generic image
Processing...
  • Games
  • Catalog
  • Develop
  • Robux
  • Search in Players
  • Search in Games
  • Search in Catalog
  • Search in Groups
  • Search in Library
  • Log In
  • Sign Up
  • Games
  • Catalog
  • Develop
  • Robux
   
ROBLOX Forum » Game Creation and Development » Scripters
Home Search
 

Re: Rolling Dice Reveal Physics Bias

Previous Thread :: Next Thread 
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 10:23 AM
I have rolled a die 200 times in my Dice Roll Simulation place locally. I have removed all pips for the purpose of testing bias. And I got these results:

Roll History String: 32246246565342653132412123651536562644556211455346221242656423466366666466611354534661213664355443614114446431111163353656511465453616451126426156233264456143226644562561364445552336315664541246536636

Face/Roll/Pct.:

1 2 3 4 5 6
29 24 28 37 31 51
14.5% 12% 14% 18.5% 15.5% 25.5%

It appears that the side with 6 pips (back face) appears to be biased.

The sides/pips (for reference):
1: Front
2: Left
3: Top
4: Bottom
5: Right
6: Back

What do you guys think?

~blocco

Report Abuse
DrAgonmoray is not online. DrAgonmoray
Joined: 29 Jul 2008
Total Posts: 17428
09 Aug 2013 10:24 AM
I think you need to roll it more times
Report Abuse
Trioxide is not online. Trioxide
Joined: 29 Mar 2011
Total Posts: 32902
09 Aug 2013 10:25 AM
Yeah, roll it at least 10 000 times, then you get de results.

Change your font: http://www.roblox.com/Forum/ShowPost.aspx?PostID=109051027
Report Abuse
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 10:47 AM
I'm going to roll it to 1200 and re-post results
Report Abuse
gijsbel11 is not online. gijsbel11
Joined: 07 Feb 2009
Total Posts: 4223
09 Aug 2013 10:48 AM
if you flip a coin 100 times, there is a change that it lands 100 times on 1 side, but that doesn't mean it actual does 100% of the time. I think you need to do more rolls.
Report Abuse
18cwatford is not online. 18cwatford
Joined: 08 Oct 2009
Total Posts: 5410
09 Aug 2013 10:54 AM
But the chance that you flip a coin 100 times and land on the same side, seems like a good reason to say its going to keep lading on that side
Report Abuse
Clayguy247 is not online. Clayguy247
Joined: 03 Sep 2008
Total Posts: 3394
09 Aug 2013 11:01 AM
^ x2
Report Abuse
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 11:09 AM
    Roll History String:
    
    322462465653426531324121236515365626445562114553462212426564234663666664666113545346612136643554436141144464311111633536565114654536164511264261562332644561432266445625613644455523363156645412465366365411665452614466645661531243432416111613213442311534452456153542621245436335523622636336254245316416463116126251111254211632524543444553143546124651332363116444421346133236465634661321311554652365452631661214253353315432234524362651621562452253611335544365325426465121232114236535141331544262243632152145425442312663433613263131162164546416356433533512111132335511413511135616422153664362316413351645255314425122664642662231235342211111155363224564456522445614411462615213346532665252244326613662313566115216454661411641322333531155611366661126233361525626336322645443366241666363323451362462446415631632212424565511225614524211516343164224112152432244143232665536235626224514212233425262114212355411231456666323123541152643134452331155531524131153416142333633234325215612124164216223354614416452331465261631253353346535222226531155424434244432116644361452622316261224624121656242622332215436351665361531544514354256263234532123116415412243556531614125656642335166613561513244463653116165635634323562

Look for trends if you want.

    Roll Data, grouped:
    
    1        2        3        4        5        6
    211      197      203      193      174      222
    17.58% 16.42% 16.92% 16.08% 14.50% 18.50%

The rolls *should* tend to 200
The percents *should* tend to 16.67%

There is still a bias of 6, and 5 is not rolled very often. 1/6 (opp. sides) are highest. 3/4 (opp. sides) tend toward the middle. 2/5 (opp. sides) are the most different, with 2 being the second closest to the middle.

~ blocco
Report Abuse
stravant is not online. stravant
Forum Moderator
Joined: 22 Oct 2007
Total Posts: 2893
09 Aug 2013 11:35 AM
@blocco

The above stats are within the ranges are within the variation you could expect to see given an unbiased random number generator as well. They don't actually prove a bias.
Report Abuse
Prehistoricman is not online. Prehistoricman
Joined: 20 Sep 2008
Total Posts: 12490
09 Aug 2013 11:40 AM
OK, well, that's great.

So... how would one add parts to the die (pips) without affecting the die's results?
Report Abuse
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 11:42 AM
@stravant: That's good to note

Now I'm rolling with pips (no bodyforce) to 1200
Report Abuse
Debranched is not online. Debranched
Joined: 18 Mar 2013
Total Posts: 266
09 Aug 2013 11:44 AM
That's what I call real randomness.
Report Abuse
dop5k5 is not online. dop5k5
Joined: 18 Mar 2013
Total Posts: 42
09 Aug 2013 11:45 AM
codepad/utJNmzO9


eh
Report Abuse
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 11:48 AM
@dop: I'm using the physics engine though, looking for bias to make my dice as fair as possible
Report Abuse
stravant is not online. stravant
Forum Moderator
Joined: 22 Oct 2007
Total Posts: 2893
09 Aug 2013 11:49 AM
You should roll about 10,000 times if you actually want to more prove the amount of bias to those faces that you are speculating.
Report Abuse
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 11:50 AM
I'll roll 10000 when I'm asleep

For now, 1200 will be fine
Report Abuse
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 12:06 PM
PIPS W/O BODYFORCE:

History:

632526653544531666625332246654514332542261624165213323645213315656334224655145323634462145465232421661211511564342214155336563332224421665614122215214326611552613222346555354252613511561331556114363455466346144251633632526143534526561526512324611114535632246526435416124612513245231124621534245135116112431213131351143135655555621136513414135535232355533516245421333133136153636132611446651121421624523523366654625363353143245335414433524434654562122134145615441534655462366316324522636255326631456432351626162155465242334646424123423155235456331464513233632566166314152345463424223361115236522664456224511355632511363135255111362212561334212664561462664515613364466654161536332561454255453323214425545513255211243655615143312225261351565646154234551623436116636453425335216312241115664134442652432642466453243115316354236523121623242232641635255255332543365653323342511223415665234511215364162262125235413662524115124461326121113613362462441562222231531234644326352446252535611626245222262634552445556662351231612226132642236612156551531431321662635166655552336536223422266151114135625246213146125631154455321431455241112553223464546423536334626342655461441225413366623133463431214566153154611631542

Grouped Data:

1 2 3 4 5 6
201 211 207 171 208 202
16.75% 17.58% 17.25% 14.25% 17.33% 16.83%
Report Abuse
morash is not online. morash
Joined: 22 May 2010
Total Posts: 5834
09 Aug 2013 12:20 PM
Your data seems different there then your other rolls

@gijsbel11: There is also the chance, that if the physics are right, a coin can also land on neither side and stay on the edge of the coin which would make flipping a coin not a 50/50 chance but rather a 49.9/49.9/0.2 chance.
Report Abuse
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 12:31 PM
^ If you have rotation going in all directions, it's more like 49.99999/49.99999/0.00002
Report Abuse
Xtreme101 is not online. Xtreme101
Joined: 03 Jan 2009
Total Posts: 4385
09 Aug 2013 12:34 PM
Instead of looking for biased pips, why not look for unbiased ones? By that, I mean the lesser-generated.
Report Abuse
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 12:49 PM
What exactly do you mean, Xtreme
Report Abuse
DrAgonmoray is not online. DrAgonmoray
Joined: 29 Jul 2008
Total Posts: 17428
09 Aug 2013 02:03 PM
He means look for ones that are biased towards not being picked.

like they're rolled less than others
Report Abuse
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 02:06 PM
Four, the side with four pips
Report Abuse
NXTBoy is not online. NXTBoy
Joined: 25 Aug 2008
Total Posts: 4533
09 Aug 2013 02:11 PM
Where is the randomness introduced in your die rolling simulation? If I throw the same dice with the same state each time, I'd expect the same result every time.
Report Abuse
blocco is not online. blocco
Joined: 14 Aug 2008
Total Posts: 29474
09 Aug 2013 02:11 PM
I add random RotVelocity and random positions within a certain range
Report Abuse
Previous Thread :: Next Thread 
Page 1 of 1
 
 
ROBLOX Forum » Game Creation and Development » Scripters
   
 
   
  • About Us
  • Jobs
  • Blog
  • Parents
  • Help
  • Terms
  • Privacy

©2017 Roblox Corporation. Roblox, the Roblox logo, Robux, Bloxy, and Powering Imagination are among our registered and unregistered trademarks in the U.S. and other countries.



Progress
Starting Roblox...
Connecting to Players...
R R

Roblox is now loading. Get ready to play!

R R

You're moments away from getting into the game!

Click here for help

Check Remember my choice and click Launch Application in the dialog box above to join games faster in the future!

Gameplay sponsored by:
Loading 0% - Starting game...
Get more with Builders Club! Join Builders Club
Choose Your Avatar
I have an account
generic image