<pre>6/28/2019 at 3:08
Tiur, the Gnosis
Everyone
June Refresh
The refresh is here! Why'd it take so long? I diverted my goals, as I'll explain below.
Two new coffer types
Shadow
------
shadow_shield
flair_shadow
obsidian chrysalis
carckling storm cloud minipet
muffling vox
shadow phylactery - NEW!
Spirit
------
spirit_shield
flair_light
crystal statue chrysalis
a fragile glass frog minipet
enriching vox
spirit phylactery - NEW!
Pity - Salvage
----------------------------------------------------------------
One of my greatest frustrations with balancing the prizes is the standard deviation. Say I want a given prizebox to be worth 30 credits, but I want it to have a chance of giving out awesome 2000cr prizes... to counteract that, I need to add prizes worth 5cr, and adjust the odds so that the average is 30cr. Sometimes this means the odds for the awesome stuff are stupid low, sometimes it means the odds for the low stuff is too high... but in math terms, the standard deviation is always too high.
Finally, I have an idea to fix it. Starting with these last few days of the coffers, I'm adding a pity prize.
Every time you open a coffer, if your total value is less than the average by the pity prize value or more, you get the pity prize. For example, if you've won 1000cr but odds say you should have won 1200, as long as the pity prize is worth less than 200cr, you get one (only one). Next chest you open, if that's still the case, you'll get another one.
The minimum number of chests to start rolling the pity prize is 10. If I set it any lower than that, it would be too obnoxious to balance AGAIN.
Essentially, I can now add prizes that specifically only go to people with bad luck.
-T
PS. We're starting high here, you need a 200cr deficit to see a piece of salvage via pity.
Penned by my hand on Gosday, the 19th of Midsummer, in the year 481 MA.</pre>
0
Comments
I've got an unreasonable amount of morphic metal however!
Every coffer is a different table, so you need 10 on a specific element before the pity prizes open up. Which, with 6 types of tables, maybe I made an oopsie there. Otherwise you're beating averages, so it wouldn't have given you one anyway!