R4tt3xx
I want to believe
- Joined
- Apr 10, 2005
- Posts
- 2,176
- Location
- South Africa
- Society
- Freelancer
- Avatar Name
- Alexis Sky Greenstar
I agree, I will have to find a way to log that information, decide on a budget and compare it with other results.How are you testing your results? I mean, how do you determine if one approach is superior to another? Hit rate? Return with/without multipliers? Are you getting above the currently excepted avg. hitrate/return? By how much and what was the sample size? Giving that info would go along way to establish that your approach is doing something. Actually, even lower than random would show that your approach does something.
Can you share any of that?
I am thinking of pitting the different arrangements against each other. Say for example a large factor like 54 vs 0.618 using a md1..
The reason for using phi as a factor is the fact that both the distributions of the game levels and mining loot are moving averages and phi fits that.
I like the way that the circular lattice seems to be more accurate as it uses a lot more irrational numbers than the square lattice while using the same amount of points, it also does not go outside the lattice of the finder...
Last edited: