Re: Binary sequencer with data structures Re: Input binary values by mouse

Thanks for your insights!

> Am 19.02.2026 um 19:30 schrieb João Pais <[email protected]>:
> 
>> I had a go with data-structures and hacked something up with the great examples in the PD help (particularly 04.append.pd and 09.sequencer.pd from 4. data.structures). Find attached my WIP
>> 
>> I am struggling to get the duration of the sequences right. If somebody with knowledge of data-structures, and the examples could have a look I would be very thankful. Currently I just have two „lanes“ of data, each lane triggering an oscillator on and off, this is just for debugging. 
>> 
>> What works:
>> 
>> The UI for the most part, but I am not sure if my „trick“ with forcing the y position of segments is advisable.
>> The onset of segments is correct and the tempo can be adjusted correctly
> - if you change the y creation parameter in [pd stuff] to match the height of both lanes, you don't need to click there to "move" the scalars.
> 
I am aware of that and would consider this implementation a feature, since a newly created event can immediately be seen, but that is just me :)
> - or use any of the externals to detect mouse clicks on the canvas to add the scalars in the correct x+y positions (iem or else should have something)
> 
Will look into that
>> What does not work:
>> 
>> -the duration (length of drawn segments in my UI) is not exactly linked to the sequenced time. Hence an overlapping segment in the UI can still cut-off before the next segment begins or vice versa. It depends on the tempo that is set. I am not sure how I can link the length of the bars (the z variable in my struct) to the time set by the tempo numbox. The delay simply delays the next pointer lookup, which is the next start of an event. I somehow need to link the length of the polygon drawn at that pointer to the delay. In my first attempt I simply multiplied the length by the delay-time (as in the rest of the sequencer patch) but that does not work correctly.
> I would myself use a different concept from the ground up:
> 
> - y defines the frequency (you can adjust the canvas y value to adjust the overall graphic)
> 
> - instead of having 2 general osc~, make a poly or clone with enough "voices", so that each has a parallel sound generator, which is activated by the list of parameters (specially y). z defines the delay/duration to switch off the generator. together with the previous suggestion, this will open you to many more options.
> 
The sequencer will be used to control 8 relays on a RPI4, the osc~ were just a quick debugging solution to post on the forum...
> - if you're not changing anything during the playback, probably you don't even need the sequencer patch, it's enough to read all scalar data and feed it to a [text sequence] object. Playback goes through the [text].
> 
Thanks for that suggestion, I will investigate this seems indeed easier.
>> -I would like to implement a timeline i.e. a hslider that increments position along the x axis, but again I have no idea how I would link the time set by the numberbox to the position so it actually corresponds to the drawn polygons…
> in my patches (mainly on my click tracker patch) I use data structures to add a timeline (struct with a symbol, and drawsymbol+draypolygon), and also a playback line (vertical line with x, that gets fed from a [line]). You can see all that if you download the click tracker, but you might not be able to decypher the patch.
> 
> I'm not sure what's better, to make the timeline stable and redraw/rescale everything else, or to make the scalars stable and redraw/rescale the timeline. I myself prefer the first (specially when connected to the playback line), but I don't use that on a patch where I change tempo often.
> 
>> It could also well be that this is not possible this way, then a short message would also be great :)
> it's possible :)
> 
> joao
> 
>> Thanks
>> 
>> Simon
>> 
>> 
>> 
>>> Am 18.02.2026 um 23:20 schrieb Simon Iten <[email protected]> <mailto:[email protected]>:
>>> 
>>> Hi list,
>>> 
>>> I am looking for a way to input binary values via mouse input. 
>>> 
>>> Basically I want a sequencer like structure to turn on and off a bunch of relays on a RPI.
>>> 
>>> It would be nice to use something like an array for each output, so the user can draw in the state of the output over time.
>>> 
>>> For that it would be great to only allow inputs that are:
>>> 
>>> 1) in range (so mouse-movement outside the object does not draw outside the visible array, but constrains to the max or min)
>>> 2) round the values to only integer input, in my case 0 and 1.
>>> 
>>> An array would be nice since it would allow very fast (no clicking needed) entry of many points and the state of each output would immediately be visible…
>>> 
>>> Maybe there is some external that allows something like that?
>>> 
>>> Thanks for any hints
>>> 
>>> 
>> 
>> 
>> ---
>> [email protected] <mailto:[email protected]> - the Pure Data mailinglist
>> https://lists.iem.at/hyperkitty/list/[email protected]/message/3GSEBM3ODAGK3ZMDNDJ75NPBPZ7X4SBK/
>> 
>> To unsubscribe send an email to [email protected] <mailto:[email protected]> mailing list
>> UNSUBSCRIBE and account-management -> https://lists.iem.at/
> -- 
> Audre-Lorde-Str. 26
> 10997 Berlin (Deutschland)
> Tel +49 30 74921288 | Mob +49 176 98476769
> [email protected] <mailto:[email protected]> | https://www.facebook.com/jmmmpais
> ---
> [email protected] <mailto:[email protected]> - the Pure Data mailinglist
> https://lists.iem.at/hyperkitty/list/[email protected]/message/DF4UY5VTXME3CLUNMMU7BF47SQ64ZBZF/
> 
> To unsubscribe send an email to [email protected] <mailto:[email protected]> mailing list
> UNSUBSCRIBE and account-management -> https://lists.iem.at/

---
[email protected] - the Pure Data mailinglist
https://lists.iem.at/hyperkitty/list/[email protected]/message/ALAJGS4P3ILCQFYPDD2WHVNDZU5KYD4Q/

To unsubscribe send an email to [email protected] mailing list
UNSUBSCRIBE and account-management -> https://lists.iem.at/