How to analyze loot

[kalanen]

Originally Posted by falkao

Not sure if I've got it right, but do you see a relation between crafting loot and hunting loot, so that the gaps there can be applied to the hunting ones?

What I exactly did, was first killing 200+ longu matures, and pasted 20% transparent gray line on vertical position 1px equalling 1pec in 500px high PS doc - a nooby visual presentation. Darkest area (~2 years ago) was around ~200px and had 2 other "slots" below it and 2 above it (all below 50ped cap).

I also did some basic filter clicking, did similar pasting job and when I stretched filter document to match longu document (aligning darkest areas), areas matched almost perfectly. "Almost" propably due rounding issues of tiny tt of filters and occacional small tt items in longu loots.

I moved on to doing the same with lyst to see if caim level X upper limit and lower limit falls inside cap area but I was low on peds, found shit and abandoned whole study

So how about area above 500px in stretched filter doc? Do longu globals match those values and can it be used to predict mob loot values up to the max? Those were my original questions.

In general, I'm a bit sceptic about the integrity of recorded globals as data for this kind of analysis; tt value of SIB/L stuff seems to come out of nowhere so there's many global values that dont fall into "natural" slot.

Also I dont understand why big ones have been left out. Notice how often "a half" of a big one drops pretty soon after the big bang? And "a half" of the halved one? All the way down to the no looter...

[falkao]

Originally Posted by kalanen

What I exactly did, was first killing 200+ longu matures, and pasted 20% transparent gray line on vertical position 1px equalling 1pec in 500px high PS doc - a nooby visual presentation. Darkest area (~2 years ago) was around ~200px and had 2 other "slots" below it and 2 above it (all below 50ped cap).
...

ok, now I understand. Interesting approach btw, rememebers me to some kind of spectral plot. So for every loot you had a rectangle 20% gray starting from 0px to (loot)px? Pasting them over each other makes areas darker. The figure you've got from Longu was proportional in length with that of filters.

Originally Posted by kalanen

In general, I'm a bit sceptic about the integrity of recorded globals as data for this kind of analysis; tt value of SIB/L stuff seems to come out of nowhere so there's many global values that dont fall into "natural" slot.

Crafting loots might be indeed difficult to analyze when residue is involved for example.

Originally Posted by kalanen

Also I dont understand why big ones have been left out. Notice how often "a half" of a big one drops pretty soon after the big bang? And "a half" of the halved one? All the way down to the no looter...

when I have enough data I'll try to do some autocorrelation analysis.

[falkao]

Originally Posted by falkao

ok, now I understand. Interesting approach btw, rememebers me to some kind of spectral plot. So for every loot you had a rectangle 20% gray starting from 0px to (loot)px? Pasting them over each other makes areas darker. The figure you've got from Longu was proportional in length with that of filters.

Now I know what the plot is. It is nothing else as another representation of the survival function, where the probability at x loot corresponds to a grey value. The higher the prob the darker the area from x-1 to x.

[kalanen]

Originally Posted by falkao

ok, now I understand. Interesting approach btw, rememebers me to some kind of spectral plot. So for every loot you had a rectangle 20% gray starting from 0px to (loot)px? Pasting them over each other makes areas darker. The figure you've got from Longu was proportional in length with that of filters.

It was actually set of 100px * 1px 20% opacity horisontal lines plotted on 100px * 500px doc. Y position equalled loot's value in pecs.

In your another excellent thread I suggested "seed value" from where loot is modified to hunter's final reward, to be 5pecs * stamina. That's because somewhere around y 200px was very dense dark area of lines with very little spread. Other "slots": *0.5, *0.25, *2 and *4 had clear limits too but spread was higher.

My idea for whole loot system (L/SIB items exluded, they come from (L)ootius) is simply dmg inflicted * (0.5pec - TT of skills gained). The rest is just stacking *2 and /2 modified attached to cycles: mob, area, maturity etc.

...with few oddball assumptions:

- certain amount of no-loots are "misses" by avatar
- set of multipliers are capped off from avatars ability to loot at certain times
- there's additional multiplier attached to mob (notice different global animations)

[Noodles]

Originally Posted by falkao

Now I know what the plot is. It is nothing else as another representation of the survival function, where the probability at x loot corresponds to a grey value. The higher the prob the darker the area from x-1 to x.

Wouldn't it be an awful lot like a histogram? Maybe I'm screwing up the visual in my brain . . .

[falkao]

Originally Posted by Noodles

Wouldn't it be an awful lot like a histogram? Maybe I'm screwing up the visual in my brain . . .

You won't see any frequencies only boxes and the color is steadily increasing.

[Botchla]

The basic condition for all avatars within the Entropia Universe is the same, meaning that you have the same chance to succeed as everybody else. Each individual can influence his/her success by gathering more experience and by using appropriate equipment, but still there is also the general economy of the virtual universe affecting your experience.
The cash economy, upon which the Entropia Universe is based, is dynamic and this also affects the in-world systems, such as the loot system.
Due to several reasons, we cannot reveal the exact details about how these systems work. We can only reveal as much as that we have not implemented any changes, which could affect the return rate calculated as a percentage.

[falkao]

Originally Posted by Botchla

The basic condition for all avatars within the Entropia Universe is the same, meaning that you have the same chance to succeed as everybody else.

Indeed, chances are different with different skills. I don't mean with that higher skills mean higher loot, but your ability to succeed will be better.

Originally Posted by Botchla

Each individual can influence his/her success by gathering more experience and by using appropriate equipment, but still there is also the general economy of the virtual universe affecting your experience.
The cash economy, upon which the Entropia Universe is based, is dynamic and this also affects the in-world systems, such as the loot system.

With the data we have, I do not even see some autocorrelation between times and loot, e.g. times where mean loot is higher or lower.

Originally Posted by Botchla

Due to several reasons, we cannot reveal the exact details about how these systems work. We can only reveal as much as that we have not implemented any changes, which could affect the return rate calculated as a percentage.

I second that.

[Tussi]

Originally Posted by Botchla

We can only reveal as much as that we have not implemented any changes, which could affect the return rate calculated as a percentage.

This is IMHO true for the loot system.

But for some reason (bug? intentional?) the hitrate seems to be lower than expected (be it with (L) or (R) stuff).

So instead of changing the return % something nerfs the hitrate which in the end gives (at least me) a lower defacto return rate per ammo spend.

Some detailed stats on this will come (when i find the time).

Tussi

[Botchla]

Originally Posted by Tussi

This is IMHO true for the loot system.

But for some reason (bug? intentional?) the hitrate seems to be lower than expected (be it with (L) or (R) stuff).

So instead of changing the return % something nerfs the hitrate which in the end gives (at least me) a lower defacto return rate per ammo spend.

Some detailed stats on this will come (when i find the time).

Tussi

Intentional