An exploration of randomness.
  • 266 posts
  • Page 10 of 18
elysium5 wrote:
@aeronautic
I would like to point out Matty's response to that post:

http://www.dominating12.com/forum/?cmd=topic&act=view&id=879&page=9#post-17881

as well as all of Matty's other posts on the subject.

My comment was directed in general, however, I would also like to point out that without directly explaining why my_rand is an inferior program to -'insert blank program _________' and also explaining which available program is superior and why, continuing to argue this is moot.

Vexer acknowleged that here may be something to your suggestion and Matty explained why it would likly make no difference at all, so until this is actually programmed and tried, there is not much point in discussing this further.

That being said, and no offence to Vexer, if a variance in the program is written and applied to this one, don't you think some of these (for money) gaming sites or scientists who work on probability equations and programs would have come up with somthing better than mt_rand?
"Bad Deadpool... Good Deadpool!"
aeronautic wrote:
@elysium5
Yes I believe that if there was a better way to program a fuction then the experts would have done so. However, I believe the vulnerability of the function/s is the computer itself! Which is why I initially suggested that it not be allowed to compare its own results!!
I am right on board with this and highly respect everything the programmers of this site do and are able to do.
Hyd yn oed er fy mod Cymraeg , dim ond yn siarad Saesneg, felly yr wyf yn gobeithio y bydd y cyfieithu yn gywir.
aeronautic wrote:
On a completely different part of the subject and back-tracking quite a bit to my original suggestion.
Lottery computers are able to process consecutive Raffle Numbers in the millions to terminals all over Europe without duplication or delay.
What would be wrong with having another computer that generates the random dice with mt_rand or whatever is best, which stores them consecutively by the billion in the order they were generated ready for a call for a dice from the main computer and its program? They can just be fed to the distribution part of the program in the order Attack, Defence, which ever next, or which ever way they are currently distributed.

The fact that the main program needs dice for many different players at any one moment and that they are told to be allocated individually to Attack then Defence consecutively, randomises the already randomly generated dice stored and the program and main computer have no programming to compare what it has decided, as it is now only programmed to distribute consecutively from an external sealed database.
Just a thought for the next cuppa!
Hyd yn oed er fy mod Cymraeg , dim ond yn siarad Saesneg, felly yr wyf yn gobeithio y bydd y cyfieithu yn gywir.
elysium5 wrote:
Well, I'm not by any means a computer programmer or an expert on the subject. If there is some varience involved in how a compute compare's its own results, I would like to know more on how this occurs and find some information that would explain the best way for a computer to compare its own results when using the mt_rand function and have the least amount of varience possible while producing said results.

I have so far not been able to find anything that explains how these variences occur depending on the computer comparing its own results and how to affect a change in those results.

Any information you could find on the subject would be appreciated. I would like to learn more.
"Bad Deadpool... Good Deadpool!"
aeronautic wrote:
Well I haven't got a clue why computers do what they do, but their abilities are man made and they seem to follow patterns...
It has come to my attention recently (just about the time I decided to retire) that there are other patterns that I was unaware of and certain people are/were.

I won't give details here in case the info is correct and it is used for an advantage. Please feel free to PM me if you are a programmer or admin and I will discuss this further with you.

I would also like a programmer to set up a live 1v1 deathmatch (private game (in secrecy)) with a 3 minute counter with no points awarded or deducted, so that we can pile troops next to each other and then put it to the test for accuracy, will this be possible?
Hyd yn oed er fy mod Cymraeg , dim ond yn siarad Saesneg, felly yr wyf yn gobeithio y bydd y cyfieithu yn gywir.
Vexer wrote:
I've tested that very scenario dozens of times already. Don't believe me? Check the game log of my test game 49797. Scroll down a bit and you'll see some massive attacks. If you calculate the results you'll see they are very close to a 54% attackers advantage, as they should be.

With the number of dice rolls on the site at any given time I find it hard to believe that you could find a pattern even if one existed. If you did find a pattern it's by coincidence. Humans were programmed by nature to find patterns, even when they do not exist. Ever found a face in a cloud?
aeronautic wrote:
@Vexer, completely not what I am talking about! Not dice patterns at all.
Want to know more?

p.s. Your game log only shows 2 recent games.
Hyd yn oed er fy mod Cymraeg , dim ond yn siarad Saesneg, felly yr wyf yn gobeithio y bydd y cyfieithu yn gywir.
Vexer wrote:
I'd rather you share it here instead of a private message so that the whole community can evaluate it. It there is indeed some pattern that can be abused then it will be easiest to confirm and fix if everyone knows about it.
aeronautic wrote:
I'm not sure if it can be fixed unless there is something in the program code that causes it!
From what the programmers have stated in this thread, it doesn't look like anything has been placed in the code that could cause this.
They only make reference to 'mt_rand' and as long as there are no time or period attributes added to the code neither in the generation nor the distribution part of the code, it is perhaps computer processor generated, if it even exists at all.

If you want me to state it here, fine!

I was told about a 'time bias' by 'westchester' who wanted to bring it up in this thread, but all the while was exploiting it. I told him at the time to report his discovery regardless of his claimed sources.

I watched the player exploiting his bias knowledge and true to his word, he was able to make attacks on any amount of troops at set times during a turn and gain the majority of kills each time. Not just in 1 game but many.

The player has since been banned for other reasons and can no longer exploit it nor bring it to bare.

These are all his words not mine!
He said he had received some prior training from Vexer due to his poor playing techniques. During this time, he was told, (at least this) "if you ever initially lose 2 in an attack, stop".
This, he then discussed with another member of staff Crystal, who suggested the reason was 'time bias'!
He said, he was told of a 35 second change over in bias during either a 3/5/7 minute timer (I don't recall which one he first discussed, but think it was the 5 minute timer).
He claimed that depending on turn order position, he could place his attacks inside these 35 second pockets and gain an attack bias.
According to his theory / beliefs, if he attacked in the first 35 seconds of the timer and lost 2, he could wait until the next 35 second pocket and continue the attack and use that same pocket each time. He also discovered that when in an odd player amount game, the order of bias switched every turn, so for example, the first attack could be in the 1st or 3rd pocket, the next attack would be in the 2nd or 4th pocket.

As I stopped playing live games prior to this belief being confided in me, I have not been able to test it, but would welcome the staff or programmers setting up some games to put it to the test, which I would gladly take part in.

Just so that you know, I hold no faith in this theory / belief due to the discussions above, but the fact remains that what I saw in his final few live games, he was winning winning winning and gaining half the map very quickly, much to the other players disbelief! he even boldly told players in chat that he had knowlege of dice bias. You should still have access to his game logs and will see that he would attack 4 with 4 or less, 6 with 6 or less or 8 with 8 or less and still get victories time after time, too much to call luck / coincidence and his rank was rising fast.

Okay, you can stop laughing now, it took me a while to stop laughing at first until I saw what he was doing with his beliefs.
Just test it and then you can put it to bed!

I advise caution with allowing this post to stay on view if you find some substance to this!
Hyd yn oed er fy mod Cymraeg , dim ond yn siarad Saesneg, felly yr wyf yn gobeithio y bydd y cyfieithu yn gywir.
Vexer wrote:
There is no substance to this. If any one player duplicates this then it would be mere coincidence. He got a lucky streak and attributed it to his timings.

I wish he had copied and pasted what I said to him instead of sharing all this bs.

There is no bias in the dice related to time. It is true, however, that if you roll now you will get a different roll than if you roll 1 second from now. I suggested to Westchester that if he was getting a bunch of bad rolls to stop attacking for a few seconds. I told him that he would get a different dice roll than he otherwise would have, which is true. But that does not mean that he would get a better roll, just a different one. The dice are random, period. You cannot predict what the dice are going to be. The real reason why I told him this was to slow down his attacking so he would stop and think before proceeding in hopes he would stop attacking before leaving himself and his victim easy prey for the next player. He was a horrible suicider.

Putting delays into your attacks is equivalent to changing up your roll or having your lucky girlfriend blow on the dice. It's superstitious. You still can't predict what you are going to roll but it's something we do. If your bad rolls continue you tell yourself that you didn't roll it right--or in this case that your timing wasn't right. It's nonsense. Random is random...unpredictable.
Thorpe wrote:
I have also told players to click...pause...click...pause

It just lets you control your impulse to keep losing and makes you think you are doing better...but who knows ...lol
95.5% of the time you kill a players cap before your 2nd turn in... you fail or die next
aeronautic wrote:
As I said, I never believed a word of the bs and actually laughed out loud when he claimed to know about a time related att/def bias.
I, like Vexer, also believe that time can have no relation to the result apart from the luck factor of which numbers appear at that precise moment.
I also fully trust that no time attributes have been placed in the program.

However, the fact remains that he 'got lucky' in every attack of every game since claiming to know the times.

Let's for a moment put aside any misquotation and disregard programming, randomness, luck, voodoo and whatever else that can affect virtual dice and assume that something made him look for a bias in this way and by some fluke, he came across something that just happens!?
Bare in mind that he claimed to be finalising the time pockets for other timer settings and therefore would also have been testing these in some games.

Do the staff and players here think we should
(a) Check his last 20 game logs?
(b) If seen to be bias, openly test the theory?
Hyd yn oed er fy mod Cymraeg , dim ond yn siarad Saesneg, felly yr wyf yn gobeithio y bydd y cyfieithu yn gywir.
elysium5 wrote:
I checked his last 10 games and even though he won a few, It was because of one lucky move he made in the last game and bad moves by other players in the others. He certainly did not have favourable dice.

The real secret to getting favourable dice is to do what I do; stand on one foot with your left index finger touching your nose and that same elbow sticking staight out to the side - ***IT MUST BE EXACTLY PARRALEL TO THE FLOOR - and then click the mouse with your right pinky toe every three mississippi's. Everything else is just superstitious gibberish.
"Bad Deadpool... Good Deadpool!"
Cireon wrote:
That means it must be working!
“This is how humans are: We question all our beliefs, except for the ones that we really believe in, and those we never think to question.”
- Speaker for the Dead, O.S. Card