Those not-so RNG moments

mpqr7

I just pulled two Green Max Pun covers in the same few days, which is too bad because he's not championed and I already have 5, so they'll both go to waste.

broll

I got 2 bagman covers in a row from PvP this weekend. I think that brings up to a whopping 4 total in all 262 days of playing.

kobu

I tend to notice when I pull the progression reward 3 and 4* from heroic tokens so they go right into my queue next to the others. Also the time I pulled three yellow Thing covers in a row opening LTs.

Calnexin

alphabeta wrote:

Maybe worth reading a bit about the mathematical concept of randomness if you are going to comment on an RNG and the examples you sight are the definition of random outcomes especially on a sample size that isn't even worthy of being called a sample.

Maybe worth brushing up on your vocabulary before you tinykitty on someone's math skills. I cited your post, but I sight your arrogance.

JackTenrec

AlphaNik wrote:

To be clear (to my knowledge) when it says predictable fashion it does not mean you can predict the exact results just than in the long run the numbers will get in line with statistics. As far as I know RNGs tend to generate streams. As someone ever experienced an entire line of tiles of the same color coming down after a board shake? I mean how are the odds of that happening with a true RNG?

You're confusing determinism (can something be reliably reproduced over and over again) with randomness. Computer-based random numbers are not truly random because there's a reproducible element to the randomness -- a starting point generally known as a "seed". You can predict the output of a computer-generated sequence of random numbers (that is, the sequence is deterministic) if you know the seed value.

That being said, accepted random number generation algorithms are used because they're demonstrated to be statistically random. That is, if you don't know the seed value (and don't have a way to reverse-engineer it from the generated numbers) the values are just as apparently random as, say, dice rolls. When you see streaky behavior, that's a function of all randomness. People tend to see streaks as "broken" behavior because we have a terrible intuitive understanding of what actual randomness is.

dsds

There are many different scenarios for random. Even what everyone thinks is a completely random physical dice roll, may be influenced by weight distribution, even the surface of the table. Nothing is truly the random we expect. I guess as long as the distribution of covers is the percentage that is listed on the token over a large/full sample size, then nothing is broken.

They could probably code something in there so that you don't get the same cover in the same day. But then others would just complain that there are no more streaks which are also expected in random and that they can't complete a character as quickly because you can't get the same character on the same day.

You can't please everyone in the game.

On a side note, I've always wondered if it would be easier on server resources if they had a sticky random. For instance, when drawing the same token twice, they could have a code in there that has a 50% chance of the same repeat cover. This way, it doesn't have to go through all that code to produce a new draw. Random number between 1 and 2. If 1, proceed with [name of token] draw sequence. If 2, repeat cover draw of last [name of token] draw sequence. This way it skips all the rules for drawing that specific token. Such as for a event token draw, you have to have very specific percentage distribution so the draw would probably have a ton of code in it to get the correct percentage listed on the token.

TeamStewie

Pulled a red Carnage cover with an heroic tokent after just after getting one in progression. Still doesn't make me want to roster his awfulness.

smkspy

TeamStewie wrote:

Pulled a red Carnage cover with an heroic tokent after just after getting one in progression. Still doesn't make me want to roster his awfulness.

A 5-0-5 Carnage is great for his crash.

TeamStewie

smkspy wrote:

TeamStewie wrote:

Pulled a red Carnage cover with an heroic tokent after just after getting one in progression. Still doesn't make me want to roster his awfulness.

A 5-0-5 Carnage is great for his crash.

Well I only have 2 covers and TBH I would rather roster my 1 Hulkbuster cover than 2 Carnage.

carrion_pigeons

AlphaNik wrote:

Quote from the first site I get in google searching "Random number Generator".

Perhaps you have wondered how predictable machines like computers can generate randomness. In reality, most random numbers used in computer programs are pseudo-random, which means they are generated in a predictable fashion using a mathematical formula. This is fine for many purposes, but it may not be random in the way you expect if you're used to dice rolls and lottery drawings.

It's not a conspiracy theory it's just the way RNG works. To be clear (to my knowledge) when it says predictable fashion it does not mean you can predict the exact results just than in the long run the numbers will get in line with statistics. As far as I know RNGs tend to generate streams. As someone ever experienced an entire line of tiles of the same color coming down after a board shake? I mean how are the odds of that happening with a true RNG?

Random number generators are indistinguishable from "true" randomness. There's a common perception that there's actually any such thing as true randomness, but there isn't. Random doesn't mean numbers get pulled from some sort of ether where everything is evenly distributed and everything occurs at the statistical average all the time.

If I roll a dice, and I have accurate, precise information about all the forces acting on them between the time they leave my hand and the time they stop moving, I can predict the roll with perfect accuracy. The only thing that makes rolling dice random is that I *don't* have that information, and that the results can be organized into categories which are normally distributed.

It doesn't matter how they generate their numbers. Until someone comes up with either a way to reproduce a result or else to predict a result at a level of confidence that falls outside a normal distribution, there's no meaning to the question of how computer generated randomness is different from "true" randomness, because there's literally no difference in the first place.

If you think there's an increased rate of featured characters in heroics/LTs, I won't say it's impossible. But there are plenty of people who have hoarded their tokens and posted their hoards who ought to be able to say yay or nay on the topic, and looking at their results I think the answer is way more than likely to be nay.

TeamStewie

Just got Jean Grey from a heroic token. Will definitely roster her

WEBGAS

STOCKED 190 TACOS in order to open them asap a favourite cover will show....Hey That Yellow Quake is just what I needed

Opened all 190 tacos.......Quake it's still there

It' s not random.....is demoniac!