Talk:Brain-meltingly-hot chicken wings

Received 2 adv, 21 str, and 27 ench from a single consumption. --Magered 03:06, 4 July 2007 (CDT)

You now take hot damage when eating these, 5-7 from what I've seen. —Preceding unsigned comment added by Seras (talk • contribs) on 13:51, 8 February 2008

Just ate 15 on a multi and got the range as 5-8. How many wings should I eat before removing the "needs spading" note? --CheezyBob 14:34, 2 March 2008 (CST)

the general "sure" figure is generally an order of magnitude bigger than that. try 100, 1000 would be better. --Evilkolbot 15:06, 2 March 2008 (CST)

But isn't that for things that we want distribution knowledge of to a tenth of a percent (like drop rates)? It's very rare when a food/booze has it's actual distribution spaded out with precision; or at least rare when such information gets posted to the wiki, especially like Yiab does monster drops. We just want to know the minimum and maximum, not a precise mean. --Flargen 15:16, 2 March 2008 (CST)

Assuming a flat distribution, then how likely is it that we observe a range of Y when the real range is X? For each datapoint, the odds are Y/X. So for N datapoints, the odds are (Y/X)^N.

The hardest to detect scenario is when X is just 1 greater than Y, so we want N large enough that (Y/[Y+1])^N is small, say about 1%. We can solve and find that we want:

           2 
 N >  -------------
      log10(1+1/Y)

In this particular case, Y is 4, so we need N>20 for 99% certainty. 15 gives us 1-(4/5)¹⁵=97% certainty, though. Note that finding food adventure ranges was much harder, because the distribution wasn't flat at the ends. (Getting the min might be only 10% as likely as any other value, for instance.) That mechanic doesn't seem to be favored much in NS13, though. --Starwed 15:54, 2 March 2008 (CST)

• alrighty then I'll keep at it with that multi for a few more days and if I fail to see any different range, I'll remove the needs spading note. --CheezyBob 16:20, 2 March 2008 (CST)