English


uiWebPrevious12uiWebNext

#1 Report | Quote[en] 

From time to time I argue with ppl because they „believe“ different things about Ryzom than me ^^
Many things are easy to prove or to disprove. But not all of them.

Myth 1:
Accurate Attack does more than it’s (english) description says, it also reduces the parry rate of parry enemies.
My opinion: Yes, I’m pretty sure this is correct.

Myth 2:
If many ppl use the Gubani Wheel at the same time, they have a lot higher chance to win something rare.
My opinion: I was not able to prove or disprove this yet.

Myth 3:
The smaller the homin (highest overall skill lvl), the higher is the attack range by mobs.
My opinion: There is something about it, but I’m not sure.


What do you think or know about this myths?
Do you know other myths to investigate further?

Thank you :)

#2 Report | Quote[en] 

My 2 dappers worth ...

1) Accurate Attack finds a way to hit more often regardless of defense mode, whether it's predicting a Dodge target's movement or aiming for weaknesses in a Parry target's defenses. Different means, but same results.

2) More spins means higher odds of someone getting a better prize, but no higher than the odds of one homin achieving the same results with the same number of spins. However, 10 homins can get 100 spins 10 times faster than a lone homin can.

3) I've never noticed. Mobs treat Widget and I the same as near as I can tell despite me being twice her "size".

---

Do not assume that you speak for all just because you are the loudest voice; there are many who disagree that simply have no desire to waste words on you.

#3 Report | Quote[en] 

1. My experience is that accu influences the dodge-adv_dodge relation as well. I as well as my alts used accurate attack to make training on stingas less tedious. The effect can be tested by disabling accurate attack from the stanza. The hit rate then drops remarkably.
2. I don't think so. If the random algorithm is implemented properly (and I ran a couple of tests in fight and crafting corroborating this assumption), the process should be ergodic, meaning that average over space equals average over time. Honestly, I did not test this hypothesis on the wheel proper (would be hard to gather a sufficient number of participants).
If randomization is implemented properly, each wheel turn should be independent from the previous or following ones no matter wether they are happening at a time or subsequently. I shall test a 2 player situation and note the win rate playing alone and together once I have murdered enough eggs for the next 500 tokens :).
3. No. I studied attack ranges as a beginner, going so far that I approached and followed KP in the roots. to my experience, the attack ranges never changed over my level progress, and remained the same for lower level alts.

---

Daomei die Streunerin - religionsneutral, zivilisationsneutral, gildenneutral

#4 Report | Quote[en] 

1. To my view ... that is what accurate attack should do, increasing your accuracy, could not be accomplished w/o, in effect negating adversary's dodge / parry.

2. No comment, don't use the wheel.

3.a Not sure I understand ... if we talking smaller by toon height, I have not observed any real difference. The aggro range in the Bunny Tools chart has been around for about 12 years and have not rec'd any zorai or trykers comments on this topic.

3.b The part 'The smaller the homin (highest overall skill lvl)," ... wouldn't smaller by skill level mean lowest overall skill ?

3.c I haven't observed any mobs targeting homins by skill level, but I would say once engaged, they have a tendency to "go to damage". If a tank is engaged doing 1150 damage, they do have a tendency to ignore the tank and go to the mage doing 2700 if the tank does not hit frequently enough. This tendency can be pretty much negated entirely by having the mage pod at 20-25m away.

---

#5 Report | Quote[en] 

There are many myths and misconceptions about AI behavior, and this reminded me of various misunderstandings I've heard led a few people astray, as well as a few (possibly flawed/erroneous to the point of bordering on "myth") observations;
Fyrosfreddy
3.c I haven't observed any mobs targeting homins by skill level, but I would say once engaged, they have a tendency to "go to damage". If a tank is engaged doing 1150 damage, they do have a tendency to ignore the tank and go to the mage doing 2700 if the tank does not hit frequently enough.

As near as I can tell, mobs will go after whoever is altering the HP balance of a fight to their detriment. Mobs have fairly short attention spans so they tend to go after whoever has swung the balance most recently, with some adjustment for how dramatic those swings were, though not as much as you'd think. (If I can tank plods 2 at a time with a pair of q90 daggers and a CC skill around 120, there has to be more to it than DPS!)

The most common way to tilt the HP balance against a mob is to deal damage to them, whether by nuking or tanking. Tanks are generally effective at holding a mob in part because they have more HPM than casters... when they hit. Depending on amps and stanzas, most casters average about 35-40 HPM. Sure, nukers have better DPS than tanks, but the HPM disparity means that, assuming the nuker gets resisted about as often as the tank gets parried/dodged, the mob will stick with the tank most of the time. Things change a bit when there is enough of a level disparity between the tank and the nuke to affect hit percentages but it's still not totally about DPS; speed seems to play a larger role than it gets credit for. Moving on....

What many people miss is that up you make the tank's HP bar grow, you are also altering the balance of HP to the detriment of the mob. Instead of subtracting from one side of the equation, you're adding to the other. Think of healing as "negative damage" if that helps things make more sense. It's actually a bit worse for healers since heals are never parried/dodged/resisted, and are generally for more points than many melee/ele attacks. That gives them a pretty high DPS, which in turn raises the odds of the mob ignoring the tank and shifting it's attention towards the squishy if the tank misses too often. Against a solo mob, that isn't too big a deal as the mob can only be in one place at a time. However, social mobs can decide they want to split their attention. Next thing you know, half the spawn is now trying to chew on the pod while the tank has to decide between dragging the rest of the spawn onto their teammate(s) or just letting the pod deal with it on their own.

Given the complex interplay between level, effective HPM, and DPS (both theoretical and actual), I can see how one might think that mobs treat high-level homins differently from low-level ones. However, I also believe that people mistake correlation with causation all the time, and I've seen enough "exceptions to the rules" that I have a hard time believing that there is a simple answer here.

Now for the 64,000 dapper question; do the above observations and theories have a strong basis in fact, or am I merely making my own myths in an attempt to try to understand something I don't fully grok? Or is there no telling one way or the other and thus a need to go blow stuff up in order to find out?

---

Do not assume that you speak for all just because you are the loudest voice; there are many who disagree that simply have no desire to waste words on you.

#6 Report | Quote[en] 

Gidget
There are many myths and misconceptions about AI behavior, and this reminded me of various misunderstandings I've heard led a few people astray, as well as a few (possibly flawed/erroneous to the point of bordering on "myth") observations;
Fyrosfreddy
3.c I haven't observed any mobs targeting homins by skill level, but I would say once engaged, they have a tendency to "go to damage". If a tank is engaged doing 1150 damage, they do have a tendency to ignore the tank and go to the mage doing 2700 if the tank does not hit frequently enough.
As near as I can tell, mobs will go after whoever is altering the HP balance of a fight to their detriment. Mobs have fairly short attention spans so they tend to go after whoever has swung the balance most recently, with some adjustment for how dramatic those swings were, though not as much as you'd think. (If I can tank plods 2 at a time with a pair of q90 daggers and a CC skill around 120, there has to be more to it than DPS!)

I agree that frequency of hits supersedes damage done, and can confirm that even low level dagger tanks are able to hold much stronger foes. While I succeeded in holding much more opponents using circular when dealing with social herbies, I never managed to hold more than 2 hornchers no matter which weapon used (and even when having taunted all of them) while I well recall to have held at least 3 kinchers, so there may be a species dependency. Torbaks seem to always be trying to turn to the healer and back to the damage dealer (no matter whether melee or nuker) often resulting in ineffective change of target.

Btw., up to 40 hpm is only available for a caster when constantly spamming enchants, otherwise max hpm will be 21.4 (60/2.8) hpm for a single rocket cast, and 20 hpm for a double. But my experience did at least not systematically show that a mob strayed from a tank when it managed to dodge/parry often, rather it seemed to take the hostile intent for the action.

Yet I did not fully understand what you mean with "HP balance". Do you mean the HP left to any party, so that a severely wounded mob would act differently? I cannot confirm that from my observations. If you mean that a healer is perceived like a damage dealer (which perfectly makes sense), I agree.

---

Daomei die Streunerin - religionsneutral, zivilisationsneutral, gildenneutral

#7 Report | Quote[en] 

Daomei
Yet I did not fully understand what you mean with "HP balance". Do you mean the HP left to any party, so that a severely wounded mob would act differently? I cannot confirm that from my observations. If you mean that a healer is perceived like a damage dealer (which perfectly makes sense), I agree.

I used the word "balance" because of the visual in my head. Put the party's HP on one side and the mob's HP on the other...


Whether you nuke a mob for 3k damage or heal a tank for 3k HP, the scales are tipped three thousand points in your favor. And subtracting a negative number adds while damage subtracts, so I think of heals as "negative damage" since it adds HP. Dealing negative damage is still dealing damage though, so yes, healers are treated the same as tanks and nukers.

As for caster HPM, my thinking is that the listed cast time may be 2.8s for singles and 3.0s for doubles, but then you multiply HPM by ( 1 + ( Amp_cast_speed_bonus / 100 ) ). A set of +97% Sup Zun amps can cast a double-missile in just over 1.5s (3.0/1.97=~1.52) and 60/1.52=39.47; just a hair under 40 HPM with good amps and no time credits. Since a lot of people use small time credits on big spells or "grind" amps that are closer to +87% than +97%, actual HPM is often a bit lower, but still well above 20. That's still a bit faster than most 2h weapons, but slower than a decent 1h weapon (especially Spear). Spamming enchants doesn't really affect speed since the cast time becomes a cooldown yet leaves the time between spells unchanged.

Speaking of enchants, I recall getting absolutely raked over the coals for saying that enchants actually did stop you for a split-second. A few people were not so kind about insisting that that was just an illusion caused by the animation and/or server lag. However, my experience is that if you are being chased by a mob that you can barely outrun, the gap between you may increase 1m every few seconds but will narrow by about 1m instantly if you fire off an enchant. Do I have a freaky internet connection or do you actually stop for a split second when you fire off an enchant?

---

Do not assume that you speak for all just because you are the loudest voice; there are many who disagree that simply have no desire to waste words on you.

#8 Report | Quote[en] 

Gidget
As near as I can tell, mobs will go after whoever is altering the HP balance of a fight to their detriment. Mobs have fairly short attention spans so they tend to go after whoever has swung the balance most recently, with some adjustment for how dramatic those swings were, though not as much as you'd think.(If I can tank plods 2 at a time with a pair of q90 daggers and a CC skill around 120, there has to be more to it than DPS!)

Just spent 30 minutes typing a response only to have it go poofie ... seems Ryzom forums don't support Lazarus add on :(.

Many moons ago, killing plods was much slower than it is today:

a) Cats were in very short supply, especially 250s.

b) Enchantments hadn't come into play as yet.

So in a plod hunt, ya had much more time to chat, observe, discuss and examine / evaluate options. Of course the real challenge to working plods was / is dealing with their "special skill" of cloning. There were many things that we observed affected clones peeling off on mage pods. These observations were undertaken in the early days of Ryzom right after the Weapons Adjustment patch.

1. Obviously circular attack with a count = to the number of clones was essential.

2. As the tank's job w/ plods was primarily as a meat shield, making sure you don't miss was more important than doing damage. So accurate attack was a key action component and generally given preference over Inc Damage.

3. I'm sure many old timers will remember that, at one time, all 2H weapons had the same damage as far as damage and hit rate. You can see some of the old values in the Crafting Ranges Tool threads as they have been modified several times over the years. As with most things Ryzom, there is not very many "what is the best ?" answers ... only "what is the best in which situation ?".

4. When the weapon stats were 1st adjusted, the patch notes included explanations. Obviously dev's goal was not to have 98% of the player bases all using the same weapon. Post patch, Mace and Axe now gave more damage but had a lower hit rate. While DPM was the statistic in most players held in highest regard when choosing a weapon, the oft ignored stat was that higher hit rates used more creds and therefore increased healer's load.

But as the goal of the patch was to provide more specialized selection of weapons, the idea was to still maintain a balance and not have a "best weapon" in the game. The notations on pike were less obvious and at first glance it is the poorest weapon as it has the lowest DPM. So what was pike's + side that the dev's figured made it a compelling choice (other then advantage offered by it being "pointy"?

According to the patch notes and other information released post patch ...

a) The increased reach of pike was to be an advantage ... the implication was that this kept the mob off the tank and at a greater distance, it would do less damage making it easier on the healer. I was never able to figure a way to observe / confirm this as to magnitude of effect or even if it was real.

b) The benefit of this feature that was directly observable, was in "holding" the clones and keeping them from peeling off to the mage pod. When the clone would peel off and start for the mage pod, circular attack, gave the tank another shot at getting / maintaining their attention because the weapon had a longer reach.

c) The pike also had a slightly higher hit rate than other 2H alternatives so the combination compounded the tank's ability to catch it before it got away.

Back then, never found pike to be much of an issue with 2 or 3 plods, 4 required attention and 5 was a challenge ... if you could hold 5 for more than 30 seconds, you did very well.

5. While I don't know what the 'mechanic' is, it certainly seemed that there was a benefit to rotating the primary target. I dunno if the tank's hit % is lower with secondary targets , whether again, the claimed distance advantage figured in to the %, it certainly seemed that if you changed primary targets you had less clones peel off to the pod.

6. These observations were made shortly after that weapon patch and as we rotated who served as tank and what weapon was used, we did come to a consensus that we had less "peel off's" with pike than any other 2H weapon.

7. However, distance to the mage pod seemed to be the primary factor in how often the cloned peeled off. With most spells in use having a range of 42 - 43m that became the default starting point .... but any distance above 25m was certainly workable.

8. Another factor also seemed to be team size / kill rate. We seemed to get more clones the faster we killed and the fewer mobs left in the spawn. As the years have gone buy and I have participated in Plod teams, the teams are smaller, the kills are slower and so the amount of clones appearing is much lower than it was when we were chasing our 1st masters.

9. Finally another factor that was noticed is one we still see today w/ every NPC Hunt. When killing the cats at Aen, the guard cats are much easier for the tank to hold when a)

a) The tank is 20-25m away from mages.

b) The tank healer and damage mages do not hit the mob before the tank gets a shot or 2 in for each cat. This goes back to the rotating targets observation above ... when a new plod appeared, always found it best to "hit the new guy" and get their attention ... you already have the other one's attention.

Now one thing that must be mentioned I'll again equate with the Crafting Ranges Tool. If you look at the CR Tool threads, after that 1st weapons patch, there were several later changes to weapon properties and stats. Most of them were documented in patch notes, some were not. Other than where noted otherwise, these observations for Plods come from experiences and observations from over a decade ago.

Now plod killing is much changed with smaller teams, cats, enchants, etc which have provided much less opportunity to observe cloning behavior. Last time I participated we had teams of 4 - 5, and we almost never saw more than a single clone, rarely would have a 3rd plod to deal with. Holding the clone(s) (2 - 3 total) was no problem even when I tried pistol and CC.

So in recent years, not having advanced any 2H skills (208 and 228 left as I recall) much in 5 years or more ... can't really speak to what may have changed which was not described in patch notes. So, when speaking of plods, I certainly can not speak from experience **post merge** as not enough "plate appearances" so o speak post merge. If things have changed, I didn't notice in the few instances that I have participated since then.

Edited 2 times | Last edited by Fyrosfreddy (7 years ago)

---

#9 Report | Quote[en] 

Fyrosfreddy
Now plod killing is much changed with smaller teams, cats, enchants, etc which have provided much less opportunity to observe cloning behavior. Last time I participated we had teams of 4 - 5, and we almost never saw more than a single clone, rarely would have a 3rd plod to deal with. Holding the clone(s) (2 - 3 total) was no problem even when I tried pistol and CC.

If you want to see clones, pull them with bomb-ele. There have been enough times where I've pulled the entire pod of five with bomb-ele and the team has killed all twelve of them. However, when I pull them like a sane homin would, cloning is nearly unheard of.

---

Do not assume that you speak for all just because you are the loudest voice; there are many who disagree that simply have no desire to waste words on you.

#10 Report | Quote[en] 

Having tested with about 8000 tokens so far, I want to add a preliminary report of my experiences with the wheel. First: no exceptional reward won (no title, yubo etc.).

Win rate varies slightly, between 22 and 27%. Most of the rewards are tokens, causing a token reflux of 21-52% averaging to 35-40% ca. Other rewards like cats (in packs of 20), mostly of q150, seldom at q200, are rare, most of other rewards are boost flowers q100 (dex int con, rarely will or balance), very seldom a set of 10 fireworks A-J q100.

I tested with solo play and simultaneous two players (well, a displacement of some ten or hundred microseconds between Diwu and me) . There was no significant aberration. Initially, it seemed that Diwu fared about 10% better in win-lose and over 20% better in mass when playing in Fyros, raising suspicion of race bias. But regression tests over all of Atys (wheels in FH, Zora, and Yrkanis, and Pyr again) could not support that assumption. Overall, we fared more or less equally inside the error margins.

So far, the null hypothesis that the wheel is working randomly is not challenged.

Last edited by Daomei (7 years ago)

---

Daomei die Streunerin - religionsneutral, zivilisationsneutral, gildenneutral

#11 Report | Quote[en] 

Have not been around for a while, as far as tanking a mob and holding aggro, player and mob HP doesn't have anything to do with it
Aggro is damage based, it's an aggro table that we can't see in-game


I can tank any boss with default attack and taunt in the same macro
Can also tank multiple mobs with default attack and circular without them attacking other team members, the key is not to heal or nuke until the tank has hit the mob. This is why plods and other multiple mobs go for healers or nukes.

Taunt resets aggro table so the tank goes to the top again, if tanking multiple mobs the members of the team need to target the same mob the tank is targeting

The most basic mistake people make is healing the tank before he has control of a mob or multiple mobs.

As for plod clones, as far as I can remember, kill and quick as you can and don't loot, once they start cloning just kill them in the order they clone, never kill the plod that just clones last, not looting is always the key

---




Peace is a lie, there is only passion.
Through passion, you gain strength.
Through strength, you gain power.
Through power, you gain victory.
Through victory, your chains are broken.
Ma-Duk shall free you.

#12 Report | Quote[en] 

Gidget
If you want to see clones, pull them with bomb-ele. There have been enough times where I've pulled the entire pod of five with bomb-ele and the team has killed all twelve of them. However, when I pull them like a sane homin would, cloning is nearly unheard of.

Are we talking social behavior or actual clones ?  If there's already a pod of 5, then that's not a "clone".    When speaking of "clones", at least how I use the term.  You have a pod of 5 plods and the tank is dragging the target 40-50m away from the "spawn point".   In the middle of the fight the plod "clones itself" and you are now fighting 2 plods ... 3 ... 4 ... or even 5.  They didn't "spawn", you are quite a good distance away from the spawn point.  Each of the super herbies has a special skill and this behavior is unique, in my experience,  to plods.

---

#13 Report | Quote[en] 

Fyrosfreddy
Are we talking social behavior or actual clones ?  If there's already a pod of 5, then that's not a "clone".

I've pulled 5 a fair distance from the spawn and our party killed all 13 of them.

---

Do not assume that you speak for all just because you are the loudest voice; there are many who disagree that simply have no desire to waste words on you.

#14 Report | Quote[en] 

Donno if this fits in the myth thing, but it is my own personal belief when melee fighting mobs.

Where your character is standing matters.

At times, my char will get into a miss/parry/dodge chain where the mob can't be hit. Moving forward or back or slipping to the side a few steps can change that and the hits start piling up.

I've also developed the opinion that being at a slightly higher elevation, terrain wise, to the mob helps a lot. That's why I always try to pick the side of a hill to pull my mobs to for solo fights.

#15 Report | Quote[en] 

Thats a good question, I have noticed that I do dodge more often when moving.
uiWebPrevious12uiWebNext
 
Last visit Friday, 29 March 01:16:40 UTC
P_:

powered by ryzom-api