FUMBBL :: Online Blood Bowl League

Joined: Apr 11, 2004

Posted: Jun 01, 2005 - 18:56

Am I the only one slighty concerned that theres a possible way to cheat? Predicting your dice rolls is cheating and with his information I'm pretty sure anyone could win almost all of their games.

The way I see it, its easier to just think of the random generator as something that plucks a random number from the air. As soon as I start thinking further than this I get this strange crackling sound in my head and it hurts. Wink

Its random enough for most of my games anyway.

_________________
Nuffle Sucks!!!

CircularLogic

Joined: Aug 22, 2003

Posted: Jun 01, 2005 - 18:57

@BMMs calculations:
Actually the margin is even smaller, because you rounded down the amount of 6s available and neglected the fact, that such a densitiy of 6s is very likely to be derived from a string containing more then 1/6 events of "6".

Just a teaser... Wink

Penix

Joined: May 25, 2005

Posted: Jun 01, 2005 - 19:01

For me, it's not random enough. I've seen here more 1111 than in my whole life... (if they were for the opponent i wouldn't be reading this....xD)

_________________
Why do I foul?
'cause I can

Mezir

Joined: Aug 02, 2003

Posted: Jun 01, 2005 - 19:09

I'll back BMM's calculations up. I'm also willing to believe that Rynkky knows exactly what he's talking about.

I'll repeat my two keywords again. "Large" and "enough". I think that sums up BMM's conclusion quite well enough.

Being able to predict what number will be rolled via something that extracts the seed and the position from JavaBB, on the other hand, is more disturbing.

_________________
Build a man a fire and he's warm for a day; set a man on fire and he's warm for the rest of his life.

Idolen

Joined: Aug 02, 2003

Posted: Jun 01, 2005 - 19:48

Doesen't anybody play just for fun anymore ?

_________________
1f u c4n r34d th1s u r34lly n33d t0 g37 l41d

CircularLogic

Joined: Aug 22, 2003

Posted: Jun 01, 2005 - 19:54

I agree, Mezir, that´s what´s bugging me the whole time.

I have thought of such a program myself as the ultimate cheat (undetectabe, since you don´t mess around with what is rolled) and sometimes wondered if some of my opponents had such a tool (when they made some strange blocks, resulting in pushes only, just before blitzing my blodge-ballcarrier and roll a POW or when all the double-bothdown happend with the only guy in the whole team who has block).

In any case, Rynkky should send the program to christer/SkyJunkie - and to me of course Wink

Rynkky

Joined: Aug 03, 2004

Posted: Jun 01, 2005 - 20:24

CircularLogic wrote:

Rynkky wrote:

I do get your point. And you're right here. But as I said in my previous post distribution is pretty even for 100 rolls in Java random generator. Thus in one BB game (1000 rolls you said?) the distribution is close to even as well in most cases (please note I did not say every). Thus making my original claim valid.

To make your claim valid, the amount of each number must be exactly the same for each number in EVERY string. This is a prerequisite for the claim, that past results affect the next events.

I am not really sure that I understand your point here. Are you saying that taking any number of sequential digits (divisible by six) from the pool of 2^32 number must have even distribution? Well that is not the case, because you can't have even double if that is what you meant.

I really find it weird how you neglect to notice the fact that if any amount of numbers has almost even distribution but NOT homogenous distribution it still can be estimated how propable after getting a group of ones is to get another group of ones or something else.

E.g. We have 600 numbers who have even distribution (i.e. 100 ones, 100 twos, etc.). Now, if I understood you correctly, it is possible to estiamte something if I get six ones? What if we have 599 numbers which are almost evenly distributed (variance being 1% so there will be 94-106 ones, 94-106 twos, etc.) now we cannot estimate anything? To me situation is still same, there is just larger error margin. BUT I am not mathematician so I may be wrong but thinking this case even with common sense says that it is still fairly easy to pretect what may happen.

CircularLogic wrote:

Regarding the hack/no-hack discussion:
Decryting packages not intended for me and using the information, isn´t this part of the thing called hacking?
From a gamer´s point of view you are clearly using a hack -> gathering information you should not have according to the game.

What if you don't need to decrypt anything? Everything is delivered to you on a golden plate. Is it still hacking?

And as you stated it is all matter of the point of view..

BadMrMojo

Joined: Aug 02, 2003

Posted: Jun 01, 2005 - 20:38

Rynkky wrote:

CircularLogic wrote:

Regarding the hack/no-hack discussion:
Decryting packages not intended for me and using the information, isn´t this part of the thing called hacking?
From a gamer´s point of view you are clearly using a hack -> gathering information you should not have according to the game.

What if you don't need to decrypt anything? Everything is delivered to you on a golden plate. Is it still hacking?

And as you stated it is all matter of the point of view..

Can we just settle on calling it cheating?

(No, I'm not accusing anyone of anything, just trying to cut to the point. Regardless of what you call it, using such a thing would obviously be completely and totally against the rules and spirit of the league.)

_________________
Ta-Ouch! of BloodBowl
Condensed Guide for Newbies

Rynkky

Joined: Aug 03, 2004

Posted: Jun 01, 2005 - 20:40

CircularLogic wrote:

I agree, Mezir, that´s what´s bugging me the whole time.

I have thought of such a program myself as the ultimate cheat (undetectabe, since you don´t mess around with what is rolled) and sometimes wondered if some of my opponents had such a tool (when they made some strange blocks, resulting in pushes only, just before blitzing my blodge-ballcarrier and roll a POW or when all the double-bothdown happend with the only guy in the whole team who has block).

In any case, Rynkky should send the program to christer/SkyJunkie - and to me of course Wink

I think that main problem is not the prediction. It is prevention of the prediction. Currently it is very easy to make a predictor that gives you numbers that will be rolled next. BUT to prevent making such program is not a trivial task. I can think of one secure solution (there is other, but that would not be secure in the same degree):

Centralized random number generator servers like the proxy servers. That would of course increase the data flow in net BUT noone could twiddle with the numbers or if anyone did, opponent would notice it very easily.

JanMattys

Joined: Feb 29, 2004

Posted: Jun 01, 2005 - 20:43

Am I the only one thinking that this thread is completely absurd????

I mean...

Let's assume I and you play a BB game.
Let's assume I always roll 6s for blocks-dodges-gfits-pickups-passes and I always roll ones for armor rolls.
Since Armor rolls are 2d for everyblock, I can assume the number of armor rolls and the "other" number of rolls will be quite the same.

In this case I would have rolled (say) 5 hundred 6s and 5 hundred 1s.
Is this ok? Apart from the very unlikely distribution (out of 1000 rolls), the average die is 3.5 as it *should* be.

Well... don't you think I will win this game? I think I will win this game FOR SURE. No matter if I never break armor... I will always succeed in everything, so I will have an AWESOME defence and an unstoppable offence.

I can't see the point behind all this math and Java talk. Rynkkyn says that once the seed is given, if I roll "good rolls" they will be taken away from the pool of possible results, thus giving my opponent a higher chance of "bad rolls".

This is NONSENSE. Because good or bad rolls do not exist in fumbbl "as an absolute". I really don't care if I roll ones for injury rolls... if I can have my sixes when I need them. So (imho) this chattering is useless. It's the INGAME IMPORTANCE of the roll that determines if that "1" you roll is crucial or not. The distribution of the numbers (unless a VERY high amout of rolls are ones) is quite meaningless in predicting "who will have an advantage".

_________________

Rynkky

Joined: Aug 03, 2004

Posted: Jun 01, 2005 - 20:47

BadMrMojo wrote:

Rynkky wrote:

CircularLogic wrote:

Regarding the hack/no-hack discussion:
Decryting packages not intended for me and using the information, isn´t this part of the thing called hacking?
From a gamer´s point of view you are clearly using a hack -> gathering information you should not have according to the game.

What if you don't need to decrypt anything? Everything is delivered to you on a golden plate. Is it still hacking?

And as you stated it is all matter of the point of view..

Can we just settle on calling it cheating?

(No, I'm not accusing anyone of anything, just trying to cut to the point. Regardless of what you call it, using such a thing would obviously be completely and totally against the rules and spirit of the league.)

Aye. Cheating is what I'd call it as well.

And my apologizes to you BMM. I hope that you read my post carefully (the one where I criticized your theories) I was not making fool of you, only of the facts given. I really do not have anything against you but after reading my post someone could think that it was personal, which was not the case. I even mentioned that there was lot of wrong facts with faint of truth AND by saying that "person who wrote that obviously had a glue but lost it" I meant that reading those theories I really noticed that you know alot about computer random generators but when speaking of Java your information was not correct.

thesquig

Joined: Apr 11, 2004

Posted: Jun 01, 2005 - 20:49

Anything that could prevent cheating in anyway shape or form is a great thing IMO. I'm not sure of how possible it is to cheat but I know that it anoys to think about anyone doing it. Also its sad to say that protection against cheating is needed purely by the fact that people disconnect when they have bad luck.

It is cheating, either way you look at it. Imagine a tabletop game where someone rolls a huge amount of dice at the start of the game and records them on a piece of paper. Then he uses these numbers for his dice rolls during the game. Its cheating.

_________________
Nuffle Sucks!!!

Azurus

Joined: Aug 02, 2003

Posted: Jun 01, 2005 - 20:50

Well done BMM for the number-crunching post. I was going to post something almost identical when I got back (didn't have time when I made my first post), thanks for saving me the effort. Wink

As for cheating, I'd think it would be fairly obvious to see. A player seeing the numbers in advance would rarely fail an important dice roll, and could instead simply soak up the '1's on 2d-blocks, RRs or whatever. Since the 'luck' meter actually measures success, cheating in this way would likely result in ludicrously high luck scores, unless the cheater was trying really hard to balance it back. That sort of thing get's noticed after a while. (And hurrah! An actual use for the luck-o-meter! Twisted Evil

)

_________________
*This is a public safety announcement. Azurus is a cynical, sarcastic idiot. Please ignore any and everything he may say. Thank you for your attention.*

Jinxed

Joined: Jul 04, 2004

Posted: Jun 01, 2005 - 20:55

All I want to add is that I (=me!) am Jinxed.

None other.

Thanks to Mr_QB and Ludicruz to point it out. Very Happy

_________________
Nuffle sucks

DonKosak

Joined: Apr 06, 2004

Posted: Jun 01, 2005 - 20:56

I can now confirm that we have a highly skilled Javaprogrammer among us. I've tested the program in a localhost non-Fumbbl game and it surely predicts the coming rolls. Rynkky deserves credit for proving that it is possible, and as I believe that he wouldn't have posted in this forum if he wanted to cheat, he deserves credit for not doing so too.

I totally agree, that using the program here on Fumbbl would be cheating. I have deleted the program from my computer.

I sincerely hope, that SkiJunkie and Rynkky will get in contact - it could make the client an even greater tool and Fumbbl even funnier.