Broken random generator or just bad luck?

[CumeriTarenes]

note on the first pic I even casted focus to increase my spell focus

Also you can see the mob hitting me very often in a row, it was very laggy this day, that's why. Maybe lag influences miss rate...

well, and the third pick...it was a marksman who was allone at this time



oh, and the last two pictures is just my little dwarf doing some climbing and acrobatic

[InsektO]

That's what i've been talking about with some of my realm mates lately (Reki & Denethor particularly). IMHO the random number generator seems to be (at least) not working correctly and should be checked... That could explain those incredible multiple blocks/evasion-in-a-row.

[urgit]

Two pictures:

Pic 1: My hunter using SoW evades only 1 attack (bad luck ?)
Pic 2: My conjurer evades 7 attacks (good luck ?)

Anyway I died in both cases

[LuthienNenharma]

I think that it is normal now.
Easy targets are able to evade very much spells. Mobs and enemy, always the same. Sometimes they resist 5 spells.

Only group hunting and lvling helps a bit.

[Angelwinged_Devil]

how can a "random" number generator be broken? it produces random numbers, try to perspectivize about it , I did

[Comp]

Originally Posted by Angelwinged_Devil

how can a "random" number generator be broken? it produces random numbers, try to perspectivize about it , I did

Actually if you are not careful with random number generators they really aren't as random as one would expect. Random number generators are based on a seed value (well, pseudo-random one's are). Two basic ways to generate random numbers are by measuring some phenomenon that is expected to be random and then compensate for some sort of possible biases that may exist in the process, and generating random numbers based on a seed or a key value. The random number generated when using a seed is only as good as the algorithm used to generate the number. Regardless, the randomness isn't always - so....random if you aren't expecting the same number to be generated back-to-back as that is a possibility.

Is it broken.....no, maybe not implemented correctly but not broken

[Torg_Snowflake]

I Don´t have the pic, I´ll just tell what happened...

I was defending Menirah. When a group of reds closed in too much, They were alltogether and surounded by mobs, so I took miy chances and Shot a nice Terror (with the +10 concentration buff).

I MISSED every single one of the 11 reds...

I HIT all the 7 purple mobs...

Sometimes I can´t believe my luck...

[Tyr]

No one has proved anything here yet. The sample data is far too small and there is a bunch of unknowns. For one each attack is evaluated for hit or miss so you could have no evades or lot in one battle and under the laws of probability that is fine. It is over the 1000s of attacks that is the true test.

Look at it is way if you were to roll 2 dice over enough rolls it should turn out like this:

2 = 1/36 (2.7%)
3 = 2/36 (5.5%)
4 = 3/36 (8.3%)
5 = 4/36 (11.1%)
6 = 5/36 (13.8%)
7 = 6/36 (16.6%)
8 = 5/36 (13.8%)
9 = 4/36 (11.1%)
10 = 3/36 (8.3%)
11 = 2/36 (5.5%)
12 = 1/36 (2.7%)

But if you only roll 36 times (the number of possible combination) you will not.

There is also a few other details missing any passive, active, or item that have a hit chance increase. The effect that level has on hit chance.

There is just no way for us to know all the detail to tell if this is working correctly.

I do not think the random number generator is broken but from seeing how bad they are doing math there could still be something not working correctly. For evidenced of that just buy more than one item at a time from any vendor the numbers don't add up right.

[Angelwinged_Devil]

Originally Posted by Solarus

Actually if you are not careful with random number generators they really aren't as random as one would expect. Random number generators are based on a seed value (well, pseudo-random one's are). Two basic ways to generate random numbers are by measuring some phenomenon that is expected to be random and then compensate for some sort of possible biases that may exist in the process, and generating random numbers based on a seed or a key value. The random number generated when using a seed is only as good as the algorithm used to generate the number. Regardless, the randomness isn't always - so....random if you aren't expecting the same number to be generated back-to-back as that is a possibility.

Is it broken.....no, maybe not implemented correctly but not broken

some random number generators take the current time, then through some preprocessors they multiply it up high using output from various hardware and then divides it down again.

[valterbla]

I think I am so lucky...