Moderated Coin Flipper

[Thermal]

P1.One in six thousand is based on an American nickel, which has the highest thickness to diameter ratio of all American coins. Change name to Nickel Flipper.

P2. This is consequently an American centric interpretation of coin flipping, and hence needs to have the National Anthem played in full before each flip. Kneeling posters will be publicly shamed.

P3. Who flips nickels? What kind of cheapskates are doing the flipping here?

P4. ????

C. Profit

 

[jimbob]

Ok... your wishes have been fulfilled.

Coin Flipper V4

• allows choosing the RNG
• allows setting the Edge probability 1= no Edge case to 1/10,000,000,000... defaults to 1/6,000
• from 10 to 10,000,000 coin flips can be done in one go... to do more just go again
• Except on the TRNG only 10 to 10,000 flips at a time.

[IMGW=400]http://GodIsADeadbeatDad.com/CoinFlipper4/images/pseudo.png[/IMGW] [IMGW=406]http://GodIsADeadbeatDad.com/CoinFlipper4/images/crypto.png[/IMGW]

[IMGW=400]http://GodIsADeadbeatDad.com/CoinFlipper4/images/true.png[/IMGW]
​

You still have not explained how using something that is unpredictable but inherently determistic tells people anything about processes that are predictable but based on the combined effects of random events, for example pressure in a piston based on 10^26 individual molecules in random trajectories.

 

[theprestige]

Ok... your wishes have been fulfilled.

Coin Flipper V4

• allows choosing the RNG
• allows setting the Edge probability 1= no Edge case to 1/10,000,000,000... defaults to 1/6,000
• from 10 to 10,000,000 coin flips can be done in one go... to do more just go again
• Except on the TRNG only 10 to 10,000 flips at a time.

Interesting how all three sets converge on 50/50. But you may want to check your code; it seems you might have a slight bias towards tails.

 

[Leumas]

Interesting how all three sets converge on 50/50. But you may want to check your code; it seems you might have a slight bias towards tails.

Why are you so concerned what a useless app that eschews having anything to do with reality, does or does not do?

And I think you need to check out the meaning of the word "converge"...

Hint:... converge does not mean oscillate.

... But you may want to check your code; it seems you might have a slight bias towards tails.

Prove it.

 

[theprestige]

Why are you so concerned what a useless app that eschews having anything to do with reality, does or does not do?

Entertained would be a better word.

And I think you need to check out the meaning of the word "converge"...

Hint:... converge does not mean oscillate.

And yet your running averages do indeed converge on a 50/50 split between heads and tails. What starts out as a very swingy oscillation damps down over time, as each successive flip becomes less influential to the overall trend towards a 50/50 split.

Prove it.

All three of the v4 results you presented have Tails slightly above 50%, and Heads slightly below 50%. I think this is a sufficient reason to believe your code might have a pro-Tails bias. I'm not going to try to prove whether it does or not (I'm not that "concerned"). Just saying you might want to check your code.

 

[Leumas]

You still have not explained how using something that is unpredictable but inherently determistic tells people anything about processes that are predictable but based on the combined effects of random events, for example pressure in a piston based on 10^26 individual molecules in random trajectories.

I am glad to see that you do not deny that randomness is inherent in the natural reality.

But I think you need to look into good PRNGs and TRNGs and how they can certainly be nondeterministic...

Moreover the True RNG in the app uses TRUE RANDOM data from a hardware that uses atmospheric noise.

Nevertheless... don't you think it is fun to be able to try out what Nx10⁶ coin tosses can do in simulation when there is no way you could have tried out the same physically.

 

[Leumas]

Entertained would be a better word.

Ah... so it is not a useless worthless app after all... huh?

And yet your running averages do indeed converge on a 50/50 split between heads and tails. What starts out as a very swingy oscillation damps down over time, as each successive flip becomes less influential to the overall trend towards a 50/50 split.

You really need to learn more about probability and probability distributions and what they mean.

All three of the v4 results you presented have Tails slightly above 50%, and Heads slightly below 50%. I think this is a sufficient reason to believe your code might have a pro-Tails bias. I'm not going to try to prove whether it does or not (I'm not that "concerned"). Just saying you might want to check your code.

Yes... a hasty generalization fallacy.... I suggest you look at what this fallacy means...

 

[jimbob]

I am glad to see that you do not deny that randomness is inherent in the natural reality.

But I think you need to look into good PRNGs and TRNGs and how they can certainly be nondeterministic...

Moreover the True RNG in the app uses TRUE RANDOM data from a hardware that uses atmospheric noise.

Nevertheless... don't you think it is fun to be able to try out what Nx10⁶ coin tosses can do in simulation when there is no way you could have tried out the same physically.

Oh if you look at my posting history, you will see that I have argued that intelligent life is only here due to random events.

Or indeed that truly random quantum events are the most parsimonious explanation for there being *any* structure in the universe, as I don't see how one gets anything other than infinite uniformity from a deterministic expansion from a uniform starting point for the big bang.

10^6 is small, I've run Monte-Carlo simulations at work for ion implant trajectories before now. And they're still "bumpy".

 

[jimbob]

Oh if you look at my posting history, you will see that I have argued that intelligent life is only here due to random events.

Or indeed that truly random quantum events are the most parsimonious explanation for there being *any* structure in the universe, as I don't see how one gets anything other than infinite uniformity from a deterministic expansion from a uniform starting point for the big bang.

10^6 is small, I've run Monte-Carlo simulations at work for ion implant trajectories before now. And they're still "bumpy".

So yes, ultimately I think things happen because acausal events are fundamental to the universe, and that that is the only way to avoid the "turtles all the way down" question. Basically, random events in a meaningless universe.


Or **** happens because **** happens.

 

[theprestige]

Ah... so it is not a useless worthless app after all... huh?

There's nothing useful or worthwhile about someone accidentally stepping into an open manhole. But it is, on occasion, hilarious to third-party observers. Just because your app has no real value, that doesn't mean your antics around it aren't fun to watch.

You really need to learn more about probability and probability distributions and what they mean.

I'm pretty sure I need no such thing.

Yes... a hasty generalization fallacy.... I suggest you look at what this fallacy means...

I'm not making a generalization, though. I'm saying there are some indicators that there might be a problem with your code.

 

[Thermal]

... converge does not mean oscillate

Is this really the ridiculous distinction you are relying on? Really?

You flip one coin. The odds of it coming down in a 50/50 proportion are dead zero.

Flip another. It's 50/50 that you will arrive at an event 50/50 ratio, and 50/50 that it will not (ignoring edge landings for the moment).

Flip a third. The you will have either 100%, 33 1/3%, or 67 2/3% distribution, and so on and on.

As the flips increase, the average will get closer and closer to a reliably consistent +/- 50/50, drifting back and forth hitting dead on 50/50 and back and forth while all the time, as theprestige correctly notes, getting closer and closer to being exactly 50/50, but like a limit that approaches zero, it will never stay there consistently, which he appropriately calls convergence.

I am confident that every member posting and reading is aware of all this, and has been since primary school. Are you seriously saying that this is the intellectual tour de force that you've hitched your argumentative wagon to?

 

[jimbob]

There's nothing useful or worthwhile about someone accidentally stepping into an open manhole. But it is, on occasion, hilarious to third-party observers. Just because your app has no real value, that doesn't mean your antics around it aren't fun to watch.


I'm pretty sure I need no such thing.


I'm not making a generalization, though. I'm saying there are some indicators that there might be a problem with your code.

And an easy way to check it would be to run another ten runs of the V4, treating each run as a single event and working out the odds of the distribution of heads and tails assuming a binomial distribution of ten events. If you felt a bit more funky, one could look at the initial 4 runs all favouring tails as indicating a prior probability.

 

[jimbob]

Is this really the ridiculous distinction you are relying on? Really?

You flip one coin. The odds of it coming down in a 50/50 proportion are dead zero.

Flip another. It's 50/50 that you will arrive at an event 50/50 ratio, and 50/50 that it will not (ignoring edge landings for the moment).

Flip a third. The you will have either 100%, 33 1/3%, or 67 2/3% distribution, and so on and on.

As the flips increase, the average will get closer and closer to a reliably consistent +/- 50/50, drifting back and forth hitting dead on 50/50 and back and forth while all the time, as theprestige correctly notes, getting closer and closer to being exactly 50/50, but like a limit that approaches zero, it will never stay there consistently, which he appropriately calls convergence.

I am confident that every member posting and reading is aware of all this, and has been since primary school. Are you seriously saying that this is the intellectual tour de force that you've hitched your argumentative wagon to?

With something that is definitely not random but completely deterministic even if we don't cannot predict the output numbers.

 

[Leumas]

With something that is definitely not random but completely deterministic even if we don't cannot predict the output numbers.

How do you know it is deterministic and not random if you cannot predict it?

I thing you need to define random.

Moreover... do you know what a TRNG is? Do you know what the T in TRNG stands for?

 

[Leumas]

And an easy way to check it would be to run another ten runs of the V4, treating each run as a single event and working out the odds of the distribution of heads and tails assuming a binomial distribution of ten events. If you felt a bit more funky, one could look at the initial 4 runs all favouring tails as indicating a prior probability.

No... no... that won't do... how can one continue to be CONCERNED when one's concerns can be alleviated by just EXPERIMNETIMNG with things instead of doing bare assertions and hasty generalizations in order to remain CONCERNED.

And why 10 runs only... why not 100 or 200 and not just at 10M but also at 10K and 100K and 1M and do it 100 or 200 times for each and then SEE FOR ONESELF stuff that one cannot do in real life.

 

[dudalb]

When it comes to flipping coins, that app ain't no Harvey Dent.

 

[jsfisher]

Entertained would be a better word.

Yeah, that is a good word for it. We lead very boring lives, I guess.

...
Just saying you might want to check your code.

Oh, so very true. Real mathematics and computer math don't always agree, and program bugs can sneak in all sorts of non-obvious ways.

I wonder if Leumas knows which of the intervals [0, .5) and [.5, 1) has more elements (not that that necessarily matters to his applications).

 

[jimbob]

Sigh, if you perform 10 runs, for v4 each of 10k and each of those 10 runs favours tails, then there would have been a 1 in 2^10 chance that it was a fair program evenly balanced between heads and tails.

You can also look at the binomial distribution to check what the odds are if it's some other set, and assuming that there's a reason from previous observations to be slightly suspicious that it favours tails.

As to your other question, if you know the seed values, you can repeat the calculation. That is not what happens with a genuine random event.

One can only assign a probability of occurrence within a particular timeframe

 

[Leumas]

As to your other question, if you know the seed values, you can repeat the calculation. That is not what happens with a genuine random event.

And when you do not... then you cannot... which is what happens in genuine random events.... no?

So when you use a random seed... it is random.

 

[Leumas]

When it comes to flipping coins, that app ain't no Harvey Dent.

Deliberately so... out of deference to my god... Batman.