Kha wrote:Same progression as 1 year ago, they have all data needed.
But it seems hard to code unfortunately.

Well it can't be that hard to code if mincers have the right formula (apart from named being hardcoded to resist charms etc).

Mincers and ments used the same formula on live bar the fact that mincers can spam it as it's an insta.

Genedril wrote:

Kha wrote:Same progression as 1 year ago, they have all data needed.
But it seems hard to code unfortunately.

Well it can't be that hard to code if mincers have the right formula (apart from named being hardcoded to resist charms etc).

Mincers and ments used the same formula on live bar the fact that mincers can spam it as it's an insta.

As you can see :

Blue wrote:For some reason we use a different formula for Mentalists. Genjiro is on it to fix charm behavior. Getting test results and evaluating them takes time.

They choosed to ignore provided data, and do different formula for both classes, (for some reason). The result is very effective :
Nothing. Both class are still bugged, and will probably always be.
Nice to see testers giving week of datas for nothing (for some reason #quote blue)

Wasn't supposed to show that, but i don't even care anymore.

Here is the result of weeks of tests :
https://docs.google.com/spreadsheets/d/18zMii8sq8-DoQy97PB27jfwoDz94uP2Sopkb38Iwf1g/edit#gid=971951514

You shouldn't be able to modify the document. Some videos to proof those were shared to them aswell + txt combat logs.

They just ignored that. And we are still (for both classes) with a bugged formula.

Kha wrote:They choosed to ignore provided data, and do different formula for both classes, (for some reason). The result is very effective :
Nothing. Both class are still bugged, and will probably always be.
Nice to see testers giving week of datas for nothing (for some reason #quote blue)

Not saying it's not bugged; Just that it's not beyond them to code given that mincers appear to be correct.

Whether the have the inclination or time is a different matter ofc .

Ohh, and they hard coded the charming of named to 'no' as well didn't they which isn't like 1.65.

{edit} props to those providing test data btw. I can't bring myself to login to live (for at least 6 years).

Kha wrote:Wasn't supposed to show that, but i don't even care anymore.

Here is the result of weeks of tests :
https://docs.google.com/spreadsheets/d/18zMii8sq8-DoQy97PB27jfwoDz94uP2Sopkb38Iwf1g/edit#gid=971951514

You shouldn't be able to modify the document. Some videos to proof those were shared to them aswell + txt combat logs.

They just ignored that. And we are still (for both classes) with a bugged formula.

We are not ignoring it, I am simply still not yet able to reverse engineer the correct formula matching the logged values.
The issue is still open -> https://www2.uthgard.net/tracker/issue/2489/

As I have stated multiple times, we cannot simply use a switch-case with those values and code something ugly like:

Code: Select all

if (Minstrel.Level = 50 && Mob.Level = 51 && Skill.Level = 50)
{
  resist = 15.40%
}
else if (Minstrel.Level = 50 && Mob.Level = 52 && Skill.Level = 50)
{
  resist = 28.15%
}
...

A mathematical formula is needed that utilizes all known (and maybe still unknown) factors and variables
which then results in as close as possible float values (= resist chance) we have seen during the tests.

No formula has been found yet that could match multiple test values.
This could mean:

a) we are not able to find the correct formula (doh!)
b) wrong or not enough test data
c) missing variables like hidden modifiers or stats that influence charm resist chance

Since you shared the data, anyone feel free to find me the correct formula if you can and I will impliment this asap.

Kha wrote:https://docs.google.com/spreadsheets/d/18zMii8sq8-DoQy97PB27jfwoDz94uP2Sopkb38Iwf1g/edit#gid=971951514

Sharing is caring, these are your tests, why would you not share it if you feel like it?
these are pendragon tests i guess?
nice and intensive testing. But i think it lacks detail in the part you are actually claiming is bugged. lvl50 instrus+skill+rr.
99% is written in some parts of the table without any testing (0/0charms)
The rr5+ one is missing the most important value, the one where resists hit the 99% mark like they do for other tests.
higher than rr5 is speculation in your sheet, or also missing the 99% mark.
I think you should test that, since that is very important.

@Genjiro
if finding a formula is too difficult or not possible, you can also just use the tabled data. the data is discontinues, which makes finding a nice formula very tricky.

read a table in, format it as an array that you can use it in the code once, and then just point to the correct index of the array every time you need the resistchance as a floating number, dependent on minst level, moblevel, skill.
if there are hidden modifiers, this would lead to an incorrect result. but the current one is even more wrong id say and it is not sure if there is a hidden modifier at all, and if so if it will ever be found.

pweet wrote:@Genjiro
if finding a formula is too difficult or not possible, you can also just use the tabled data.
read a table in, format it in a way you can use it in the code once, and then just point to the correct index of the table to get the needed resist rate dependent on minst level, moblevel, skill.
if there are hidden modifiers, this would lead to an incorrect result. but the current one is even more wrong id say and it is not sure if there is a hidden modifier at all, and if so if it will ever be found.

That won't do. A formula that calculates correct (based on live test data) resist chances is required to disprove the current formula and therefore justifies a change.

Why do you need a 2nd formula for that?
Insert the values of the current uthgard charm formula and see if it matches the live results. I guess it doesnt and hence we are discussing.
My suggestion of using the tabled data circumvents the need of a formula. You just use the correct values of the live tests.

That would require a test of every combination though, that would be testers work i guess
It is just a work around if the formula cant be found, which i agree would be a nicer approach.

pweet wrote:Why do you need a 2nd formula for that?
Insert the values of the current uthgard charm formula and see if it matches the live results. I guess it doesnt and hence we are discussing.
My suggestion of using the tabled data circumvents the need of a formula. You just use the correct values of the live tests.

A hardcoded/db table with fixed/parsed values to pick from is not the correct way to solve this.

Even IF such table would be acceptable it would require values for every "PlayerLevel-MobLevel-SkillLevel" combination there is.
There are 45 player levels, 70 skill levels and up to 90 mob levels the table would need to have values for each combination.

Therefore a formula is required.

I agree, see my last edit

Also, live sadly limited charming to a max. mob level of 59 (red con in general) which for Classic/SI would require to guess
resist chance values for level 59+ for all PlayerLevel-SkillLevel combinations.

A formula could be scaled or would scale itself when entering 59+ values.

Hmm didnt know about the 59 cap introduced on live and hence the missing data. Sorry for stepping in here with limited knowledge
Just had a look at the data and wanted to help.
I will leave it to the charm experts again

pweet wrote:

Kha wrote:https://docs.google.com/spreadsheets/d/18zMii8sq8-DoQy97PB27jfwoDz94uP2Sopkb38Iwf1g/edit#gid=971951514

Sharing is caring, these are your tests, why would you not share it if you feel like it?
these are pendragon tests i guess?
nice and intensive testing. But i think it lacks detail in the part you are actually claiming is bugged. lvl50 instrus+skill+rr.
99% is written in some parts of the table without any testing (0/0charms)
The rr5+ one is missing the most important value, the one where resists hit the 99% mark like they do for other tests.
higher than rr5 is speculation in your sheet, or also missing the 99% mark.
I think you should test that, since that is very important.

I didn't shared the tests because they kinda asked me not to. But now .. who cares.

Well, it was agreed that 99% is the max resist. (while uth use 100% resist at many points).
See yourself, make a level 6 minstrel, and go charm a level 59 ellyl (humanoid). It will work on live, with 99% resist rate average and it's always been like this. (On uth a level 6 minstrel can't charm more than level 10, if 11 it will be 100% unbreakable breakpoint.)
So, when a RR hit 99% on a level 50 mob let's say ... the lower RR doesn't need to be test for this mob level : it will be 99% aswell.
Same for 5% resist (which on a reasonable scale seems to works the same, if a 30+5 instrument get 5% resist for a level 40mob (just an exemple) no need to test level 40 mob for 30+6 instrument. It will be around 5% aswell. Unless you go for a wild 100.000 pulse test on each mob. But 1.5 to 3K took me weeks to do already.

That being said, you should undertand the datas a bit better.
Look for example : (level 7 minstrel with 6+0 instrument charm level 54 mob on one of my test : https://www.youtube.com/watch?v=3LrHuy_u37Q ) But i can provide level 6 minstrel vs 59 mob (or more if live wasn't capped at 59 the past 5 years). No level cap before, and no 100% resist ever. It now says : you can't charm this mob if lvl 60+).

It's a freaking solid fact, they just had to change 100% to 99% max in formula. Easy no ?
While looking for the perfect scaled one, it was a step forward(same resist rate for both classes, that's why mentis where able to charm lvl 70+ before they get raped by their own pet). This only difference would have made the 1.65 fun charms possible. Level 10 minstrel farming a few orange before getting insta killed by pet. Level 50 minstrel trying insane charm on RVR, to sometimes miserably fail or sometimes make awesome moves. The real 1.65

Here now on uth, if you are 50+15 you can charm level 60 with 90ish% resist. If you are 50+14 you'll face a solid unbreakable 100% resist on a 60+ mob, no matter what. Level 50+18 however, can charm 61 pets Drastic as example.

Let's get back to mentalists now, the main topic : they seems to get an even more unfair formula. Unable to charm a yellow/orange sometimes ... Look at what they can do on live with only 6+0 in the spec line....
I was ****** for my class, but am really mad for them ... Some datas provided would have helped making their life easier for the past year or so... Even if not the perfect one yet.

But different formula for reasons ... Cheer mentis !
I can understand they want the perfect formula they'll probably never have for a final implant. But with everything they provided (and i helped somehow) it was easy to find a fair modification to avoid nerf, especially for ments, for the past .1or even 2 years ?

Now you know every detail Pweet, and probably better understand why mentis and minstrels are pi**ed for the past year, espcially since 99% max resist on pulsed charm data was already provided during BETA... 2 years ?

Well if your main complain is that there is a 100% cap instead of a 99% cap i think the formula works pretty well.
With 99% this only triggers even more macro minstrels.
Maybe not live like then with 100%, but it has no influence for honest players in RvR, and makes the live for staff easier to prevent macro abuse.
As a trade of, to me this sounds like a better deal than having the possibility to charm a lvl59 mob as a lvl6 minstrel for 1sec just for fun.

pweet wrote:Well if your main complain is that there is a 100% cap instead of a 99% cap i think the formula works pretty well.
With 99% this only triggers even more macro minstrels.
Maybe not live like then with 100%, but it has no influence for honest players in RvR, and makes the live for staff easier to prevent macro abuse.
As a trade of, to me this sounds like a better deal than having the possibility to charm a lvl59 mob as a lvl6 minstrel for 1sec just for fun.

You maybe are right with this approach, however with this way of thinking : better close the server to avoid cheaters then, should i remind you how many players were publicly banned already and the 10 times more banned behind the scene before it started to be public ? Cheaters exist and will always exist. 1.65 charm REAL behavior can't be a cause, and i assume will never be since it will probably never happen on uth, perfect formula is out of the table.

99% max resist is an established fact which can bring fun, on a game (if we forget the competitive RvR scene we are familiar with, fun is the purpose of this game, and any game...to begin with.) I explained it as an example of an easy proofed thing to modify from the start for them, aswell as some easy modification to mentalist charms to avoid lot of disappointment for an entire class, they are f***** a lot more than minstrels i'll agree.

It's been a long fight, for charm class to enjoy the 1.65 fun charms again, i believed in it a-lot, and tested what i could test.
But i'll face it. It's surely a lost fight. However i hope miracle can happen for them even if i don't play anymore myself.

Edit : thanks to Genjiro, i know for a fact he tried at some point, i didn't came with those test idea on my own, who knows he may try again...