The Nightbird
Elite
- Joined
- Feb 8, 2006
- Posts
- 3,841
- Location
- Chicago, USA
- Society
- Project-Asylum Corp.
- Avatar Name
- Revan The Nightbird Evenstar
__My theory is based on time and short "cycles". I won't reveal everything here but I want to leave some clues for fun and perhaps see if anyone else has caught the same patterns I have.
__My main assumption is that deposit locations are not pre-generated and stored in a database (when a bomb is droped, a search is done in the database to find if there are any deposits within the finder radius).
__First, what is a cycle? Well, a cycle is just a word that I thought was fitting to describe this phenomenon.
To put this more simply, say the cycle is 12 seconds long.
1, 2, 3, 4, 5, 6, 7, 8, 9, [10, 11, 12], 1, 2, 3, 4, 5, 6... etc
__Then if you drop a bomb/probe on seconds 1-10, you don't hit anything. If you drop on seconds 10-12 (the loot window), you find something. Sounds good?
__The difficulty then lies in finding what the total length of the cycle is, (I used 12 seconds just as an example). Finding the length requires a very methodical method. To make it worse, the cycle length is always changing (albet slowly), so you have to adapt on the fly.
__Why do I think this is a reasonable method for MA to control loot? That is because of the loot window. The loot window size, like in the example above, determines the returns of a person dropping randomly. If the window was [6 to 12], then that person would hit alot and profit, if the window was [11 to 12], then that person is gonna cry if he persists too long. With the change of a single number, by making the window [0 to 1] or [.5 to 1] or [.75 to 1], MA can create good and bad periods with no formula changes or server load.
__To penalize overlapping, every drop is blacklisted by the server (it records the drop location, the radius from that location of the find, or the radius of the finder if nothing is found). Dropping too close to a blacklisted spot results in a penality (for example, a window of [0,1] becomes [0.75,1]), with the penality being more severe the closer you are to a blacklisted spot. Blacklisted spots expire after a period of time and there is a limit to the list of blacklisted spots - when full, a new one will bump the oldest one off.
__Finds that are generated by a drop are stored in a database, making it possible to find unclaimed resources.
last notes:
1. cycle length is not likely a whole number in seconds
2. I did not reveal everything such as the method here, and will not for obvious reasons
3. my method is hardly perfect anyways, it's very hard adapting to cycle changes but I can boast of 9 finds with no misses in between and 80+ finds with up to 2 misses in between.
4. I won't prove my method to anyone for the same reason that I won't share my method, as there is some information that I've refrained from divulging that would become obvious should someone witness my method.
5. Cycle lengths are short, under 1 minute for certain.
6. Finding out the system is fun, and I am still having fun refining my system.
7. This theory can also be applied to hunting, but cycles changes occur much faster there so hunt with teammates on the same mob to kill and loot very quickly might help reveal the cycles because it gives more data points
8. Accuracy to the 0.1 seconds is enough (and necessary)
9. I don't do this for profit, but for the challenge of solving a puzzle. I won't say no to profit though
__My main assumption is that deposit locations are not pre-generated and stored in a database (when a bomb is droped, a search is done in the database to find if there are any deposits within the finder radius).
__First, what is a cycle? Well, a cycle is just a word that I thought was fitting to describe this phenomenon.
For those with a math background, what I think governs mining loot is a sine (or cosine, doesn't really matter which) function based on time. It goes from 1 to 0 to -1 to 0 then back to 1 every X seconds. If you drop a bomb/probe at a time Y, and sin(Y) is between 0.5 or 1 (I call this the loot window, more on this later), then you find a claim. Why is this a cycle then? Because sin(Y)=sin(Y+X)=sin(Y+2X)=sin(Y+3X) etc etc... so if you know X, and get a hit on Y, then you can get hit after hit after hit.
To put this more simply, say the cycle is 12 seconds long.
1, 2, 3, 4, 5, 6, 7, 8, 9, [10, 11, 12], 1, 2, 3, 4, 5, 6... etc
__Then if you drop a bomb/probe on seconds 1-10, you don't hit anything. If you drop on seconds 10-12 (the loot window), you find something. Sounds good?
__The difficulty then lies in finding what the total length of the cycle is, (I used 12 seconds just as an example). Finding the length requires a very methodical method. To make it worse, the cycle length is always changing (albet slowly), so you have to adapt on the fly.
__Why do I think this is a reasonable method for MA to control loot? That is because of the loot window. The loot window size, like in the example above, determines the returns of a person dropping randomly. If the window was [6 to 12], then that person would hit alot and profit, if the window was [11 to 12], then that person is gonna cry if he persists too long. With the change of a single number, by making the window [0 to 1] or [.5 to 1] or [.75 to 1], MA can create good and bad periods with no formula changes or server load.
__To penalize overlapping, every drop is blacklisted by the server (it records the drop location, the radius from that location of the find, or the radius of the finder if nothing is found). Dropping too close to a blacklisted spot results in a penality (for example, a window of [0,1] becomes [0.75,1]), with the penality being more severe the closer you are to a blacklisted spot. Blacklisted spots expire after a period of time and there is a limit to the list of blacklisted spots - when full, a new one will bump the oldest one off.
__Finds that are generated by a drop are stored in a database, making it possible to find unclaimed resources.
last notes:
1. cycle length is not likely a whole number in seconds
2. I did not reveal everything such as the method here, and will not for obvious reasons
3. my method is hardly perfect anyways, it's very hard adapting to cycle changes but I can boast of 9 finds with no misses in between and 80+ finds with up to 2 misses in between.
4. I won't prove my method to anyone for the same reason that I won't share my method, as there is some information that I've refrained from divulging that would become obvious should someone witness my method.
5. Cycle lengths are short, under 1 minute for certain.
6. Finding out the system is fun, and I am still having fun refining my system.
7. This theory can also be applied to hunting, but cycles changes occur much faster there so hunt with teammates on the same mob to kill and loot very quickly might help reveal the cycles because it gives more data points
8. Accuracy to the 0.1 seconds is enough (and necessary)
9. I don't do this for profit, but for the challenge of solving a puzzle. I won't say no to profit though
Last edited: