Jump to content


RNG is not random at all – losing streaks, strange non-penetrations etc explained


  • This topic is locked This topic is locked
270 replies to this topic

zen_monk_ #1 Posted 11 August 2017 - 09:53 PM

    Sergeant

  • Clan Diplomat
  • 22641 battles
  • 205
  • Member since:
    05-08-2011

Ever wondered why in some games your shots go straight at the center, and in others wide from the target?

 

Why in some games you penetrate exactly what you should, and in some can’t pen 10mm?

 

Why you have winning streaks, and then a long row of humiliating defeats where you can’t make a difference?

 

 

 

It’s RNG, you’ll say. Well not quite. RNG here is not random at all. There is a “now you’ll win” setup, and “now this team will lose” setup.

 

What I have here is the actual patent of MM for WoT, used also in WoWs. (probably in WoWp too).

 

http://www.google.com/patents/US8425330?dq=8,425,330&hl=en&sa=X&ei=OIaMUeWBAuLIigK32ICYDw&sqi=2&pjf=1&ved=0CDQQ6AEwAA

Publication number

US8425330 B1

Publication type

Grant

Application number

US 13/472,945

Publication date

Apr 23, 2013

Filing date

May 16, 2012

Priority date

May 16, 2012

Also published as

US20130310181

Inventors

Victor Kislyi, Ivan Mikhnevich

Original Assignee

Wargaming.net, LLC

 

In there, this is the most interesting part that explains all the questions from the start of this post:

 

"According to another aspect, the matchmaking server may store a win/loss percentage for each user (or vehicle) at a given battle level. As the player's win/loss ratio decreases, the player becomes more likely to be placed in battles having battle levels at the lower end of the allowable range, whereas as the player's win/loss ration increases, the player becomes more likely to be placed in battles having battle levels at the upper end of the allowable range. Thus, when a player has been repeatedly put into too many difficult battles, the balancing is done in favor of easier battle sessions, thereby encouraging the player by providing an easier game environment. Similarly, when the player has been repeatedly put into too many easy battles, the balancing is done in favor of harder battle sessions, thereby keeping the player challenged instead of letting the player become bored with easy games."

In here Viktor Kisyl clearly says that MM system is designed  to prevent randomness.

 

The exact details of what they do in “now this team must lose” battles are not published (eg, the handicap to the penetration, failing to hit point blank, when exactly it this scenario triggered etc). THAT is the thing they tweak.  But the maker of this game explicitly said:

 

"...when a player has been repeatedly put into too many difficult battles, the balancing is done in favor of easier battle sessions, thereby encouraging the player by providing an easier game environment. Similarly, when the player has been repeatedly put into too many easy battles, the balancing is done in favor of harder battle sessions, thereby keeping the player challenged..."

 

This way WG prevented unicums of 100% and bozos of 0%. You can move up from 50% with being extra good, playing in good platoons all the time, but this Patent will hit you eventually. On the other side, even if you AFK, you won’t be 5%. “providing an easier game environment” or “keeping the player challenged”.

 

That’s all there is. It is NOT random at all. What your equipment do, how your gun behave, MM, the works. Not random but tweaked + or -.


Edited by zen_monk_, 11 August 2017 - 09:56 PM.


zen_monk_ #2 Posted 11 August 2017 - 10:00 PM

    Sergeant

  • Clan Diplomat
  • 22641 battles
  • 205
  • Member since:
    05-08-2011
Tin Foil my *edited*, read the text and the Patent.

Edited by Jbnn, 15 August 2017 - 06:23 AM.
This post has been edited by the moderation team due to inappropriate remarks.


Lil_Dimitry #3 Posted 11 August 2017 - 10:01 PM

    Private

  • Player
  • 2318 battles
  • 43
  • Member since:
    12-07-2014
It is indeed true that you get put in harder situations if you're good, you have 48% wr, it has nothing to do with WG rigging the game to screw you over, it's most likely the other way around

Velvet_Underground #4 Posted 11 August 2017 - 10:01 PM

    Major

  • Player
  • 22243 battles
  • 2,766
  • Member since:
    12-19-2014

Wow, yet another crayon-eater who doesn't even understand (or didn't read) his "evidence":facepalm:

 

View Postzen_monk_, on 11 August 2017 - 10:00 PM, said:

Tin Foil my *edited*, read the text and the Patent.


The cherry-picked part of the partent you waste of oxygen posted is solely about influencing the likelyhood of being top or bottom tier according to ones performance and has nothing whatsoever to do with RNG manipulation or winning/losing streaks. Learn to read and comprehend what you are talking about.



Cannes76 #5 Posted 11 August 2017 - 10:04 PM

    Second Lieutenant

  • Player
  • 55236 battles
  • 1,125
  • [2PDD] 2PDD
  • Member since:
    04-16-2011
This is really old news, and no, this is not why you are failing at WoT. Your poor results are a product of the fleshbag that sits 40cm from the monitor, banging the keyboard and mouse with a hammer, instead of activating the brain to overcome his opponents on the battlefield.

Pansenmann #6 Posted 11 August 2017 - 10:05 PM

    Field Marshal

  • Player
  • 30880 battles
  • 10,180
  • [WJDE] WJDE
  • Member since:
    08-17-2012

for a zen monk you seem to be quite picky about this stuff.

 

 

Meditate about this I must.



HaZardeur #7 Posted 11 August 2017 - 10:08 PM

    Warrant Officer

  • Beta Tester
  • 30940 battles
  • 753
  • Member since:
    08-14-2010

View Postzen_monk_, on 11 August 2017 - 10:00 PM, said:

Tin Foil my *edited*, read the text and the Patent.

 

I´m more into Latex, thanks.

jabster #8 Posted 11 August 2017 - 10:24 PM

    Field Marshal

  • Beta Tester
  • 11936 battles
  • 17,543
  • [WSAT] WSAT
  • Member since:
    12-30-2010

View Postzen_monk_, on 11 August 2017 - 09:00 PM, said:

Tin Foil my *edited*, read the text and the Patent.

 

One might first suggest you reread the patent and understand what it actually says and then understand what a patent is. Besides those minor points, top post.

StaryWilk #9 Posted 11 August 2017 - 10:24 PM

    Private

  • Player
  • 16798 battles
  • 35
  • [WFTTE] WFTTE
  • Member since:
    11-14-2011

Man everyone with 10k+ battle playing since early know this,

they are too scared to admit it.

Those win streaks followed by losing streaks... experinced since 2011.

But i dont mind it,putting me in bad situations  help to improve myself.


 

Edited by StaryWilk, 11 August 2017 - 10:28 PM.


ZlatanArKung #10 Posted 11 August 2017 - 11:05 PM

    Captain

  • Player
  • 1529 battles
  • 2,439
  • [T-D-U] T-D-U
  • Member since:
    12-20-2014

View Postzen_monk_, on 11 August 2017 - 10:00 PM, said:

Tin Foil my *edited*, read the text and the Patent.

You didn't,  so why should he?



Janduin #11 Posted 11 August 2017 - 11:37 PM

    Private

  • Player
  • 11208 battles
  • 33
  • [PTO] PTO
  • Member since:
    04-08-2012

That patent just talks about a skill-based mm, nothing about nerfing certain aspects of the tank if you're winning alot.

I'd bet WoT has used skill-based mm in some game mode. Easily explained. I'd bet you couldn't even make a tinfoil hat if you really tried.



Jigabachi #12 Posted 11 August 2017 - 11:42 PM

    Field Marshal

  • Player
  • 17109 battles
  • 14,775
  • [T-D-U] T-D-U
  • Member since:
    08-12-2011

View Postzen_monk_, on 11 August 2017 - 10:00 PM, said:

Tin Foil my *edited*, read the text and the Patent.

You first.

You also forgot the replays and the collected data that supports your claims.

 

But hey, what would be a month without a clueless bob mentioning the patent...

 

 



jabster #13 Posted 12 August 2017 - 05:52 AM

    Field Marshal

  • Beta Tester
  • 11936 battles
  • 17,543
  • [WSAT] WSAT
  • Member since:
    12-30-2010

View PostJanduin, on 11 August 2017 - 10:37 PM, said:

That patent just talks about a skill-based mm, nothing about nerfing certain aspects of the tank if you're winning alot.

I'd bet WoT has used skill-based mm in some game mode. Easily explained. I'd bet you couldn't even make a tinfoil hat if you really tried.

 

The part of the patent the OP quoted is about modifying tier place if you're winning or losing too many games in a session. It's a stretch to call that skill based MM.

Homer_J #14 Posted 12 August 2017 - 06:19 AM

    Field Marshal

  • Beta Tester
  • 24832 battles
  • 23,271
  • Member since:
    09-03-2010

View Postzen_monk_, on 11 August 2017 - 10:00 PM, said:

Tin Foil my *edited*, read the text and the Patent.

 

How about you read the patent, I've read it several times and it contains at least four contradicting methods which cannot possibly be implemented at the same time.

1ucky #15 Posted 12 August 2017 - 06:56 AM

    Staff Sergeant

  • Player
  • 63012 battles
  • 416
  • Member since:
    11-05-2013
This patent is not about RNG, it is about MM.

It's very simple, too.
Says that it tries to vary players being low tier or high tier.
It actually even helps players that suck as low tiers.
(By making them high tiers more often, - not by rigging RNG rolls!)

But nevertheless, it's all irrelevant, simply because it's outdated.
This was filed for patenting in Spring of 2012 and we are in Summer of 2017 now.
The MM has been reworked/changed extremely noticeably, in between.
People are still complaining every day about the change.
So it makes no sense to argue about the old MM, because evidently that's not what we're having anymore.

But anyway, even if the new MM had some leftover similarities with the old one, it still has nothing to do with RNG.
It's just MM, nothing more than that.

Btw, if you're disappointed because you love conspiracies and blaming others for your being competitively challenged or unusually underdevelopped as a person, you could also google "programming RNG" and discover that RNG in games is technically speaking never truly random.

There's multiple mathematical algorithms that try simulating something very similar to randomness, but it's never 100% the same, and that's the best thing any programmer (or anyone else) can do.

So you could freak out over it, thinking that the World of Tanks' RNG is therefore also bound to be imperfect, i.e. somehow biased or rigged.
Have fun with that.

Edited by 1ucky, 12 August 2017 - 07:08 AM.


Phobos4321 #16 Posted 12 August 2017 - 07:23 AM

    Lieutenant General

  • Player
  • 40493 battles
  • 6,486
  • Member since:
    09-27-2011

thought it was never proven to be actually in use

 

if it was would we have 65% or 33% wr players ?  shouldnt they be averaged if the patent was working ?



jabster #17 Posted 12 August 2017 - 07:25 AM

    Field Marshal

  • Beta Tester
  • 11936 battles
  • 17,543
  • [WSAT] WSAT
  • Member since:
    12-30-2010

View Post1ucky, on 12 August 2017 - 05:56 AM, said:


Btw, if you're disappointed because you love conspiracies and blaming others for your being competitively challenged or unusually underdevelopped as a person, you could also google "programming RNG" and discover that RNG in games is technically speaking never truly random.

There's multiple mathematical algorithms that try simulating something very similar to randomness, but it's never 100% the same, and that's the best thing any programmer (or anyone else) can do.

 

Well you can use a non-deterministic source for randomness. Does that make it truely random, hard to say as you quickly bump into the question of what that actually means and how would you know if you had one?

 

View PostPhobos4321, on 12 August 2017 - 06:23 AM, said:

thought it was never proven to be actually in use

 

if it was would we have 65% or 33% wr players ?  shouldnt they be averaged if the patent was working ?

 

The patent doesn't set out a method which 'balances' win-rates to average.
 

1ucky #18 Posted 12 August 2017 - 07:42 AM

    Staff Sergeant

  • Player
  • 63012 battles
  • 416
  • Member since:
    11-05-2013

View Postjabster, on 12 August 2017 - 07:25 AM, said:

 

Well you can use a non-deterministic source for randomness. Does that make it truely random, hard to say as you quickly bump into the question of what that actually means and how would you know if you had one?

I don't know about that, but for all intents and purposes, the RNG that's implemented in WoT is random enough for me.

And I think it's random enough as well for practically anyone who's not a conspiracy theorist.


Edited by 1ucky, 12 August 2017 - 07:44 AM.


HunterXHunter8 #19 Posted 12 August 2017 - 07:45 AM

    Warrant Officer

  • Player
  • 30288 battles
  • 762
  • Member since:
    04-27-2013

View PostPhobos4321, on 12 August 2017 - 07:23 AM, said:

thought it was never proven to be actually in use

 

if it was would we have 65% or 33% wr players ?  shouldnt they be averaged if the patent was working ?

 

you can platoon in OP tanks to beat the system.  3 unicum players in obj 140 will decimate players. 

 

as for 33% they are either incredibly bad or bots



a_noob_in_his_ #20 Posted 12 August 2017 - 07:48 AM

    Sergeant

  • Player
  • 41204 battles
  • 297
  • Member since:
    09-29-2013

so many people leaping to WG's defence

 

admirable white knighting

 

i can usually tell within the first few shots of a game whether i'm going to do well or not

when the tank doesn't change, i don't change (much), but my RNG varies wildly

games where i bounce or miss over half my shots and games where everything works well

almost as if my RNG+/- has already been set for that particular battle before i start

 

call it "confirmation bias" if you like

or continue to blow more smoke up WG, thinking they could/would never do anything like that

 

keep giving them money if it makes you feel better

 

peace






2 user(s) are reading this topic

0 members, 2 guests, 0 anonymous users