Duamutef wrote:Premium may indeed make a difference; it may actually be what's causing the bug.
And yes, it's possible to get Fayzee. At least, it should to be. You can double check by opening "startingset.txt" in the card editor and checking her rarity; unless it's more than 100, you should be able to get her. Looking at the file I have, it looks like her rarity is 60, which is uncommon but not super rare.
Given the size of the starting set, I'm guessing you'd statistically have to buy about 700 cards before you'd have better than even odds of having any given rarity 60 card. (I don't know how to do the actual math; the way rarity works is it's basically a chance that you'll have to "re-roll" if the card gets picked. The game picks a random card, then generates a random number. If that number is below the card's rarity, it will pick another random card and try again. So if a card's rarity is 0, you will always get it if it's the first card picked. If it's rarity is 80, there is an 80% chance that the game will chuck it and go for a different card. Fayzee is 60, so she'll get hucked 60% of the time. However, every card that isn't Fayzee also has a chance of getting tossed, which gives you that many more chances to get her, so it's not a straight ratio...)
I looked at my blog today, noted that I had referred to this calculation somewhat ignorantly, and decided to start getting to grips with it. Showing my work in case anyone wants to check. Code used to calculate total commonality from file will be provided on request.
The "commonality", or percent chance that you will keep a card, varies from 2 to 100 and is 40 for Fayzee. There are a total of 326 cards in the starting set. Their total commonality is 24485 (average about 75 per card).
When the game generates a card, first there's a [LABEL 1] one in 326 chance of getting Fayzee, then a 40% chance of keeping her, and 60% chance of re-rolling, in which case there's a [GOTO LABEL 1].
This is an infinite sum. The intermediate form is: 1/326*0.4 + 1/326*0.6*1/326*0.4 + 1/326*0.6*1/326*0.6*1/326*0.4 + ...
Proper form for dealing with an infinite sum: 0.4 * 1/326 * (1 + (0.6*1/326) + (0.6*1/326)^2 + (0.6*1/326)^3 + ...)
= 0.4 * 1/326 * (1/(1-(0.6*1/326)))
= 0.4 * 1/326 * 1.00184388
= 0.00122925629 or about 1/813.5, which is close to the 1 in 700 estimated above. However, I have yet to account for the fact that about 25% of other cards will be rejected, which increases this chance to about 1 in 651. This last bit isn't quite rigorous because I don't know if I can use that average that way.