Question: Punishing the Pattern Recognition of Players

I have loads more reading to do ...
You certainly do, and it is going to take a lifetime, dspecially if you focus on the "wrong" things. The PRNG is what it is, and there is nothing you can do about it. Understanding how it works internally, while interesting, will lead to nothing
Im going to like this forum... I have a lot to learn, so please proceed to teach and advise...
No thank you, I think I am done humoring.
So I am already seeing something that I recognize.. LCG, So Ma wants me to try it, fail and deposit more peds into the system, that has been the cycle from day 1....
MA doesn't want you to fail. MA wants you to succeed. MA doesn't care if you deposit. MA wants you to cycle as much ped as possible.
 
What I'm wondering about is whether the waves:
- are an unintended result of mathematical RNG
- are intentional

As in, if they are intentional, it could be meant ans an incentive measure to make the lootgame seem less boring?
 
What I'm wondering about is whether the waves:
- are an unintended result of mathematical RNG
- are intentional

As in, if they are intentional, it could be meant ans an incentive measure to make the lootgame seem less boring?
Impression I got with waves is it is intentional to avoid being classed as gambling.
 
You certainly do, and it is going to take a lifetime, dspecially if you focus on the "wrong" things. The PRNG is what it is, and there is nothing you can do about it. Understanding how it works internally, while interesting, will lead to nothing

No thank you, I think I am done humoring.

MA doesn't want you to fail. MA wants you to succeed. MA doesn't care if you deposit. MA wants you to cycle as much ped as possible.
so f(x)=x*0.96 ?
 
OK Question then, how is this not classified as gambling ?
 
redacted...... sorry for the post, more research is needed.. My sheer incompetency astounds even me.
 
Last edited:
I see quite often in threads like this that learning how it works will not get you something extra.
Imo and for me, it's not the goal, the goal is reduce the inefficiency by understanding how things work
but also for the fun in general.

I havn't done any bigger tests in mining for a very looong time, so not 100% sure how it works now (and
didn't in the old days either :D) but I have tested some stuff lately.
One is mining directly after another player, and I mean seconds after his attempt is done. My ROI was the
same as it use to be, and same resources as I use to get in that area.
The second was to mine in same small area just minutes after the previous run, with double drop at hits
as I always do. Result is on my profile pic, filling up on all three runs I did there.

So I guess keep testing and figuring things out, sometimes its rewarding. ;)
 
best field to test is in craft. easy to get a consistent data set

1. each craft has same TT value in so result is always a function opf the same input value
2. it is convenient at 5 second per loot instance and can accept long runs

i will start to collect Enhancers returns to create an histogram, and post a 2000 click data
parameter are qr100 , maxed BP and quantity slider
 
I figured out my own way of dealing with the game's diminishing return failsafe... Just stop cycling ped....
 
I think that this game employs a system that punishes pattern recognition..

Read my thought process below this..

The game switches from linear to exponential progression when the player makes too much peds.. And then scales again when the player "cracks" that sequence or scales down when the player is not obtaining the needed rewards..

The game literally builds an equation using the actions of the players. x^(m)+x^(m-1)+c eg x coord^2+y coord^1+height = 0
The game builds a trend line to match the players actions..


Below is the old post... Sorry for the rambling..

I perform an action that the game has not seen before and it gives me a reward fo it. (C)
I perform the same action and the game starts to recognise the pattern.. (X)+(C)
Third time, the pattern is locked and game knows what the pattern is and scales the pattern up.. (2X) (C)
Now the player gets shafted ... Fourth time... (X^2)+(X)+(C)

I think its using simple math mainly the quadratic equation to break patterns... Even if you do random stuff it, will try and find a pattern that you as the player are not aware of, and break it..

I saw the reference to the Sack's spiral right from the beginning of the game, even the global animation is a variation of the Sacks spiral math. There is also a reference to another profession in mining (click click). The game wants players to find these patterns, in order to wait for them to be abused and then scale the system up an exponent...

If you pickup that pattern change, you can abuse it for a while, and the game scales the pattern up again...

I am sorry If I am not explaining this properly.

The game has a system in-place that breaks either intentional or unintentional patterns.

Let me plot it quick

Move 0 starts at 0,0 (0,0)
Move 1 50 (50,0)
Next move can be 100, or 2x+0 =110 (110,0)

The game can now either reward that sequence or expand the sequence by adding an exponent x^2+x+0 50^2+50+0 or (2550,0), both sequences are valid, the one carries on using the current base, the other adds an exponent...

Think of it as a freeway entrance or difficulty scaling, or even a level up in an MMO. The game gets harder.. OH It's dynamic difficulty scaling .....

You have made too much ped today, how about we challenge you, you can earn more ped, but when you loose, we will scale back... so that you will always be within x loot...

I quite like reading about your maths formulas. Was it you about 8+ years ago provided the formula for pet levels. I lost that formula on an old pc, but it worked perfectly.

Rick
 
I quite like reading about your maths formulas. Was it you about 8+ years ago provided the formula for pet levels. I lost that formula on an old pc, but it worked perfectly.

Rick
Nope, I am a miner...
 
OK Question then, how is this not classified as gambling ?
Because it's a game of skill so they can avoid the extra tax etc. They still have to prove it's not entirely chance based which could explain why Sweden citizens can't buy strongboxs?.
 
Last edited:
I figured out my own way of dealing with the game's diminishing return failsafe... Just stop cycling ped....

This is an interesting conclusion.

Does your perceived pattern reset after some time not cycling?
 
You are getting old when you realise that nobody remembers R4tt3xx
 
You are getting old when you realise that nobody remembers R4tt3xx
I remember too, and back then I remember his theories at least gave the impression of something that could possibly, perhaps, maybe bear fruit. And that thorough tests had been run. This thread however, is a complete implosion of that credibility lol
 
I see quite often in threads like this that learning how it works will not get you something extra.
Imo and for me, it's not the goal, the goal is reduce the inefficiency by understanding how things work
but also for the fun in general.

Don't get me wrong. I am not saying "Don't test things" or "Don't try to understand how things work under the hood". This is what I spend most of my time doing in this game as that is what is enjoyable for me.

What I am saying is, there is no point in reading research papers or college presentations on state of the art PRNG methods. It is a complete waste of time if you are trying to learn anything about how the game works.

We have no way of seeing the output from the PRNG and as such we can't learn anything about it. We can also assume that it is not composed of state of the art methods, this is a 20 year old game. Sure they could change things, but I don't see why they would have unless there was an issue somewhere.
For the purposes of learning about the game, it is probably best to assume that the PRNG does what it should; output seemingly random numbers

What is interesting is gathering the data we can actually see, and try to learn what we can about it.

You are getting old when you realise that nobody remembers R4tt3xx

A lot of us still do remember, I think. I think that most stay away from the discussions, because they think it is a waste of time. I just enjoy entertaining these discussions as they tend to teach me new concepts. Learning is fun, even if it has no relevancy in the game
 
Don't get me wrong. I am not saying "Don't test things" or "Don't try to understand how things work under the hood". This is what I spend most of my time doing in this game as that is what is enjoyable for me.

What I am saying is, there is no point in reading research papers or college presentations on state of the art PRNG methods. It is a complete waste of time if you are trying to learn anything about how the game works.

We have no way of seeing the output from the PRNG and as such we can't learn anything about it. We can also assume that it is not composed of state of the art methods, this is a 20 year old game. Sure they could change things, but I don't see why they would have unless there was an issue somewhere.
For the purposes of learning about the game, it is probably best to assume that the PRNG does what it should; output seemingly random numbers

What is interesting is gathering the data we can actually see, and try to learn what we can about it.



A lot of us still do remember, I think. I think that most stay away from the discussions, because they think it is a waste of time. I just enjoy entertaining these discussions as they tend to teach me new concepts. Learning is fun, even if it has no relevancy in the game
If they haven't changed much then the most important thing in mining is (still) the bell curve. ;) Possible the 2D grids too then. :p

One reason to test stuff is to see if some theroies are correct or not. Other times just to try them out.
 
Don't get me wrong. I am not saying "Don't test things" or "Don't try to understand how things work under the hood". This is what I spend most of my time doing in this game as that is what is enjoyable for me.

What I am saying is, there is no point in reading research papers or college presentations on state of the art PRNG methods. It is a complete waste of time if you are trying to learn anything about how the game works.

We have no way of seeing the output from the PRNG and as such we can't learn anything about it. We can also assume that it is not composed of state of the art methods, this is a 20 year old game. Sure they could change things, but I don't see why they would have unless there was an issue somewhere.
For the purposes of learning about the game, it is probably best to assume that the PRNG does what it should; output seemingly random numbers

What is interesting is gathering the data we can actually see, and try to learn what we can about it.



A lot of us still do remember, I think. I think that most stay away from the discussions, because they think it is a waste of time. I just enjoy entertaining these discussions as they tend to teach me new concepts. Learning is fun, even if it has no relevancy in the game
Ditto to all of this. Even as a researcher whose job is to tease apart something meaningful out of block boxes IRL I can say there are a lot of red herrings out there I see people chasing in part because they aren't familiar with statistics or how to do formal analyses, but I usually try not to get too involved (especially comments like "What if Pi changes?").

The key thing if anyone really wants to nail something down is to collect data that isn't confounded (logs are unstructured not really usable). Unless someone can show actual data, it's not really testing, and that's a pretty good metric to weed out pet theories or people chasing shiny new terms from much more solid things.

A lot of the fun for me with mining isn't just the actual mining, but a lot of the testing with formal scientific experimental design I've been able to do. Most of that isn't public for good reason, but it's also fun to provide some of the more minor things to the community.
 
Ditto to all of this. Even as a researcher whose job is to tease apart something meaningful out of block boxes IRL I can say there are a lot of red herrings out there I see people chasing in part because they aren't familiar with statistics or how to do formal analyses, but I usually try not to get too involved (especially comments like "What if Pi changes?").

The key thing if anyone really wants to nail something down is to collect data that isn't confounded (logs are unstructured not really usable). Unless someone can show actual data, it's not really testing, and that's a pretty good metric to weed out pet theories or people chasing shiny new terms from much more solid things.

A lot of the fun for me with mining isn't just the actual mining, but a lot of the testing with formal scientific experimental design I've been able to do. Most of that isn't public for good reason, but it's also fun to provide some of the more minor things to the community.
Yes thats a good approach to keep certain things privat and not public, at least for those who are out for profit.
Could use the rule what works in general could be shared, but what works in a specific way could be kept privat.

I like doing small tests though, quite often to see if it is possible to figure out mechanics and so and not so much about
return, even though that happens too.
The downside is that statistics quite often just show the average, not the specific. Just look at DPP, the number used
for e.g a specific handgun only occurs once in the entire damage cycle, so DPP is for sure not constant per interaction.
I did a very very small test just some minutes ago in PvP area, took 8,5ped probes with me. Got a HoF of 296. :D I guess
it could be called luck by some. :p
 
The downside is that statistics quite often just show the average,

Average is all one really needs. I don't even track my loot anymore. I know what I can expect to lose, what I can expect in returns, and how many units of certain items I can expect from certain mobs on an hourly basis. With what I'm currently primarily working with.

When I'm down bad I don't really feel any ill out of it, when I hof it was only a matter of time.
 
Average is all one really needs. I don't even track my loot anymore. I know what I can expect to lose, what I can expect in returns, and how many units of certain items I can expect from certain mobs on an hourly basis. With what I'm currently primarily working with.

When I'm down bad I don't really feel any ill out of it, when I hof it was only a matter of time.
Average is all some need, fixed that for you. ;)
It all depends on what you are out for. I'm not out to do a profit, I'm here to have fun and killing time.
I stoped doing depos in june, and also done some withdrawals in the autumn so I have to play in another
way now then what I used to do, so small tests should happen more often now due to that.
There are certain parts in my testing I want to see way more precise outcome per interaction than the
average.
 
This is an interesting conclusion.

Does your perceived pattern reset after some time not cycling?
No, it seems that suprise suprise, the pattern is on a timer. At x time during the day on x server at x location the pattern is y... This makes the game frustrating and a joy to play... When you see it, or when you think you see it.It's all just so irrational..
 
A lot of the fun for me with mining isn't just the actual mining, but a lot of the testing with formal scientific experimental design I've been able to do. Most of that isn't public for good reason, but it's also fun to provide some of the more minor things to the community.
Now this I find interesting..

How do I phrase this without sounding like an @$$.. If the miner that you are getting your data from does not know how to mine a set area and does not adjust to seen changes on a set server, it will appear to be random.

I think it all comes down to knowledge and technique. Nail those things down, then see what the statistics say.

King, do you mind if I pm you sometime ?
 
I love ChatGPT, it knows a lot about 747 jumbos...
 
Don't get me wrong. I am not saying "Don't test things" or "Don't try to understand how things work under the hood". This is what I spend most of my time doing in this game as that is what is enjoyable for me.

What I am saying is, there is no point in reading research papers or college presentations on state of the art PRNG methods. It is a complete waste of time if you are trying to learn anything about how the game works.

We have no way of seeing the output from the PRNG and as such we can't learn anything about it. We can also assume that it is not composed of state of the art methods, this is a 20 year old game. Sure they could change things, but I don't see why they would have unless there was an issue somewhere.
For the purposes of learning about the game, it is probably best to assume that the PRNG does what it should; output seemingly random numbers

What is interesting is gathering the data we can actually see, and try to learn what we can about it.



A lot of us still do remember, I think. I think that most stay away from the discussions, because they think it is a waste of time. I just enjoy entertaining these discussions as they tend to teach me new concepts. Learning is fun, even if it has no relevancy in the game
Agree. I think the loot calculation mechanism at its core is the same since its inception. And it had to be proven to not be gambling to be permitted. So skills and efficiency play a part in that. Today even more skills play a part in that. When you hear the advice of experienced players talk about maximising loot opportunities they are correct because that is how you average out the loot calculations. If you take aside the variables that affect loot today at the core is a system that you have the same chance of success whether its your first mob on log in, last mob before log out or any of the many in between. There is not building of loot or multiplier in a personal balance. I asked that question and answer was you have same chance of success at each loot opportunity.
 
I love ChatGPT, it knows a lot about 747 jumbos...
Your logic is flawed. ChatGPT knows nothing. It must scrape source data and index it to build accurate answers. It might have a cache of common responses. But it knows nothing. This is where your approach to your theories get unstuck. You are humanising ChatGPT which is just code.
 
EU is very similar to Buddhism... we know the Effect, we look for the Cause.

@rattexx: there is noreal need to understand if time series are made by an AI, a randomizer or the priestesses of Lootius
look at the series, that esplains the logic of multipliers. that is the last part of the equation... in loot 2.0 it is PED IN--- PED out so the SEED of the multiplier is the total cost of kill and that number is not the same "total decay soffered by player" but is something LESS that that. (armor is ALMOST participating say devs)
The complexity is just because we can not deriva a linear formula from the data series (to me that means that there is a very nice randomizer).
some events in shared loot are different, ther is a random delivery of the "weapon" or "matriarch blood" to one player when sharing the loot, so one takes more than he injected compared to others.

the item i clicked most is damage enhancers, about 300.,000 click events, and i use a 95% return excel model and the results of craft, despite not costant, converge to the "success %" of the craft machine.
from that observation i "trust in teh algorithm" and assume that MA tells the truth on average return.

From that point of view, we can all say "they lie, i am cursed" and more, but a company can not lie to the gambling commission so EU is a skill based game. POINT.

From these assumptions... thre comes the same advice that ALL players give: manage bankroll, rise looter, rise efficiency, rise dpp.

if you find a black magic formula for this, dont publish it anyway, use it to manufacture 10m ped then post the pic of 1m usd withdrawal :)
 
EU is very similar to Buddhism... we know the Effect, we look for the Cause.

@rattexx: there is noreal need to understand if time series are made by an AI, a randomizer or the priestesses of Lootius
look at the series, that esplains the logic of multipliers. that is the last part of the equation... in loot 2.0 it is PED IN--- PED out so the SEED of the multiplier is the total cost of kill and that number is not the same "total decay soffered by player" but is something LESS that that. (armor is ALMOST participating say devs)
The complexity is just because we can not deriva a linear formula from the data series (to me that means that there is a very nice randomizer).
some events in shared loot are different, ther is a random delivery of the "weapon" or "matriarch blood" to one player when sharing the loot, so one takes more than he injected compared to others.

the item i clicked most is damage enhancers, about 300.,000 click events, and i use a 95% return excel model and the results of craft, despite not costant, converge to the "success %" of the craft machine.
from that observation i "trust in teh algorithm" and assume that MA tells the truth on average return.

From that point of view, we can all say "they lie, i am cursed" and more, but a company can not lie to the gambling commission so EU is a skill based game. POINT.

From these assumptions... thre comes the same advice that ALL players give: manage bankroll, rise looter, rise efficiency, rise dpp.

if you find a black magic formula for this, dont publish it anyway, use it to manufacture 10m ped then post the pic of 1m usd withdrawal :)
The problem with the black box, is that it is filled with black boxes, that are filled with black boxes...

Heres something else that makes my brain hurt..

https://mathworld.wolfram.com/VoronoiDiagram.html ... Not going there...

Pity I am not a hunter, so I do not understand what you are saying for the most part..

For the most part, the algorithm as you call it, is nothing more than a simple math domain that exists around the avatar and is brought into existence when a tool is equipped. Each trigger "location" on the matrix will return good loot when acted on at the right time.

Skills now here is where it gets tricky, modify the layout of the matrix, increasing its size, activating a single coordinate pair, etc. It also may be possible to use the value of skill itself as the matrix and build that around an avatar, essentially it's labeling the points that the finder can scan, assigning each one a label or number and seeing if there is a pattern there.

Also when using this I can see areas that should be "hits" turn into skill gains instead.
 
Last edited:
i think "professional skills" have zero impact after maximizing a tool (a finder need max depth to be used or loot is diminished and converted in faster progress)
i dont think place is really relevant if not for the "loot table" of the drop. in some areas you get different materials.

for mining i just regret that loot events are few because of extraction time so i seldom drop more than 200 bombs and the data set is very scarce
 
I am also looking into set theory.. Thanks guys, your fault..

I think that the game is very irrational when it comes to sets, and I have setup a probability matrix using coordinates as the criteria. The sheet tells me by how much % a certain number is a factor of another number, I even have averages set up and can set a custom number set.

The idea is a bit of a hybrid of resources in the ground and dynamic resources. I plan to have the new matrix select the coord pairs to examine their irrationality.

I understand that most of you are not miners and therefore may not 100% understand the relevance.
 
Back
Top