Page 5 of 12 First ... 34567 ... Last
  1. #61
    Quote Originally Posted by Syn3rgy View Post
    I got curious after reading this, so i as well did the same experiment you did, 100 trys at 50% pass rate and failed 83% of the time, so i did it again, 100 tryed at 50% pass rate and failed 79% of the time... Something not adding up here....

    Regardless if its left to chance or not, weve done 3 tests now and havent gotten any better than a 79%, i will try again once i get more power cells, would be nice if gree chimed in on this.... Dont like spending my money on something that states is a 50% chance which is really about a 20% chance, theres a word for that... Its called false advertisement
    The #'s have always been bogus

  2. #62
    Lolsies's Avatar
    Join Date
    Dec 2013
    Posts
    729
    I think the experimental glitch is happening because the 1/2 probability is adding up.

    Example: If you fail two times-
    1/2 x 1/2= 1/4

    So you now have 3/4 chances of failing and 1/4 chance of succeeding.

    I'm not sure, just a guess on how this thing works.

  3. #63
    Has there been a recent update? I saw in message in wc saying the server was going to down for maintenance for an update of new buildings and new units? The server went down for a 20 minutes and came back up... but didnt see any changed?

  4. #64
    Drifting death's Avatar
    Join Date
    Jul 2013
    Location
    Death to Ming
    Posts
    452
    Quote Originally Posted by Lolsies View Post
    I think the experimental glitch is happening because the 1/2 probability is adding up.

    Example: If you fail two times-
    1/2 x 1/2= 1/4

    So you now have 3/4 chances of failing and 1/4 chance of succeeding.

    I'm not sure, just a guess on how this thing works.
    You could be right, but if the odds are advertised as 50% they should not be doing that. Slot machines are luck rolling machines with odds built in they are designed to pay at a set frequency.
    That after a set amount of money is spent the machine increases the odds of a win so to maintain its payout odds set by law.
    It looks like Gree needs to tweak the pay out system so the win fail ratio match the odds advertised it would require them to log player attempts and make sure by tweaking the odds, that the player gets the advertised odds on a 100 try average. So on average they win approx 50% out of every 100 attempts. At the 50% level.

    The law of averages says if you flip a coin a million times the results of heads or tails will approximately be very close to 50:50, the problem is when you only flip it 10 times you don't get as close a split but it would still be closer than the odds players are getting in the game and over a hundred flips again things should average out nearer the 50:50 split.

  5. #65
    Lolsies's Avatar
    Join Date
    Dec 2013
    Posts
    729
    Quote Originally Posted by Drifting death View Post
    You could be right, but if the odds are advertised as 50% they should not be doing that. Slot machines are luck rolling machines with odds built in they are designed to pay at a set frequency.
    That after a set amount of money is spent the machine increases the odds of a win so to maintain its payout odds set by law.
    It looks like Gree needs to tweak the pay out system so the win fail ratio match the odds advertised it would require them to log player attempts and make sure by tweaking the odds, that the player gets the advertised odds on a 100 try average. So on average they win approx 50% out of every 100 attempts. At the 50% level.

    The law of averages says if you flip a coin a million times the results of heads or tails will approximately be very close to 50:50, the problem is when you only flip it 10 times you don't get as close a split but it would still be closer than the odds players are getting in the game and over a hundred flips again things should average out nearer the 50:50 split.
    I have updated this. Thanks to everyone for raising this point, very interesting.

  6. #66
    Agreed I have failed going from a 10 to 11, 9 times now, and going from 11 to 12, six time. All on the same one. All I total I have tried the same exp over 800 times. Sometimes failing at the 95% 3 or more times in a row. All I can say, is that it is a good thing they don't go I to negative numbers lol

  7. #67
    Lolsies's Avatar
    Join Date
    Dec 2013
    Posts
    729
    Quote Originally Posted by The_Fortum View Post
    Agreed I have failed going from a 10 to 11, 9 times now, and going from 11 to 12, six time. All on the same one. All I total I have tried the same exp over 800 times. Sometimes failing at the 95% 3 or more times in a row. All I can say, is that it is a good thing they don't go I to negative numbers lol
    One person in my alliance started a new account in another world. He tried to upgrade a level 1 three times with 10 power cells each. All attempts failed haha.

  8. #68

    watch battle

    Wondering when the watch battle will happen been waiting forever

  9. #69
    Watch battle? Did they disable it? It has always worked for me

  10. #70
    Drifting death's Avatar
    Join Date
    Jul 2013
    Location
    Death to Ming
    Posts
    452

    Commanders

    Quote Originally Posted by Lolsies View Post
    I have updated this. Thanks to everyone for raising this point, very interesting.
    They can also do the same for commanders as they seem to drop very badly far poorer than odds would normaly give if you, take elites commanders with 50% chance of getting an epic the game should log your attempts, on the same 100 try average system is used, so you get the advertised chance of them dropping.

    If the game logged every time a player gets a commander which it does, at that point it should generate what the commander is based on the last 100 drawn odds system.
    That way gree can garantee the odds but still give random commanders. If a player is due an epic then the game picks at random an epic so that is linked to that elite commander he has won or purchased the game only rolls the chance as the player gets the commanders. so the commander the player has in there inventory are fixed before he gets them the commander he gets is pre set. So when they click on it the system just loads it into the players active list the commanders could sit in a players name in a non active commander list so opening the commander just moves it from non active to active.
    The game would need a list for every type of commander so junior senior elite and any in between but the commanders are pick up front they would only take up one space each in a players inventory as they are not a stack of unknown commanders from different pots of commanders but just a list of non active commanders in each category.

  11. #71
    Odds are easy to do, no need to log attempts. Every time a person trys, say an exp at 50%, generate a random number for the player. Then generate 50 random numbers out of 100, if the players number matches one of the 50, the exp passes.

    If my math is off please fell free to correct me, not enough coffee yet

  12. #72
    Lolsies's Avatar
    Join Date
    Dec 2013
    Posts
    729
    Quote Originally Posted by The_Fortum View Post
    Odds are easy to do, no need to log attempts. Every time a person trys, say an exp at 50%, generate a random number for the player. Then generate 50 random numbers out of 100, if the players number matches one of the 50, the exp passes.

    If my math is off please fell free to correct me, not enough coffee yet
    Coffee is not needed. Your explanation to the problem does look correct.

  13. #73
    Drifting death's Avatar
    Join Date
    Jul 2013
    Location
    Death to Ming
    Posts
    452
    Quote Originally Posted by The_Fortum View Post
    Odds are easy to do, no need to log attempts. Every time a person trys, say an exp at 50%, generate a random number for the player. Then generate 50 random numbers out of 100, if the players number matches one of the 50, the exp passes.

    If my math is off please fell free to correct me, not enough coffee yet
    Hi nice solution, does that work for items with multiple odds like commander items? and is it self correcting so to guarantee that the odds follow a statistical average, like slot machines do to garantee the correct odds are payed out, I don't think a simple maths check can.

    I may be wrong but, any maths check would have to disregard a results that were doubled up or had numbers that came out more than 2 times as it shows the numbers are not as random as they should be. If I rolled a 100 sided dice 50 times the odds are I would not get the same roll twice so looking for results that do to verify my result is random seems odd in a way it should be the other way round my result should not be repeated to show its random. If it is then re roll but this may require the server to run the roll multi times before getting a random result but this on its own still doesn't garantee the results are following the odds. And may require a lot more server power than a simple loged results ratio check to keep the results honest.
    I see no method of electronicly making random that's truly random and is sticking close to statistical averages with out checks to verify it. Hence some type of log and ratio check so why go to the effort of trying to generate a random number when the goal is to give results that follow odds just make the game check results so they are within 5% margin of error correct. If the result makes it greater than that, its disregarded so items with say 30% chance of epic if it gets a result that pushes it to only 24% of the results so far it throws it out. Or automaticly picks epic. Doing 1 calc rather than risking having to do hundreds to get a result seems more sensible use of server power.
    Last edited by Drifting death; 01-31-2015 at 01:25 AM.

  14. #74

    Join Date
    Sep 2012
    Posts
    261
    I wiped out all enemy units in the cross server and lost... wtf.

  15. #75

    Join Date
    Apr 2014
    Posts
    110
    Exps are def doable. I have all 12s in w27 which is a newer world compared to most without any coin. I had to hit a trillion npc 24/27s to get PCs but thats besides the point. Things I noticed are odds seem to change depending on event and whether youre attacking or not. I never tried exps in war events, or when I was attacking someone. I almost always failed. Arms races seemed to give me the best odds (much closer to 50%). I got 3 exps from 11 to 12 in consecutive attempts during an arms race. I never try to make any exp with a 50% chance jump 2 levels because your chances decrease. For example to go from level 8 to 9 is 50%, and to go from 9 to 10 is 50%. Therefore to go from 8 to 10 consecutively, you have a 25% chance. To go from 8 to 11 is worse... Yes I agree if you fail 1 at 50% you should theoretically pass another. If I failed one 50% I tried another 50% right after. This seemed to work a lot better for me. It is absolutely better than just spamming the upgrade button and blowing through 1k PCs on 1 exp.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

Log in