The exact cost of all Galaxy Gates (excluding ABG)

12/19/2016 18:53 Palladium<3#1
Intro:
Over the summer I made a little program which build hundreds of virtual gates per second, so with a large sample I am able to accurately determine the cost of any gate given the number of parts. Many thanks to [Only registered and activated users can see links. Click Here To Register...] for helping me with my first project. I hope this is useful in terms of planning for a few of you.

Assumptions:
  • Each spin has a chance of 13% to hit a part/multiplier (although when you spin with a full nano, it may be between 13-17, I do not know how much it actually is, depends on how BP programmed it. i.e the extra 4% might become ammo straight away).
  • That the cheapest method of building gates is to use multipliers straight away (x2) until it gets up to 6 parts away from being completed and then it will hit the multiplier.This program spins 1 spin at a time.
  • That my random number generation is truly random, I used a cryptographically secure pseudo-random number generator which I believe is 'random enough'.

Results:
Delta:
Number of gates built: 100,000
Minimum spins required: 1166
Maximum spins required: 3494
Sample standard deviation: 265
Average spins required: 1966
Epsilon:
Number of gates built: 100,000
Minimum spins required: 804
Maximum spins required: 2994
Sample standard deviation: 236
Average spins required: 1522
Zeta:
Number of gates built: 100,000
Minimum spins required: 907
Maximum spins required: 3298
Sample standard deviation: 248
Average spins required: 1706
Kappa:
Number of gates built: 100,000
Minimum spins required: 1042
Maximum spins required: 3514
Sample standard deviation: 258
Average spins required: 1844
Lambda:
Number of gates built: 100,000
Minimum spins required: 257
Maximum spins required: 1785
Sample standard deviation: 153
Average spins required: 683
Hades:
Number of gates built: 100,000
Minimum spins required: 257
Maximum spins required: 1673
Sample standard deviation: 153
Average spins required: 683
Kuiper:
Number of gates built: 100,000
Minimum spins required: 782
Maximum spins required: 2875
Sample standard deviation: 237
Average spins required: 1538
Code:
Written in C#, you can copy/paste this directly into a console application, if you want- although it's just a bunch of nested loops :P
12/19/2016 19:20 *-Linkin™-*ITA#2
Good job sir
12/19/2016 19:42 Skrux#3
Seems a bit higher than what I would expect, I'm probably getting super cheap gates somehow then xD

Good work anyway, appreciated!
12/19/2016 20:43 Řoβч966#4
Good, now if you're wrong you'll burn in hell <3 :D
12/19/2016 21:44 Snow❄️#5
With all do respect, this doesn't accurately prove anything. All you did was use samples to figure out averages and deviation. The title says "exact cost" , clickbait on Epvp :P

This is more like the average cost. Good job however, it still helps people know approx. How many is needed. However, these results are a little higher than the true average as multiplier tricks arent included.

Regards,

Snow
12/19/2016 22:01 ptipois#6
french players already make this little program for gates
[Only registered and activated users can see links. Click Here To Register...]
you can also see how many npc for uri you need, npc/pvp you need for next rank, a comparatif of npc/ship etc etc
12/19/2016 22:26 Palladium<3#7
Quote:
Originally Posted by *-Linkin™-*ITA View Post
Good job sir
thank you!
Quote:
Originally Posted by Skrux View Post
Seems a bit higher than what I would expect, I'm probably getting super cheap gates somehow then xD
Good work anyway, appreciated!
thanks! See assumption #1, this is all assuming you have an empty nanohull, I don't know what the % is with a full nanohull.
Quote:
Originally Posted by Řoβч966 View Post
Good, now if you're wrong you'll burn in hell <3 :D
Let's hope I'm not wrong then ;)... Source code is there for anyone in doubt.
Quote:
Originally Posted by Snowman- View Post
With all do respect, this doesn't accurately prove anything. All you did was use samples to figure out averages and deviation. The title says "exact cost" , clickbait on Epvp :P

This is more like the average cost. Good job however, it still helps people know approx. How many is needed. However, these results are a little higher than the true average as multiplier tricks arent included.

Regards,

Snow
Well, this is true of course but the reason why I think it IS appropriate given the assumptions that I have taken, is because of the very large sample.
For example- if I were to repeat this again I would get the same average value. This is because the sample s.d = (parent s.d)/root(N) where N is the number of data points.
For me N=100,000, so the s.d of the 'exact value' =0.0391 and this is approximately 0, hence why the title is 'exact'.

Sorry if you interpenetrated it as clickbait- not my intention :rolleyes:
12/19/2016 23:40 Arschwasserkanalpirat#8
;)
12/20/2016 00:23 Palladium<3#9
Quote:
Originally Posted by Arschwasserkanalpirat View Post
[Only registered and activated users can see links. Click Here To Register...]
What is this?
Where do these figures come from?
Why is the cost of Hades/Lambda different (both 45 parts)?
Why is Delta cheaper than Kappa?
Or am I missing something here?
12/20/2016 00:40 Snow❄️#10
Quote:

Well, this is true of course but the reason why I think it IS appropriate given the assumptions that I have taken, is because of the very large sample.
For example- if I were to repeat this again I would get the same average value. This is because the sample s.d = (parent s.d)/root(N) where N is the number of data points.
For me N=100,000, so the s.d of the 'exact value' =0.0391 and this is approximately 0, hence why the title is 'exact'.

Sorry if you interpenetrated it as clickbait- not my intention :rolleyes:
So you would have an exact average value. I got it. Well said reply btw. Thanks for the info

Regards,

Snow
12/20/2016 01:11 Arschwasserkanalpirat#11
Quote:
Originally Posted by Palladium<3 View Post
What is this?
Where do these figures come from?
Why is the cost of Hades/Lambda different (both 45 parts)?
Why is Delta cheaper than Kappa?
Or am I missing something here?
Donīt know why it doesnīt show the image.

Itīs from a thread out of the official forum.


The costs are the total costs - rewards ! So that it is what you have to pay
pay for each gate after getting the gate-reward and the uri from npc
12/20/2016 01:18 Arby#12
This picture is also quite a close approximation as well. Still, many thanks for using some good ol' math for these things

[Only registered and activated users can see links. Click Here To Register...]
12/20/2016 03:35 Snow❄️#13
Quote:
Originally Posted by Arby View Post
This picture is also quite a close approximation as well. Still, many thanks for using some good ol' math for these things

[Only registered and activated users can see links. Click Here To Register...]
Lmao where did you find this? I thought this was lost in time :P

Some1 posted here long ago, when all used to say use x2 multiplier only . Good work ARby
12/20/2016 16:38 MankarCamoran#14
Quote:
Originally Posted by Snowman- View Post
Lmao where did you find this? I thought this was lost in time :P

Some1 posted here long ago, when all used to say use x2 multiplier only . Good work ARby
Just beeing curious.. what do you say now? I still think x2 multiplier and 1 spin at a time xD
12/20/2016 19:47 Snow❄️#15
Quote:
Originally Posted by MankarCamoran View Post
Just beeing curious.. what do you say now? I still think x2 multiplier and 1 spin at a time xD
Yes, to save the most uri. That way is correct. However, if you don't wanna waste all day spinning gates, I usually do 5 spins at a time and (x2) multiplier. IF I have even more spins, and i dont care too much. I do 100 spins until im about 60% done, then switch back to 5 spins.