Forum Discussion
9 years ago
TLDR: I do think there could be an issue with their RNG.
Some background:
Real random numbers on a computer aren't possible so an algorithm is used. This algorithm generates a sequence with (apparently) zero correlation between terms. After a certain length of sequence it is not possible to spot any pattern in the terms (i.e. they seem random).
However, it is a known problem that some basic RNG algorithms aren't truly random and in fact, can require a long 'burn in' before the numbers are truly random. Prior to this burn-in one can have long sequences (hot or cold streaks) that occur more commonly than would be expected from a true random process.
I suspect this may be happening in the game. This would mean both sides above are correct - there are indeed cold streaks which don't seem random (and shouldn't occur as much as they do) but the long term averages work out as they should. Both these things can be true in this situation.
If I am right then the best thing to do when you get a cold streak is to go and do some other action in the game (e.g. leave cantina and do a challenge or something to reset the RNG sequence). I find this does tend to break the cycle of the cold streak. Of course, if you have a success then keep going as the 'bad' RNG may be favouring you in such cases.
Of course, all this could be confirmation bias, but a bugged RNG is certainly quite possible and would display the properties that people do seem to observe (and given how many other bugs are in the game it wouldn't surprise me at all!)
See here for an example of a RNG with these issues: https://en.wikipedia.org/wiki/Mersenne_Twister
Some background:
Real random numbers on a computer aren't possible so an algorithm is used. This algorithm generates a sequence with (apparently) zero correlation between terms. After a certain length of sequence it is not possible to spot any pattern in the terms (i.e. they seem random).
However, it is a known problem that some basic RNG algorithms aren't truly random and in fact, can require a long 'burn in' before the numbers are truly random. Prior to this burn-in one can have long sequences (hot or cold streaks) that occur more commonly than would be expected from a true random process.
I suspect this may be happening in the game. This would mean both sides above are correct - there are indeed cold streaks which don't seem random (and shouldn't occur as much as they do) but the long term averages work out as they should. Both these things can be true in this situation.
If I am right then the best thing to do when you get a cold streak is to go and do some other action in the game (e.g. leave cantina and do a challenge or something to reset the RNG sequence). I find this does tend to break the cycle of the cold streak. Of course, if you have a success then keep going as the 'bad' RNG may be favouring you in such cases.
Of course, all this could be confirmation bias, but a bugged RNG is certainly quite possible and would display the properties that people do seem to observe (and given how many other bugs are in the game it wouldn't surprise me at all!)
See here for an example of a RNG with these issues: https://en.wikipedia.org/wiki/Mersenne_Twister
About SWGOH General Discussion
Discuss and share your feedback on Star Wars: Galaxy of Heroes with fellow players.
78,031 PostsLatest Activity: 26 seconds agoRelated Posts
Recent Discussions
- 26 seconds ago
- 4 minutes ago
- 5 minutes ago
- 11 minutes ago
- 35 minutes ago