Warder Proc Rate?

[Tipop]

Sorry, no evidence, just my observations. I've seen procs go off "between" swings. (Yes, I have my hits and misses turned on.)

However, that is why I preface things with "I think" or "it seems", so people won't take what I say as absolute truth. Just mentioning my observations. I will rarely say "This is how it is, I guarantee it".

[Chubaka]

On a side note, it occurs to me that giving the pet proccing weapons might actually lower the proc rate from the spell procs. This happens with PCs as well. PCs can only get one proc per hand per combat round. The primary proc of the weapon gets checked first, then the secondary (from augments), and finally the spell proc. The spell proc can only go off if the primary and secondary proc fail their check, and therefore the spell proc rate is lower then it would be if the PC had a weapon with no procs on it.

I have to dissagree with this.  At least in regards to weapon proc and aug proc (I have no spell procs).  I have had my Pitchakta 150dd and my 150dd aug on that weapon fire simultaneously.  They pop onto my combat spam chat window at the same time.  If I remember correctly, they have had the same time stamp in my log.

Oh wait.  Just had a revelation (maybe). BF - double attack.  I suppose one proc could be on the main hit and the second proc could be on the double attack?

[Coprolith]

I have to dissagree with this. At least in regards to weapon proc and aug proc (I have no spell procs). I have had my Pitchakta 150dd and my 150dd aug on that weapon fire simultaneously. They pop onto my combat spam chat window at the same time. If I remember correctly, they have had the same time stamp in my log.

From your magelo i see you are using a Savage Lord's Pitchatka with a delay of 17. Even with moderate haste item + Celerity, your hasted delay will be below 10, that is, less then 1 second. You have to realize that the time stamp in your log only has a resolution of 1s. With a hasted delay below 1s two subsequent combat rounds can therefore show up in your log with the same time stamp. For instance, at max haste (100%), the time interval between combat rounds is 0.85s. If the first round takes place at say 1.08s then the next is at 1.93s. Both times get truncated down to 1s in the log. (If timestamps get rounded to the nearest integer instead of truncated it can still happen. 1.51s and 2.36s both get rounded to 2s).

What you see in the chat window isnt a good measure at all; the timing of the lines is affected by lag. I see mobs hitting me after they died all the time in the chat window, but never in a log.

Oh wait. Just had a revelation (maybe). BF - double attack. I suppose one proc could be on the main hit and the second proc could be on the double attack?

Nope, double attack doesnt change the fact that you can only proc once in a combat round. This is exactly why proccing weapons are so important for beastlords. A monk for instance has innate double attack, triple attacks at 60 and AAs for quads. He can swing his weapons about twice as often as we do, thus getting twice the melee output from the same weapons. But he gets only one chance to proc each combat round just like we do so he still only procs 2x per minute and his proc dps is the same as ours. In short, our proc dps is a bigger portion of our total weapon dps then for a monk. With weapons like the ED, that's a very big deal.

/hugs

[Chubaka]

Gotcha...........sounds good.

[tkyn]

This might work for a test weapon on the thing Tipop mentioned about combat swings.



I can equip it, and attack with it, but nothing happens. Not a damn thing, try it out. Maybe you could stick a proc aug on it and see if it works then. Weird item.

[Derrict]

By the way... I don't think procs go off when you swing. If I am on autoattack and within range of a mob, I can proc even when I haven't made an attack that second.

Sorry, this can't occur unless you're seeing server or client side lag.  You can only have 1 proc fire per combat round per primary or secondary weapon.

If you have a double attack round on your primary, you can have either the natural proc, augmented proc, or an imbued proc fire once.  The same for the offhand weapon, thus, a max of 2 procs per complete round.  This was tested by myself and several others on the warrior forum; the only cases where it seemed to have several procs within a combat round is due to the log file rounding of seconds and having 2 complete swings within the same round.

[Derrict]

1. How many procs per minute our warders average with their proc buffs.
2. If given a weapon, do they use it's proc rate or the same rate as the buff procs?
3. Does dex affect pet proc rates the same as player proc rates?

I posted parses on .com but since the forum is down I can't make reference to it.

1. From my memory, approx 6-7 ppm when the pet is buffed with arags, dexterity, and rellic.

2. Imbued proc decreases when you equip it with a proccing weapon(s)

3. Yes.  Casting dexterity *or* infusion of spirit will increase the imbued proc rate but casting both had no further increase (with arags and rellic already on the pet).

[Xarilok]

Someone should parse the proc rates of mage toys, rellic, and rellic + toys.  Here is why:

My old main was a shaman, I used lifetap proccing weapons all the time, since life=mana to me.  My main weapon for this was the 15/23 mace that procced a 75 poing lifetap, auged with a 50 point lifetap.  I parsed it out to 2PPM for the main proc, and 2PPM for the aug'd one, with only a few % of error, plus or minus.

I swung approx 17% of the minutes I was fighting, due to constantly casting canni4, torpor, or dots.  During the time I was NOT casting, and COULD proc, I did so at a rate of 1 proc per 5.1 seconds.  If I procced at this rate for the entire minute, I would have procced 11.8 times, but since I was only actively swinging for 17% of the minute, my 2 procs per minute occured in a condensed time.

From this, I concluded that the proc time counts while you have autoattack on, and don't switch targets, whether or not you are swinging.

Therefore, I would think, based on my old parses, that *IF* you give a pet two proccing weapons, it will NOT lower the innate proc rate of the spell cast on him, as the proc rates per swing will go up to compensate for the fact that there are less available swings to proc on, due to other procs occuring at that time.

[Tastian]

" I swung approx 17% of the minutes I was fighting, due to constantly casting canni4, torpor, or dots. During the time I was NOT casting, and COULD proc, I did so at a rate of 1 proc per 5.1 seconds"

I used to do the same with my shaman, but the 17% you are listing sounds quite low.  Cani V is 2.5 casting time with 3 sec refresh.  I'd often get 2 swings in during spell recovery.  Torpor was 3.0 for me and again 2.5 recast.  Dots took a bit longer depending on what I was doing, but I only cast 1 disease every 2 mins and maybe 3 poison during that time (again depending).  Epic was a big time eater, but again it lasts over a minute and if it's not running it's duration then you aren't using it.  

I've got aug'd out weapons and max WA and simply put I don't proc after each cast.  As I recall I tested proc rate with my shaman by chain casting khura's as one point.  It has a 26 second base casting time and 2.25 refresh.  If you chain casted it with the standard 2 procs per minute you'd basically have to have your weapon proc after every single cast of it finished.  

Rellic has been parsed and I'm working on parses of pet and pet with weapons (including toys).  From what I've found so far though the pet attacks so much that the overlap in procs isn't enough to cause a reduction in rellic procs from weapon proc.  I'll try to get sometime to finish up some tests this weekend and post results.

[Kylaz]

I swung approx 17% of the minutes I was fighting, due to constantly casting canni4, torpor, or dots.

It really often happen to me that I get a proc right after a cold nuke hits.
From what I understand, Proc rate seems independant with weapon speed/haste, just a basic #procs/minutes.

Did anyone parse proc rate, with auto-attack on, but chain casting whatever spell is available, to see if it is linearly dependant on the amount of time spent meleeing ?

[Coprolith]

It really often happen to me that I get a proc right after a cold nuke hits.
From what I understand, Proc rate seems independant with weapon speed/haste, just a basic #procs/minutes.

ROFL, what is this, the new "I-haven't-parsed-but-I-notice..." - flavour of the month? We've already been over this twice in recent weeks

Procs are still tied to combat rounds. There is no clock ticking that tells your weapon "its time to proc now". The game calculates a % chance to proc per combat round in such a way that it is independent of delay/haste. But when you reduce your number of combat rounds by simply not attacking but casting spells instead, you're also reducing your number of procs. That basic #procs/min is the rate you get when you do nothing but melee.

/hugs

[Tastian]

Ok here is what I'm going to do.  First I'm going to give 3 completely different situations.  Then I'm going to offer up how I think the system works.  After that I'm just going to pound my head into my desk and weep openly.  Here goes...


Situation #1 -  ARCHERY  -   For those of you that don't know there is NO form of auto attack archery currently in the game.  There is no "on" mode for archery there is no way for the game to honestly know whether the shot you just fired was your first or your 50th in a row unless it was somehow tracking every shot you'd made recently.

Situation #2 -  DUELING  -  Duel someone, Target them, Turn on auto attack and then joust them.  Do you get a proc EVERYTIME you get into melee range with them?  You don't even have to duel someone for this.  Target mob_X, be sure you are out of melee range and turn on auto attack.  Now wait 5 minutes, or two minutes or 30 seconds and step into range.  Did you proc right away?  If so, do this 5X more, did you proc first go every single time?  If there were some "on" mode counter going it'd recognize you had been in "autoattack" mode for 5 minutes with no proc, clearly this "proc timer" would have registered this by now.  If not then you would have to tell me that the game is tracking every melee hit you do and that it's not counting the time auto attack is on well not scoring hits.

Situation #3  -  AE'N  -  Warriors can get "nice" exp solo'n by using PB AE off an earthshaker.  This is a weapon that procs a PB AE earthquake spell.  They remove any haste items, get malo'd/slowed by a shaman, then round up a bunch of monsters and fire off rampage which hits every mob and off of most of those mobs they get PB AE procs to fire off.  Again if there were some "proc timer" the game would simply go "oh my cannon_fodder has proc'd 17 times this SECOND".

What likely happens is that a weapon/aug/whatever is given a calculated X value based upon weapon_mod'd_proc_rate, haste, etc and on a combat round the good ole RNG spits something out if it's <= X then you get a proc, if it's not then you fall through that proc check into other proc checks if you have any.  

There are just too many cases where a "proc timer" doesn't make sense.  I've had 3 hour sessions where my LDoN aug fired more than my weapon augment (not by much, but it did).  I've had a lot of variations in my procs.    I think people hear the averaging that parsers and people doing weapon comparisons talk about and misinterpret it.  If there really was some sort of "proc timer" then procs wouldn't be nearly as random as they are.  There's no way you'd go 3 minutes without a proc if the game was making you proc 2X per minute.  There's also no way the game would let me fire off 5 LDoN procs in under 15 seconds if this "proc timer" existed.

[Tastian]

Ok I took some more time and ran a few quick tests in PoJ.  First I had 305 dex, I have WA5.  I had standard proc rate weapon main hand with an extra aug'd LDoN proc.  Offhand I left a weapon that only had an aug'd LDoN aug.  I *should* have proc'd 7.5 times per minute on average ((2 main + 2aug'd_main + 1 off)*1.5).  

First mob was some prisoner rat with a weird name.  I simply cast slow, turned on auto attack, I left autoattack on and just cast celerity. (8.0 casting time without spell haste).  At one point I did paragon.  If I got interupted on celerity I waited for a combat round then recast, etc.  

Results - 3 MH procs  7 aug'd procs  1 OH proc in ~5.5 minutes.  ~13% of the procs I should have had.

Next mob was a poor rat that I happened to also notice in the cell.  I targeted him and turned on auto attack.  I stood there with him targeted with autoattack on for over a minute, then I cast slow, mob tash'd me (oh no I'sa gonna die!!!!!) then came running over, NO PROC, I cast celerity, NO PROC, I cast fero on myself (can't be too careful!!), NO PROC, I cast celerity again and finally got a proc after that one finished.  I do the same basic cast melee, cast melee, cast melee thing.  Until he runs then I crit nuke him into the floor for 1485 (HE TASH'D ME!!!)

Results - 3 MH procs  1 aug'd proc  1 OH proc  in ~3.5 minutes.  ~18% of the procs I should have had.

Final mob is a minotuar guy just down the hall.  I move into his room, same basic deal.  I turn auto attack on then cast, pull, cast, melee, cast, melee, cast, melee.  Once he turned to run I just left attack on and proc'd a couple times to kill him (hey he didn't tash me or else I would have nuked him down too!!).

Results - 4 MH procs 2 aug'd procs 4 OH procs in ~4 minutes.  ~33% of the procs I should have had.

Total results!!  10 MH procs 10 aug'd procs 6 OH procs in ~13 minutes.  Or ~26.6% of the procs I should have had.  If I had just been chain casting celerity then I would have been melee'n ~21% of the time btw.  With the other spells I used, the extra melee and interupts these numbers (% of fight time and % of expected procs) are VERY close especially with such a small sample and a crazy RNG (guess that makes it good? hehe).

These are small sample sizes, but you are talking about something that is supposedly a timer.  I could have kept this up forever and there is just no way to get back to the ~7.5 procs per minute I should (and usually do about have) because you would need to have multiple procs per round.  Spell casting DOES LOWER PROC RATE.

Also this is just something to throw in.  I have CF3/FoA and of course 305 dex.  During this time I actually had a 6.7% MH crit rate and a 5% OH crit rate.  

The RNG is streaky people and the human mind works in funny ways sometimes.  Before I end this post and hopefully see the last of this discussion involving "seems" (bring data/parse/whatever and I'd be very happy to continue any discussion, things change, but please give me reason to go through this stuff over and over and over again) I just wanted to post a small snipit from an old log.

[Sat Jun 05 21:12:19 2004] A festering rat has been burned by Rujarkian bile.

[Sat Jun 05 21:12:24 2004] A festering rat has been burned by Rujarkian bile.

[Sat Jun 05 21:12:28 2004] A festering rat has been burned by Rujarkian bile.

[Sat Jun 05 21:12:35 2004] A festering rat has been burned by Rujarkian bile.

[Sat Jun 05 21:12:50 2004] A festering rat has been burned by Rujarkian bile.

Yes that is PROC, 5 seconds, PROC, 4 seconds, PROC, 7 seconds, PROC, a whopping 15 seconds PROC again.  Things like this will jump out at a lot of people and stick in their minds.  Truth is I'd gone 3 minutes without a proc before that barage, then went another 40 seconds before my next one.  My point is things like the above with procs are easy to see especially if you have your chat windows set up a certain way.  Don't let yourself be fooled by what you "think" you see though.

[Coprolith]

Before I end this post and hopefully see the last of this discussion involving "seems" (bring data/parse/whatever and I'd be very happy to continue any discussion, things change, but please give me reason to go through this stuff over and over and over again)

Amen, and thanks for taking the time once again to sort this out T. You'd think that after 5 years of dispelling EQ myths involving "it seems" people would know better then to believe their perceptions of the RNG, but here we are.

If you fight less, you proc less. This is such a simple and intuitive concept that you'd think no one could possibly miss it, but apparently they can.

Perhaps a 'proof-by-intimidation-using-technical-lingo' approach will work.

The basic formula that governs your proc rate is the chance to proc per combat round which the game calculated from the simple expression:

proc_chance_per_combatround = C(dex) * hasted_delay (1)

where is C(dex) is a multiplier which depends on dex only. The number C is prolly looked up from a table, or is itself calculated with a simple formula.

The average number of procs you get per minute (ppm) is then

ppm = #combat_rounds_per_min * proc_chance_per_combat_round (2)

I think what confuzzles people is this: all those parses on proc rates are done against mobs like the katta banker, that is, mobs that dont fight back. You hit auto-attack, turn of the monitor, go to bed, and in the morning you have a large sample of procs. The key thing is that this is strictly melee, 100% of the time. In that case, and that case only, the number of combat rounds per minute is

#combat rounds per min = 60/hasted_delay (3)

If you substitute (3) and (1) into (2) then hasted delay factors out and you get

ppm = 60*C(dex) (4)

that is, the proc rate is a constant number depending only on dex. But when you're casting spells, or get out of range/cannot see the mob, or even simply turn of attack in the middle of the fight for a few secs, you are not meleeing 100% of the time, so (3) is not valid and therefore (4) is not valid. Instead, if f is the fraction of the time that you are meleeing, then instead of (3) we get

#combat rounds per min = f*60/hasted_delay (5)

and

ppm = f*60*C(dex) (6)

Note that ppm is still independent of hasted delay, and it still has the same dependency on dex. But its lower then the theoretical max because f can only take values between from 0 to 1. If you are casting half the time, you're also only meleeing 30s of every minute, so f=0.5 and the ppm is cut in half.

/hugs

[Tastian]

"Amen, and thanks for taking the time once again to sort this out T. You'd think that after 5 years of dispelling EQ myths involving "it seems" people would know better then to believe their perceptions of the RNG, but here we are."

Nps, the only thing that is starting to bother me is that I am now spinning my wheels.  I'm already working on collecting data and a few other tests, but I keep coming back and re-inventing the wheel because various people "seem" to notice changes.  When I first pointed out that warder's no longer lose damage from getting 1hs weapon at various levels I started with ~3 hours of parsing and followed it up quite quickly with 6 more.  I spoke with other people and ran various tests.  When you were testing str's worth and it's impact on attack you ran different tests you parsed, etc.  People that keep argueing fact with "seems" and "opinions" are driving me nuts atm.  

Understand that I'm not even asking for a conclussive game altering fact.  "Lastnight I logged 6 hours and well using weapons my pet did less than the number I constantly see people quote HERE IS MY LOG".  Maybe you were fighting in VT, maybe the pet's hit rate was down, maybe you summoned your lower level pet and he was hititng for less and bashing for less.  People need to realize that even if what they think they are observing is accurate there are still many different possible explainations and it's very hard to offer up an answer sometimes without any form of data at all to go on.  

Again I'm not looking for 12 hours of parses before you even chime in or posts filled with lamda's sigma's and % signs, but could ya atleast let me see what you are basing this "seems" on.  I feel like I'm starting to get to the point in a "discussion" with someone where they start argueing facts with opinions.  If that's all we are doing then I'm going to start going "well it seems the exact opposite of what you just said" and save myself some time lol.