Here's my hypthesis; disprove it

[Cerberus]

First off I'm going to start off by saying that this hypothesis clearly breaks Occam's razor, however, I feel that it could explain why looting can be so inconsistent.

Assumption: The loot works without memory
Assumption: When checking individual runs, loot behaves very irrational; much unlike what you'd expect from a simple "You have an 60% chance of getting a loot worth 120% of your TTin, and 40% to get a no looter"; or similar.
Assumption: TTout goes to roughly 90% of TTin under a longer timeframe

To explain the hypothesis, I'll keep it as simple as possible:

What I propose is that once in a given timeframe, a fictional number is drawn. This could either be related to you personally (once every 100 attempts has been done looting hunting [or even getting loot from the system at all, which would incorperate all systems]) or on a more global level (every time the server processor usage goes to 45,34894% or similar).

This number, in turn, signifies which subset of the loot table will be used when your avatar attempts to loot something. So, to keep it absurdly simple, it could be something as:

Table #4:
0 0 0 115% 0 65% 0

Table #1:
20% 20% 25% 0 15% 0

With the only rule that the sum of the attempts go to an estimated 90% TTout (which my example doesn't follow; lol).

This would give an added layer of control, as the probability of getting a profit margin of 300% on Chirpies would go from tiny to absurd. It would also allow modifications of the loot payout very easily. So for instance, if someone gets a huge profit margin, it would be quite easy to see why, and fix that subset, instead of rewriting the entire loot randomization code.

---

As far as I can see, the only plausibility problem this hypothesis has is that it disagrees with Occam's razor. But I see as that is justified, as the added layer of control and ease of debugging for developers this theory suggests outweights the implementation problems.

What I want now is for someone to break this theory. Partially or completely, I'll be satisfied either way.

 

[Rumsponge]

"Entropia is dynamic.". Therefore, it is virtually impossible to prove or disprove any theory regarding "the loot pool".

However, I will take a stab at disproving your hypothesis.

A) You give MA your money so you can play EU.

B) Anyone who sends MA money has lost the ability for rational thought.

C) Although there is always a random chance, it is highly unlikely that anyone who has lost the ability for rational thought will produce a viable hypothesis for anything (with the possible exception of the existence of his own belly button lint).

D) QED your hypothesis must be considered to be nothing more than the unfortunate ravings of a lunatic.

 

[XeroX2]

B) Anyone who sends MA money has lost the ability for rational thought.

I don't think that's true.
However it has a scaring tendency to be true.

 

[Cerberus]

"Entropia is dynamic.". Therefore, it is virtually impossible to prove or disprove any theory regarding "the loot pool".

I disagree.

Many theories suffer from two problems:
1) Inreproducable results. Either they only work "sometimes", or data frequently directly contradicts the theory in question
2) The level of complexity makes them possible, but not plausible. After all, it's a company with a game we're discussing, not a space federation.

As such, it'd be nice hearing from people if this theory falls into either of these categories.

 

[Thorns Rose]

Can't disprove that...

Sounds really reasonable to me. I have thought of different ways and ideas for loot drops, I even entertained the idea that depending on whether the land is privately owned and if they sink money into that area and mobs also.

 

[humanwoman]

well, if they use a set of tables with pre-generated random numbers, instead of just generating numbers "on the fly", many people would have noticed these patterns

like, after dropping about 10000 bombs, you would go through the same table more than once and see that your loot distribution has parts that repeat themselves

doesn't happen to me, all number sequences are random, the only thing which is not random is the average, which is set by MA. that's how they control your return in the long run

 

[milduser]

Understanding the "Game Theory" is your first step to understanding the Loot Theory (if there is one .

 

[aridash]

so a loot table has to be generated for every single player? for how many loot actions would this be? 10, 100, 1000, 1000000? this seems to contradict the assumption "without memory" since a table is clearly a form of memory. and predictive, so potentially open to expolitation (spot a pattern, know whats come up).

on that last point, i wouldnt dismiss it though, as i think such patterns can emerge.

 

[Infinity01]

"Entropia is dynamic.". Therefore, it is virtually impossible to prove or disprove any theory regarding "the loot pool".

However, I will take a stab at disproving your hypothesis.

A) You give MA your money so you can play EU.

B) Anyone who sends MA money has lost the ability for rational thought.

C) Although there is always a random chance, it is highly unlikely that anyone who has lost the ability for rational thought will produce a viable hypothesis for anything (with the possible exception of the existence of his own belly button lint).

D) QED your hypothesis must be considered to be nothing more than the unfortunate ravings of a lunatic.

LOL
I felt your post is one of the most irrational answer...now go figure !!!

 

[Zavy]

I can't disprove your theory, but I wonder if graphic settings affect loot. Higher settings would = more bandwith use.
Could be a factor, but like your theory how can we prove it?

 

[Futurama]

Assumption: The loot works without memory
...
Assumption: TTout goes to roughly 90% of TTin under a longer timeframe

Doesn`t these 2 assumptions disprove one another? The second assumption basicly means that that system has some kind of memory, otherwise we would see big tt profits from some ppl (i`m not talking about just 1 lucky hof for a 2 day old avatar). To keep the average tt return at some predetermined level the system has to know if you are getting too much or not enough.

 

[Mikeemooo]

I'm not sure I'm quite following this.

So instead of the server generating random figures, you're suggesting that there's actually a massive hard coded 'loot payout table' with random figures on it? So the developers could then adjust the payouts on this huge table if they wished?

I'll let you answer this before typing any more, because I may be understanding this wrong..

 

[Zorander]

So if I understand correctly, your theory is that there are multip loot payout tables and at some unknow interval a new table is selected for your avatar at random.

That's great... I can't disprove it.

Also... you can't prove it, lol.

In addition, even if it is correct, it does no one any good to know since the interval that a new table would be randomly chosen is unknown. How would you map those tables? How would you know when one table stopped being used and another started?

Your theory boils down to 'Loot is random'

Now if you crack when the tables are changed, that could be useful as you'd be able to go hunt/mine/craft a bit, figure which table you're on and decide if you want to keep going or wait for a switch.

Without that little bit of information, and mapping out the tables, your theory is slighty less than useful.

 

[Mel]

This theory has been bantered about for years...."I think, therefore I loot (poorly, well, or wtf)"

 

[Cerberus]

I'm not sure I'm quite following this.

So instead of the server generating random figures, you're suggesting that there's actually a massive hard coded 'loot payout table' with random figures on it? So the developers could then adjust the payouts on this huge table if they wished?

I'll let you answer this before typing any more, because I may be understanding this wrong..

I just had a good nights sleep, and realised how badly that goes against Occam's razor.
The problem as I see it, is that if there was just one algoritm/matrix/database for loot returns, there'd be more consistency in the numbers. After a very high amount of loots, one would eventually land at same % of individual loot returns (ie 0,25% of the time, 30% TTin is received).

Idk, maybe that's what keeps popping up after a while. I merely suggest this, as there seems to be a consensus that those kind of figures never happen, which would be very weird for just one algoritm, but definietly possible if one was jumping around a myriad of different matrixes.

Anyway, take it with a grain of salt. I'm merely suggesting it, to see if it would be even worthwhile looking further into it (ie the theory doesn't get shot down instantly).

well, if they use a set of tables with pre-generated random numbers, instead of just generating numbers "on the fly", many people would have noticed these patterns

like, after dropping about 10000 bombs, you would go through the same table more than once and see that your loot distribution has parts that repeat themselves

doesn't happen to me, all number sequences are random, the only thing which is not random is the average, which is set by MA. that's how they control your return in the long run

It wouldn't go through numbers linearly, there's no need for that.

Doesn`t these 2 assumptions disprove one another? The second assumption basicly means that that system has some kind of memory, otherwise we would see big tt profits from some ppl (i`m not talking about just 1 lucky hof for a 2 day old avatar). To keep the average tt return at some predetermined level the system has to know if you are getting too much or not enough.

If if tell you can buy a ticket from me for 1$ from my massive pile, whereas half of them give 30% of what you paid and the other half gives 150%. After buying a million of these tickets, you're bound to end up at 90% of what you originally spent on the tickets; no memory used.

In addition, even if it is correct, it does no one any good to know since the interval that a new table would be randomly chosen is unknown. How would you map those tables? How would you know when one table stopped being used and another started?

Your theory boils down to 'Loot is random'

Now if you crack when the tables are changed, that could be useful as you'd be able to go hunt/mine/craft a bit, figure which table you're on and decide if you want to keep going or wait for a switch.

Without that little bit of information, and mapping out the tables, your theory is slighty less than useful.

I'd go on with collecting data and trying to "crack it", if the standing hypothesis doesn't get shot down. If I suggested it's Lootius will that I loot how I do, I'd obviously not sink a few thousand PED into it if someone calls it batshit insane from day one.

 

[kalanen]

breaks Occam's razor

Don't worry: Occam's razor is just something that noob uses as excuse to call the first "logical conclusion" in his mind built on premises (which typically are only assumptions) as "theory".

There's never the simplest solution. Only one which is closest to the way one have used to think.

Oh, and grats for not using term "loot pool".

Keep it up

 

[Cerberus]

Don't worry: Occam's razor is just something that noob uses as excuse to call the first "logical conclusion" in his mind built on premises (which typically are only assumptions) as "theory".

There's never the simplest solution. Only one which is closest to the way one have used to think.

Oh, and grats for not using term "loot pool".

Keep it up

Yes of course. Just like any form of research, we're not lookin for a definitive answer, but rather one that as closesly as possible explains what's happening. I would however disagree on your point regarding Occam's razor. Most people when trying to figure out a system from predetermined standpoint, usually does such with such a bit too much disregard for the complexity (or in our case, the programmers). We are after all playing a PC game, and it feels like a lot of unneccessary work for a duo of programmers to base something as simple as a random generator on a sine wave plotted over the X,Y coords.

 

[Aydonis]

They have a human counterpart assigned to each of us that sits watching what we see on their own computer. They are sitting there with a set of D&D dice at the ready to roll while we are attacking a mob, dropping a bomb, crafting etc, that's why there is lag to generate your loot sometimes, cause one die fell off the table and they have to roll that one again to figure out if we get no loot, some loot, good loot, great loot.

Throughout the day the office has constant raffles using the peds of those that run around throwing away money, or performing actions that generate no loot (repairs, shooting at the sky, healing yourself or other random people, logging out, or falling asleep at the keyboard without claiming a resource so the claim disappears, etc), and if you're counterpart wins, then they decide to hand out a hof, ath, or discovery to you depending on how big the pools get up to.

I like your banana, so I stole it.

 

[Aydonis]

This theory has been bantered about for years...."I think, therefore I loot (poorly, well, or wtf)"

lol, so in my case I guess its "I no think, therefore I no loot"

 

[Futurama]

If if tell you can buy a ticket from me for 1$ from my massive pile, whereas half of them give 30% of what you paid and the other half gives 150%. After buying a million of these tickets, you're bound to end up at 90% of what you originally spent on the tickets; no memory used.

True, but imho it still would mean that the return rate range would too wide, too unpredictable and too unfair. Even in such a loterry you describe - there will someone who will always get 30% ones (don`t confuse it with the whiners who run around saying he`s getting only 30% return, while in reality he killed only a few mobs and forgot he had a global right before that) and some who will make nice profit just because they are luckier. In EU it could also possibly mean exploits etc.

From my own experience - the fluctuations aren`t that high really which could be because its not so random after all...

 

[Mikeemooo]

True, but imho it still would mean that the return rate range would too wide, too unpredictable and too unfair. Even in such a loterry you describe - there will someone who will always get 30% ones (don`t confuse it with the whiners who run around saying he`s getting only 30% return, while in reality he killed only a few mobs and forgot he had a global right before that) and some who will make nice profit just because they are luckier. In EU it could also possibly mean exploits etc.

From my own experience - the fluctuations aren`t that high really which could be because its not so random after all...

Fluctuations are easy to reduce without needing memory. When you have enough people using a system it evens itself out very nicely. These things are easily balanced which is why MA have a balancing team...

http://jsfiddle.net/k3y6N/17/

Big fluctuations at first, but leave it running long enough and the levels start to even out for each player.

That one runs slow if you leave it too long, so here's one without the damage/loot messages:

http://jsfiddle.net/k3y6N/18/

I've had this running 5 minutes, and everyone is on around 80-110% payout, and a few people have had some lucky hofs...

The chance of someone ALWAYS getting low returns is very very very very very very very low.

When I do 200-mob argonaut runs, I can come away with 60% or 200%. Even in entropia it doesn't quickly even out..

 

[Cerberus]

True, but imho it still would mean that the return rate range would too wide, too unpredictable and too unfair. (...)

No, not at all.

 

[Futurama]

Fluctuations are easy to reduce without needing memory. When you have enough people using a system it evens itself out very nicely. These things are easily balanced which is why MA have a balancing team...

http://jsfiddle.net/k3y6N/17/

Big fluctuations at first, but leave it running long enough and the levels start to even out for each player.

That one runs slow if you leave it too long, so here's one without the damage/loot messages:

http://jsfiddle.net/k3y6N/18/

I've had this running 5 minutes, and everyone is on around 80-110% payout, and a few people have had some lucky hofs...

The chance of someone ALWAYS getting low returns is very very very very very very very low.

When I do 200-mob argonaut runs, I can come away with 60% or 200%. Even in entropia it doesn't quickly even out..

I didn`t look at it this way, you could be right. But still there are too many factors that could be influencing it, we can`t know for sure.

What about the hofs after longer period of inactivity? It has happened on quite a few occasions when someone who has been inactive for a longer period of time, but use to play and cycle a lot, comes back with a nice hof. Is this just pure luck or he "earned it" because his returns back in the days weren`t so good? That could mean that there are some kind of memory... Example: Prodiq came back at SGA and hit a nice hof (or two?) quickly.

 

[Xandra]

Hi,

What I want now is for someone to break this theory.

Impossible, because those that could are under non-disclosure.

But think of it, why in *** would someone code such? What would be the benefits? Would it be accepted by a juristic research due to someones claim that this system might be a scam? Would it enable it to fine-tune this system in a way that the output (loot) wouldn't top the available resources? Would it make it possible for the known occurrences where obviously special mobs at a special place would loot special things?

There's a lot of things for MA to consider when making a loot distribution system.
So any try to put up a loot theory should be to analyze these requirements at first:

1. It has to be assured that not more money is looted then is available.
2. It has to be assured that no rogue employee can just alter an database entry to feed it's "little brother's avatar" with ATH's.
3. It has to be assured that at anytime it can be proved that any participant has the same chances, should a judge demand such.
4. It has to be assured that it's possible to rise the probability to loot for mobs and areas temporarily, to meet the reality that we experience in events, for instance.
5. And it has to be assured that:
   • skill matters, to prevent the accusation of gambling, and
   • that the big payers are rewarded accordingly, to keep them
6. And, last but not least, this has to run on hardware not requiring huge data centers for the few 10K Entropians ...

Agree? Then test your theory against this premises. I leave this as an exercise to you ;-)

A few explanations:

• Most of this can be achieved just with the use of a few hardware RNG's that are not this expensive anymore these days, and
• an intelligent use of very few variables that can be altered on-the-fly.
  These variables could contain "LootQuote", "maxLootValue", "AreaQuote", "MobQuote", "MobMaturityQuote", maybe even "MobLootTableNumber".
• #5, "skill matters" is easily to reach with the current "wasted HP determines max. Loot" and with the current Loot tables, that reserve any high-MU loot to the ones only that are able to kill the big mobs.
• The usage of multiple RNG's, together with the possibility to fine-tune these variables, and the loot tables, is enough to explain close to any loot we're talking about.
• Random occurrences (even pseudo-random ones as we'd need to use here, necessarily) are notorious to create strange patterns - see the dates of the recent nuclear catastrophes.

We just need to think like we'd be MA coders. We'd have to reach some defined goals, and we'd have to do it in a way that isn't too expensive.
This doesn't leave a lot of ways to go. Judge yourself!

And have a good time!

PS:
For sure MA is renown for their quality of coding, since years. ;-)
So I'd not preclude any most strange, lunatic algorithms. Maybe the current loot distribution is still encapsulated in a 16bit .dll, written back in 2002 by a long buried acid fiend?

 

[Mikeemooo]

I didn`t look at it this way, you could be right. But still there are too many factors that could be influencing it, we can`t know for sure.

What about the hofs after longer period of inactivity? It has happened on quite a few occasions when someone who has been inactive for a longer period of time, but use to play and cycle a lot, comes back with a nice hof. Is this just pure luck or he "earned it" because his returns back in the days weren`t so good? That could mean that there are some kind of memory... Example: Prodiq came back at SGA and hit a nice hof (or two?) quickly.

When you have so many people with different behaviours, you'll always come across things like this which make you wonder if there's something else going on - but it can all be put down to chance.

For example, I was surprised that I didn't see Megavolt on the top highest loots list for argonauts. He's had OVER twice as many globals as anyone else on the server. 889 tracked globals. The next person on the list has 365 globals. But when you think about it more, he's done maybe 2% of the total argonaut globals ever done on the server, so the fact that he's not on the top list isn't all that insane. A HUGE percentage of the globals are done by <1 year people, so a noob getting a 15k argo loot isn't very surprising. There's a reasonable chance that someone could recently switch to argonaut from another mob then find himself with a 15k and will put it down to him having switched.

When it comes to people returning to the game and getting a big uber: Entropia is one of those games where people go away and come back regularly. People generally don't like to close all ties because of the time/money invested, but like to take breaks because of frustration or money issues. I'd suggest that there's a lot of "coming back", so the chance of one of the people on the ATH list having recently returned to the game isn't too low. For every person who's come back and had a nice big ATH, there's a whole bunch of people who got a nice big ATH without recently coming back.

 

[Mikeemooo]

Because I like pictures:



1) Completely random chance
2) Reasonable return per player
3) Some with big lump sums, others without..
4) 90% global payout rate
5) No memory

I can't see a reason why MA would choose to implement some complicated 'memory system' when just like any other casino-style game, it can all be done with very simple maths.

 

[Cerberus]

I can't see a reason why MA would choose to implement some complicated 'memory system' when just like any other casino-style game, it can all be done with very simple maths.

'Cause it justifies a player doing /em pogo for exactly 2m12s then running straight NNE until the first mob. If that mob is turned with it's back towards the player, and the player during the same day crafted exactly 86,4 PED of springs, then that player will ATH.

 

[Futurama]

'Cause it justifies a player doing /em pogo for exactly 2m12s then running straight NNE until the first mob. If that mob is turned with it's back towards the player, and the player during the same day crafted exactly 86,4 PED of springs, then that player will ATH.

Wait a sec, legion hasn`t even posted in this thread yet

 

[Mega]

For example, I was surprised that I didn't see Megavolt on the top highest loots list for argonauts. He's had OVER twice as many globals as anyone else on the server. 889 tracked globals. The next person on the list has 365 globals. But when you think about it more, he's done maybe 2% of the total argonaut globals ever done on the server, so the fact that he's not on the top list isn't all that insane. A HUGE percentage of the globals are done by <1 year people, so a noob getting a 15k argo loot isn't very surprising. There's a reasonable chance that someone could recently switch to argonaut from another mob then find himself with a 15k and will put it down to him having switched.

The "big" Argonaut loots tend to be Scavenger, Raider, Hunter or greater maturity, whereas I hunt Young to Guardian. Nexiuz hit the most recent 16k Argo which I think was a Raider. There was a recent case of an 11k Argo young, which was the largest I have ever seen.

 

[humanwoman]

It wouldn't go through numbers linearly, there's no need for that.

It doesn't matter how exactly you go, if you go through the same set of tables, there will be repetitive patterns, and they do not exist

So your theory is plain wrong

Unless you go through these tables randomly, which doesn't make any sense because these tables should be a replacement of a random number generator in the first place