Developer-Notes--3

falkao

Stalker
Joined
Oct 25, 2005
Posts
1,993
Location
Italy
Society
Freelancer
Avatar Name
Marc falkao Falk
It seems as the discussion goes in the direction of gambling vs. non gambling. The problem is that this discussion is useless as the term gambling is not defined. Sure, there are some laws dealing with gambling but their purpose is to regulate something and this has nothing to do with what is discussed over here.

Furthermore, there do exist some very generic definitions of gambling like gambling is when the outcome of an activity id based on chance. However, this would also hold true for some sports like darts as a player can't hit the bull in every throw, or sailing where you can't control the wind.

Wouldn't it be better to simply accept that there is no person loot pool and that the return rate (not the loot) depends on skills and how eco you play?

Nonetheless, what should be discussed is if loot distribution should be changed or not. As already mentioned, it has a very high variability and therefore it will lead to very different results per ava. Some might accept it but some will start complaining about it and the latter ones are the most heard, at least over here.
 

Fifth

Elite
Joined
Jan 26, 2012
Posts
2,721
Avatar Name
The "Fifth" Entity
Nonetheless, what should be discussed is if loot distribution should be changed or not. As already mentioned, it has a very high variability and therefore it will lead to very different results per ava.
Different mob classes have different variability rates. That's where it is "very different".
I believe you are trying to solve a problem that has been already solved.

What to do with the players who can't figure it out?
Idk, give them a Field Handbook of Loot maybe?

Well, u see, it's not really possible because:
  1. It kinda defeats the whole idea of the game - you are supposed to go and figure it out on your own.
  2. Besides, trying to do it any other way won't really work - boring stuff, ppl will simply dismiss it:TL;DR (and go whine on the forums instead :D)


No need to fix the system, it's not broken. What is broken is the size of the playerbase > hence too high (average) cost to play > hence small playerbase... vicious circle.
 

Jimmy B

Indioni Jones
Joined
Dec 2, 2005
Posts
12,256
Location
UK
Society
Skillin' Villains
Avatar Name
Oni
Different mob classes have different variability rates.

Do you have data to back this up?

Sure, if you spend 1000 Ped per hunt killing small mobs you'll have smoother returns per hunt than if you spend 1000 Ped per hunt killing big mobs. But I've not seen any hard evidence that 1000 kills on one mob can be less volatile than 1000 kills on another mob.
 

bettlejus

Stalker
Joined
Apr 20, 2011
Posts
1,674
Location
Top Hunters
Society
ROUA
Avatar Name
Joan Xperia Dárc
Do you have data to back this up?

Sure, if you spend 1000 Ped per hunt killing small mobs you'll have smoother returns per hunt than if you spend 1000 Ped per hunt killing big mobs. But I've not seen any hard evidence that 1000 kills on one mob can be less volatile than 1000 kills on another mob.

Well, Jimmy, take this


1000 ped on longtooth - you kill about 50 LTs and can have -500 ped return
1000 ped on Foul - you kill about 3000 Fouls, you cannot posibly have -500 ped after this

So I think that is correct
 

Jimmy B

Indioni Jones
Joined
Dec 2, 2005
Posts
12,256
Location
UK
Society
Skillin' Villains
Avatar Name
Oni
1000 ped on longtooth - you kill about 50 LTs and can have -500 ped return
1000 ped on Foul - you kill about 3000 Fouls, you cannot posibly have -500 ped after this

Yes, that's what I said.

But there need not be any difference in loot volatility between Fouls and Longtooth for that to be the case, the difference you speak of there is down to how long it takes (or more to the point how much it costs) to kill each mob. I took, maybe mistakenly, Fifth's comment to be suggesting that some mobs have different relative loot variance than others over the same number of kills. Which might be true, or might not, I've not seen any study of it.
 

wizzszz

Banned
Joined
Dec 31, 2006
Posts
5,202
Location
Germany
Society
Jurai Blood
Avatar Name
Nicholas wizzszz Wolf
It seems as the discussion goes in the direction of gambling vs. non gambling. The problem is that this discussion is useless as the term gambling is not defined. Sure, there are some laws dealing with gambling but their purpose is to regulate something and this has nothing to do with what is discussed over here.

There is no discussion "gambling vs. non-gambling".

I only use it to disprove all the silly theories people come up with - after all it is the only real fact we have.

Furthermore, there do exist some very generic definitions of gambling like gambling is when the outcome of an activity id based on chance. However, this would also hold true for some sports like darts as a player can't hit the bull in every throw, or sailing where you can't control the wind.

Those are not "generic definitions", but actual law texts, and they are pretty much the same in every country that doesn't permit gambling activities on the free market.

I have no idea why everyone is trying to marginalize this all the time.

Dart and sailing are not gambling, for a simple reason:
You can train it and become better.

That's the definition of "skill based games" and exactly the opposite of gambling.


Wouldn't it be better to simply accept that there is no person loot pool and that the return rate (not the loot) depends on skills and how eco you play?

You think this is a matter of "unwilling to accept something"?

You couldn't be more wrong, and actually i am surprised to see such an argument from you - you love numbers and statistics so much and then you resort to wild guesses...


The loot system MUST have a memory, otherwise it could not produce the results we see - and i will for sure not rate some silly "dev notes" from MAs inept staff higher than my own perception & experience, no f'ing way.

I have absolutely no idea why everyone else appears to have no trust in his/her own observations, yes even misinterprete their own data to make them match what some dude from MA said when he was bored.


Nonetheless, what should be discussed is if loot distribution should be changed or not.

Nothing to discuss there - it should be changed, it is one of the main reasons why people quit this game galore.



And btw - where are dev notes #4? Or did they get scared?
 

Jimmy B

Indioni Jones
Joined
Dec 2, 2005
Posts
12,256
Location
UK
Society
Skillin' Villains
Avatar Name
Oni
I only use it to disprove all the silly theories people come up with - after all it is the only real fact we have.

The only fact we have on this topic is that the Swedish authorities ruled in 2008 (or whenever it last was) that at that time EU did not require a gambling license to operate. In many cases you are using assumptions (not facts) derived from that fact to attempt your disprovings. That's one of the reasons why we keep ending up getting drawn off-topic into gambling discussions in these threads.
 

tamlin

Prowler
Joined
Jun 4, 2010
Posts
1,151
Location
Somewhere in EU
Avatar Name
tamlin
[personal loot pool]
The loot system MUST have a memory, otherwise it could not produce the results we see
While I think this statement is easily falsified if taken as-is, I'm willing to entertain alternative ideas. Could you outline the points leading you to this conclusion?

If intended to suggest that the loot system has a per-avatar memory it contradicts my own personal experience with the system, but maybe we're talking about two (slightly?) different things?
 

wizzszz

Banned
Joined
Dec 31, 2006
Posts
5,202
Location
Germany
Society
Jurai Blood
Avatar Name
Nicholas wizzszz Wolf
The only fact we have on this topic is that the Swedish authorities ruled in 2008 (or whenever it last was) that at that time EU did not require a gambling license to operate. In many cases you are using assumptions (not facts) derived from that fact to attempt your disprovings. That's one of the reasons why we keep ending up getting drawn off-topic into gambling discussions in these threads.

Assumptions?

So it is not a logical (and perfectly valid) deduction that EU is not gambling because the authorities did rate it "not gambling"?

You and some others here argue that it could be gambling nonetheless - THAT is an assumption.


While I think this statement is easily falsified if taken as-is, I'm willing to entertain alternative ideas. Could you outline the points leading you to this conclusion?

Easily falsified? If it's so easy, please go ahead and falsify it.

If intended to suggest that the loot system has a per-avatar memory it contradicts my own personal experience with the system, but maybe we're talking about two (slightly?) different things?

I just want to get rid of the term "personal loot pool", because people tend to read only buzz words and react accordingly.

MA themselves have underpinned that with terms as "your luck will turn sooner or later" - or do you think this is a lie as well?




There is no point in argueing anyway, when people question EVERYTHING just to "win" a debate.
If you question things like gravity or the intelligence of the swedish authorities the debate is over.
Immediately.

Call it "wizzszzs law", in the style of "Godwins law".
 

tamlin

Prowler
Joined
Jun 4, 2010
Posts
1,151
Location
Somewhere in EU
Avatar Name
tamlin
Wouldn't it be better to simply accept that there is no person loot pool and that the return rate (not the loot) depends on skills and how eco you play?
While the former seems to be a point of contention, why I won't touch that subject, the latter is simply false when viewing the system as a whole.

Skills have absolutely no meaning whatsoever in EU when it comes to tt returns - they are only a measurement of how much money you have spent and allows you to play a higher stake game. The absolute proof for this fact is the gateway HoF. It could be further backed up by f.ex. very noticeable mining hits (ATH's) by very unskilled avatars, but that's really not required. That is, unless one thinks "less skilled gets more tt return". :)

For hunting the game changes a bit, but that doesn't falsify the verifiable facts.

Sure, I also know that outside the promotional loots the game changes quite a bit, but for that part to matter the least, must we not also accept that we the Very Large Multipliers(tm) / promotional loots are decoupled from "the system" when discussing it, even that it de facto is an integral part?
Are we prepared to accept this?

Iff this can be accepted, that the VLM's/PL's are really "rogue" elements inside the system, only then can I accept that Skills matters for tt returns - but still only for some rather limited and specific instances.
 

FallenAngel

Stalker
Joined
Dec 9, 2006
Posts
1,968
Society
Free!
Avatar Name
Fallen Yours Angel
While the former seems to be a point of contention, why I won't touch that subject, the latter is simply false when viewing the system as a whole.

Skills have absolutely no meaning whatsoever in EU when it comes to tt returns - they are only a measurement of how much money you have spent and allows you to play a higher stake game.

Falkao said returns (not loot), skills effect your cost to do certain activities, if i hunt spiders my TT returns will be similar to Star's only my costs will be higher so my net returns worse.
 

tamlin

Prowler
Joined
Jun 4, 2010
Posts
1,151
Location
Somewhere in EU
Avatar Name
tamlin
So it is not a logical (and perfectly valid) deduction that EU is not gambling because the authorities did rate it "not gambling"?
Correct. Many are of the opinion that Lotteriinspektionen did an incomplete and mislead "investigation" into EU, and therefore reached an incorrect conclusion. What that led to was a "false positive" (or false negative, depending on POV).

If you watch the sun for 10 minutes at noon, do you claim to have enough data to assert that "On planet earth, it is always day. Everywhere."? Of course not, but that's the kind of false logic many believe is behind the claim that EU isn't gambling.

Easily falsified? If it's so easy, please go ahead and falsify it.
Oh, we're playing that game now - not answering questions flipping the issues around to make others do the work. And here I thought we could have a civil discussion. Silly me.

As you sidestep the important part, I see no reason to spend time on providing such facts. Should you reevaluate your decision to provide the requested information I will reconsider providing falsification.

I just want to get rid of the term "personal loot pool", because people tend to read only buzz words and react accordingly.
Oh! I was under the impression you argued there is such a thing. As I never used that term myself, consider it done! :)

MA themselves have underpinned that with terms as "your luck will turn sooner or later" - or do you think this is a lie as well?
Have they? I have seen this repeated by support drones, but MA themselves have on multiple occasions clarified that EUSO had wrong information and had since been updated, for a wide range of subjects.

So the question becomes: Do we consider EUSO to be MA? If we do, then MA has, by its own admission, on many, many occasions been telling its customers lies.

But even if that wasn't the case; would I think MA lies to its customers? If they think even remotely they have anything to gain from it, absolutely!

If you question ... the intelligence of the swedish authorities the debate is over.
Oh. So me living in Sweden, having first hand experience with the incompetence (and stupidity) of authorities that I can attest to, makes you close the debate? Interesting way to end what you call a debate - "No fair, you know how this works! I'm taking my toys and leave now!". OK. Whatever you think it was about winning or losing, you have now lost by default.
Congratulations to the winners, whoever you are.
 

wizzszz

Banned
Joined
Dec 31, 2006
Posts
5,202
Location
Germany
Society
Jurai Blood
Avatar Name
Nicholas wizzszz Wolf
Oh, we're playing that game now

Look, you acted as if it was disproven already - but acting like that aint enough, either you can do what you've claimed or you can't.

I am 100% sure you can't, and i am 100% sure that you simply think your arguments are valid while they are infact not when examined thoroughly.

And, all that's left amongst other stuff i don't really want to comment on is this:

You had a big mouth, subsequently you've been challenged to put up proof and you wimped out with the rather lame excuse that i'd be "playing games".

And, as if that wasn't enough already, you wimped out when reading the preconditions under which i would enter a debate.

What's left to say? After all your post was nothing but heaps of lukewarm nothing.




It works always like that - people simply want to jump over the basics to reach the point where they draw false conclusions from their false basic assumptions. A mistake that could have been avoided by checking which basics we want to argue and deduct from. But these basics are wrong - and that's exactly WHY people want to skip that inconvenient part.

If so many people think MA is wrongfully rated "not gambling", why on earth NOBODY ever called them and asked them to reconsider their decision, to make a more in-depth examination of how EU works??
 

Fifth

Elite
Joined
Jan 26, 2012
Posts
2,721
Avatar Name
The "Fifth" Entity
Different mob classes have different variability rates.
Do you have data to back this up?
No, no solid data whatsoever, only a general observation:

- Take 5k ped ammo and hunt, say, small Longu. You'll end up with your initial 5k plus-minus ~25%
- Take 5k ped ammo and hunt Longtooth. You will end up with significantly more, or significantly less.


Yes, your probably right when you say it's not actually the variability that's different, it's the requirement for the size of the bankroll that's different. Probably, maybe...

However, for the Average Joe who doesn't have infinite resources there's no difference. For him both "has a bigger loot variance" and "requires a bigger bankroll" will look exactly the same. He will, most likely, go broke before he can find out if he would have reached the magical average 90% return at some point... or not.



PS: I do believe the variability itself is different as well (a logarithmic curve that goes up with mobs threat level) but this is very hard to prove, so: Not proven, not quilty, just dismiss it. :)
 

falkao

Stalker
Joined
Oct 25, 2005
Posts
1,993
Location
Italy
Society
Freelancer
Avatar Name
Marc falkao Falk
...Skills have absolutely no meaning whatsoever in EU when it comes to tt returns - they are only a measurement of how much money you have spent and allows you to play a higher stake game. The absolute proof for this fact is the gateway HoF. It could be further backed up by f.ex. very noticeable mining hits (ATH's) by very unskilled avatars, but that's really not required. That is, unless one thinks "less skilled gets more tt return". :)...

I guess there is a basic misunderstanding. It is rather obvious that a lower hit rate will lead to higher kill costs or that lower extraction skills will lead to higher extrication costs.

What MA did state is, that those additional costs are not tracked and not compensated for. This implies that loot is independent from it.

To relate all this to ATH’s is flawed as those are very rare events. Sure, we have them but there will be only a very limited number of avas to get one (1 out of 10k?). Do you really think it is meaningful to base playing style on it?
 

falkao

Stalker
Joined
Oct 25, 2005
Posts
1,993
Location
Italy
Society
Freelancer
Avatar Name
Marc falkao Falk
...
The loot system MUST have a memory, otherwise it could not produce the results we see - and i will for sure not rate some silly "dev notes" from MAs inept staff higher than my own perception & experience, no f'ing way.

Dear Wizzszz,

everyone is free to believe what he likes to believe but he is also responsible for his own believing.

For the first time, MA is communicating something about loot. It can be true or false. To falsify it we would need quite a lot of data, something that is not achievable easily. Perception is for sure rather useless here. On the other hand, we have studied the system in the past and all the data we had so far suggested a random system.

For instance, the double bombing experiment did show that return rate is lower and it did correspond to the expected reduction in hit rate when assuming that a find in an area has a Poisson distribution. Furthermore, the time between globals per avatar follows an exponential distribution, something that you would expect when dealing with a random system. It is however rather unclear if ATH’s do belong to this process or if they are triggered by something else.

Nevertheless, all those systems don’t need a memory but if you observe such a system and do record data, then many do think that they are able to break it. For instance, take a fair coin. The probability to get heads is 1/2. If you had 4 heads in a row, what’s the probability that the fifth toss will also be heads?
Well it is still 1/2.
Sure, many will state that the probability of the series is 1/32 as five independent events will have the probability of 1/(2^5) and therefore it must be more probable to observe tails in the fifth toss, but a series of 4 heads followed by one tail has the same probability. Hence, both possibilities for a series with five tosses ( 5 heads or 4 heads plus tail) do have the same probability of 1/2 and the information of 4 heads in a row is useless to predict the fifth toss.

On the other hand it is true that recorded data might influence perception. Let’s assume that a global has the probability of 3/100. To get two globals in a row is a rare event and would have the probability of 9/10,000. Hence, if one has recorded a global then to record a further one is quite improbable as 9/10,000 is quite lower then 3/100 * 97/100 = 291/10,000.
Similarly, if you had a series of bad loots, say no global in 100 kills, then the probability of the series 100+1 “no globals” is still more probable then 100 “no globals” plus 1 “global” as (1-3/100)^101 is greater than (1-3/100)^100*3/100 as you could easily verify.
Nevertheless, it is however quite improbable to observe a series of 100 “no globals” as the probability of at least one global in 100 loots is larger than 95% (1-(1-3/100)^100 = 95.2%) and hence the alternatives to 100 “no globals” as 49ng-1g-50ng or 1g-99ng etc.. are in sum quite more probable.

Hence with a random system one is able to perceive what he likes to perceive.
 

Jimmy B

Indioni Jones
Joined
Dec 2, 2005
Posts
12,256
Location
UK
Society
Skillin' Villains
Avatar Name
Oni
PS: I do believe the variability itself is different as well (a logarithmic curve that goes up with mobs threat level) but this is very hard to prove, so: Not proven, not quilty, just dismiss it. :)

I believe it is different too. My instinct is that higher HP mobs actually have lower relative loot variance over the same amount of kills on a smaller mob, but not enough to offset the lower kill rate when considering variance over fixed spend rather than kills. Not sure if you were suggesting the same or the opposite hehe, but just gut feeling really anyway.

and i will for sure not rate some silly "dev notes" from MAs inept staff higher than my own perception & experience, no f'ing way.

If you question things like gravity or the intelligence of the swedish authorities the debate is over.

So, OK to state MA's staff are inept, not OK to suggest swedish authorities may not be perfect. Noted.

For the record I wasn't necessarily saying the swedish authorities are incompetent anyway, more that their requirements for EU not to be classified as needing a gambling license may not match what your requirements for EU not to be considered gambling are.
 

wizzszz

Banned
Joined
Dec 31, 2006
Posts
5,202
Location
Germany
Society
Jurai Blood
Avatar Name
Nicholas wizzszz Wolf
So, OK to state MA's staff are inept, not OK to suggest swedish authorities may not be perfect. Noted.

And, what exactly is your problem with that?

I know how inept MAs coders are, because their field of work matches parts of my own expertise.

And, i do know how anal retentive authorities are here - if the swedish authorities are only 10% of that, there is no way that they have misclassified EU.

Sweden is no banana republic, there are clear guidelines, the examiners have to write a detailed report WHY they classified EU the way the did.

And it's meanwhile the 100th time i wrote this, but it has been conveniently ignored 100 times by those who support the "EU is gambling theory" - i really hope the next dev notes include a statement on that - no doubt it will shatter this silly little assumption of yours into millions of tiny pieces.

For the record I wasn't necessarily saying the swedish authorities are incompetent anyway, more that their requirements for EU not to be classified as needing a gambling license may not match what your requirements for EU not to be considered gambling are.

There are laws, they are written down and available to the public.
The definition of gambling is, with minor exceptions, identical in all western countries.
(and no, there is no exception big enough to cover what you claim)


We are even on our way to have these minor differences removed, in the process of uniting europe further.


You claim that this is not the case, and that swedish gambling laws are somehow from the moon, completely different to all the other civilized countries, yet you have provided nothing, absolutely zero to back up your (imo rather ridiculous) claim.



My swedish sucks, otherwise i would have completely dismantled this argument already.

But why bother, you will still say that swedens authorities are too dumb to apply these law texts properly.
 

wizzszz

Banned
Joined
Dec 31, 2006
Posts
5,202
Location
Germany
Society
Jurai Blood
Avatar Name
Nicholas wizzszz Wolf
Dear Wizzszz, ...

Wow, do you really think you have to lecture me about stochastics?


I used the term "has a memory", assuming that anyone out there with a bit knowledge on stochastics would recognize it as the exact opposite of a (non-memory) even like i.e. tossing a coin.

Because this is exactly what i wanted to express:
That the events (looting mobs) are NOT independant, as opposed to (independant) "toss a coin multiple times" events.
And actually, i did that only to get rid of the term "personal loot pool", because people appear to have different associations connected with that term.


Your rather lengthy post conveniently ignores something:
Random system = gambling = stupid swedish authorities = MA lying = EU gets prohibited in many states soon = ....

Want me to carry on?


I don't know what "patterns" you talk about, but here some of mine:

Let's say you do 100 mining runs, all with enough bombs to go on until you hit a global.

And record the return all the time.
The return rate pretty quickly converges to 2/3rd

So far a random system can produce these results.


Now you hit a global
And the global size EXACTLY matches the missing amount of PEDs thats needed bring you to 90%.

And you get the same pattern 98 out of 100 times....

NO random system is able to produce THAT pattern.
Only a loot system with a memory is able to produce that.
 
Last edited:

Jimmy B

Indioni Jones
Joined
Dec 2, 2005
Posts
12,256
Location
UK
Society
Skillin' Villains
Avatar Name
Oni
I don't know what "patterns" you talk about, but here some of mine:

Let's say you do 100 mining runs, all with enough bombs to go on until you hit a global.

And record the return all the time.
The return rate pretty quickly converges to 2/3rd

So far a random system can produce these results.


Now you hit a global
And the global size EXACTLY matches the missing amount of PEDs thats needed bring you to 90%.

And you get the same pattern 98 out of 100 times....

Finally, thanks. Now we have something falsifiable. That's what I was asking for 40 pages ago.
 

wizzszz

Banned
Joined
Dec 31, 2006
Posts
5,202
Location
Germany
Society
Jurai Blood
Avatar Name
Nicholas wizzszz Wolf
Finally, thanks. Now we have something falsifiable. That's what I was asking for 40 pages ago.

You asked for records...

However, the 2 out of 100 runs follow the same pattern, just replace the 66% and 90% by 50% and 66%.


Can't wait for you to "falsify" this and suggest a stateless "no-memory" algorithm that is able to produce such results, or to subsequently admit that this is not possible without "tracking" losses, at least over single runs.
 

Skaffen

Old Alpha
Joined
Nov 23, 2005
Posts
778
Breeze

You asked for records...

However, the 2 out of 100 runs follow the same pattern, just replace the 66% and 90% by 50% and 66%.


Can't wait for you to "falsify" this and suggest a stateless "no-memory" algorithm that is able to produce such results, or to subsequently admit that this is not possible without "tracking" losses, at least over single runs.

I'd be surprised if this relationship holds - my take from many posts is that there are negative and positive returns over any given period, surely many would notice this 'cap' you suggest?
 

wizzszz

Banned
Joined
Dec 31, 2006
Posts
5,202
Location
Germany
Society
Jurai Blood
Avatar Name
Nicholas wizzszz Wolf
I'd be surprised if this relationship holds - my take from many posts is that there are negative and positive returns over any given period, surely many would notice this 'cap' you suggest?

What "cap"? Many others have already shown that ~90% long term is indeed what you can expect if you don't act silly.

And when you make your runs long enough you will see that pattern (you gotta stop after one of these "payback globals" that bring you back to 90% and start a new run) - if you cut these runs into 200 or 500 PED "slices" you will have exactly what you describe: Runs with negative and positive returns.

There are a few minor prerequisites though, that are needed if you want to reproduce my results (nothing special, really) - but i prefer to not share them in public. It's nothing that really has an impact on what i've outlined above.


In mining, it is very easy to see that pattern, because instead of several sources of decay (armour, fap, gun, amp, sights, scopes, ammo) you have only finder decay and bombs.
(Finder decay is so minimal on OF-105, i did not include it in my calculations)
 

Jimmy B

Indioni Jones
Joined
Dec 2, 2005
Posts
12,256
Location
UK
Society
Skillin' Villains
Avatar Name
Oni
You asked for records...

However, the 2 out of 100 runs follow the same pattern, just replace the 66% and 90% by 50% and 66%.

Can't wait for you to "falsify" this and suggest a stateless "no-memory" algorithm that is able to produce such results, or to subsequently admit that this is not possible without "tracking" losses, at least over single runs.

OK, I asked for records, but the main thing is to have something that can be tested and corroborated by others.

I don't myself have any data that could corroborate or falsify this, as I've only ever tracked returns on a run-by-run basis rather than a loot-by-loot basis. Maybe Noodles and steffel's data from the mining loot study can be used though.

I presume we're talking unamped mining here? I've had consecutive globals with mid-range amps which wouldn't fit in.
 

wizzszz

Banned
Joined
Dec 31, 2006
Posts
5,202
Location
Germany
Society
Jurai Blood
Avatar Name
Nicholas wizzszz Wolf
OK, I asked for records, but the main thing is to have something that can be tested and corroborated by others.

I don't myself have any data that could corroborate or falsify this, as I've only ever tracked returns on a run-by-run basis rather than a loot-by-loot basis. Maybe Noodles and steffel's data from the mining loot study can be used though.

I presume we're talking unamped mining here? I've had consecutive globals with mid-range amps which wouldn't fit in.

Yes, unamped, and no enhancers (didn't exist back then anyway).

It is possible though that the "payback" comes in several smaller globals, this indeed happens, but not very often.
Those consecutive smaller globals nonetheless add up to the expected 90% return rate.

And, i forgot to mention another thing - if you go much below that 66%, you will mostly get one or more bigger claim (usually VI - XI in unamped ore mining) that brings you back (or very close) to the 66% line - sometimes these can become big enough to trigger swirlies, but are not the big "paybacks" i mentioned earlier.



And not 100% of the runs are like that, a small percentage appears to be "predetermined losses", and of course you get sometimes a "welcome back gift" from MA after a longer break, somewhere around 1k - these do not follow the usual pattern, too, of course.



This is all not important though, and actually i've already said too much here - but an algorithm that follows the pattern in 98 out of 100 runs MUST have a memory.
 

falkao

Stalker
Joined
Oct 25, 2005
Posts
1,993
Location
Italy
Society
Freelancer
Avatar Name
Marc falkao Falk
...I don't know what "patterns" you talk about, but here some of mine:

Let's say you do 100 mining runs, all with enough bombs to go on until you hit a global.

And record the return all the time.
The return rate pretty quickly converges to 2/3rd

So far a random system can produce these results.


Now you hit a global
And the global size EXACTLY matches the missing amount of PEDs thats needed bring you to 90%.

And you get the same pattern 98 out of 100 times....

NO random system is able to produce THAT pattern.
Only a loot system with a memory is able to produce that.

I'll get the following return rate distribution from n=85 consecutive runs till a global.

e75_rrtillglobal.jpg



RR = return rate, unit 1 = 100%
Median is 98%, min = 57%, and max = 1330%.

The median is close to 90% but there shouldn't be any variance about it, when assuming that your hypothesis holds true.

And here the resepective distribution of the number of drops till a global, which follows an exponential distribution with mean 86.

e75_dropstillglobal_799120.jpg
 

wizzszz

Banned
Joined
Dec 31, 2006
Posts
5,202
Location
Germany
Society
Jurai Blood
Avatar Name
Nicholas wizzszz Wolf
I'll get the following return rate distribution from n=85 consecutive runs till a global.

You missed the entire point - it's not "how long will it take until i hit a global", it's about the size of that global (in PED).

The size of that global matches prior losses (corrected by a 10% fee for MA, if you record runs from payout to payout) - this is, in my humble opinion, proof enough that the loot system indeed has a "memory".

And unless you can come up with some kind of algorithm that mimics this behaviour without the use of a "memory", i will keep claiming that MA in fact "tracks" loot on a personal basis (= loot is avatar based, with the help of a "pool" where you accumulate those payouts, if not overall, at least on a "per run" basis).
 

falkao

Stalker
Joined
Oct 25, 2005
Posts
1,993
Location
Italy
Society
Freelancer
Avatar Name
Marc falkao Falk
..It is possible though that the "payback" comes in several smaller globals, this indeed happens, but not very often.
Those consecutive smaller globals nonetheless add up to the expected 90% return rate.
....

All data I have seen so far doesn't have any bias and is random. However, there is one thing that is unclear to me. Is the loot an area produces independent from what has spent on it or is there some sort of a limit (PED spent on area) that must be reached before it drops higher loot classes? With this MA would have the guarantee that payout is always lower then spent PEDs.

In the latter case, the area would have some sort of memory but the loot system can still be random.
Unfortunately this hypothesis can’t be verified as we would need many new areas to check if you can get a global on first hit.

Nevertheless, although in the case of memory per area, it will not influence your return rate but you have to play for longer time as you have to feed the area first. Furthermore, if something like that exists, then the area must be very large, otherwise mining would be very tricky as dry areas need to respawn and this must be considered when feeding an area. Therefore, I don't think that something like that exists but I can't exclude it.
 

falkao

Stalker
Joined
Oct 25, 2005
Posts
1,993
Location
Italy
Society
Freelancer
Avatar Name
Marc falkao Falk
...if you record runs from payout to payout) - this is, in my humble opinion, proof enough that the loot system indeed has a "memory".

that's the first figure and RR of those runs is very volatile. You've stated the contrary.
 

Uvas

Prowler
Joined
May 10, 2008
Posts
1,047
Society
Toxic Templar of Fallen Knights
I, for one, would be delighted to see the run data on those 98 out of 100 globals that EXACTLY evened out the returns to 90%.

Otherwise someone is STILL talking out of his butt. Grandiose sweeping conjectural statements without a shred of evidence.

:laugh:
 
Top