EPTWTF!!! Random isn't random
| Forum Index > Closed |
|
EAGER-beaver
Canada2799 Posts
| ||
|
gLyo
United States2410 Posts
| ||
|
Mandalor
Germany2362 Posts
in your case it's terran, in my case it's toss | ||
|
MaGnIfIcA
Norway2312 Posts
| ||
Liquid`Drone
Norway28702 Posts
nope. :p | ||
|
MyDoom
1274 Posts
 | ||
|
Degussa
Singapore108 Posts
| ||
|
gg_hertzz
2152 Posts
Especially when it's concerning computers. | ||
Bill307
Canada9103 Posts
 . | ||
|
ProudCappi
United States366 Posts
| ||
|
Refrain[FriZ]
Canada4337 Posts
| ||
|
Degussa
Singapore108 Posts
worse come to worse just pick the race haha | ||
|
RamenStyle
United States1929 Posts
| ||
|
gg_hertzz
2152 Posts
Someone, else I mean. | ||
|
ShabZzoY!
Great Britain760 Posts
i believe thats 1/729 chance the whole point of random is anything may happen, its designed to have equal odds but you could have same race for 8 games i suppose(more chance than most lottery :o) | ||
Excalibur_Z
United States12237 Posts
On December 25 2004 12:33 EAGER-beaver wrote: This has really irked me since the beginning of sc. It seems like random actually means terran %90 of the time. Since I random virtually every single game, if I want to be zerg or protoss, I have to manually select them at the start of every game. Anyone else notice how random is extremely biased towards terran? Yeah, just like coin-flipping is extremely biased towards Heads. Heads comes up 90% of the time! That's not random! | ||
|
8882
2718 Posts
On December 25 2004 12:36 Mandalor wrote: I think you always get the race you don't want or it seems like that in your case it's terran, in my case it's toss it's one of the Murphy's rules | ||
Liquid`Jinro
Sweden33719 Posts
On December 25 2004 12:36 Mandalor wrote: I think you always get the race you don't want or it seems like that in your case it's terran, in my case it's toss Same. Wtf I get p and p and p and p :O When I random I want to get zerg or terran since I play only p matchups normally =[ | ||
|
Excalibur_Z
United States12237 Posts
On December 25 2004 13:02 ShabZzoY! wrote: me and my friend had zvt in rvr 3 times in a row i believe thats 1/729 chance the whole point of random is anything may happen, its designed to have equal odds but you could have same race for 8 games i suppose(more chance than most lottery :o) No it's still a 1/3 chance =) Again I'll use coin-flipping as an example. If you get Heads 3 times in a row, the chances of that happening aren't 1/8, they're still 1/2. | ||
|
FrEaK[S.sIR]
2373 Posts
| ||
|
littlechava
United States7221 Posts
| ||
|
Excalibur_Z
United States12237 Posts
On December 25 2004 13:10 FrEaK[S.sIR] wrote: Its 1/9 chance...both sides of it are randomized.. How do you figure? If you flip the coin again you'll have an equal chance of it hitting Heads or Tails, it won't be 1/8 in favor of either side. | ||
|
paper
13196 Posts
On December 25 2004 13:05 Excalibur_Z wrote: No it's still a 1/3 chance =) Again I'll use coin-flipping as an example. If you get Heads 3 times in a row, the chances of that happening aren't 1/8, they're still 1/2. uhh. no. getting heads three times in a row is .5 x .5 x .5 = 0.125, or 12.5%, or 1/8 | ||
|
Orlandu
China2450 Posts
On December 25 2004 13:11 Excalibur_Z wrote: How do you figure? If you flip the coin again you'll have an equal chance of it hitting Heads or Tails, it won't be 1/8 in favor of either side. It's because it's sequential. Indivudally you would be right. But you have to group them as a whole, because sequentially means it's part of a bigger series of coin flips. | ||
|
EntiTy.
United States80 Posts
| ||
|
Excalibur_Z
United States12237 Posts
On December 25 2004 13:20 Orlandu wrote: It's because it's sequential. Indivudally you would be right. But you have to group them as a whole, because sequentially means it's part of a bigger series of coin flips. Okay, but how is that indicative of the results of future flips? It wouldn't be scientifically sound to use the sequential flip results as evidence for how future flips will turn out, so isn't that information irrelevant? | ||
|
gg_hertzz
2152 Posts
On December 25 2004 13:24 Excalibur_Z wrote: Okay, but how is that indicative of the results of future flips? It wouldn't be scientifically sound to use the sequential flip results as evidence for how future flips will turn out, so isn't that information irrelevant? Wouldn't it? | ||
|
ShabZzoY!
Great Britain760 Posts
anyway, i have same problem as FA, when i get bored with toss i pick random and then get toss again :S | ||
|
Excalibur_Z
United States12237 Posts
Not on such a small scale, no. If, on the other hand, you were collecting statistics, you could identify the actual randomness based on tens of thousands of flips. But if you got 3 Heads in a row, that's not enough to draw any type of concrete conclusion. | ||
|
SoL.Origin
Argentina2400 Posts
| ||
|
Orlandu
China2450 Posts
On December 25 2004 13:24 Excalibur_Z wrote: Okay, but how is that indicative of the results of future flips? It wouldn't be scientifically sound to use the sequential flip results as evidence for how future flips will turn out, so isn't that information irrelevant? It's not about the results you're getting. It's about the math behind it. In flipping one coin, there is a 1/2 chance of flipping a heads because there are only two possible outcomes, heads or tails. When flipping two coins and looking at them as a sequence, then there are two possiblities for the first one, which we covered. Assume you get heads. For the next flip, there are two possibilties: heads-heads, or heads-tails. Assume you get tails, two possibilities: tails-heads, tails-tails. So there are 4 possible outcomes of that sequence. There is only 1 outcome that is the result you want, heads-heads. So, 1 out of 4 gives you a 1/4 chance. Apply similar logic to a third round and you get 1/8 chance of getting heads 3 times in a row. | ||
|
gg_hertzz
2152 Posts
Isn't it all just a series If true then excute this Else excute this ? | ||
|
Danka
Peru1018 Posts
| ||
|
gg_hertzz
2152 Posts
On December 25 2004 13:37 Orlandu wrote: It's not about the results you're getting. It's about the math behind it. In flipping one coin, there is a 1/2 chance of flipping a heads because there are only two possible outcomes, heads or tails. When flipping two coins and looking at them as a sequence, then there are two possiblities for the first one, which we covered. Assume you get heads. For the next flip, there are two possibilties: heads-heads, or heads-tails. Assume you get tails, two possibilities: tails-heads, tails-tails. So there are 4 possible outcomes of that sequence. There is only 1 outcome that is the result you want, heads-heads. So, 1 out of 4 gives you a 1/4 chance. Apply similar logic to a third round and you get 1/8 chance of getting heads 3 times in a row. Yeah, what he said. | ||
IntoTheWow
is awesome32277 Posts
On December 25 2004 13:36 SoL.Origin wrote: I learned a few days ago from playing race wars that when 8 ppl are in the game and all go random, its always 2 of a race, 3 of another, and 3 of the other one. You never get like 7 terran and 1 zerg for example. Yeah i noticed that too playing rw also. The oddest thing we got was a 5 player race war going 3t 1z & 1p | ||
|
Excalibur_Z
United States12237 Posts
On December 25 2004 13:37 Orlandu wrote: It's not about the results you're getting. It's about the math behind it. In flipping one coin, there is a 1/2 chance of flipping a heads because there are only two possible outcomes, heads or tails. When flipping two coins and looking at them as a sequence, then there are two possiblities for the first one, which we covered. Assume you get heads. For the next flip, there are two possibilties: heads-heads, or heads-tails. Assume you get tails, two possibilities: tails-heads, tails-tails. So there are 4 possible outcomes of that sequence. There is only 1 outcome that is the result you want, heads-heads. So, 1 out of 4 gives you a 1/4 chance. Apply similar logic to a third round and you get 1/8 chance of getting heads 3 times in a row. Okay, I understand that. What I don't understand is how you're applying sequential logic to individual games. Everyone has had their suspicions about BW's randomizer. Some say they get Terran 3 times in a row, some say Protoss 4/5 times, some say they've never got Zerg. So without larger sets (i.e., thousands of games) to draw from, how can any conclusion be reached about the randomizer and the math behind it? | ||
|
SChasu
United States1505 Posts
| ||
|
Nal_Testie
Canada1257 Posts
In 1on1's, it is indeed random. And is always random. The chance of a mirror matchup however in RvR is extremely rare and almost never happens. But, it occasionally does. In 2on2's, if all four players random. And you are z/z your opponents are t/p (automatically). If all four random and you and your ally get z/p, one of your opponents is T. And the other may be z/t/p. Likewise if you are t/t they are z/p. Etc.. etc... In games with 8 players, for race wars. All players can random without ever having to fear that you will have a team of five protosses, three terrans etc... It is always 3 of one race, 3 of another, and 2 of the other. Unless one person picks and fucks it all up. So in any game including 1on1, it is indeed random. There are days when you will random zerg almost for an entire day, random terran, or random toss. But they are rare instances. I didn't read any of the other posts. So if anyone else pointed this out, good job. If not, what a bunch of fucking morons. Happy Holidays. | ||
|
Orlandu
China2450 Posts
On December 25 2004 13:48 Excalibur_Z wrote: Okay, I understand that. What I don't understand is how you're applying sequential logic to individual games. Everyone has had their suspicions about BW's randomizer. Some say they get Terran 3 times in a row, some say Protoss 4/5 times, some say they've never got Zerg. So without larger sets (i.e., thousands of games) to draw from, how can any conclusion be reached about the randomizer and the math behind it? Oh I wasn't saying anything about BW, I was just commenting on the coin logic. I guess I'd have to look into the whole BW-Random debate before I could make an opinion. EDIT: Personally I don't feel that it is truly "random", because from my experience with computers, as someone mentioned, true randomness isn't an option. Most likely Blizzard's algorithm is based off of a timer, so you could possibly have an algorithm where if you start the game very quickly right away each time, you might get the same race much more often because of the timer. That's IF the timer starts each time a game is created/or someone joins or whatever. There's lots of possiblities. I don't think any of them are truly random though, more dependent on some variable like the timer, and when you actually start the game would be my best guess. So I think I see what you're saying here, and the only comment I can really say is, I personally don't feel BW is truly random. EDIT2: But as far as our needs and concerns go, you could call it random =] | ||
|
Excalibur_Z
United States12237 Posts
On December 25 2004 13:52 Nal_Testie wrote: In 1on1's, it is indeed random. And is always random. The chance of a mirror matchup however in RvR is extremely rare and almost never happens. But, it occasionally does. So in any game except 1on1, it is indeed random. There are days when you will random zerg almost for an entire day, random terran, or random toss. But they are rare instances. I'm confused... | ||
|
Nal_Testie
Canada1257 Posts
| ||
|
Excalibur_Z
United States12237 Posts
On December 25 2004 13:53 Orlandu wrote: Oh I wasn't saying anything about BW, I was just commenting on the coin logic. I guess I'd have to look into the whole BW-Random debate before I could make an opinion. Oh okay, so the coin logic and the BW randomizer aren't necessarily related (disregarding the obvious internal-clock that all computer randomizers use)? | ||
|
Orlandu
China2450 Posts
On December 25 2004 13:58 Excalibur_Z wrote: Oh okay, so the coin logic and the BW randomizer aren't necessarily related (disregarding the obvious internal-clock that all computer randomizers use)? I just edited my post, but yeah I think it would be pretty wrong to relate the two arguments. I could be wrong, but I don't see why there would be a relation. Computers just don't function the same way as a coin toss. | ||
|
Excalibur_Z
United States12237 Posts
On December 25 2004 13:53 Orlandu wrote: Most likely Blizzard's algorithm is based off of a timer, so you could possibly have an algorithm where if you start the game very quickly right away each time, you might get the same race much more often because of the timer. That's IF the timer starts each time a game is created/or someone joins or whatever. There's lots of possiblities. I don't think any of them are truly random though, more dependent on some variable like the timer, and when you actually start the game would be my best guess. The thing is, internal-clock randomizers go down to the... what is it... millisecond? nanosecond? While I suppose it would be possible to create some kind of script so you'd always get the same race whenever you chose Random, it would be nearly impossible for a human to have that kind of timing, assuming someone was even able to figure out which cpu-tick translated to which race was selected. | ||
|
Nal_Testie
Canada1257 Posts
| ||
|
Orlandu
China2450 Posts
On December 25 2004 14:01 Excalibur_Z wrote: The thing is, internal-clock randomizers go down to the... what is it... millisecond? nanosecond? While I suppose it would be possible to create some kind of script so you'd always get the same race whenever you chose Random, it would be nearly impossible for a human to have that kind of timing, assuming someone was even able to figure out which cpu-tick translated to which race was selected. Probably. I'm not really sure how it works or how it would work. But I don't think it functions as what we would call "luck" like the coin toss. I guess you could call it random as far as our needs and concerns go, but I don't think it would be truly random. I dunno, I could be totally wrong on that, but it just seems that way to me. I don't know nearly enough about computers to make a factual statement like that =] | ||
Hot_Bid
Braavos36379 Posts
this way you can almost always figure out at least one of your opp's races. if you and your partner get... z/z, t/t, p/p, your opponents are the other two races any other combo, at least one of your opponents is the race you are not it's impossible to conclude anything when you say something like "oh, it randomed to terran for me five times yesterday! its biased!" first your sample size is way too small, and the nature of random allows for stuff like terran five times in a row. i'm pretty sure the more than one random player in a game bias is right, someone test it! | ||
|
Nal_Testie
Canada1257 Posts
You fucking idiot. I was saying what I was saying as fact. Not even speculation or a hypothesis. IF 8 players random there will be 3/3/2 If 4 players random there will be 2/1/1 IT WILL NEVER EVER EVER BE DIFFERENT Stop pissing me off. | ||
|
Ack1027
United States7873 Posts
Haha. | ||
|
gg_hertzz
2152 Posts
You might get T 4/5 times and think that there probably is a pattern, but it could also be random and you would not know it. How can you know if it's random or not? You could get T 99/100 times and it could still be random. How could you know? | ||
|
ShabZzoY!
Great Britain760 Posts
| ||
|
LogaiN
Sweden1073 Posts
| ||
|
gg_hertzz
2152 Posts
| ||
|
SoL.Origin
Argentina2400 Posts
| ||
|
HnR)Pride
Canada297 Posts
| ||
|
Cambium
United States16368 Posts
90% TvZ -__- | ||
|
NuclearAntelope
United States1369 Posts
On December 25 2004 13:05 Excalibur_Z wrote: No it's still a 1/3 chance =) Again I'll use coin-flipping as an example. If you get Heads 3 times in a row, the chances of that happening aren't 1/8, they're still 1/2. If you flip a coin 3 times, you have these possibilities: H = heads, T = Tails HHH HHT HTH THH TTT THT HTT TTH how is it 1/2 chance to get HHH? | ||
|
Redan
Poland7 Posts
1/9 * 1/9 * 1/9 = 1/729 Chance to get ZvT three times in a row Simple? | ||
|
Liquid`Drone
Norway28702 Posts
and mirror matches are very very rare in 1v1 matches I remember when me and testie played on coulee in the blizzard map tourney he claimed that I was zerg and I was protoss and I instantly "knew" that he was either terran or made the claim to confuse me, because if he was zerg he would not think that I was zerg, and he wouldn't be protoss cause I was protoss. I believe I then said "well im terran and its not a mirror matchup so you're toss then" to confuse him back though. as for the other ongoing discussion, excalibur and everyone else are talking about different things if you have flipped heads 3 times in a row, the chance of getting it a fourth time is not 1/16, it's 1/2 but flipping heads 3 times in a row in the first place only happens 1/8 of the time. likewise if you've randomed terran 5 times in a row it's still 33% likely that you will random terran in your next game. I've checked my replay folder before and I think I had 35% zerg 34% protoss and 31% terran. the slight lack of terran is however explainable with me normally only saving wins, and winning sliightly less with terran. | ||
|
Nal_Testie
Canada1257 Posts
| ||
|
Alpha
France1495 Posts
On December 25 2004 15:09 Redan wrote: That's correct but mmmm if u count Zvt = TvZ it's more likely to happen1/3 * 1/3 = 1/9 Chance to get ZvT 1/9 * 1/9 * 1/9 = 1/729 Chance to get ZvT three times in a row Simple? | ||
|
Nal_Testie
Canada1257 Posts
| ||
|
SuperCrazyMan
43 Posts
All in all, who gives a shit. We could all use some variety. | ||
|
J1
Canada579 Posts
| ||
|
Nal_Testie
Canada1257 Posts
33.3333333 33.3333333 The last .00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001% goes into those times you don't want to play a certain matchup and it fucks you, giving you that matchup out of spite! | ||
|
ironchef
Canada1350 Posts
Although i do find the random race assignments testie mentioned quite interesting. | ||
|
gg_hertzz
2152 Posts
On December 25 2004 15:27 SuperCrazyMan wrote: Jesus Christ, calm down people. If what Testie said is true, which I do personally believe, Blizzard obviously implemented bias in the randomization algorithm to prevent morons complaining "omg random is fucked up cuz we all terran!!!!!! gayyyyyyy" Apparently, they wanted it to be an even distribution of races when random is selected by more than one person. All in all, who gives a shit. We could all use some variety. calm down, from what? Testie's the one throwing a tantrum that only Elton John can appreciate. But if you like suckin' Testie's nuts (pun fully intended) then so be it. | ||
Liquid`Daaman
Sweden1225 Posts
it's built in to decrease the chances that you get the same race for both players. had it been let alone it would of course be 1/3 that u get either zvz,tvt or pvp (since no matter what the first player gets, 1 of the 3 races drawn by player2 would mena they got the same one). however, I believe blizzard edited this so that it would only occur 1 in 27 times! | ||
|
aseq
Netherlands3983 Posts
Can't anyone remember having played something like rP-rZ versus rP-rZ ?? About the thing being random for a single player, it uses the system clock. I now and then play vs racine, and if i mess up my split i restart. You get 1 race for about 30secs-1min, then the next, then the next again. Try out plz? I think they're just using a few bits out of the system time. Kinda silly, but who restarts the game that often anyway? I don't think the regular 10 games-in-a-row of one race is non-random, or influenced by previous games, but i am not 100% certain here. And plz stop talking about the coins, the probability you get the same matchup vs some other dude 3 times is: 1 (you,any race) x 1 (him,any race) x 1/3 (you2) x 1/3 (him2) x 1/3 (you3) x 1/3 (him3) = 1/81 The chance of you and him getting a particular matchup (like t-z) 3 times is : 1/3 ^ 6 = 1/729 (replace the 1's in the previous example by 1/3) just like the probability you get heads 3 times is 1/8. edit: t: z without the space results in some stupid smiley | ||
|
Mora
Canada5235 Posts
me and my friend had zvt in rvr 3 times in a row i believe thats 1/729 chance the whole point of random is anything may happen, its designed to have equal odds but you could have same race for 8 games i suppose(more chance than most lottery :o) He didn't say. "Damn, we had zvt times in a row, it was a 1/729 chance to get it again!". he said "damn, the chances of getting zvt 3 times in a row are 1/729". The only part of that should be in debate is whether TvZ and ZvT are the same thing. (and they aren't if he specifically got only zerg or only terran in all 3 games). | ||
|
aseq
Netherlands3983 Posts
| ||
|
dt
Germany126 Posts
| ||
|
karelen
Sweden2407 Posts
| ||
|
tranCe[RaGe]
United States420 Posts
| ||
|
HooTie
United States177 Posts
| ||
|
Bill307
Canada9103 Posts
On December 25 2004 13:39 gg_hertzz wrote: I really don't see how random can actually be random. If it could be, then BW would be the most complex instrument in the solar system. Isn't it all just a series If true then excute this Else excute this ? There are LOTS (infinitely many, in fact) of mathematical functions that generate sequences of random numbers. E.g. set your calculator to Radians mode (as opposed to Degrees mode) and enter any number other than 0 into this function, in place of x: f(x) = sin (4 * x) Now take the result ("f(x)") and enter it into the function again, in place of x. Do this over and over and you'll get a random sequence of numbers between 0 and 1. Of course, the functions used in real random number generators are much more complex. With this function here, if you get a number that's REALLY close to 0, then the next number will also be really close to 0, but slightly farther away. A proper random number generator should not be this predictable. Furthermore, this function does not result in uniformly-distributed numbers, i.e. you are more likely to get some numbers than others. A proper random number generator should return uniformly-distributed numbers. And yes, such functions do exist. For more information or examples, use Google .Oh, and last of all, computers usually generate the initial value based on the computer's internal clock, e.g. they might use the number of milliseconds that have elapsed since midnight and generate a sequence of random numbers starting from that value. Alternatively, the computer can use a given value, and this is what SC will do when you play a replay: it will start with the same value that the actual game started with. This will result in the exact same sequence of random numbers, allowing the game to be played back correctly. | ||
|
gg_hertzz
2152 Posts
| ||
|
PuertoRican
United States5709 Posts
somedays i can get protoss like 3 times in a row, but then it will switch back to terran. the random thing can mean random, but in the 7 years ive played, random is terran favored. | ||
|
Juiyo
Canada174 Posts
when we both go random i could guess ppl race by there color..On Lost Temple Ladder Maybe im a magician.. | ||
|
Azmo
Sweden83 Posts
On December 25 2004 14:11 Nal_Testie wrote: ... You fucking idiot. I was saying what I was saying as fact. Not even speculation or a hypothesis. IF 8 players random there will be 3/3/2 If 4 players random there will be 2/1/1 IT WILL NEVER EVER EVER BE DIFFERENT Stop pissing me off. do you have proof? if it's unlikely that the 2 random players in a 1:1 have the same race, and we have similar effects for games with more than 2 players, then, it makes no sense to me for this to be programmed specifically for certain ammount of players in a game, but instead an effect of a general algorithm. my point is, it makes no sense for it to be impossible to get more than 4 players of the same race in an 8 player game (with all random players, yes..) also, i am quite certain i have been in a 6 player racewar game where there were 4 of one race (everyone left the game ofcourse). that was one game of several hundered, however, since some games end early because of lag or other problems, this would not be valid statistics in any way, however, they do show the existence of the possibility of 4 of one race in a 6player game, unless i'm insane and imagined it :-/. | ||
|
Meat
Netherlands3751 Posts
| ||
| ||
