Defense Skills; OR, everything you ever wanted to know about Evade but were too lazy to search for

Aske said:

Nice graphs and data.

Questions i have are (maybe someone did experiments) whether anything else besides evade and mob type influences mob hit rate. In specific:

* do you get hit less if you move a little bit (e.g. walk backwards) as mob hits you? (without running faster than mob)

* do you get hit less if you damage mob -- it might be a coincidence but to me it always felt like once i start fapping (instead of hitting mob) i get hit more often.

* what happens if 2 of same mobs hit you at the same time -- do you get hit 2x more on average?

* Does successful mob hit rate depend on how heavy you are?

Click to expand...

Those are all good questions, but i haven't directly tested any of them. I know that some here will say that #2 is a definite yes, but my feeling is that you don't -- and i do a lot of fapping while hunting.

Also, the same for #3.

Number 1 i feel might be true, but i think it's more likely an unintended programming defect if so (for example, you occasionally are momentarily out of the mob's hit range when it attacks if you back up).

These are all quite hard to test because of the "dynamic" variation in the hit rate over seconds and minutes and maybe more.

Aske said:

Nice graphs and data.

Questions i have are (maybe someone did experiments) whether anything else besides evade and mob type influences mob hit rate. In specific:

* do you get hit less if you move a little bit (e.g. walk backwards) as mob hits you? (without running faster than mob)

* do you get hit less if you damage mob -- it might be a coincidence but to me it always felt like once i start fapping (instead of hitting mob) i get hit more often.

* what happens if 2 of same mobs hit you at the same time -- do you get hit 2x more on average?

* Does successful mob hit rate depend on how heavy you are?

Click to expand...

Great questions - I have wondered this as well. In fact, I do believe that if you hit the mob, it disrupts the attack of the mob. I will walk backwards when fapping, especially with Aurli and I do think it makes a difference. Makes sense that it is more difficult to hit a moving target (you).

Also, I posted this in a thread a while back but not many people seemed interested... it fits here so...

I discovered this when fighting hogglo. Once the beast in on you, imagine that the hogglo is the center of the clock dial. Move left or right in a circle around the hogglo and there is almost always a spot in the circle where you take 1 dmg hits. Camp that spot and let the bullets fly. It works on 9 out of 10 hogglo I fight. It can take a while to find that sweet spot, but it is obviously worth it. When I use this tactic on Aurli, they seem to miss more and thus more evade gains. Has anyone else experienced this? I encourage others to try it.

Rayne Jade said:

In fact, I do believe that if you hit the mob, it disrupts the attack of the mob.

Click to expand...

From animation of the mob this seems to be so, however if you look at yourself from 3rd person view as mobs hit you -- it looks like your attacks are disrupted too, while they are not.


With hogglo I don't think I can last long enough to find that 1dmg spot - i might try it later on, on some easier mob.

Doer... great work man; like always I might add.
tx

Aske said:

From animation of the mob this seems to be so, however if you look at yourself from 3rd person view as mobs hit you -- it looks like your attacks are disrupted too, while they are not.

Click to expand...

True... I have observed the animation from 3rd person as well.

With hogglo I don't think I can last long enough to find that 1dmg spot - i might try it later on, on some easier mob.

Click to expand...

Should be noticed on any mob... but would be more difficult to notice on very small ones. I discovered this on hogglo, but I also have seen it with Argos and Aurli. I am curious to see if you notice it as well.

I have always suspected that hitting a mob can interrupt its attack, because when I'm fisting (which hits really rapidly), mobs seem to hit me less often. Could just be an illusion b/c of the multiple small damage messages I get making it seem like it's longer between mob hits. I don't know... maybe it's worth doing some sustained testing on this.

I think their attacks also affect us. The first time I found my self in a tant swarm, I couldnt get my fap to work, they kept hitting me and it wouldnt go off.

High hit rates do affect the mobs hit rate too, comparing a powerfist to a sword is a huge differnce.

I think evade is a very misunderstood profession. Doer did a great work, but he had the chance to compare data for only few mob level and few evader level. All low level.

I can tell from my experience that a higher level of evade dose more than what is recorded in this thread. I suspect - and I am willing to participate in testing in future - that things are more complicated.

EU is skill based. We all have skills. But also mobs.

If is a fight between 2 avatars, this is obvious.

My impression undocumented is that evader level has a direct influence to the opponent HA. In the same time, the mobs deal attack using a "weapon". Mob weapon is similar to a SIB normal weapon. It dose maximum damage for normal skill, and it behaves badly as the skill is lower. Try using H400 with level 1 pistoleer and you will see what I mean.

If my claim is true, evade will cause mobs to fail some atacks until a threshold level is reached. After that, the misses are increased and the minimum damage is affected. Also critical hit ability is lowered.

All this is mob dependent.

I've noticed that on Longu Stalker first. It was impossible for me to tank one. It used to kill me whatever I did. I chipped from level 25 evader some alertness. Reached level 6600 alertnes from 2k or so. The result was that longu stalker could not kill me anymore as it used to. I've put that on the fact that he misses me and I manage to fap, and I kill it fast, but is more to it. Because is not missing me. His reload is changed, and that is something else.

Same thing with Kreltin and Aurli.
I used to have huge problems with adj fap in dome 4. After I chipped up and boosted my evader level to 40 I managed to solo a hunter in gremlin. It takes a long time to kill one, so you see things killing it.. like the fact that in gremlin you get many 1 hits. This never happened while I was at lower evader level. Now I am level 42 evader and can kill majestic in ghoul+5b with UR125 if it dose not crit me. Impossible some time ago.

Would help to know the real influence of evade, so testing it is important. I am a noob yet, but some ubers with level 60-70-90 could help us understand more about how this works.

My assumptions may not be correct, but I'm sure is something more to it than just "mob misses more". Like with our HA... a higher HA dose more than just "miss less". Linear interpolation in efficiency tools is a great help figuring out the economy of normal UL weapons, but is not final or accurate.

If you want to start a testing session, I will do my best to participate and share any info you think is needed and relevant. Working in team would make things easier.

BlackHawk said:

I think evade is a very misunderstood profession. Doer did a great work, but he had the chance to compare data for only few mob level and few evader level. All low level.

...

Click to expand...

This is pretty much as I have seen the situation myself for some time. So I second this.

But... add to this the damage being dealt to the mob too. I believe this has an affect. If you were to chip out all your handgun skills, you would find they would start hitting you harder again.

Evader Level + Dmg/sec = Affected Mob HA in loose terms.

BTW, Level 34 Evader and Level 55 Laser Pistoleer seemed to be the start of Hogglo misses FYI

Hunt Hogglo a while.

You will notice that they hit you more often when you are FAPping then when you are shooting at them. This seems to definitely indicate that shooting them disrupts their attack.

That is why having someone else FAP you while you hunt them can be a lot more efficient in several ways.
1. reduces kill time by not having to FAP your self.
2. reduced armor decay due to less hits
3. reduced ammo/weapon decay cost due to less HP to remove for killing a mob (the mob has less time to naturally heal)

Cheers
Richard
(ingame Siam)

The theory that doing damage causes more misses needs some testing first before I can agree with that. Since I dont experience it myself, and it also doesnt seem logical, I considder it to be untrue for the moment. I also think it is very unlikely that evade has any influence on the damage a mob does. I have never seen any proof of a mob that didnt have a 50%-100% random distribution.

Aske said:

* what happens if 2 of same mobs hit you at the same time -- do you get hit 2x more on average?

Click to expand...

If you don't shoot one of the mobs, over time, them do same damage. Asuming same mobs, same maturities.

If you shoot one mob, and the other hits you.. well.

We asume value X as "total damage inflicted on your arse by a mob you're shooting at". We asume value Y as "total damage inflicted on your arse by a mob you're not shooting at".

Soo, in the above situation, resulting damage will be : X+Y+X = 2X + Y more than double the damage (and decay) inflicted by one mob at which you shoot.

If we asume three mobs, two aggro, u shoot the third. U kill the third, then you have two mobs, u kill the second and finally u shoot the last.

Then we have:

Y+Y+X+Y+X+X = 3X + 3Y

So, again, is more than double and the difference increases alot.

So in the situation when you are aggroed by 12 mobs and you shoot the 13th, the damage taken and decay on your armour is already >>>>>>> than what would result from killing 13 mobs, one at a time.

Observations:

- fapping increases dmg & decay taken on armour (aside from decay on fap itself)

- taking damage increase defense skills

- X and Y change value with your skills, from decrease number of hits taken

Is up to you to measure and observe what you prefer

Witte said:

The theory that doing damage causes more misses needs some testing first before I can agree with that. Since I dont experience it myself, and it also doesnt seem logical, I considder it to be untrue for the moment. I also think it is very unlikely that evade has any influence on the damage a mob does. I have never seen any proof of a mob that didnt have a 50%-100% random distribution.

Click to expand...

Easy enough test.

Get an Atrox Young alone in full shogun (+plus plates maybe) and shoot it with an M2100.

Measure the time to kill and measure the hits taken (which should mostly be 1.0s to avoid having to FAP and nullify the test.)

Repeat this test until you feel you have enough data.

Then use a more powerful weapon lets say by 4 times, maybe a Breer P2a or P3a and hunt the same level of mob and measuring again the time to kill point and the number of hits landed and again repeat to match the same data points as the first test.

Divide the number of hits taken by the time taken to kill in seconds for each mob.

You should hopefully see the hit per second ratio will be reduced for the higher dmg/sec weapon. However, your current Evader level could skew the results somewhat which is not something we could logically factor in or out without involving the same test over a wider selection of avatars.

This however, maybe too low a mob for this test and might be wiser to aim for an Atrox Gaurdian level in Angel.

More HP, more Mob potency and more scope for dmg/sec in several weapons to test against.

Or no armor with imp fapper behind...

AkiranBlade said:

Easy enough test.

Get an Atrox Young alone in full shogun (+plus plates maybe) and shoot it with an M2100.

Measure the time to kill and measure the hits taken (which should mostly be 1.0s to avoid having to FAP and nullify the test.)

Repeat this test until you feel you have enough data.

Then use a more powerful weapon lets say by 4 times, maybe a Breer P2a or P3a and hunt the same level of mob and measuring again the time to kill point and the number of hits landed and again repeat to match the same data points as the first test.

Divide the number of hits taken by the time taken to kill in seconds for each mob.

You should hopefully see the hit per second ratio will be reduced for the higher dmg/sec weapon. However, your current Evader level could skew the results somewhat which is not something we could logically factor in or out without involving the same test over a wider selection of avatars.

This however, maybe too low a mob for this test and might be wiser to aim for an Atrox Gaurdian level in Angel.

More HP, more Mob potency and more scope for dmg/sec in several weapons to test against.

Click to expand...

Ok so you also think the amount of damage done determines the damage you recieve.

IMO the test you propose isnt realy covering everything. There should be a test vs a high HP mob where:
1) You dont to anything
2) use a low damage high firerate weapon (Mann MPH)
3) use a high damage low firerate weapon (Marber)
4) use a high damage and high firerate weapon.

Only problem is that option 4 will kill the mob pretty fast. It should also be tested vs several mobs at several evader levels to be sure about the results.

I've not read the whole thread, so hope I'm not missing anything vital or re-iterating something that's already been stated.

From my experience, it seems a mob's hitrate is based on a discrete rather than continuous function. There are a number of "tiers", with the first being "always hit", and then maybe 3-5 more tiers until it gets to a (complete thumbsuck) 3-5% hitrate.

So, every maturity for each mob will have certain Evader/Dodger level "thresholds" at which the hitrate moves from one tier to another.

So you might find a scenario as follows (these values aren't real):

Mob: Muluk-Hir Clan Warlord
Always Hit: Evader Level 0-45
Hit 75%: Evader Level 45-55
Hit 50%: Evader Level 55-65
Hit 20%: Evader Level 65-80
Hit 10%: Evader Level 80-120
Hit 3%: Evader Level 120+

I think it's changed now, but 6+ months ago Muluk-Hir Clan Warlords would indeed hit every time against someone with Evader Level less than 45. As soon as one passed the 45 threshold, they'd start missing. I believe it works the same way for every mob.

For the majority of mobs I guess the "Always Hit" and "Hit 75%" tiers are both Level 0->0.

Maybe players with really high Evader/Dodger levels can give some feedback on hitrate of Daspletor Stalkers, Araneatrox Stalkers, etc.

EDIT: It could also be possible that there's only a single threshold Evader/Dodger level determining when a mob is able to miss an avatar and not get a 100% hitrate; and then a continuous function is used from there.

I do still favour a discrete function. One example of why I believe this (although this example may no longer be valid): From discussions I had with a couple friends, there appeared to be a big jump between the hitrate of Hogglo Youngs for avatars with less than 40 Evader and those with more.

Witte said:

Ok so you also think the amount of damage done determines the damage you recieve.

Click to expand...

No, I think the Damage Per Second value is key, maybe not damage done per hit but over a time frame.

All you have to do is go on one Spider hunt with a team of 8 or more Level 40 Snipers and you will see how useless the Spider becomes at hitting people.

In fact a massive team of polish guys did a video which can be seen on Utoob where they were all power fisting spiders (about 20 of them) and they could easily tank the spider without them all dieing and they were all naked!

By that token it could even be the number of hits * dmg or some quotient like that.

AkiranBlade said:

No, I think the Damage Per Second value is key, maybe not damage done per hit but over a time frame.

All you have to do is go on one Spider hunt with a team of 8 or more Level 40 Snipers and you will see how useless the Spider becomes at hitting people.

In fact a massive team of polish guys did a video which can be seen on Utoob where they were all power fisting spiders (about 20 of them) and they could easily tank the spider without them all dieing and they were all naked!

By that token it could even be the number of hits * dmg or some quotient like that.

Click to expand...

The problem with this kind of observations is (like you already suggest) that we don't know what causes it. Do the high number of people close to it cause the spider to hit less? Does the fact is switches target very often causes it? Does the damage done to it causes it? Does the number of hits causes it? Or maybe a combination of all of these? All plausible explanations. It might even be that the observations are incorrect, as they are just impressions and not properly measured (I fisted spiders too in large teams and we all died several times). Overall, there is quite a lot to it to test this properly.

AkiranBlade said:

This is pretty much as I have seen the situation myself for some time. So I second this.

But... add to this the damage being dealt to the mob too. I believe this has an affect. If you were to chip out all your handgun skills, you would find they would start hitting you harder again.

Evader Level + Dmg/sec = Affected Mob HA in loose terms.

BTW, Level 34 Evader and Level 55 Laser Pistoleer seemed to be the start of Hogglo misses FYI

Click to expand...

One of the problems is that there is a large variation from minute to minute on mob hitrate. I had ample chance to observe this while working as a fapper for hire. One example that really stands out was when i was fapping Jimmy B on Chompers. He was getting hit far less than i would have expected based on his Evader level for a good 15-20 minutes. Then, in just one or two mobs, he was back to the expected range.

This complicates good tests of these very nuanced issues greatly...

By the way, i've had hogglos miss me while below level 31 evader. It was a very isolated event, but happened on more than one occasion.

I appreciate everyone sharing their observations. Unfortunately, most of this will remain conjecture until someone mounts a truly heroic extensive test to get reliable answers.

I will be repeating the tantillion Evader test again soon, now that my level is enough higher to be meaningful. I still have access to a level < 1 Evader for the tests, and i'm a level 32. Anyone in between or significantly higher that would be willing to follow the same protocol to contribute a data point is much appreciated. It would be nice to get enough points to prove or disprove Avery's theory of discretized levels of effective evasion, so we can proceed knowing how to treat future test results. My feeling is that it's a continuous function, but let's test.

Edit: I forgot to address a comment by Black Hawk. The way that HA is treated in PVP is completely different than against mobs. In PVP, a person with 10/10 HA will never miss another player in their sights, regardless of the player's Evader standing, while in normal hunting situations the best hit rate one can expect is about 92%. The (untested) hypothesis is that Evader level affects the curve between 0 HA and 10 HA. In any case, it's treated independently and thus isn't useful for analogy when discussing Evader and mobs' "HA". See Recoda's thread on the effect of sights and scopes for my testing results i use as a basis for these statements.

Doer said:

One of the problems is that there is a large variation from minute to minute on mob hitrate. I had ample chance to observe this while working as a fapper for hire. One example that really stands out was when i was fapping Jimmy B on Chompers. He was getting hit far less than i would have expected based on his Evader level for a good 15-20 minutes. Then, in just one or two mobs, he was back to the expected range.

Click to expand...

Jup, I totally agree on this, it would indeed take a heroic data capture to ensure a reasonable average data spread.

Doer said:

By the way, i've had hogglos miss me while below level 31 evader. It was a very isolated event, but happened on more than one occasion.

Click to expand...

So did I, but it really becomes "par for the Korss" (pun most definately intended!) at the Levels I intimated

Some interesting posts to read...I've got a WoF meeting to go to so I'll reread in more detail later.

On being hit affecting hitrate, I feel its certainly worth some testing. My intuitive guess (ie. completely untested) is that being hit shortly before an attack affects the chances of success. I don't think its dmg/sec based, more just based on how recently you were hit. I feel my own hit rate is slightly lowered when I'm mobbed, even if I'm being hit for 1.0 by them all. Likewise mobs seem to miss more when being hit by a team, although admittedly thats a bit hard to judge as you're facing the mob much less.

Thx will read into it if i have more time

My evade tests @ camp

I'm a reletively new player so not sure what I say will help but here goes...I hunt the mobs @ camp daily not leaving anything out but molisks . The regen is more disgusting than that of bibos so no molisks until i get a higher dps gun. I've done a few experiments with armour configuration to get evade so I could reach my target of 500 now currently working on 600. I've noticed in my tests that evade goes up in spurts u get them in quick succession after being hit like hundreds of times then after that nothing for quite some time. I did a test with Goblin + 4B plates this was the most eco for all maturities of bibo found @ camp so doing it on daikiba was a bit silly. I did something called being a "tank" where u just soak up all the damage. I got the idea from cihrpies of just receiving 1dam, I'm sure ppl remember them but with an average of 1 damage, techinically not over protecting to accellerate evade growth. As I could gather I think evade points are awarded after a certain amount of hits taken to HP. Anyways I used this armour combination & after 30mins of being hit the decay was "a lot" as a direct result of me tanking with 16 mobs . So I guess general hunting and moving through bibos would of been better for that test. Thou for next time I will probably just hunt snables in goblin for an amazing 0.43 pec decay per hit & see what happens to my evade then. At this current level of evade I can evade most mobs with a high % maybe 85% to sometimes 100%, thou I think an average of about 75% taken from 50-100% of hits is more likely, as sometimes a snable male young can land hits better than a daikiba old. . I'm guessing that with another 100 to my evade this would increase this % to around 75-100% so 87.5% average, but thats only guess work & strictly observation, not sure how you guys do all your calculations to work it all out? Well thats all I will add to this thread ty all and take time out to debunk any of my comments . PS I also think that getting evade is a lot faster when using armour than fapping.

~Mark~

Selling line breaks @ boxes.

I edited the first post to point first-time thread visitors to the Entropedia entry on Evader. That article has been updated to include the findings of the PvP thread on defense skills as well as this one (which it already contained). That makes it a good summary of what is known about defense professions and skills in EU.

I think evade and dodge tests for avatars need to be re-done since VU10 may have changed some things (and we have the wonderful hit/miss messages now)

If you want to test things again, send me a PM and I'm sure we could find some people and do some rigorous testing in game again.

Hijacker27 said:

I think evade and dodge tests for avatars need to be re-done since VU10 may have changed some things (and we have the wonderful hit/miss messages now)

If you want to test things again, send me a PM and I'm sure we could find some people and do some rigorous testing in game again.

Click to expand...

Thanks Hijacker. I haven't tried any PvP since VU10. Have you observed anything that suggests it's changed?

I was thinking the other day that it would be really easy to test with a launcher: load up 5 avatars ranging from 0 to 50+ dodger in the ring and have different avatars take turns blasting the lot, with each player counting hits/misses from his chat. Parallel processing at its best.

hit 20 lower than evade/dodge skill

Well done for the research. Reading it there is one question that comes to my mind. If hit your targer with 100% accuracy when your hit lvl is 20 higher than your target's evade/dodge lvl... will u miss 100% when your hit lvl is 20-lvls lower than your target evade/dodge.

vaseka said:

Well done for the research. Reading it there is one question that comes to my mind. If hit your targer with 100% accuracy when your hit lvl is 20 higher than your target's evade/dodge lvl... will u miss 100% when your hit lvl is 20-lvls lower than your target evade/dodge.

Click to expand...

No. Better read it again.

http://entropiawiki.com/Info.aspx?chart=Activity&id=11

% = 0.5 ( hit - dodger - 20 ) + 100

Nice old thread, but I see it with very different eyes today then when I read it the first time.

Does the cost compared to benefit justify going over lvl 25 ish pro standing in evade/dodge?

Sircus said:

Nice old thread, but I see it with very different eyes today then when I read it the first time.

Does the cost compared to benefit justify going over lvl 25 ish pro standing in evade/dodge?

Click to expand...

25 is not a big deal, 50-70 it is