Hi all,

I use [tabwrite array1] to plot some values ​​(points) with index from 1 to n, and this works fine.
If I then plot again other values, always with index from 1 to n, the previous ones are progressively deleted. Is it possible that instead all the points remain, in order to overlap more graphs on the same canvas?

Thanks,
a.

@atux You can plot more arrays (separate arrays) on the same graph by selecting "last graph" in the properties window when you create the array........

Or you can do some calculation with [tabread] and [tabwrite] if you actually want something else (see below).
But an array can only have one value per index.
More on array manipulation........ all_about_arrays.pd
David.

@whale-av said:

@atux You can plot more arrays (separate arrays) on the same graph by selecting "last graph" in the properties window when you create the array........

Thanks, all clear, it works fine.

Instead I ran into a problem. I state that the following happens in Purr-Data on Ubuntu.

In a patch I generate a list of numbers, for example the first one you see below (I omit the patch that generates it, it is irrelevant). Now I would like to plot this list: it would be enough to modify it by hand in this way:

[; array1 1 48 2 69 43 49 17 83 21 34 20(

then, clicking on it, the graph appears. This works correctly.

Here is the problem:
in the patch you see below I would like to do this step automatically. Why does \ at the beginning appear in the last message?
If I delete it by hand, then it works.
Yet in print it turns out without \ , as it should be ...

Thanks,
a.
array1.pd

@atux A semicolon is a reserved character, so cannot be used as part of a list.

Putting a semicolon into a message can be done with a message [addsemi( but then adding your list would have to be done by dripping each float using [list-drip] and feeding the dripped atoms through an [add2( message.

It's easier to do this.....

David.

Thank you.

Going back to the question of overlapping graphs, in this sample patch I would like to overlay 10 graphs of random numbers:

array2.pd

What I was looking for, was to use a single array message like:

[s array $1 (

passing the values ​​to $1 from a [counter] ...but I haven't been able, it doesn't work.
So in that patch I created an object for each array in this way:

Same thing for messages to clear graphs, I was looking for something like:

[; array$1 const 0 (

but it only worked by creating one for each array.
If the graphs were 100, it would be difficult to do all this.

Regarding the quality of the graph: I have seen that no points are created but segments (evident in the image of your last post).
So, to draw a point cloud, remaining in pure data, soon I'll try with "data structures" , using for each point a circle of minimum possible radius. I just started reading some tutorials.
Otherwise of course I export the data (x, y, z) and use an external software. But I'm curious to see what I can do by always remaining in Purr-Data.

P.S.:
Purr-Data, when I use [prepend something] it always gives me a warning that says:

[prepend]: this object is a legacy replacement for prepend in cyclone library that is considered deprecated in pd-l2ork. Please use [list prepend] and [list append] instead.

So I should replace a single object with two objects:

[list prepend something]
|
[list trim]

In the patch above I tried to do this replacement, and it no longer gives me the warning.
It seems to complicate things.

Bye,

a.

@atux Yes..... [prepend] has been added to vanilla as [list prepend]..... a while ago now.
Yes they are bars..... not points..... but if you size the graph correctly for your data then they will effectively be represented as points on the display.

I often solve these problems through messages. It's a bit "old school" as so many new objects have been added that can do the same. But I am not sure that they are always more efficient.

I mentioned above using messages to resolve the semicolon problem, and I have included [list-drip] in the zip below although you probably already have it....

Sticking with your original method...... here is the solution........ this.zip
David.

@whale-av said:

Sticking with your original method...... here is the solution........ this.zip

I added a [counter] to send the numbers to [r counter].
Everything works fine.

Thank you very much.

Bye,
a.