Orb of Chance Community Log
Updated
@dokkich: Topaz Rings currently have no (known) Unique form. If needed, you can check the Unique Compendium for a list of the possible uniques. ;) |
![]() |
btw i would give a ratio like "5:27" instead of "1:5.4".
If the the ratio is adjusted to the lowest probability,that would be 1/27 = 3.7% for uniques and 21/27 = 77.8% for magic items which fits amazingly well to 1418/1814 = 78.2%. And it also fits to my earlier calculations. But again it doesnt fit to the gathered data.... we just dont get enough uniques. Edit: Lets take a closer look at the rares: If the probability of getting a rare is 1/5.4 = 0.1852, the standard error is 1814*sqrt((0.1852*(1-0.1852))/1814)=16.54 The 99% confidence interval: 0.1852*1814 - 2.58*16.54 = 293.28 0.1852*1814 + 2.58*16.54 = 378.63 If the gathered data is right, the 1/5.4 probability of getting a rare is outdated/wrong with more than 99% certainty because the 391 rares we got are not inside this interval. It would help me a lot if someone from ggg could confirm this ;) In consequence all my earlier calculations are wrong because i relied on this value. If we had around 10 uniques at 3600 orbs used we could estimate the unique probability because we could assume that the uniques are approximately normally distributed (n*P*Q > 9). So keep going :D Last edited by jaankoo#1089 on Sep 26, 2012, 9:34:37 AM
|
![]() |
The math hurts my brain. :(
Also, just like I said in the first post, it's not likely we'll get entirely accurate data here especially with the unique rate. If every person in the game posted their every chance orb, we might get a bit closer, but there's no way that's going to happen. As it is, a very small minority come here to post and an even smaller minority comes only to post if they get a unique (something I've explicitly asked not to, but people will do it anyways). The numbers here are fun to keep track of and someday might come close to accurate, but as it is, we're nowhere near that point right now. ;) |
![]() |
Got 1 rare and 4 magic on Onyx amulets.
|
![]() |
" Im just doing the math because im curious how accurate we can become. For magic an rare items we have more than enough data to get valid results.(i still claim that 1:5.4 is the wrong probability for rare items) But, as you said, for the unique rate we need much more data. The probability for uniques is just so small. I can say that the probability for rare items lies between 19.66% and 23.45% with 95% certainty. The probability for magic items lies between 76.27% and 80.07% with 95% certainty. Last edited by jaankoo#1089 on Sep 26, 2012, 6:03:59 PM
|
![]() |
Based on some of the Dev posts I've read, I have a strong feeling that tracking Unique rates won't matter. It sounds like some Uniques ("build enabling" as they've been referred to) have a better chance of dropping. The drop rate is supposedly the same as the Orb of Chance rate, so that'd mean that each unique may have a different probability. In any case, I have no way of knowing this for sure unless the devs want to reveal more info - which they don't seem to want to do. :)
As for the Rare rate, I've personally used ~300-400 Orbs of Chance and added them to this log (with one Unique from those). I seemed to get Rares on a 1/5 basis, but that's just gut feelings not hard math. My results are mixed in the log on the first post along with the rest of the community's results, so I assume just use those for data assessments like yours. Last edited by MonopolyLegend#6284 on Sep 26, 2012, 6:42:06 PM
|
![]() |
14 orbs on gold rings, sapphire rings and onyx amulets.
2 rare, 12 magic. IGN: N_Bohr
| |
Used 12 chance orbs on a dusk blade in the hope of getting the %50 ES increase. So far:
Magic:10 Rare:2 "Minions of your minions are your minion's minions, not your minions." - Mark
|
![]() |
2 chain belts
1 magic, 1 rare |
![]() |
4 rare, 2 magic. Got lucky
RIP Bolto
|
![]() |