Hi,

when I want to open a new sequenced composition to mix it with a composition that is already playing there is a short gap of silence as the composition loads. This is the only problem I have ever had with pure data. Is there any way that this problem could be eradicated in the next version? Or is there a better way of doing things that I am missing.

One way I have though about would be to have all of the files for the entire set open from the start, but some of the sends and receives are the same and therefore I would not be able to play the track properly as it would control other parts that I did not want to add in until later. I am willing to change all of the sends and receives to be unique if this is the only way, but it would be very useful if someone could try and get around this problem by changing the source code.

Any help would be much appreciated.

Yeah, unfortunately the only way to avoid it is to just have everything open before hand. You can easily avoid the [send]/[receive] problem by using $0 in all of the names, and you should probably just get in the habit of always doing it. $0 gets assigned a unique variable when a patch or abstraction loads, so if you do something like , only a [r $0-volume] in the same patch will receive it, so there will be no crosstalk.

Also, if you're going to be loading a bunch of patches that aren't going to be used for a while, be sure to [switch~] them off until you need them to avoid cpu overload.

thanks!

Quick question about $0: exactly how 'local' is it? If I use it in a subpatch, is the random number localised to that subpatch only (and therefore inaccessibe from the patch that contains it)?

subpatches are just like separate rooms in the 'house' that is your main patch. they share all the attributes of the parent, including $0 and any other creation arguments.

abstractions however, are more like your TV, fridge and oven. When you turn on the TV, the oven doesn't automatically fire up as well. They have their own $0 value.

i often wish there was a way to 'name' subpatches with a $0-like method, it would be very useful. I know that [namecanvas] can send messages to individual subpatch canvases, but some general way of localizing subpatches from their parent would often be handy.

no-one has a nice workaround for this? (preferably vanilla)

Nice analogy mod

actually, it's pretty dodgy. your rooms don't really share the all the attributes of the house. but let's just ignore logic here for a bit

Yeah it was actually a little tenuous after re-reading, but ah well.

You could say the electricity circuit is all under $0 because if the fuse goes then it's all screwed. But then upstairs might have to be another abstraction ...

@mod said:

i often wish there was a way to 'name' subpatches with a $0-like method, it would be very useful. I know that [namecanvas] can send messages to individual subpatch canvases, but some general way of localizing subpatches from their parent would often be handy.

no-one has a nice workaround for this? (preferably vanilla)

What do you mean by this? I mean, what kind of behavior are you looking for?

i'd like to take some abstractions i have, and just use them as subpatches.

however, if i use 2 or more of the same abstraction as subpatches, then the $0 values will be the same in each case.

i have made a little hack that involves using just one very simple abstraction like this:

[inlet]
|
[t b]
|
[f $0]
|
[outlet]

that way, if i embed that abstraction inside a subpatch, it gives a unique $0 variable that i can use for various local routing options inside the subpatch.

but, it would be nice to have a way to do it without ANY abstractions.

Hmm...I can see how you would make [send]s with that using message boxes, but how are you using that $0 for a [receive]? Could you post an example?

here, it works brilliantly.

as long as you connect the send and receives with a patch cord, that is.

umm...there's some sort of drawing board i need to be getting back to.

http://www.pdpatchrepo.info/hurleur/diy-dolzero.pd

here, it works brilliantly.

as long as you connect the send and receives with a patch cord, that is.

umm...there's some sort of drawing board i need to be getting back to.

http://www.pdpatchrepo.info/hurleur/diy-dolzero.pd