frenzy, backstab and kick) are all roughly going to have the same distribution with most hits landing in the middle of the chart.

The the rest of the damage for us rogues (piercing and w/e your offhand is) is going to land pretty near the middle of the chart (seemingly as though our atk was what they based current mobs AC on maybe?) yet the distribution of hits for the other classes are all further to the right.


Imasu
Imasu is offline
Registered Member

Join Date: Dec 2007


 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #10

Highest I capped out at during raid burns tonight was 5717 for a brief time. I can see how a ranger hitting 7k+ atk (IF thats really what they're reaching) would skew the DI so far to the right on the charts given where im at with what I can achieve. Now I wonder what other rogues and the other classes are really hitting under raid settings.


brogett
brogett is offline
EQ Community
Round Table Member

Join Date: May 2004


 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #11

I started work investigating this, but ran out of steam. However see

http://www.thesafehouse.org/forums//...9&postcount=93

In particular the graphs are quite informative.

I decided to take a different approach, given I had access to a G11 keyboard with macros. Instead of doing long long parses under fixed conditions, I instead varied the conditions at known rates and looked at the graphs over time. This gives a good visual cue as to what's going on.

For example, use "increase mitigation" every 5 minutes on an arena dummy and then extract out all the min (DI1, remembering to remove any fero gear) or max (DI20) non-crit hits with a bit of basic perl / awk parsing.

The notion of a "cap" is basically wrong. These distributions are probabilistic and have specific shapes. As we increase attack or decrease mob AC we skew our distributions, but it's pretty much impossible to ever get to the situation where *EVERY* hit is DI20. Hence the cap really is just an asymtote which you gradually approach. I prefer to think therefore of attack as having diminishing returns.

Unfortunately it's clear that the mitigation of arena dummies is nowhere near the same as mitigation of even trash mobs, let alone raid bosses. You can get a rough ballpark figure though just by looking at distributions and trying out various "increase mitigation" figures on the arena dummy until you get something looking about the same. 30-50 should about do it in my experience, but I haven't done this in some time.

Anyway if you make any findings please share them with us. :-)


Imasu
Imasu is offline
Registered Member

Join Date: Dec 2007


 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #12

I wont lie I was sort of trolling this technical stuff hoping to lure you out of hiding, Brogett.

Considering you've a lot more experience with sorting through this stuff, yet dont have the time to dedicate to it being but one rogue, if we could come up with some tests to run ill reacquaint myself with the test server and get to generating as much data as we need.

As I see it the unknowns are what peak (burn situation) atk values the various classes are achieving. I suspect that with equal atk values most of the classes are falling in line with one another. What im assuming is that the buffed/situational atk values are varying wildly. So then how to go about collecting this info considering even here on the rogues class forums ive only been able to really pull TWO peak atk values for our own class and one of them id take with a grain of salt because I dont think they were taking me seriously. :P

Is this an exercise in futility and its better to just move on to another class thats equal or better than us at our primary role yet has much less risk as well as other roles they can perform? Say it aint so~


Catweazel
Catweazel is offline
Veteran

Join Date: Dec 1969


Catweazel's Avatar

 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #13

There is less work rolling a new class (say a ranger) from level 1 to 90 with 1000AAs than there is in trying to prove Rogues would benefit from (say) +500ATK.

At least one of those options provides you with a completely overpowered toon with tons of utility to solo, molo, group or raid.


Bargai_Stromm
Bargai_Stromm is offline
Registered Member

Join Date: Jul 2006


 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #14

Catweazel said:
There is less work rolling a new class (say a ranger) from level 1 to 90 with 1000AAs than there is in trying to prove Rogues would benefit from (say) +500ATK.

At least one of those options provides you with a completely overpowered toon with tons of utility to solo, molo, group or raid.
I didn't say that adding 500 Attack would not benifit rogues, it would. What I am trying to say is that increasing attack is not a silver bullet.

The other classes are not where they are because of their Attack ONLY. Attack is only a part of it, a small part of it. In the case of rangers, they do not have a single significant advantage in any one area, they have a series of minor advantages multiple areas that reinforce each other. If you want to claim they have a significant advantage, the only thing you can point at is the synergy that their tools develop, not the tools themselves. You won't beat that just by increasing Attack. MORE is needed than just Attack.


Catweazel
Catweazel is offline
Veteran

Join Date: Dec 1969


Catweazel's Avatar

 
Re: Call for ATK mechanics info, stats, calculations and observations
Reply With Quote   #15

I was just using it as an example - there is a shedton of work to try and prove what Imasu is looking for and then what?

Just read the current whinefest on EQLive for an insight into perceived problems. It's not so bad that players (classes other than rogues) generally disagree that rogues are underpowered. What is bad that the dev quote from Eli did not even acknowledge there is a problem any more.

It's interesting (interesting as in watching the special Olympics) to see Hatshe now pushing for elimination of Nature's Fury and then wait a month or two to see if Rangers are still over powered.

It's lolworthy enough to take a few months off.




Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump







 
All times are GMT -7. The time now is 08:00 PM.
All posts are
© Copyright 2004-2006 The Safehouse Network