Randomness of Rollable Tables

Hey Goblintrain, The way that rollable tables work is that each table item's ID is loaded into an array. Then we choose a random number up to the number of items in the array, and that's the item that is chosen. When you use the weighting, it simply creates a bunch of "duplicate" entries in that array. So functionally speaking, a table with 10 items each with a weight of 1 and a table with 5 items each with a weight of 2 are equivalent. So you could equally weight each item with a weight of 10,000, but you're just creating more work for the computer to do the roll and the outcome should be the same. Regarding that thread the other day, at this point there's an easy way to test to see if rollable tables are random. Create a a simple table with 5 items, and weight each item with a weight of 10,000. Then do 10 or 20 rolls from the table and you should see that you get results from the whole table. As I said, this would be equivalent functionally to manually creating a table with 50,000 items in it, it's just much easier to see the distribution of the results this way. I've done a complete review of the code on the client side and I did not find any bugs or issues which were causing random tables to not be random. My guess is that if you do more than a few dozen rolls on your table, you will see results from the whole spectrum of possibilities. Sometimes randomness does not mean "equal distribution" when the sample size is small. That said, we did find a bug in the API which was affecting its ability to roll tables with more than around 9,000 entries (or weights that in total exceeded 9,000), but it was only affecting the API. We have since pushed out a fix for that bug as well, so all should be good to go. Let me know if that helps clear it up for you.