Bernoulli Gate

[Schville]

Anybody else experiences that the two LEDs below the outputs are not lighting up?

And I wonder what the change-switch in the upper left does. Benard says when set to CHNG the module determines if the trigger is sent to the same output again or to the other one. But what does the module do if the switch is set to OUT? Have used a scope and a trigger counter to try if I see any differences, but so far I cannot see any. Could anyone explain please?

[huggermugger]

I don't have that module so I can't test it, but I'm wondering if the CHNG setting is about the likelihood of switching to the opposite output on each trigger. If so, then with CHNG turned on and the PROB value set to maximum, the outputs should alternate, like a ping pong. As you reduce PROB, the alternation becomes less predictable.

I have Hetrick's Probability module, and it has an Alternate mode that does just that.

[Schville]

That is absolutely the solution! Tested that approach with a trigger counter, the input trigger is now sent evenly to output A and output B. Thank you for the input!

[huggermugger]

Glad to help.

If you're interested in Bernoulli-type randomness, here are a few others that I use. Tossed Coins (Black Cloud) and the Probability Routers (Insomniac) are favourites.

[Schville]

Looks nice :D

Still figuring out how to implement Bernoulli Gates and don't know your used modules but I'm a big fan of the Possibility Routers and use'em a lot!

[ColinP]

N-Step Main includes a couple of simple Bernoulli gates and the documentation on them might be of interest. You should be able to apply most of the ideas discussed to Bernoulli gates in general.

https://adroitsynthesis.com/n-step/#Usi ... ulli_gates

[Schville]

Hi Colin, nice to read from you here :)

Yes, I already worked with N-Step but didn't try the Bernoulli Gates yet. Will do so!

[Schville]

Anybody else experiences that the two LEDs below the outputs are not lighting up?

Just in case anyone is reading who has the same question: If the Bernoulli Gates are used for triggers or very short gates only (gate length around minimum) the LEDs don't have enough time to light up. Increasing the gate length will make the LEDs light up as expected, but only if at least one output (Out A or Out B) is connected.

And just to explain the function of the CHNG/OUT switch: When CHANGE is selected and the probility is set to 100% the module switches from left to right evenly in a "ping pong" behaviour, as @huggermugger said. Decreasing the probability slightly will force an output to repeat itself once, decreasing it further will make the repetition occur more often while a setting of 0% probabilty will make the current output repeat until eternity.
Is the switch selecting OUT a probability of 0% will always choose the left output, while 50% will try to route it evenly between the left and the right output and 100% will always choose the right output.

Made a little patch to "prove" the equality of probability 50% in OUT-mode:

Screenshot 2025-05-29 124656.png (170.52 KiB) Viewed 2732 times

The upper row of the trigger counter is purely the A output of the Bernoulli Gate, the lower row is purely the B output. Since the distribution is 57% to 42% by a total amount of 3.978 triggers we see it's nearly a 50:50 chance

Repeated this experiment with the doubled amount of Bernoulli Gates and Trigger Counters and the result is almost a perfect 50:50

Screenshot 2025-05-29 125533.png (559.77 KiB) Viewed 2732 times

The Trigger Counter counted 3.293 triggers by the A Output (49,53%) against 3.355 triggers by the B Output (50,45%), which makes a total of 6.648 triggers.