Going at a rate of 40k per attempt(for simplicity), nam's efforts would equate to approx 6,200 attempts* (little more but i like to keep this simple).
a 1% chance would on average take 100 attempts to be successful, if we assumed he will get one in the next 100 attempts (or roughly 6300-1) it would be around a .0015% chance. Any roulette wheel that ran at that kind of rate would get promptly shut down by the authorities.

*Note, for the roulette analogy, comparing to 1 spin is a total fallacy, it should be compared to a lifetime of betting the same number that is generally expected to hit at a rate of approximately 39-1. Which at 6000 times would expect to have 150 "wins" or discoveries, so to complete the analogy it would be talking about winning 0 times instead of 150, which any mathematician would say is statistically significant (although not automatically conclusive).

Not only is Nam's question legitimate here, but it should be getting echoed a thousand times over. I don't expect to know the exact odds of something when Dan deliberately does not reveal them, but I also don't think it unreasonable to ask if the odds or better or worse then 1/5,000

C'mon Namalak, you don't seriously believe Dan has it in for you and isn't letting you get a discovery. You've been playing this game long enough to know that Dan doesn't give a s@$t about you, or any 1 particular person in the game. Everyone gets treated the same. Bad luck is just bad luck, and incredibly bad luck is just incredibly bad luck.

C'mon Namalak, you don't seriously believe Dan has it in for you and isn't letting you get a discovery. You've been playing this game long enough to know that Dan doesn't give a s@$t about you, or any 1 particular person in the game. Everyone gets treated the same. Bad luck is just bad luck, and incredibly bad luck is just incredibly bad luck.

Actually I'd rather have the CT Lab items have a high but known total cost. The "discoverer" can be the person who put the most RP but whatever it hardly matters who. This way it still remains a legion effort but as we add more items and it becomes "more" impossible (?) or let's says 0.0001 to like 0.00001 or whatever chances, well I'd rather just have the cost be the average of the cost per roll times the current probability - even if it's like 100m RP or more - otherwise as noted some legions will randomly get it pretty quick while others seem to never be able to find it - the "deviation" is high enough with the latest items that I think it needs to be re-thought but I agree with Nam's assertion that it's pretty ridiculous.


As for the RNG... http://www.php.net/manual/en/function.mt-rand.php .

Many random number generators of older libcs have dubious or unknown characteristics and are slow. By default, PHP uses the libc random number generator with the rand() function. The mt_rand() function is a drop-in replacement for this. It uses a random number generator with known characteristics using the » Mersenne Twister, which will produce random numbers four times faster than what the average libc rand() provides.

Also:

http://tjl.co/blog/code/followup-php-rand-vs-mt_rand/ , it's pretty clear visually that one isn't that "random" !

For heaven sakes replace rand() with mt_rand() now !!