Legends of Elveron
Forums for Legends of Elveron

Home Guide Forum Archive
FAQ SearchRegisterLogin



Post new topic Reply to topic  [ 118 posts ]  Go to page Previous  1 ... 4, 5, 6, 7, 8

All times are UTC + 1 hour [ DST ]

Author Message
 Post subject: Re: Formulas
PostPosted: Fri Jun 08, 2012 14:19 
Offline
Site Admin
User avatar

Joined: Fri Mar 12, 2004 01:04
Posts: 2162
Location: Sweden
I don't know, maybe you could use this info.
http://www.elveron.com/elveron/game/dat ... ?races=yes
http://www.elveron.com/elveron/game/dat ... gnames=yes


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Wed Sep 26, 2012 13:45 
Offline

Joined: Wed Feb 24, 2010 09:41
Posts: 104
New KDimp formula:

KDimp bonus = M * ( 1 - exp (-X/(1400*max(L,500)+15000))

With M the max imp bonus, X the number of points invested and L the land.

Notice that the latest changes (removing Rimps) have doubled M from (15%,20%,10%,10%,10%) to (30%,40%,20%,20%,20%). The max() operator is also new. It only has an effect while you have less than 500 land, however, and will cause you to have slightly less than double the bonus you had before the change. When you're above 500 land, you effectively get a bonus twice as much as you would have had before the change.

Disclaimer: this formula has not been confirmed by Zip or anyone from the Dev team. I got it myself, by looking at the bonus for different M, L and X. Use at own risk.

Edit: min(L,500) has to be max(L,500), of course. Changed this.


Last edited by Rogerus on Wed Sep 26, 2012 13:55, edited 2 times in total.

Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Wed Sep 26, 2012 13:52 
Offline
User avatar

Joined: Wed Jul 06, 2005 02:17
Posts: 2758
Location: Finland
Rogerus wrote:
New KDimp formula:

KDimp bonus = M * ( 1 - exp (-X/(1400*min(L,500)+15000))

With M the max imp bonus, X the number of points invested and L the land.

Notice that the latest changes (removing Rimps) have doubled M from (15%,20%,10%,10%,10%) to (30%,40%,20%,20%,20%). The min() operator is also new. It only has an effect while you have less than 500 land, however, and will cause you to have slightly less than double the bonus you had before the change. When you're above 500 land, you effectively get a bonus twice as much as you would have had before the change.

Disclaimer: this formula has not been confirmed by Zip or anyone from the Dev team. I got it myself, by looking at the bonus for different M, L and X. Use at own risk.


Nice job at figuring it out quickly, too bad the new kimp formula is off by nearly half of what we agreed on. So don't crunch your strats until zip has fixed the formula.

I asked for the formula to be made public, dunno if we can do that. You'll figure it out pretty quickly anyway, so I see no reason not to tell it though.

_________________
www.ceekayed.com


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Wed Sep 26, 2012 14:00 
Offline

Joined: Wed Feb 24, 2010 09:41
Posts: 104
If you use close to the same formula as before, I will find it out quickly indeed :). I have to admit the max() operator was hard to get to, those non-standard functions are pretty annoying...

A logical and straightforward change, by the way, would be to make it:
KDimp bonus = M * ( 1 - exp (-0.5*X/(1400*L + 15000))
This being equivalent, of course, to M * ( 1 - exp (-*X/(2800*L + 30000))

That's kind of what I expected the formula to be. With M doubled, but the required number of points X also doubled. This would make it about the same as it was with Kimps and Rimps.
Taking efficiency of split Rimps/Kimps into account, you'd probably want something like -0.65*X. The current formula (displayed above) is a huge buff to KDimps. But I'm sure these things have come to the table in your discussions as well.


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Wed Sep 26, 2012 14:26 
Offline

Joined: Sat Jun 09, 2012 16:29
Posts: 67
It is fairly straightforward to figure out the new kimp formula. I got exactly the same as Rogerus and am 100% sure it is the correct formula. Anyway, the easiest way to fix it would be to double the constant from 1400 to 2800. It will scale very similarly to the old formula (slightly less needed), which is probably what you're looking for. The 15000 constant doesn't really matter, it becomes inconsequential when land size is large. The max(land,500) seems redundant as well.

EDIT: Anyway, I think this should be brought over to the Changes thread so it gets more attention.


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Fri Sep 28, 2012 18:34 
Offline
User avatar

Joined: Wed Apr 13, 2005 21:25
Posts: 1856
Location: Michigan
Hero Mortality =.95/sqrt(exp+16)

It's accurate within 0.02%.

Edit: For uncommon heroes, divide this number by 2.


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Mon Oct 15, 2012 16:30 
Offline

Joined: Wed Feb 24, 2010 09:41
Posts: 104
Does someone know if Hero Mortality is the same for relic quests as it is for regular attacks?
Also, is hero mortality for relic quests also reduced by academies (e.g. 0 mortality with >25% academies?) - I'm asking because acads don't increase XP gains on relic quests, based on their description.


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Sun Oct 21, 2012 16:19 
Offline

Joined: Wed Feb 24, 2010 09:41
Posts: 104
Corruption formula estimate:
100% * (1 - exp ( - land / 8415))

It's accurate up to .03% right now up to 2K land. I encourage others to find the exact formula. It might involve adding a constant to land.


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Mon Oct 22, 2012 20:28 
Offline
User avatar

Joined: Wed Apr 13, 2005 21:25
Posts: 1856
Location: Michigan
Rogerus wrote:
Corruption formula estimate:
100% * (1 - exp ( - land / 8415))

It's accurate up to .03% right now up to 2K land. I encourage others to find the exact formula. It might involve adding a constant to land.
It's a little bit more complex than that. . .

The actual corruption formula is something like:
    = A*atan((ld/B+C)*D)+E

rokushiki, also has a good estimate of the corruption formula. Maybe you could ask him for his list of data points?


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Sat Oct 27, 2012 16:07 
Offline
User avatar

Joined: Wed Apr 13, 2005 21:25
Posts: 1856
Location: Michigan
I got permission to post this:

corruption = 300 * (atan(totalLand / 8000 + 1) * 2 / PI - .5)

Not sure how close it still is.


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Wed Oct 31, 2012 10:04 
Offline
User avatar

Joined: Thu Mar 18, 2004 05:55
Posts: 2303
Location: Dorkland
Corruption for diamonds might starts at ~56.5x raw production. (WTF?)

_________________
We can't hear the truth
When we kneel down
In front of what we think
We cannot change


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Wed Oct 31, 2012 10:11 
Offline
User avatar

Joined: Wed Jul 06, 2005 02:17
Posts: 2758
Location: Finland
Corruption formula for secondary resources is now raw prod*36+100k. It was changed R8 last age.

_________________
www.ceekayed.com


Top
 Profile  
 
 Post subject: Re: Formulas
PostPosted: Tue Apr 30, 2013 08:34 
Offline
Site Admin
User avatar

Joined: Fri Mar 12, 2004 01:04
Posts: 2162
Location: Sweden
Explore formula, tL = total land, tU = tL + amount to explore
price = (int)(2.85714*(Math.pow(tU, 2.1) - Math.pow(tL, 2.1)) - 952.381*(Math.pow(tU, 1.05) - Math.pow(tL, 1.05)));
price = (int)(price*1.08*1.10);

Max afford formula in gold, tG is total gold this kingdom have.
long g = (int)((tG+85714.4) * (.35/(1.08*1.10)) - 333.33368 * Math.pow(tL,1.05) + Math.pow(tL,2.1));
int maxAcresGold = (int)(Math.pow(166.66666667d + Math.sqrt(g), .95238100d)-tL);


Top
 Profile  
 
Display posts from previous:  Sort by  

All times are UTC + 1 hour [ DST ]

Post new topic Reply to topic  [ 118 posts ]  Go to page Previous  1 ... 4, 5, 6, 7, 8

All times are UTC + 1 hour [ DST ]


Who is online

Users browsing this forum: Google[Bot] and 0 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  





Elveron - the online strategy game
Elveron phpBB3 style by Ulf Frisk and Michael Schaeffer
Copyright © Ulf Frisk, Michael Schaeffer 2007.
Powered by phpBB® Forum Software © phpBB Group