Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Random

Status
Not open for further replies.

Dr.EM

New Member
I tried to build a circuit which would randomely flash an LED. I guess it works, at least it looks random, of course its not really. But I was thinking about the criteria for something to be truly random. For instance, this device will flash the LED only, so the LED operates in only 2 states, on or off. To be of total randomness, it needs to be able to operate in an infinite number of states (bright, dim etc). Also, the LED will always change state at least once every 2 seconds. To be random, it could be off for a billion years and on for just a second. So I was thinking, these are dimensions, ie. time. There are supposedly numerous dimensions, some say 9, some say 11, some more. So, surely, for something to be completely random, it has to perform an action which is outputted in any of these dimensions, or multiple ones. Hmm, this is hard to explain, but imagine an equation where everything is X :lol: . As an example, something may happen in a dimension or multiple dimensions, under the rules of every other dimension, yet that something is undesigned.
 
All you need is some white noise. and drive it in a PIC and you shod get random 1 or 0 and than tralsate that on the LEDs
 
Trying to solve : X = infinite + 1
...
...
ERROR 30563. Overflow at at adress $23 (variable X)
 
Yay!!!

Etlest thats how a computer wod solve it.

On paper well:
999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999............................
 
Someone Electro said:
Yay!!!

Etlest thats how a computer wod solve it.

On paper well:
999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999............................

no wonder this thread is 3 pages WIDE.

Get a radio, convert the output to light pulses. That will make random flashes, because you never know what frequency is coming out of the radio at any given time. And add some noise to the station, and the whole thing will be 100% random.
 
What I actually did was have 2 555 oscillators, completely independant of each other, thier outputs going to a AND gate. The frequecy of one of these is varied by a triangle wave oscillator, the frequency of the other by a mixture of square and triangle waves. Its not technically random, but it looks like it is, its never the same pattern. I intentianally used cheap timing capacitors :lol:
When I put it up to audio frequency, there was a definate "noise" to it, but also some pretty obvious repeating petterns.
 
Yea but he wants it totaly random.

You cod do realy precice 0,1ms timing for you to hit a button.It will be random for shure
 
Noise idea seems possible. You could use amplitude of the peaks to control brightness. and frequency of each peak over a certain height to control duty cycle of LED. Plus if you were to amplify the noise to feed it into a controller you get additional noise and randomness.

Wait how random is white noise anyway?

This is like trying to read RAM registers at random and trying too truly randomize the process....What ever happened to that idea? Was it random?
 
White noise has an difrent level evry time.Its like you tune an FM radio to someware there isent an station.You dont hear repets and stuff like that.

You can certaly hear it repeting is you keep repeating an white noise recording.
 
Obviously I'm telling you something you already know, but you can never achieve randomness with a finite state machine. However, if you have any familiarity with abstract algebra you can achieve psuedo randomness by counting in a finite field, and if you're field is large enough you could go a life time and never see the period of your sequence finish.

So for all intents and purposes you've got a random number generator.

If you have more questions just reply I'll be happy to indulge you.

P.S. You could actually use this to make your 3d random number generator by simply counting in 3 different fields and using some software to control the light.
 
Another thing... I'm not sure the population at large actually knows what "white noise" is exactly. People seem to throw the term around all too loosely.

White noise has a zero mean Guassian pdf. It's variance is defined as No/2 and it's fourier transform is flat over all frequencies, which in turn requires infinite power.

Since infinite power is not possible, there is no such thing as white noise. Now they've figured out hand waving techniques around this such as "band limited" white noise, but thats sort of nonsense, but for practical purposes it works.
 
Thanks guys for the replies. I'm still scratching my head about the zero mean gaussian...

Dr. EM, sorry to stray away from your topic :wink:
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top