counter.pd
working with pure data after a long long span
can somebody help me in this counter why i can't reset the counter back to zero
thanks

The problem seems to be that you are sending two values to the same inlet at the same time, and the one coming from the [+] object takes precedence (?).

I take it you want the counter to repeat a fixed cycle?

The easiest way to have a count restart from zero while running is to hook up a [%] object after the [+] like this:

Another way around it more true to your original design is to delay the bang that triggers the reset message like this:

The trade off (?) is that it takes two metro ticks (i.e the counter will output 4 before resetting)

@bocanegra [trigger] should always be used instead of [delay] when the order of operation is to be controlled.

@ingox yes indeed. But in this particular case (feeding bangs from the select object back into the counter) how would you solve the problem with a trigger?

The simple solution is to avoid that feedback loop altogether by introducing a % object in side the counter loop like I first suggested.

But if you for some reason want your select object to control the counter reset, you run into some weird behavior that is only ameliorated by a crude delay. I would like to be wrong about this, so by all means give it a go with triggers

@bocanegra Just removing the [delay] should be ok.

@bocanegra This use of trigger ensures that the select output always overwrites the [+ ] output. But if the [+ ] increment causes the counter to never land on 3, then it would never reset. It's hard to believe that this was the original intent, but who knows? @maddynakedymaddy?

@ingox, if you run the patch posted by OP you will see that it doesn't reset unless you bang the [0 ( message box out of sync with the metro. This makes sense as the last message sent to the [float]'s right inlet will be passed to it's outlet on the next bang from [metro].

The issue here is that of feedback in the message domain, and if you want your select module to control when to reset the counter- that is to say if you want to feed back the result of an evaluation of the counters output to the [float] that does the counting - the delay is needed. As far as I can tell at least.

Trying out other solutions produce weird results like select showing the output to be both 0 and > 3 at the same time, while the number box is showing zero, but continues from whatever high value it reached at the same time once you unhook the [0( message - try this at home:

@jameslo Eureka!

But if the [+ } increment caused the counter to never land on 3, then it would never reset.

The select output that triggers the reset is the last one i.e anything but 0 1 2 3, so not an issue

@bocanegra Duh, you're right. I'll never learn: don't post during or after cocktail hour.

@jameslo you still win this round

@bocanegra No way, @ingox got to it first

@jameslo I didn't really get it, not even opened @maddynakedymaddy's patch. The phenomenon is really surprising. But [float] schedules the number for both connection, gets updated to zero, but again based on the previous number.

Well, [trigger] solves it thankfully.

Thankyou
@bocanegra, @ingox and @jameslow super thankyou to all of you for suggesting the solution.
using the trigger after the float object never occured to me though it works fine .
I got a solution with % while probing with the patch later on but the solution with delay and trigger is something that didnt occured to me
So a huge huge thanks and for the super discussion.