Toast - Kissassist does significantly less DPS than you manually pressing attack

[LurkMcGurk]

I'm creating this post to share my findings with the community and to nip this argument in the butt cause I am tired of having it. This post isn't an attack on Kissassist or on your skill personally, if you think it is you should take a step back from the computer and calm down before ranting to me - I am creating this post for the sole purpose of; that I am dead ass tired of being attacked over this subject and debating people about it. I personally love Kissassist, I think it's the best bot I have ever used in any game.... But that doesn't change the fact that it does less DPS than a real player can and that is how it is designed. The amount of DPS less it does is usually about 20% - 50% but when you pop burns, it does up to half the DPS of a real player (real player = someone who is manually pressing the attack button, this includes boxers). This is actually common sense throughout the macro coders portion of this community. I have tried working with Ctaylor22 to make KA do more DPS but we failed. Many people who acknowledged the fact that Kissassist does less DPS than a real player made their own macros, like EoKnuke.

Here is my very simple argument and what I will be proving with this thread: If you manually press the attack button to DPS on your characters and you don't rely on Kissassist, you will do more damage. This is most noticeable during Burns/Cooldowns where Kissassist's slowness in popping cooldowns and attacking is exacerbated.

Very simple, very straight forward and easy for anyone who disagrees with me to go out and test themselves. Here is a post I made on my website about this issue that goes more indepth about the 2 main reasons why Kissassist does less DPS. This post on Redguides is going to be more of an FAQ and show my math for the findings. ((In the future I will be updating that post on my website with more math too; once Gamparse is feeling better and once I start raiding again in a few weeks. I will cover multiple scenarios and situations proving beyond all reasonable doubt that this is the case)).

The experiment: To collect these numbers I took 2 Wizards who are identicle in every single way to Lceanium as well as 1 tank and 1 healer. I had one Wizard fully automated with Kissassist using the same exact spells as a Wizard that I played manually. Each time I pulled a mob, I waited for the Kissassist Wizard to engage then I began attacking myself. I pressed my single button to attack with all 3 Wizard Forces and 3 spells bound to it at about 120 BPM. Here are the results of my dozen or so tests:

/GU A massive tundra yeti in 62s, 1566k @25264 | ManualWizard 975k | KissassistWizard 391k | Tank + pets 160k

/GU A stonegazer cockatrice in 53s, 1251k @23610 | ManualWizard 780k | Tank + pets 142k | KissassistWizard 105k

/GU A stonegazer cockatrice in 48s, 774k @16120 | ManualWizard 307k | KissassistWizard 253k | Tank + pets 153k

/GU A petrifier cockatrice in 55s, 1592k @28950 | ManualWizard 737k | KissassistWizard 480k | Tank + pets 212k

/GU A calcifier cockatrice in 55s, 1295k @23540 | ManualWizard 497k | KissassistWizard 493k | Tank + pets 142k

/GU A tundra yeti in 60s, 1261k @21018 | ManualWizard 777k | KissassistWizard 211k | Tank + pets 129k

/GU A stoneglare cockatrice in 81s, 2201k @27177 | ManualWizard 958k | KissassistWizard 930k | Tank + pets 191k

/GU A petrifier cockatrice in 67s, 1647k @24588 | KissassistWizard 713k | ManualWizard 695k | Tank + pets 198k

/GU A calcifier cockatrice in 61s, 1619k @26547 | KissassistWizard 688k | ManualWizard 509k | Tank + pets 198k

/GU A petrifier cockatrice in 55s, 1470k @26719 | ManualWizard 813k | KissassistWizard 402k | Tank + pets 173k

/GU A petrifier cockatrice in 70s, 1083k @15478 | ManualWizard 559k | KissassistWizard 234k | Tank + pets 188k

/GU A calcifier cockatrice in 66s, 2084k @31582 | ManualWizard 1163k | KissassistWizard 506k | Tank + pets 212k

As you can see, the Wizard who was using Kissassist to attack did DRASTICALLY less DPS than the Wizard with whom I was manually attacking on throughout almost every fight. The reason for this is because he attacks at about half the speed as the Wizard with whom I am manually attacking on. Kissassist is just a program and it's unreliable, it doesn't always work as intended. It may sound obvious, but for some reason it isn't in the Redguides community, you will *always* do better if you have more control over your characters. Complete automation drastically decreases the output of that character. In Everquest, a game that's built around min/maxing this is an extremely big deal.


Cooldown/Burn DPS Difference

Note: Kissassist didn't pop burns on either of the names I tried to parse, because it doesn't always work as intended (one of the reasons it does low DPS). Therefore, I was unable to get an accurate burn parse today. I will do another experiment in a few weeks which will have dead on balls accurate math for any naysayers who think that the only reason I got the results I did was because of my timing. In the future I will be focusing almost exclusively on Cooldowns/burns because that's where it's most noticeable that Kissassist does way less DPS than you could by pressing the button yourself manually.

Truthful Addendum: When I did this math it was on the same day Burning Lands was released. Gamparse is currently experiencing issues, it looks like it's forgetting to add a 0 to the total DPS and Kissassist is also being dumb and not popping burns (which isn't very unusual, it happens off and on for me all the time). This is the worst time for me to do this experiment if what I wanted was completely 100% dead on balls accurate math. I'm openly admitting this and being truthful because I want to run this experiment again in the future with more math & because I would like everyone here who doubts me to run the experiment themselves as well. Typically I would wait until Gamparse is fixed so that I could have perfect math but I have a very busy life right now and I barely even have enough time to make this post and explain the situation in detail. It also doesn't seem like Gamparse is so broken that it's just radically displaying the wrong numbers entirely, it looks like it's just failing to count a 0 at the end of each number when the number is above one million. If it were the case that Gamparse was so radically broken that it displayed the wrong numbers entirely, I would wait until it was fixed to do the math.

F.A.Q​

Does anything else play a role in decreasing your DPS output with Kissassist? Yes, the more characters you box the slower Kissassist seems to respond over all if the character is in your background. I have this issue with Isboxer too, in fact it's much worse with Isboxer. When I box 36 characters I do about 20% less DPS than when I box 24. I think it's the round robin setting on Isboxer causing them to respond and attack slower but I am not sure. Chances are that any program/plugin you use that drastically reduces background FPS (like reducing background FPS in Options) will also reduce your overall DPS when you box above a certain threshold of characters. That's at least my personal anecdotal experience with boxing.

Does this apply to MQ2melee too and other bots that use MQ2? NO! Only Kissassist is designed to attack slower than a regular player would and pop burns slower than a regular player would. If you are using something like EoKnuke, MQ2melee or any of the other macros with MQ2 you very likely will not have this problem as they are not designed that way. In my personal experience ONLY Kissassist does dramatically less DPS, I have seen melee that use MQ2melee and they typically out DPS an actual player since MQ2melee attacks faster and better than a real player could.

How can I increase my DPS output on all of my characters while using Kissassist? All you need to do is setup a button on your characters that allows you to manually pop burns and manually attack. You should do one to manually heal too but that's a different subject. Here is more information on how to setup an attack button with Isboxer. If you are using EQBC you'll need to do something slightly different which I will get to. Here is how to setup a burn button with Isboxer. Aside from setting up a button to manually attack there is no way to increase your DPS output with Kissassist. Maskoi and Ctaylor22 both know about the problem I discuss on this page, I have personally worked with Ctaylor22 with trying to get Kissassist to do more DPS by decreasing the size of the macro, as William12 has recommended in the past. Our efforts were futile and Kissassist did not do more DPS after decreasing the size of the macro by half. Here is Noobhaxxor's video on how to setup EQBC so that you can play your characters too instead of automating them. If you aren't using Isboxer this is how you will want to setup your characters to increase your DPS.

How can we trust your math? You used Wizards to prove that Kissassist does less DPS, the most RNG class ever. What if you got lucky Arcane Fusions and only showed us the parses you wanted to see? I concur that my choice of class isn't the best and that you should not blindly trust my math. That's why I explained how you can run your own experiments and how you can reach the same conclusions that I have. I'm very anti-blindly trust anything; you should run your own numbers, you should go out there and test this and see with your own eyes the difference. It's very easy to test it, it's also very obvious once your see the difference in DPS what I am talking about. I know at least a dozen other boxers who've tested this out themselves and already know this is a problem with Kissassist. It's just the vocal minority in this community who fail to comprehend and continuously spread misinformation/argue with me about this topic.

You admit that Gamparse isn't working properly and it's patch day/expansion release day... Are you seriously such a twat that you couldn't wait a week to make this post when you COULD produce more accurate math? Yup, I am. As stated at the start, I am currently extremely busy in life and I barely even have enough time to make this post and do the experiments I did. Today is the first day in awhile that I have had free time and it just so happens to be a day when shit is broken. This isn't the first time I have ever tested Kissassist's poor DPS, I have known about this issue for years - I just never had time to slap together an informative post about it until today. I'm 100% sure you will get the same exact results I did by running your own experiments. I know because I'm not the only person to reach this same conclusion, literally every raider friend I have who uses MQ2 has also reached this same conclusion.

I still don't believe you. Maybe it's just your computer that's causing the problem or maybe you just suck at making KA INI files. After all, we've seen your KA INI Files on your website, you make your level 1 chars use spells like True North cause you're fucking terrible and everyone here hates you. Why don't you go fuck yourself? Ahh, the response I get most when I bring up this topic. Well, prove me wrong then. With actual facts and not your emotional rambling. I'm tired of repeatedly having this debate with people from Redguides - I'm borderline ready to pull my hair out because everyone who argues with me about this topic uses emotional reasons or anecdotal evidence. I don't care if your Kissassist does more DPS than the people you group with - that doesn't prove Kissassist does more DPS than someone manually pressing attack. That just proves the person you were grouped with sucks worse than Kissassist does.

If you know this is a contentious topic, why are you even bothering to make this post? Are you deliberately trying to piss people off? No. I'm sincerely tired of having people come to me asking for advice who ONLY use Kissassist to play and wondering why I tell them to stop using Kissassist to beat a certain mission or setup actual keybinds so they can control their characters along with KA; because depending on Kissassist will cause you to do less DPS. As someone who told me to go fuck myself while we were talking about this topic before said, "When your readers see that and think it's the ONLY way to play and then we have to teach to unlearn that, it gets old fast." I couldn't agree more. That's the purpose of this post. I am so sick and tired of having the same discussions with people because they join Redguides' Discord for a few weeks and are told a bunch of misinformation then they can't figure out why they're doing so poorly compared to me or other boxers they see. A select few of you in this community hate me so much for bringing up this topic that you slander the ever living shit out of me in Discord (Kaen) and people sadly actually listen to you. Then they eventually come back to me and say, "I can't figure out why I can't beat this RoS mission as a 6 box and everyone else can!" It's because you've been getting information from people who have literally no idea what they are talking about. They get offended when I say Kissassist does crappy DPS because it means they do crappy DPS. Instead of listening and understanding the problem like everyone else does they just ignore everything I have to say and then continuously rant about me in Discord.

(People who have recently argued with me about this topic/who motivated me to make this thread) @kaen01 ; @Sum1 ; @gSe7eN ; @ChatWithThisName
Note: I don't hate any of the people I @ listed, don't assume that I am pissed off at them or upset and looking to argue.

REMEMBER!!!!! KISS = Keep It Simple Stupid. The program was never designed to do good DPS, it was designed to play your characters for you well, which it does. The bot is better than most real players but that doesn't change the fact that the bot is NOT designed to min/max a character's DPS. As I said before... repeatedly.... Everquest is a game built around the concept of min/maxing a character. For someone like me who boxes raids getting the most out of every single character is extremely important. Chances are, if you are struggling with a mission or a named or something like that you can easily make your box team twice as good by setting up a few simple keybinds to manually attack, heal and pop burns. That's all this thread is intended for, to demonstrate to people that there is an extremely simple way to increase the output of each of your characters for those situations where you need it most.

 

[gSe7eN]

This is riddled with fallacies, the reader has to make awful assumptions, and is missing so much data that a conclusion is impossible to get. And before you say I'm attacking you personally, no, because I know you're already thinking that. I mostly agree with your conclusion. Not to the extent as your "test", which again is riddled with fallacy. This was a great idea, actually. But you've hidden so much data and skewed the results to your personal bias so much, the entire effort is junk. Post your INI. Post logs. This is very junk science, Almar. Make corrections, present a better experiment.

OR attack me by saying I'm attacking you personally or that I have some agenda, or whatever the excuse is. Hardly.

But I can say that one of those awful assumptions the reader has to make in order to accept any of this "data" is that you know how to tune a good Kiss ini file (they need min/maxed as well....or am I missing the entire point of the article and am just arguing for the sake of arguing?) and you OBJECTIVELY demonstrate that by saying Kiss didn't even fire during a named. Why is that? That part isn't broken in the macro, that's user error/ignorance. And if there's a fundamental concept screwed up in such a demonstrably objective way, it really throws into question EVERYTHING else. That's not even counting the other variables such as did both windows have focus? Were they both run on a clean install of Windows using the same exact hardware and same exact settings? Did you have a control? Did you try to make as many variables static as possible? The answer is clearly "No idea" and probably "no."

Come on, man. Do better. There's no need to skew the data, the conclusion will probably be the same.

Also, for the record, I am not some Kiss fan boy or have some unquestioned loyalty to it. I very rarely use Kiss any more. For me it's like the iPhone vs android debate. I love my iPhone. Kiss is Android. Millions love it and I'm happy for them, but I don't love it.

 

[EQ_Analytics]

Post more data, you come to conclusions way too quickly.

 

[LurkMcGurk]

I figured most people would get annoyed at my test results, the only reason I didn't post more data is because Gamparse isn't fully accurate right now (not counting 0s when a number goes over a million). I can assure you, once it's working properly I will drown you all with math after math proving my point =)

The reason I didn't post my INI and everything else is because there's really no need to. That information isn't at all related to the argument I am making. The experiment is very simple, all you need is to use the same spells on two characters and have one use KA while you manually attack on the second. You don't even really need a control for this test but i can understand why people ask for a control - in the future I will use one.

Here is the INI, I changed it JUST for this experiment (which is why it's irrelevant for me to even post the INI, you can use any combination of spells you want).

[DPS]
DPSOn=1
DPSSkip=0
DPSInterval=1
DPS1=Ethereal Skyfire|100
DPS2=NULL
DPS3=NULL
DPS4=Ethereal Icefloe|100
DPS5=Force of Ice|100|Weave
DPS6=Force of Flame|100|Weave
DPS7=Claw of Qunard|100
DPS8=Force of Will|100|Weave
DPS9=NULL
DPS10=NULL
DPS11=NULL
DPS12=null

I didn't skew the data though, I have no reason to. I don't see any reason to post the INI file because people are going to bitch that I have a shitty rotation, which i do, because i was experimenting not playing.

I want people to test this themselves, see it with their own eyes. Because otherwise everyone's argument is going to be the same,

1) You suck
2) not enough data

 

[gSe7eN]

I'm not annoyed with the results. I'm annoyed with the execution. I think it's fascinating as hell and I encourage you to take the community's advice on this one.

But you are wrong, the INI is a VITAL piece of information. The rotation is the same, good. But that ini is everything. And it's also about to be torn to shreds because it's written completely wrong. HOPEFULLY someone will be NICE (you know, productive, rather than wanting to destroy....I'd do it but I've been away from Kiss too long to tune that and Kiss DPS syntax annoys the shit out of me anyway) will make simple corrections to it where you can run it again, and you can supply more data, more variables, and more constants so the conclusions speak for themselves, it will be VERY cool to see what's up.

Also, I, like Almar, encourage people to also try this test, and make sure you don't put bias in manual control, try to take as many variables out as you can.....bias is a two way street. It doesn't make the experiment any more valid if you intentionally make the outcome the one you desire to put Kiss out in front. It's not an Almar mutually exclusive concept so anyone who does repeat, we demand the same exact standards or else we'll dismiss your findings too.... assuming you put forth the effort.

 

[timmy]

2) not enough data

12 hours or gtfo! :-)

i know kiss provides worse dps than i could playing a character manually and i'm ok with that. it also removes the tediousness of fishing for procs, and the numbing execution of targeting and assisting for hours at a time. it lets me play 4 dps at once. it's also always available, won't drone on over discord about politics, and won't roll against me for augs. its benefits far outweigh any performance limitations. i'm honestly amused that anyone would expect kiss to perform as well as a real player. if i had 5 friends who were willing to camp augs at 5 in the morning, i'd likely have never looked into macroquest in the first place.

i'm interested in more numbers, almar, because math and shit. i'm also following this closely because of entertainment! go get em!

 

[LurkMcGurk]

I'm not annoyed with the results. I'm annoyed with the execution. I think it's fascinating as hell and I encourage you to take the community's advice on this one.

But you are wrong, the INI is a VITAL piece of information. The rotation is the same, good. But that ini is everything. And it's also about to be torn to shreds because it's written completely wrong. HOPEFULLY someone will be NICE (you know, productive, rather than wanting to destroy....I'd do it but I've been away from Kiss too long to tune that and Kiss DPS syntax annoys the shit out of me anyway) will make simple corrections to it where you can run it again, and you can supply more data, more variables, and more constants so the conclusions speak for themselves, it will be VERY cool to see what's up.

Also, I, like Almar, encourage people to also try this test, and make sure you don't put bias in manual control, try to take as many variables out as you can.....bias is a two way street. It doesn't make the experiment any more valid if you intentionally make the outcome the one you desire to put Kiss out in front. It's not an Almar mutually exclusive concept so anyone who does repeat, we demand the same exact standards or else we'll dismiss your findings too.... assuming you put forth the effort.

Like I said in my own post, Gamparse is currently not working as well as it should so I am annoyed with the execution as well. I don't agree that the INI is a valuable piece of information though because it's irrelevant to my argument. You can use ANY spells you want to replicate my findings. So long of course as the character you are playing is using the same spells as the character you are botting. If they use the spells out of order, that's just part of using Kissassist.

I recommended other people run the experiment to rule out these variables. If my ignorance of Kissassist is to blame for my poor performance -- no matter how many times I personally run this experiment it's always going to have the same results, aye?

Everyone I know in EQ that raids all has reached the same conclusions as I did in this post. That Kissassist does less DPS in a similar scenario when compared to someone who is manually pressing the button to attack.

 

[gSe7eN]

That's a fair point about ignorance breeding the same result. HOWEVER, I counter with this: growth is reasonably expected. You can learn Kiss better from those who are well versed in it, and others can learn your method of manual control better. Everyone wins, no one is bitter. With all this data if others choose to emulate your test will only improve everything as a whole.

But without that correct INI, you've got a completely lobotomized script which isn't good for anything. Others are working on eliminating INI's entirely but for now, they're the evil price we pay.

 

[LurkMcGurk]

i know kiss provides worse dps than i could playing a character manually and i'm ok with that.

I am ok with that too, but it seems like many people don't understand that Kiss does worse DPS than an actual player - that's what frustrates me so much. I hate having to constantly argue with people about this topic. Many people come to me for advice in EQ and if they're MQ2 users I have to tell them how to setup their characters so that they can control them manually so that they can get the most out of them.

 

[LurkMcGurk]

That's a fair point about ignorance breeding the same result. HOWEVER, I counter with this: growth is reasonably expected. You can learn Kiss better from those who are well versed in it, and others can learn your method of manual control better. Everyone wins, no one is bitter. With all this data if others choose to emulate your test will only improve everything as a whole.

I agree with you, I would be glad beyond belief if I could improve my Kissassist DPS. I have gone so far, gotten so desperate to improve Kissassist's DPS that i worked with Ctaylor22 to cut out the "fat" of the macro, shrinking it down to less than half its size. Same results, same DPS. That was with William12's recommendation that KA did lower DPS because of the size of the macro.

When I talk to people about this subject (Kissassist's DPS being lower than an actual player) I get 1 of 2 responses:

1) Nuh uh - Kissassist's DPS doesn't suck - YOU suck!!!
2) Yea, that's how it was designed, KISS = Keep It Simple Stupid

 

[Elrem]

So what does your improved INI do dps wise? Is it at least decent? I think a real person should do more if actively playing, but the whole point is so you can get more out of multiple characters that you can't focus your attention on 24/7. What about RAZnuke or whatever that other wizard dps macro was? I am curious how much dps that would do in a similar scenario. Just guessing here, I am a total newb at this, but trying to help instead of fan the flames.

 

[mroutlaw]

I guess what I am trying to see from this is do you know of any macro that actually does preform all the functions kiss does and still outperforms a player playing the same toon ? I don't and that's why I am asking.

 

[Saetun]

You are absolutely right. KISS definitely suffers from the jack of all trades, master of none syndrome. But it really is one of the best things going as well. I couldn't live without it on a day to day basis. Some things though :

1) there are macros specialized for a particular class out there that do fantastic dps, while at the loss of much of the functionality of KISS.
2) MQ2EQWire is like nitrous for KISS... Even though I don't need it for my 6 box on my main comp (2700x, 32GB 3200 RAM, 1070 ti, NVMe drives), it still makes my background boxes react nearly twice as fast (I always play my foreground character manually).
3) Even at it's worst KISS still does better DPS than your average pug player who has Netflix running on a second monitor while they play...

 

[LurkMcGurk]

So what does your improved INI do dps wise? Is it at least decent? I think a real person should do more if actively playing, but the whole point is so you can get more out of multiple characters that you can't focus your attention on 24/7. What about RAZnuke or whatever that other wizard dps macro was? I am curious how much dps that would do in a similar scenario. Just guessing here, I am a total newb at this, but trying to help instead of fan the flames.

I agree with you that someone actually playing should do more, that's basically my argument. If you're not AFK botting, you should be hitting Attack, Heals and Burns when needed. you'll do better DPS/perform better as a box team.

I guess what I am trying to see from this is do you know of any macro that actually does preform all the functions kiss does and still outperforms a player playing the same toon ? I don't and that's why I am asking.

I don't know of any macro and my post isn't a rant against Kissassist, I think Kissassist is fucking fan tastic and I love it to death - I wouldn't ever want to give it up. That said, I am not willing to overlook its own flaws and it irks me, probably more than it should, when people try to suggest that Kissassist does as good damage as a regular player. It just doesn't. I'm not even saying that's a bad thing, I just want people to understand that that's how it is. I am tired of having to correct people who ask me for help and think that they're as good as they could ever be by relying entirely 100% on Kissassist to play the game for them.

 

[mroutlaw]

my thing Lurk is every time i have ever had a human in group with me they have sucked , they dont push buttons they dont do work because they are finstagrammin facebookin afking watching midgett pornj feading the dogs , kids or dealing with their spouse .... lol

 

[ctaylor22]

If I was going to try and improve the DPS in KISS. I would first figure out how much time is being spent, outside of the DPS routines, and in what routines the time is being spent.

Those numbers would tell me where to start looking in the macro to speed things up.

KISS is slow to engage mobs, but that is not because of the DPS routine, it is because of all the NON-DPS stuff kiss has to do.

 

[Casca]

I am not sure what the point of this thread is. Me manually playing my warrior, bard, enchanter, 3x clerics, druid, shaman, 2x beastlord, mage, 3x wizard, 2x berserker, 2x rogue, ranger, paladin and shadowknight would be no where near as effective as me just playing my warrior and KISS running the rest.

Also, my wizards running KISS out DPS's caster mercs and that is more than good enough for me. My ini is not even all that great and all my wizards are FTP.

 

[deathlock]

If I was going to try and improve the DPS in KISS. I would first figure out how much time is being spent, outside of the DPS routines, and in what routines the time is being spent.

Those numbers would tell me where to start looking in the macro to speed things up.

KISS is slow to engage mobs, but that is not because of the DPS routine, it is because of all the NON-DPS stuff kiss has to do.

Cut the fat is all you can really do. But keeping in mind you will loose functionality of the other sections. @LurkMcGurk : Have you tried using @eqmule 's OhShit section and fill it with DPS spells? I would be interested to see how that responds over the typically DPS section.

 

[eqtrader74]

Tests on KA10 or KA11? KA11 is definitely faster response times for DPS/Heals.

KingArthur had a good post awhile back discussing that he wound up writing custom macros for all his raiding toons to squeeze the most DPS out of them. Ultimately, custom will always beat KA in response time. And it should. If you go custom or "button mash" you are skipping a lot of the idiot proofing that has gone into KA. A custom macro doesn't need to check if a spell is memmed for each cast. KA needs to check that and hundreds of other things while it's running. That makes it a little slower but also makes it resilient.

If we or the KA code team can find more ways to thin the code for speed, great. But KA's job was never to be the absolute best at DPS. KA's strength is "learn it once, use it on any EQ toon" plus a massive amount of support + knowledge base from the Redguides community. If you have a KA problem, somebody will figure it in 10 minutes to a couple hours. If you have a problem on a custom macro, sometimes the coder themselves is on break and you wait months for somebody else to come up with a work around.

I see you had some notes on your testing methods. I'd like to see like 1 hr averages, same camp, no named. Something like the northwest corner of Yeti caves in Lcea where you can't pop a Matriarch, all mobs are same level and pull time stays consistent.

Awhile back I ran some similar tests.

110 Wizard macros vs some 10.2.6 KA ini's

I ran some quick tests on a FTP 110 wizard with rank I spells, max AA & EoK/RoS t1 gear. Not my main team, so numbers are definitely on the low side overall but give an idea percent-wise how they stack up. Group setup was War-Dru-Ench-Mag-Wiz-Wiz, no alliance. More of a chain-pull DPS test - EoK dark blues, no named. So anything in Burn, ${Target.Named} or such wasn't triggered unless it was used via AE or XTarget trigger. Tests were 45 mins to an hour+, same camp.

110 Wizard DPS - Running KA 10.2.6 for kissassist ini's

• 70K - Raznuke110 macro (Aug '18 edition) - Wizard was at 50% mana or less, most of the time. Saw a couple med breaks.
• 65K - Sinn69's ini (above) - Wizard was at 60% mana or less, most of the time. Saw one med break.
• 62K - eqmule's ini (w some AE triggers added) - Wizard was at 80% mana or more, most of the time. No med breaks.
• 55K - IHC's wizard macro - I'm gonna mark this up to user error, I had issues with the custom HUD this macro starts. IHC's zerker is outstanding, so I'll have to try this again later.
• Islandgirl's came in low, so again might be something in my setup that wasn't working for that ini.

I use eqmule's for my wizards like 95% of the time. They use Gambit over Harvest and are setup to run 24/7 without needing med breaks and killing stacks of 3+ mobs on each pull. Raz or Sinn's are great for burst missions like OT Cacti mission.

So custom macro Raznuke did win this particular test on pure DPS output, but also causes the most med breaks.

Here is Zerkers comparing 10.2.6 to KA11 (older beta 11.008) DPS:

Ran some gamparse tests on my zerkers today. Nice DPS improvements in KA11.008b. Group setup was SK-Clr-Brd-Bst-Zerk-Zerk killing trash in Overthere. Not running zerker alliance, both zerkers running the same ini w/ conditions (via updater program for KA11) and are using rank I spells/discs.

Zerker #1 - free-to-play - 110 25000AA - (EoK/RoS T1 + Conflag weapon) 3850 atk

• 78,000 DPS avg running KA10.2.6
• 103,000 DPS avg running KA11.008b

Zerker #2 - Gold - 110 31000 AA - (RoS T1/T2 + chase weapon) 4300 atk

• 104,000 DPS avg running KA10.2.6
• 134,000 DPS avg running KA11.008b

I didn't inlcude it in these tests, but IHC's custom zerker macro will still outpeform my KA11 by maybe +10-20K DPS. It has three pre-coded Burn routines and can start them up in cascading fashion on any large pulls/Named. So it has a DPS structure advantage in addition to skipping a lot of the "idiot" checks that KA needs to run.

 

[Frimp]

@LurkMcGurk Try this with your wizard and see if that helps:

[DPS]
DPSOn=2
DPSSkip=0
DPSInterval=1
DPS1=Ethereal Skyfire|99
DPS2=NULL
DPS3=NULL
DPS4=Ethereal Icefloe|98
DPS5=Force of Ice|97|Weave
DPS6=Force of Flame|96|Weave
DPS7=Claw of Qunard|100
DPS8=Force of Will|94|Weave
DPS9=NULL
DPS10=NULL
DPS11=NULL
DPS12=null

The reason I suggest this is 2 things. 1, When you weave a spell, KA will look at the % set, and weave the extra casts in that order. By putting them all to 100, it can't choose so it picks one of those 100's and casts it. onto the next task and it ignores the other 100s. 2nd, you also miss the opportunity of casting other spells that may be passed up because the % is already too low for it to acknowledge that casting event. By using DPSon=2, KA will sequentially cast all the spells in order, not missing any beats.

So give this a try and post your results.

 

[LurkMcGurk]

my thing Lurk is every time i have ever had a human in group with me they have sucked , they dont push buttons they dont do work because they are finstagrammin facebookin afking watching midgett pornj feading the dogs , kids or dealing with their spouse .... lol

Indeed, most of the people who argue against my findings always compare it to actual humans they invite to their group. I'm not trying to say that Kissassist is bad (which many probably mistake me for saying) I am just trying to say that Kissassist doesn't min/max your character's potential which means you won't be able to get the most out of them in terms of DPS.

In the group game this is barely noticeable and doesn't really matter but when we're talking about raids... It's a BIG BIG deal. Everquest is built around the min/maxing concept.

Tests on KA10 or KA11? KA11 is definitely faster response times for DPS/Heals.

KingArthur had a good post awhile back discussing that he wound up writing custom macros for all his raiding toons to squeeze the most DPS out of them. Ultimately, custom will always beat KA in response time. And it should. If you go custom or "button mash" you are skipping a lot of the idiot proofing that has gone into KA. A custom macro doesn't need to check if a spell is memmed for each cast. KA needs to check that and hundreds of other things while it's running. That makes it a little slower but also makes it resilient.

My tests were done with Maskoi's KA and whichever one comes with the newest install/update of Redguides compile.

Also I agree a million times yes with what you said. I love Kissassist and understand it's not designed to be the best DPS in comparison to regular macros. That's why I made this post, because I am really tired/annoyed by certain members of our community saying that it is great DPS and that the only reason MY kissassist does less DPS than a player is because I suck. It's just not true, even if I do suck - Kissassist still does less DPS than a real player/other custom macros.

If I was going to try and improve the DPS in KISS. I would first figure out how much time is being spent, outside of the DPS routines, and in what routines the time is being spent.

Those numbers would tell me where to start looking in the macro to speed things up.

KISS is slow to engage mobs, but that is not because of the DPS routine, it is because of all the NON-DPS stuff kiss has to do.

The slowness engaging mobs is a big factor but moreso the things that I see causing KA to do way less DPS is nameds and cooldowns. When I fight named enemies Kissassist takes 10 - 15 seconds usually to pop burns then it finally starts attacking. For Melee characters that 10 - 15 seconds wastes a TON of time in burn DPS. For my spell casters, Kissassist rarely casts enough during Improved Twincast to consume all of its charges, thus I end up losing that DPS too.

Fighting regular enemies without any cooldowns I don't have many gripes about the DPS difference. The only time the DPS difference with Kiss actually bothers me and is extremely noticeable is when I pull a named or do a group mission or something like that. The slow speed at which it pops cooldowns coupled with the slower attack speed really shaves off your DPS.

 

[ctaylor22]

DSPOn==1, cast the first spell that becomes available.
DSPOn==2, If global cool down is in effect, wait for global cool down to expire. Now Cast the Next spell that becomes available.

The slowness engaging mobs is a big factor but moreso the things that I see causing KA to do way less DPS is nameds and cooldowns. When I fight named enemies Kissassist takes 10 - 15 seconds usually to pop burns then it finally starts attacking. For Melee characters that 10 - 15 seconds wastes a TON of time in burn DPS. For my spell casters, Kissassist rarely casts enough during Improved Twincast to consume all of its charges, thus I end up losing that DPS too.

Fighting regular enemies without any cooldowns I don't have many gripes about the DPS difference. The only time the DPS difference with Kiss actually bothers me and is extremely noticeable is when I pull a named or do a group mission or something like that. The slow speed at which it pops cooldowns coupled with the slower attack speed really shaves off your DPS.

But if you could figure out where in kiss that 10 - 15 seconds is being spent. That would be very helpful for use developers.

I know where the slow to engage is coming from, when it comes to the normal NON-Burn DPS. It is as much of a timing issue, when the Main loop is returned too, and a mob enters your camp.

 

[LurkMcGurk]

DSPOn==1, cast the first spell that becomes available.
DSPOn==2, If global cool down is in effect, wait for global cool down to expire. Now Cast the Next spell that becomes available.



But if you could figure out where in kiss that 10 - 15 seconds is being spent. That would be very helpful for use developers.

I know where the slow to engage is coming from, when it comes to the normal NON-Burn DPS. It is as much of a timing issue, when the Main loop is returned too, and a mob enters your camp.

So, realistically, which one would be better DPS? DPSOn=1 or DPSOn=2?

Kissassist has too many features in this regard and I think it's unfair sometimes to blame people's INI files for doing lackluster DPS when there's no guide or explanation of what's better. An example, when I linked my INI file at least 4 people said LOLZ UR INI FILE SUCKS NO WONDER UR DPS SUCKS.... As if there's an explanation out there on how to get the most DPS out of your INI file. Including my own guides, they only say how to setup Kissassist not how to maximize its potential. Sometimes too much customization is a bad thing.

Also, if you tell me how to figure out where Kissassist is spending its time when it should be burning/dpsing I can give it a whirl. In my experience it usually waits 1 second between popping each burn and that's where the extra time comes from. Most classes have 10 burns or more so that's 10 seconds to pop them all then 1 or 2 seconds until it starts actually attacking the enemy after that. Sometimes it moves faster, sometimes it moves slower. Sometimes it even skips a burn in the list that it should pop (like my Wizards looove to skip Frenzied Devastation which completely fucks their DPS up while burning).

In some cases, like today when I was putting together my parses, Kissassist didn't pop any burns at all when pulling a named. Like this:

[2018/12/12 03:58:25] [MQ2] ATTACKING -> The Blazing Hen <-
[2018/12/12 03:58:25] [MQ2] *** Mob:(The Blazing Hen) is a NAMED!
[2018/12/12 03:58:25] [MQ2] -- Weaving: Force of Ice
[2018/12/12 03:58:29] [MQ2] -- Weaving: Force of Flame
[2018/12/12 03:58:33] [MQ2] ** Ethereal Skyfire on >> The Blazing Hen <<
[2018/12/12 03:58:33] [MQ2] -- Weaving: Force of Will
[2018/12/12 03:58:37] [MQ2] ** Ethereal Icefloe on >> The Blazing Hen <<
[2018/12/12 03:58:41] [MQ2] ** Claw of Qunard on >> The Blazing Hen <<
[2018/12/12 03:58:45] [MQ2] ** Ethereal Skyfire on >> The Blazing Hen <<
[2018/12/12 03:58:46] [MQ2] -- Weaving: Force of Ice
[2018/12/12 03:58:49] [MQ2] ** Ethereal Icefloe on >> The Blazing Hen <<
[2018/12/12 03:58:50] [MQ2] -- Weaving: Force of Flame
[2018/12/12 03:58:51] [MQ2] -- Weaving: Force of Will
[2018/12/12 03:58:55] [MQ2] ** Claw of Qunard on >> The Blazing Hen <<
[2018/12/12 03:58:59] [MQ2] ** Ethereal Icefloe on >> The Blazing Hen <<
[2018/12/12 03:59:03] [MQ2] ** Ethereal Skyfire on >> The Blazing Hen <<
[2018/12/12 03:59:04] [MQ2] -- Weaving: Force of Will
[2018/12/12 03:59:08] [MQ2] ** Ethereal Icefloe on >> The Blazing Hen <<
[2018/12/12 03:59:08] [MQ2] -- Weaving: Force of Ice
[2018/12/12 03:59:12] [MQ2] ** Claw of Qunard on >> The Blazing Hen <<
[2018/12/12 03:59:13] [MQ2] -- Weaving: Force of Flame
[2018/12/12 03:59:17] [MQ2] ** Ethereal Icefloe on >> The Blazing Hen <<
[2018/12/12 03:59:21] [MQ2] ** Claw of Qunard on >> The Blazing Hen <<
[2018/12/12 03:59:26] [MQ2] ** Ethereal Skyfire on >> The Blazing Hen <<
[2018/12/12 03:59:30] [MQ2] ** Claw of Qunard on >> The Blazing Hen <<
[2018/12/12 03:59:30] [MQ2] -- Weaving: Force of Ice
[2018/12/12 03:59:35] [MQ2] ** Ethereal Skyfire on >> The Blazing Hen <<
[2018/12/12 03:59:35] [MQ2] -- Weaving: Force of Flame
[2018/12/12 03:59:47] [MQ2] ** Ethereal Icefloe on >> The Blazing Hen's corpse <<
[2018/12/12 03:59:47] [MQ2] SLAIN >> The Blazing Hen <<
[2018/12/12 03:59:47] [MQ2] EXP: 0.00% AAEXP: 0.00%

[2018/12/12 03:18:02] [MQ2] ATTACKING -> Yeti Patriarch <-
[2018/12/12 03:18:02] [MQ2] *** Mob:(Yeti Patriarch) is a NAMED!
[2018/12/12 03:18:04] [MQ2] -- Weaving: Force of Ice
[2018/12/12 03:18:07] [MQ2] -- Weaving: Force of Flame
[2018/12/12 03:18:11] [MQ2] ** Ethereal Skyfire on >> Yeti Patriarch <<
[2018/12/12 03:18:12] [MQ2] -- Weaving: Force of Will
[2018/12/12 03:18:16] [MQ2] ** Ethereal Icefloe on >> Yeti Patriarch <<
[2018/12/12 03:18:20] [MQ2] ** Claw of Qunard on >> Yeti Patriarch <<
[2018/12/12 03:18:25] [MQ2] ** Ethereal Skyfire on >> Yeti Patriarch <<
[2018/12/12 03:18:25] [MQ2] -- Weaving: Force of Ice
[2018/12/12 03:18:29] [MQ2] ** Ethereal Icefloe on >> Yeti Patriarch <<
[2018/12/12 03:18:30] [MQ2] -- Weaving: Force of Flame
[2018/12/12 03:18:31] [MQ2] -- Weaving: Force of Will
[2018/12/12 03:18:35] [MQ2] ** Claw of Qunard on >> Yeti Patriarch <<
[2018/12/12 03:18:40] [MQ2] ** Ethereal Skyfire on >> Yeti Patriarch <<
[2018/12/12 03:18:44] [MQ2] ** Ethereal Icefloe on >> Yeti Patriarch <<
[2018/12/12 03:18:44] [MQ2] -- Weaving: Force of Will
[2018/12/12 03:18:48] [MQ2] -- Weaving: Force of Ice
[2018/12/12 03:18:51] [MQ2] ** Claw of Qunard on >> Yeti Patriarch <<
[2018/12/12 03:18:52] [MQ2] -- Weaving: Force of Flame
[2018/12/12 03:18:56] [MQ2] ** Ethereal Skyfire on >> Yeti Patriarch <<
[2018/12/12 03:18:57] [MQ2] -- Weaving: Force of Will
[2018/12/12 03:19:01] [MQ2] ** Ethereal Icefloe on >> Yeti Patriarch <<
[2018/12/12 03:19:05] [MQ2] ** Claw of Qunard on >> Yeti Patriarch <<
[2018/12/12 03:19:06] [MQ2] -- Weaving: Force of Ice
[2018/12/12 03:19:09] [MQ2] ** Ethereal Skyfire on >> Yeti Patriarch <<
[2018/12/12 03:19:10] [MQ2] -- Weaving: Force of Will
[2018/12/12 03:19:14] [MQ2] ** Ethereal Icefloe on >> Yeti Patriarch <<
[2018/12/12 03:19:15] [MQ2] -- Weaving: Force of Flame
[2018/12/12 03:19:19] [MQ2] ** Claw of Qunard on >> Yeti Patriarch <<
[2018/12/12 03:19:23] [MQ2] ** Ethereal Skyfire on >> Yeti Patriarch <<
[2018/12/12 03:19:23] [MQ2] -- Weaving: Force of Will
[2018/12/12 03:19:28] [MQ2] ** Ethereal Icefloe on >> Yeti Patriarch <<
[2018/12/12 03:19:28] [MQ2] -- Weaving: Force of Ice
[2018/12/12 03:19:32] [MQ2] ** Claw of Qunard on >> Yeti Patriarch's corpse <<
[2018/12/12 03:19:32] [MQ2] SLAIN >> Yeti Patriarch <<

In all honesty, I personally don't care if my Kissassist occasionally makes a mistake like this and doesn't pop burns or misses a burn. I expect it to, I expect all programs to occasionally have a hiccup. When I am using Kissassist I am VERY lazily playing, usually to the point of half assing it. However, in this scenario when we're discussing the DPS output of Kissassist and compairing it to other macros or a player it's a really really big deal when it doesn't pop burns or skips a burn. If so many people didn't disagree with me about the fact that Kissassist does lackluster DPS, I wouldn't even have made this post or griped about this at all. As many other people throughout the thread said - it's expected of a macro like Kissassist, a jack of all trades a master of none. It just sets off my annoyance alarm when people out there suggest I am lying or wrong about the DPS Kissassist does or the occasional hiccup it throws out there.

 

[ctaylor22]

First off the burn routine is only executed once per mob, unless you use the /burn command for every time you want burn to be executed.

Turning on BurnAllNamed(1/2) will call the Burn routine when A named mob is in camp. 1 = Use Xtarget to find named, 2 = Use MobToBurn list for names.

There is also a burn tag that can be used in the AE section to call the Burn routine.

Another thing I noticed, is the burn routine never checks for global cool downs, so if your spell gems are in global cool down none of the spells will get cast, but anything else will. The Burn routine was setup to cast all your supporting AA's and combat ability type actions and the DPS routine would handle the DPS.

You may be assuming the burn routine is continually being called once burn is called, and from what I see. That is not the case.

 

[LurkMcGurk]

I'm about to go to bed, but I imagine this thread will continue to create discussions while I am asleep...

Based on the responses so far I want to make it clear, before I am not around to defend myself, this thread isn't a gripe on my part about Kissassist. I love kissassist and everything about it. I did not make this thread to complain about Kissassist's flaws. I made this thread because of my discussion in another thread a few months ago, 54 box Theorycraft Discussion.

In that thread I said that in order to box raids well, you'll need to use a macro that does more DPS than Kissassist or actually play your characters instead. I brought this up because I was trying to persuade the person who created the thread out of using so many Beastlords on his team. My reasons for why NOT to box a Beastlord were the same reasons I brought up in this thread. They asked me to prove my point and I suggested that at some point in the future, I would sit down, make parses and post them. As you can see from this thread - that's what I did.

I'm sure some people will see this thread as a personal attack, or even an attack on Kissassist. It's a common occurance when I bring up this topic. It's not. I apologize if you see things that way, maybe it's partly my fault for not wording my argument properly, but you're misreading the situation.

So far the ONLY arguments people have used against me on this topic are ad hominem attacks, even today we saw it. Meaning character attacks, my INI sucks, my experiment sucks, I suck, the skills I chose to use suck, my approach was wrong, etc etc. These aren't actual arguments, they're just slander. All you're doing is venting to me and it's annoying. Every time I bring this subject up people just have emotional outbursts on me and fail to provide any sort of counter argument. I expect this in a political argument, not one in a video game where we're talking about boxing.

I made this thread because I am so fucking done with being personally attacked over this topic. Many of you have asked for more math and more information and I am going to provide it. Lots of it. Sometime in the next few weeks I have to setup my raid team anyway, since I bought a shit ton of RoS copies. I need to redo my INI files, figure out my raid rotations etc etc. I'm going to document allllllllllllllllllllllllllllllllllllll the math needed to prove my point and then some. I'll be updating this thread and the post on my website. I'm 100% listening to the feedback in this thread about how to improve my experiment and actually convince people of my argument.

I also want other people, who doubt me, to sit down and test this too. So that we can rule out all personal attacks on me. If we know that it's not my absolutely horrific, terrible in every way INI files or my atrocious skill at playing this game and setting up Kissassist then maybe we can move past this silliness.

 

[aspire2008]

Your results are skewed off for these reasons:

1) No human can continuously do the actions of a machine. Try playing your wizard for 30 minutes manually vs using a macro. See who focuses better on running actions of your character.

2) Looking at the DPS section you posted of your INI file, it doesn't even take the breath of scope of wizard spells and abilities. This is a user problem of using the macro to not use the macro to the full extent. Your weave is not the best way of approaching playing a wizard imo, and my opinion is based on how majority of wizards have agreed to weave their spells for maximum dps in 80% of content, i.e. trash mobs.

Lets for argument sake, we take your DPS weave vs a human doing it, then you have to refer back to #1.

3) You are posting 1 fight logs or 2 fight logs at best, I mean come on any human has an attention span to focus on on 1-2 trash mobs, try doing it for a longer interval of trash mobs. This is why I hate it when people post those logs because they are meaningless unless you are trying to figure out your burn dps (i.e. fights less than 3 minutes). I am glad that you posted Gamparse logs instead of KA DPS logs, since I find KA DPS logs even more meaningless. Although, Gamparse is completely broken atm due to all the new chat message changes to core EQ.

I invite you to use my wizard ini configuration found >> here and retest your results.

Now there are things that humans do better. For example, 2 named mobs have spawned the first one is significantly easy and you can handle it without burning discs/spells. I can't tell the macro to not use my DPS burn and save a few discs for the 2nd named vs using them on the first named. You can't fine tune those things.


If there is 10-15 secs of delay happening before your toon is starting to do stuff, then try putting more stuff in the DPS section vs scattering it in other sections. Most of the INI files I create, I always try to focus on how much I can make the "Combat" section the most robust, because that is meat of your DPS on getting the mob dead.

I'll give you an example, "Selo's Sonata" is a bard buff. I notice people put this in the BUFF section. Now if I put it in the BUFF section, my puller who chain pulls will never ever get Selo'd. And even if he didn't chain pull, it takes a few seconds for the BUFF subroutine to kick in, so he might miss it on his way out to pull again either way. I moved that completely to the DPS section. I.E. Try to make your DPS section robust.

 

[MaxD]

I know gamparse isn't working right now personally I like to compare 30 minutes or an hour of straight killing mobs. On single mobs a lucky string of crits can really skew your results.

 

[Maskoi]

Kiss only does shit dps when people use ini files posted from your website.

 

[eqmule]

This test needs to be conducted using proper scientific methods.
Gamparse is a log file reader, it will never ever give you 100% dps. Aside from the fact that having the log turned on also slows eq down significantly once the file gets big.

Mq2damageparser will give you 100% dps because it reads damage directly from the function for incoming and outgoing damage.

Now technically that doesn’t matter if u always use a flawed parser it will be flawed for all tests therefor it will work.

BUT you absolutely need to macro the keypress u push manually or this will never work.

sub main
/while (1) {
/keypress 1
/delay 6
}
/return

Run that on the char that’s not kiss controlled, it will push your number 1 key twice per second. Whatever binds you have on number 1 will fire.

Then you can tweak your ini until it outperforms the keypress macro.
And then post those results.

Also forget all the talk about trimming the fat, computers nowadays can run 10000 lines of kiss code per second without even sweating. If your wizards suck it’s because your ini is slowing kiss down. I can create a really stupid ini where everything is set to go off at 100% and no conditions whatsoever to control which nukes to fire and it too can be beat by a manual player.

 

[aspire2008]

I know gamparse isn't working right now personally I like to compare 30 minutes or an hour of straight killing mobs. On single mobs a lucky string of crits can really skew your results.

This, but I prefer even longer parses so I can say with 95% CI that my results are not due to RNG on crits and now to add to that even Luck stat crap. 1-2 hour parses will give you a rough 20-40% standard deviation from the DPS your macro settings is able to put out, longer parses will lower than standard deviation. I.E. my dps shows 230k dps after 1 hour of parsing, i can pretty much say my dps will fall between 200-260k dps most of the time.

If I parse for longer, I can tell where exactly the DPS falls in that range. Some people will be happy with 1-2 hour parses, I like to parse for longer because I can see those deviations due to RNG, not skew the data as much. Also to find out if I'm looking at backward parses to upgrades to AA's for example, especially passive AA's, if the developer has actually coded the AA, or just added the upgrade to the rank of the AA as UI fluff. To do that you would have to parse for over 24 hours.

 

[Daryldixonurmouth]

WTF stupid Argument is this? What idiot would think a Good InI could ever beat a Good player? You say "This drives you crazy" where are these people doing it? Sounds to me like YOU are the one making all the assumptions about what other people think. I don't think i've ever met a single person here that feels the way you assume we feel.

 

[Synk]

Our raids on Ragefire (almost all 12-18 boxers) are the highest dps parses I have ever seen for GoD, OoW, and DoN fights, and the entire guild uses kiss10 or kiss11. We have finely tuned .ini files, some use holys\downs, some manually hit their burns. Personally I use a blend of manually calling the burn section of KA and holies and I do fine. I think this thread can simply be summed up by saying, you should use mq2melee & EoKnuke if you want to min\max dps with Kiss11, that's pretty much it.

I will agree sometimes Kiss CAN be slow to engage (we all have manual engage keys in case it is slow), and the burn section does not fire well on it's own but having it fire on anything MQ2 considers a named is a huge mistake anyways. Whenever we get around to doing DoDh I'll try to remember to come back here. Our full burn raid DPS in GoD was around 40k, in OoW around 50k, everything was completely trivial and everyone uses kiss now (a few switched over from other bots).

 

[MaxD]

This, but I prefer even longer parses so I can say with 95% CI that my results are not due to RNG on crits and now to add to that even Luck stat crap. 1-2 hour parses will give you a rough 20-40% standard deviation from the DPS your macro settings is able to put out, longer parses will lower than standard deviation. I.E. my dps shows 230k dps after 1 hour of parsing, i can pretty much say my dps will fall between 200-260k dps most of the time.

If I parse for longer, I can tell where exactly the DPS falls in that range. Some people will be happy with 1-2 hour parses, I like to parse for longer because I can see those deviations due to RNG, not skew the data as much. Also to find out if I'm looking at backward parses to upgrades to AA's for example, especially passive AA's, if the developer has actually coded the AA, or just added the upgrade to the rank of the AA as UI fluff. To do that you would have to parse for over 24 hours.

Longer parses are better and kiss should only catch upto a real player on a longer parse because over a longer stretch of time the real player will perform more poorly. Compared to the OP's posts some of those fights are within 5%. 5% could be a bathroom break on a 2 hour parse.

 

[LurkMcGurk]

WTF stupid Argument is this? What idiot would think a Good InI could ever beat a Good player? You say "This drives you crazy" where are these people doing it? Sounds to me like YOU are the one making all the assumptions about what other people think. I don't think i've ever met a single person here that feels the way you assume we feel.

Look at the first few posts.... Look at Maskoi's response, look at Discord Chat about this subject, look at the 54 box theory crafting page I linked. A lot of people misunderstand and think Kissassist does better DPS than an actual player. I don't know why but it's completely inaccurate to say no one feels this way - otherwise this post wouldn't have attracted so many haters. People would just say, "Yea whatever we already knew that"

Our raids on Ragefire (almost all 12-18 boxers) are the highest dps parses I have ever seen for GoD, OoW, and DoN fights, and the entire guild uses kiss10 or kiss11. We have finely tuned .ini files, some use holys\downs, some manually hit their burns. Personally I use a blend of manually calling the burn section of KA and holies and I do fine. I think this thread can simply be summed up by saying, you should use mq2melee & EoKnuke if you want to min\max dps with Kiss11, that's pretty much it.

Manually popping burns makes a HUGE difference. Also, TLP is a lot more basic and probably works a lot better with Kissassist than live, Live is very very spammy and that's the main reason Kiss does lower DPS on live.

Your results ar

I'm not saying Kissassist can never out perform a player in some scenarios. People keep recommending I recreate unrealistic scenarios (like test on target dummies for 2 hours straight) which no fight in Everquest ever lasts for 2 hours. Everyone keeps suggesting I put myself into a scenario where it's almost guaranteed to have Kissassist out perform a player - completely missing the point that my argument is based on realistic scenarios. Kissassist's slower engage time, Kissassist's slowness in popping burns, Kissassist's inability to react as someone actually in control of his 6 box team could react.

Kissassist may be able to out DPS a real player in a lab (especially when that real player is being substituted by an even worse bot -_-), I have my doubts still, but not in a real setting.

Kiss only does shit dps when people use ini files posted from your website.

I'll ignore this insult just because i expect this subject to make you emotional. However, please don't push this button with me and lie - I have a lot of respect for you and don't want to go down this road with you. If you think that because I apologized to you and feel bad about bringing this topic up it means you can walk all over me, you're going to be really surprised by how much bite my bark has. Sure, you can do personal attacks and can say my INI files suck - let's all bash the guy that took days out of his life to help the community for free. Whatever, that's realllllly mature of you.

But as you've seen from multiple posts, I am not the only person who has discovered this. Literally the only counter argument to my argument so far is emotional ranting, your argument included. ((Edit: Also some people have argued that I should setup a scenario that has no basis in reality that greatly favors Kissassist doing better DPS, that's an argument too albeit a stupid one)).

 

[LurkMcGurk]

This test needs to be conducted using proper scientific methods.

Also forget all the talk about trimming the fat, computers nowadays can run 10000 lines of kiss code per second without even sweating. If your wizards suck it’s because your ini is slowing kiss down. I can create a really stupid ini where everything is set to go off at 100% and no conditions whatsoever to control which nukes to fire and it too can be beat by a manual player.

You have a point, maybe I boiled down my argument too simplistically (manually pressing attack will always out DPS kissassist) in the hopes that everyone would understand and there would be less drama. Many people didn't even read my initial post, they just came here and started bitching immediately anyway it seems.

Your test however won't work because it's conducted in a lab only setting and not a real scenario. Maybe you have a point, that Kissassist can out DPS another player if you sit there in front of a target dummy for 2 hours and let it. I don't think it will, but this isn't what I am trying to set out to prove.

My two main arguments against Kissassist's low DPS are:

1) It attacks at a lower speed in comparison to what you can attack at manually. (sometimes, randomly, it is slow to engage some mobs; how are you going to reproduce that condition in a lab on a single target for hours?)

2) It uses Burns/Cooldowns extremely slow. (Again, how will you reproduce this in a lab setting?)

Your lab setting/test only reproduces one of these conditions - and only part of it. Your test also isn't a stress test, it's not pushing Kissassist to its limit and what it is capable of, which is where my argument is. If you want me to agree with you that Kissassist can out DPS someone who is doing absolutely nothing but pushing a single button for hours on end, I can give you the benefit of the doubt and say sure, it can. But it leaves the rest of my argument completely unaddressed.

 

[surfdog21]

One love I have for KA is that if I am meleeing and the mob is being dragged around, in a crowd, and staying on the target can be a visual problem, KA keeps me on the mob and I parse higher because there are not the lapses in dps from when "you are too far away" etc., especially if the mob has knockback or the like. KA sucks me right back to the mob, every time!

 

[joojoobee]

People talk about l these amazing inis they have and don't post them. Wtf?

Please post them for others to use, contribute beyond rhetoric.

/Said with love
Joojoobee

 

[Frimp]

Joojoo, there is a forum section specifically for INI's. You will find them in there. We will not post them in random discussion threads, because it is not the correct place for them. They would be lost in obscurity and may only be found by some who actually use the search feature.

Apart from that, we may post relevant snippets for the sake of troubleshooting purposes.

 

[sikkun]

I say you guys go all out and just have an entire Turing competition out of it. Pick a fight, make some heroic characters and go all out AI vs Human and see who wins.

 

[LurkMcGurk]

Joojoo, there is a forum section specifically for INI's. You will find them in there. We will not post them in random discussion threads, because it is not the correct place for them. They would be lost in obscurity and may only be found by some who actually use the search feature.

Apart from that, we may post relevant snippets for the sake of troubleshooting purposes.

I think what he was trying to say is - so far everyone's said my INIs were the problem and that I am the problem, yet no one has offered up an alternative with proof that it fixes the problem. People have made minor corrections to my INI and suggested I go back out and restest things, because they clearly don't want to spend the time doing it themselves, and likely so they can just say "you fucked up again!" when I come back with different math.

While i do appreciate people trying to "help" me fix this issue, I sincerely do not believe that it's an issue that can be fixed by a simple INI change. If it were, I don't think this thread would have gotten half the people agreeing with me and acting like it's common sense that Kiss can't do as good DPS as a player (Post 1) (Post 2) (Post 3) and the rest of the people disagreeing with me and suggesting I am terrible at everything I do while providing absolutely no evidence to support their claims.