AI wiki: Energy Efficiency: Error

[Blasphemy]

Resource Ratio: Energy/Resource         
# of reactors   Mark 1   Mark 2   Mark 3
                  1   2500.0   2666.7   2000.0
                  2   2000.0   1600.0   400.0
                  3   1500.0   533.3   -1200.0
                  4   1000.0   -533.3   
                  5   500.0      
                  6   0.0      
Energy
Produced   :        5000   40000   80000
Resources
Used:                  2       15        40
Eff Reduction:      0.2      0.4      0.8

The wiki page says that the mark build order is 2, 1, 1, 2, 3.  As you can see it should be 2, 1, 1, 3, 2.  (Even though the resource ratio is the same for the second mark 1 reactor it is less expensive to build then the mark 3.)

http://www.arcengames.com/mediawiki/index.php?title=AI_War_-_Resource_Efficiency
Additionally if you wish to micromanage your energy production farther, a Mark II reactor is the most efficient resource-to-power wise, followed by two Mark Is, then a second Mark 2, then the Mark 3 (2, 1, 1, 2, 3). If you find yourself needing more power than this, its likely you've done something wrong (such as let your Viral Shredders replicate out of control).

... and yes, I probably have something better I could be doing with my time.

[BobTheJanitor]

I can go edit this in the wiki, but I am confused. In your Resource Ratio section, you lose me with the negative numbers. How can energy/resources equal a negative? Small numbers indicating poor return, sure. Negatives? My head explodes. It has been a decade or so since my last official math class though, so that could be the problem.

[Blasphemy]

I put those in so that you could see when there was no point in building anymore reactors.  That is why I stopped when the mark 1 hit zero.

I can send you an excel spreadsheet if you want to see the calculations.

[chemical_art]

I didn't think it was possible to get negative ratios because I thought power plants have a minimal efficiency of 20% (maybe 10%, certainly not zero otherwise dozens of MK II's on the homeplanets as backup would never work)

 

[Blasphemy]

The programs could have put a lower limit on production or it could be that subsequent reductions in efficiency are based off of the previous lvl of efficiency.

Let n be the number of mark 1 power plants.
Equation 1:  Production Eff = 5000*(1 - (n-1)*0.2)/2
Equation 2:  P2 = 5000*(1 - 0.2)/2, P3 = P2*(1 - 0.2)/2, ..., P(N+!) = PN*(1 - 0.2)/2

You could solve this recursive sequence and see if it has the desired lower limit.

Since the limit of this sequence is zero the developers must have put in a lower limit.

[Blasphemy]

You should note that since the wiki typo occurs after only one additional reactor both methods will produce the same results.

[BobTheJanitor]

I'm fairly certain that building a reactor will never produce negative energy, if that's what is being implied. It might have a certain percentage that it floors out at, or it might stop at 0, but I don't think the game will actually let you build a reactor that takes away your total energy. If it does, that would be a pretty amusing bug. Maybe Chris or Keith will stop by and let us know how it handles that.

[Ranakastrasz]

It is not -x% of original power generation. It is x% less than the previous power generation.

This means that the T3 generator generates 40k power, then 8k power, then 1.6k power. It does not suddenly use 24k power on the third reactor. while it is extremely inefficient, it does not suddenly start using power.

[BobTheJanitor]

Engineer: You see boss, we pour tons of metal and crystal into this reactor, then we plug it into the grid and it sucks up tons of juice!

Player: And then what? What does it do?

Engineer: Nothing! That's the beauty of it!

Player: *scrap engineer*

[Ranakastrasz]

Engineer: You see boss, we pour tons of metal and crystal into this reactor, then we plug it into the grid and it sucks up tons of juice!

Player: And then what? What does it do?

Engineer: Nothing! That's the beauty of it!

Player: *scrap engineer*

[KingIsaacLinksr]

Engineer: You see boss, we pour tons of metal and crystal into this reactor, then we plug it into the grid and it sucks up tons of juice!

Player: And then what? What does it do?

Engineer: Nothing! That's the beauty of it!

Player: *scrap engineer*

Win!

King

[Blasphemy]

The programs could have put a lower limit on production or it could be that subsequent reductions in efficiency are based off of the previous lvl of efficiency.

Let n be the number of mark 1 power plants.
Equation 1:  Production Eff = 5000*(1 - (n-1)*0.2)/2
Equation 2:  P2 = 5000*(1 - 0.2)/2, P3 = P2*(1 - 0.2)/2, ..., P(N+!) = PN*(1 - 0.2)/2

You could solve this recursive sequence and see if it has the desired lower limit.

Since the limit of this sequence is zero the developers must have put in a lower limit.

I put those in so that you could see when there was no point in building anymore reactors.  That is why I stopped when the mark 1 hit zero.

I can send you an excel spreadsheet if you want to see the calculations.

I'm fairly certain that building a reactor will never produce negative energy, if that's what is being implied. It might have a certain percentage that it floors out at, or it might stop at 0, but I don't think the game will actually let you build a reactor that takes away your total energy. If it does, that would be a pretty amusing bug. Maybe Chris or Keith will stop by and let us know how it handles that.

Please read posts before commenting.

[Blasphemy]

It is not -x% of original power generation. It is x% less than the previous power generation.

Then you are saying that the recursive algorithm I already posted is used with a minimum power production cap.

while it is extremely inefficient, it does not suddenly start using power.

Obviously.

[Ranakastrasz]

It is not -x% of original power generation. It is x% less than the previous power generation.

Then you are saying that the recursive algorithm I already posted is used with a minimum power production cap.

while it is extremely inefficient, it does not suddenly start using power.

Obviously.

I cant answer that unless you were to show the equation you actually used. it looks as though you keep subtracting 80% of the original efficiency for mark 3 reactors, or at least that is the obvious way you would manage to get negative power generation.

[BobTheJanitor]

Please read posts before commenting.

I obviously never do that. What do you think of the elephant?