MArcomage

Maybe it's just a fluke, but I've been noticing in the past week or two, in different games, players tend to get the same (rare) cards at the same time.

I was playing a couple games a few days ago and I got the same rare card (New Era) in 3 different games, all within a card play of each other, and one of my opponents got it twice, too. No rare fetching cards were used. It just happened.

I wrote it off as a fluke, but just now two of my opponents just drew Centaur Trainers in about a 2 minute span.

Hm. This could be combined with the thread asking if Don't be hasty is bugged. Sounds like a problem not limited to any cards in particular, though certainly more noticeable when it happens with rare cards.

pretty sure the bug is present with dont be hasty. it just happens so often. there is a limit to luck

We are using standard PHP random generator and it passed our statistic tests. There doesn't seem to be a problem in our appliaction.

I always thought the randomisation algorithm uses time as a factor since I drew the same cards( of any rarity ) one the same day in differend games but not on differend days in the same games.

No, it shouldn't use time - not according to documentation.

would it be possible to include a programm that looks if the new drawn card is the same as one of the last three (or two) and if it is, another card will be drawn instead of this one and this one will be checked again.
with this it would be really impossible to draw the same card sooner then after 3 (2) draws.

but i think this programm should not be included when a card is drawn by card-draw-cards like the scepters or flare attack ;)

We have anti-flood protection program. It does this:

- makes impossible to draw a discarded card by normal discarding
- it reduces the chance of getting same card by (1/2)^(N) where N is the current number of the cards that are same as the next drawn card. So in other words for each instance of the same card as the next drawn card should be, chances of getting the same card decreases by 50% for each one (consecutive). It's applied at all times (all types of drawing)

Do you really mean "by" that value or "to" that value? If it's reduced -by- that value, it actually becomes more likely with each successive draw as the subtracted amount decreases. C-1/2 < C-1/4 < C-1/8, etc...

I assume you meant -to- ( C=1/2 > C=1/4 > C=1/8 etc...), but, well, what does the code actually reflect, I now wonder?

my guess is, that with this system, if you draw a card you already have, a 2 sided dice is thrown. if it is a 1 the card is drawn, if not, then not.
if the card is already 2 times in hand, a 4 sided dice is thrown etc.

but anyway, as we can see it somehow doesnt really work out with this system alone, does it?

the system i proposed wouldnt influence the chance to draw one of those cards, it would just prevent them from being drawn 2 times in a row. actually this programm would be kinda meaningless since there are always some luckyshots where someone can get the same rare card two times in a row. but the probability to get such luckyshots is waaaaaaay below the quantity of luckyshots we got so far.

if you still insist, that it is just luck, well im happy with that. then it seems like i am as lucky as a laprechaun
*me getz twin phoenix and doubble chinese wall^^*

Meh, I have no love for "fake random" algorithms, especially undocumented ones (and I consider the current system basically "undocumented" since it isn't mentioned in the manual, and the only way to find it is to look at the source code, which requires some rudimentary PhP or programming knowledge to navigate). It seems a bit deceptive to me.

On the other hand, if it's just a straight random draw, then you get people coming on the forums and complaining that the RNG is broken because they got the same card three times in a row. You almost have to introduce some order to make people believe it's random, ironically enough. I've noticed in many games that people who aren't familiar with probability or statistics seem to think any "streak" is a sign of a bug.

Anyways, anecdotally, I did notice the "don't be hasty" effect back when I had it in my deck. Obviously that's not statistically significant though.

The anti-flood program exponentially decreases the chance of getting same card multiple times, in short said.

Here is the documentation to the random generator we use http://sk.php.net/manual/en/function.mt-rand.php