Jedi Archive
Thread: Official list of common Jedi Board Misconceptions
No I haven't, and thanks for that link! That's awesome! Ok, so I see how a random number could be pulled from a client system. Excuse my ignorance again, but riddle me this: I have provided a link to my mobo below. If this RNG is intel based.. what about AMD processors and boards that don't use intel chipsets? I imagine the RNG has been reproduced and used across the board, but I am brainstorming here. That's awesome, and thanks again for that link, Katmer!
Katmer wrote:
Sorry, Riddix, but you apparently haven't been keeping up to speed on the world of random number generators.
Please see http://www.intel.com/design/security/rng/rnghow.htm
The Intel 810 and later chipsets (starting with Celeron, IIRC) provide access to a thermal noise random number generator.
This meets all practical tests for "random", and is not algorithmically driven to produce its results (the results aren't dependent on a "seed"). Hence, it is precisely the sort of real-world example of randomness referenced earlier.
Hope this helps.
The Intel 810 and later chipsets (starting with Celeron, IIRC) provide access to a thermal noise random number generator.
This is only trueif you believe white noise generated from that resistor is random. Give me all the variables involved that create the noise and I will show you your seed. Again, just because we don't know the variables involved does not mean they do not exist.
Lowca_Fatal wrote:
It has happened TWICE.
Have you read anything?? Just because you haven't seen screenshots doesn't mean it hasn't happened. Are you really that nieve?
Yes, I have more than a passing familiarity with these boards, but I can't claim to see every post that someone makes.
I never claimed i needed screenshots, those can be just as misleading as what someone says.
Can you provide a link to this claim?
Xader, if you had read my enitre post and not just the first line of the part dealing with patterns you would know that I don't buy the pattern theory. I just can't say beyond a shadow of a doubt one can't be found. Just like I can't say beyond a shadow of a doubt that theres no magical man in the sky who judges people but I'm 95% sure there isn't. The point of the post was show the proof. Show me a pattern working the majority of the time and unlocking slots.
Thummp wrote:
Riiiiiiight. OK. Someone has never designed software under a deadline.
Lowca you gonna answer my question? Did you even read it?
--Thummp
Deadlines are no excuse for not doing the game justice.
Turn up the graphics in the game all the way, in the highest resolution. You then tell me they don't care about the game, and don't spend time on things.
This is where your wrong because it is not a factual statement. It is, in fact, possible to prove that it is random and has been done. Just because you don't understand the math behind it doesn't make it not true.
Lowca_Fatal wrote:
I'm not saying its not random, I'm saying its impossible to prove its not.
Xader_Vartec wrote:
This is where your wrong because it is not a factual statement. It is, in fact, possible to prove that it is random and has been done. Just because you don't understand the math behind it doesn't make it not true.
Lowca_Fatal wrote:
I'm not saying its not random, I'm saying its impossible to prove its not.
OK, Prove these numbers are random
1 5
Prove it.
What, you can't?? Wait, you don't have enough data.
100 Jedi mean nothing. You can't prove squat with 100.
Lowca_Fatal wrote:
/bump
If I see one more post of people assuming 4+1, or one more post of people saying everything is random, will make my head explode
It is random, there is no pattern
/takecover
Xader_Vartec wrote:
Lowca_Fatal wrote:
Lowca_Fatal,
Let me put this in very simple terms for you:
If you want to waste your time looking for patterns you are welcome to. But be warned that signifcant mathematical minds, ones that are better than mine and I know their creditials so I trust them, have looked at this with the available data and determined that IT IS RANDOM.
Lowca_Fatal is in the right church, but the wrong pew. =)
Statistical analysis on such a (relatively) small sample certainly cannot "prove" that the system is random. It can, however, present a CONFIDENCE INTERVAL in how likely it is that the process is non-random in nature.
To date, from all the RELIABLE data that I've seen, I don't see anything that indicates anything more likely than a random selection process.
For contradictory data, we need to very carefully examine the claims to determine if they are accurately reported in all details, as either an erroneous report or a character with a bug does not disprove the primary mechanism.
Is it possible that there is onealgorithm thatfits almost all cases, and another for a tiny fraction of the remaining population? Sure. But I consider it unlikely that the Devs would have codedit that way.
But there's a more compelling argument than simple doubt. If we see that the vast, vast majority of cases are covered and explained well by a given theory, then FOR ALL PRACTICAL PURPOSES it makes sense to set your strategy based on the assumption that your character will NOT be the exception to the rule.
Beyond a certain point, it simply doesn't matterif thereis a tiny fraction of the population that (erroneously or not) doesn't fit the model. If almost all characters do, it makes sense to simply assume that yours will as well.
This is even more the case when a rival theoryoffers no different strategy from the original. When your plan of attack doesn't change, why bother posturing about which is the correct theory?
Lowca_Fatal wrote:
Xader_Vartec wrote:
This is where your wrong because it is not a factual statement. It is, in fact, possible to prove that it is random and has been done. Just because you don't understand the math behind it doesn't make it not true.
Lowca_Fatal wrote:
I'm not saying its not random, I'm saying its impossible to prove its not.
OK, Prove these numbers are random
1 5
Prove it.
What, you can't?? Wait, you don't have enough data.
100 Jedi mean nothing. You can't prove squat with 100.
Your right. The problem is the analysis also showed that the amount of data I had ~200 data points WAS SIGNIFICANT ENOUGH to do a data analysis. That's how I know you don't know what your talking about is during data analysis part of what you analys is whether or not you have enough data points for a signficant result.
Do you want to see my spreadsheet and calculations? You know that someone else, who did this EXACT SAME WORK INDEPENDENTLY from me post his very detailed results with his math to the board. He did it at about the time I was finishing my analysis but before I had turned over the research to a friend. Once I saw his results closely matched mine I realize that I was on the right track.
You KNOW that is how REAL scientific research is done right? Independant verification of repeatable results?
Katmer wrote:
Xader_Vartec wrote:
Lowca_Fatal wrote:
Lowca_Fatal,
Let me put this in very simple terms for you:
If you want to waste your time looking for patterns you are welcome to. But be warned that signifcant mathematical minds, ones that are better than mine and I know their creditials so I trust them, have looked at this with the available data and determined that IT IS RANDOM.
Lowca_Fatal is in the right church, but the wrong pew. =)
Statistical analysis on such a (relatively) small sample certainly cannot "prove" that the system is random. It can, however, present a CONFIDENCE INTERVAL in how likely it is that the process is non-random in nature.
To date, from all the RELIABLE data that I've seen, I don't see anything that indicates anything more likely than a random selection process.
For contradictory data, we need to very carefully examine the claims to determine if they are accurately reported in all details, as either an erroneous report or a character with a bug does not disprove the primary mechanism.
Is it possible that there is onealgorithm thatfits almost all cases, and another for a tiny fraction of the remaining population? Sure. But I consider it unlikely that the Devs would have codedit that way.
But there's a more compelling argument than simple doubt. If we see that the vast, vast majority of cases are covered and explained well by a given theory, then FOR ALL PRACTICAL PURPOSES it makes sense to set your strategy based on the assumption that your character will NOT be the exception to the rule.
Beyond a certain point, it simply doesn't matterif thereis a tiny fraction of the population that (erroneously or not) doesn't fit the model. If almost all characters do, it makes sense to simply assume that yours will as well.
This is even more the case when a rival theoryoffers no different strategy from the original. When your plan of attack doesn't change, why bother posturing about which is the correct theory?
Ok, I'll concede. It is only proven within a reasonable confidence level. That was good enough for me.
Xader:
If there's proof, how about sharing it with us? Everyone (myself included) that believes there could be a pattern have asked many, many times for someone to SHOW THE PROOF that there are no patterns. Prove it.
And I'll state it yet again: Ibelieve there are patterns, but I can't figure them out. I've actually almost reached the point of not even trying anymore because it seems to be beyond me. I've changed my thoughtsseveral times on this, but still no dice.Of course, I won't sit here and argue that there ARE patterns, but I will argue that I believe there are. I'll also give very good evidence of not only WHY I believe there are patterns, but offer ways they could have been done. Even that, however, hasn't lead me to find a pattern or even a series of patterns.
And believe it or not, there HAVE BEEN folks who have unlocked based on a pattern guess. Of course, it could be pure luck, I won't deny that. However, it seems to me that if you have PROOF of even 1 occurance of a pattern providing that 5th profession, more time should be spent analyzing it to show it was "luck", or to advance the idea to find it for others. I mean, the first steam engines didn't work all the time, did they? We are still using DOS 2.0, right? There's absolutely no reason to further an idea, run it through its paces, and try to prove or disprove anything, right?
Show me your proof. You've said you have it, share it.