The definition of 25%

VelvetStrike - Illyfue wrote: »

25% usually means 1 out of 4 or 1/4 or for every 100 combines 25 succeed.

I just combined over 400 gems lv1 without fusion agent (80*25%) 2 succeeded in total.



so yeah, i juSt leave those probability chances there.

MIGHT WANT TO FIX UR DAMN RNG.

why don't u fuse 4 gems without fa for 75% chance o.o

serenity234876 wrote: »

why don't u fuse 4 gems without fa for 75% chance o.o

cause with 4 only you can not try without fa

im 40% sure the sign affects what kind of ID you get too

Isock - Eyrda wrote: »

im 40% sure the sign affects what kind of ID you get too

When you pray under your sign you get more exp for the first 4 prays,a higher chance to get items from pray,buffs after the first 4 prays,everything else is just a myth.

i m not supersticous...it gives bad luck

The level of understanding of randomness in this discussion is below the levels of statistical likelihoods.

Some random facts to consider:

1. It's highly unlikely that developers of FW, who can put together 3D worlds with hundreds of players simultaneously interacting with each other and the environment, can **** up using two of the most simple C library functions. srand() which is called at program start to seed the random generator, and rand() to generate a random number. Remember most the ****-ups in FW are due to the incompetence of PWE who run it, not the developers / game itself.

2. It's highly unlikely that the developers of FW would choose to invent their own random number generators, when there are perfectly adequate library functions to generate random numbers. However, if for some reason they didn't want to use the library functions, there are at least two other methods that they could use (and probably would, again before writing their own from scratch). They could use the operating system's built-in random number generator - /dev/urandom on Linux - or use a public domain Mersenne Twister algorithm. Both of which produce random numbers of adequate quality for an MMO game.

3. The statement that other MMOs use time as a factor in their random number generation is misleading. Many programmers choose to seed the random number generator with the time of the call to srand(), ie, by calling srand(time(0)); . This is useful because programs then will generate a different sequence of numbers each time they're executed. HOWEVER, the time that a roll (call to rand()) is made has NO EFFECT whatsoever on the result. It's a fixed sequence for each seed value. Again, this is perfectly adequate for an MMO game because a) you have no sight on the actual numbers being generated, b) you have no sight on how many random numbers are generated by other people, and c) cryptographic levels of security are not required. (As an aside, I doubt anybody would even notice if FW failed to call the srand() function and left the RNG uninitialised. There's sufficient entropy in the user input that it would effectively mask the fact that the raw random data didn't change for each time the server was started).

4. Your sample sizes are too low. In every case I've seen in this thread, too low. In fact, with a 25% probability of success, the number of successes with N rolls only tends to N/4 as N tends to infinity. (That "tends to" is a maths term with a very specific meaning).

5. 25% is 25% each roll. It does not matter how many times you've previously rolled, the odds of the next roll succeeding are STILL only 25%.

6. If you have already done, say, 3 rolls, and have had 3 failures, then the laws of probability say that you will most-likely now NEVER get your N/4 successes for N total rolls, including the 3 fails, because these 3 fails factor in as previous known values; all future values are still unknown with a probability of 25%.

This doesn't mean that FW isn't suspect in its use of random probabilities though. Just more likely something like this is an error in the description, rather than the fault of the generator. That or you simply were an unlucky assassin on this occasion!

I saved up about 800 azures, 600 solars, and 500 mists or so to combine without FAs. My end results was just over 25%, so I'd say you just got unlucky.