What's new

Gem Simulator

I'd love a 100% accurate simulation so I can get a very good idea of how much honor I need to save up to fully gem four 6-7 slot items. As of right now, depending on how far the rates are off on this simulator... I could be widely mistaken on how much I'll need.
 
upload_2018-2-22_16-47-6.png


Just so it stays on topic.
If the rates are less than 50%, then you'd be right. This simulator isn't accurate since it calculates at 50% per socket.
Funny thing is, it still takes 130 gems for me, even at a coin-flip per slot.
So you mean to tell me that your system is either 50% per slot, or that it increases the further down you go?
Because that is the only way your system would be better than this simulation.
 
Just saying here that this is not a proper simulator as this one breaks charms after only one is slotted. In our version the first 2 slots are safe!
We also don't know if the rates HK used for gems 3-7 are the same since you won't release the numbers because rng.
 
Don't forget that you receive twice more honor. That means it's already twice easier for you, even if we forget about the increased success chances we already have.
 
I don't know what are the chances on Trion or here, but believe original chances are the same. In such a case, as we have 2 differences on this server already - 2 slots with 100% chance + 2x honor --> as a result we get, all 7 slots became 4 times easier than on off. That are much better chances than I even expected, to be honest.
 
its 50% the whole way based on the numbers i have gathered from socketing
with 2 free gems to begin with, the rates are as follows
1 socket: 1 gem
2 socket: 2 gems
3 socket: 6 gems
4 socket: 14 gems
5 socket: 30 gems
6 socket: 62 gems
7 socket: 126 gems
 
Don't forget that you receive twice more honor. That means it's already twice easier for you, even if we forget about the increased success chances we already have.
Your vague answers only reinforce the notion our rates are lower than the expected 50% per gem slot.




319 Gems used.
245630 Honor spent

Jaylong similator
 
Again the problem for me is not so much the average honor cost as it is the variability in the cost due to RNG. If the average cost was the same but we could reduce variability it would be better in my opinion .
 
Your vague answers only reinforce the notion our rates are lower than the expected 50% per gem slot.




319 Gems used.
245630 Honor spent

Jaylong similator
I agree, I find it bullshit that one of the biggest issues people have had with AA since launch was not knowing your chance to fail or not fail. Now that we are on a private server and 99% of the playerbase on said server wants to know the rates they still wont tell us.

They claim things like still better than trion.... yes in certain aspects, but u cant claim to be better than a "competitor" yet still do the same dumb shit they do....
 
first 1 100%, break on failure, 50% chance, 770 honor
Average: 190.52198
Variance: 33886.29549983478
STD.Dev : 184.08230631930593
Honor: 146701.9246


first 2 100%, break on failure, 50% chance, 770 honor

Average: 126.88467
Variance: 14573.798726977979
STD.Dev : 120.72198940946086
Honor: 97701.1959

first 3 100%, break on failure, 50% chance, 770 honor
Average: 78.85918
Variance: 5369.806667794798
STD.Dev : 73.27896470198523
Honor: 60721.5686

first 1 100%, no break on failure, 50% chance, 770 honor
Average: 13.9955
Variance: 11.96439939399548
STD.Dev : 3.458959293486334
Honor: 10776.535


first 1 100%, no break on failure, 50% chance, 4000 honor
Average: 14.00458
Variance: 12.08881991179821
STD.Dev : 3.476898030112216
Honor: 56018.32


first 1 100%, no break on failure, 25% chance, 3000 honor
Average: 26.03599
Variance: 72.14341615404493
STD.Dev : 8.493728048039031
Honor: 78107.97


first 1 100%, no break on failure, 20% chance, 2000 honor
Average: 31.96853
Variance: 119.78349747410421
STD.Dev : 10.9445647457587
Honor: 63937.060000000005

In the spirit of discussion, I tried some other test cases to get the averages. Overall, it is easy to manipulate the average honor cost but the key difference between breaking all gems on failure and preserving gems is you can control the standard deviation of the chances (reduce RNG). While that's obvious, I thought numbers would illustrate it better.
 
Still doesn't help people that much, RNG is a pain in the ass even with 2 safe slots, I couldn' t get passed 4 on 150k+ honor... Please explain that Ariess ;)

You'd be surprised how much it actually does help. Not as much as we would have liked, but it does. 1 gem at 100% means that you don't have to get that second slot based on RNG.
 
Back
Top