Hey guys, fairly new to pd but trying to get stuck in as best I can.

I'm following this tutorial:

on setting up continuous musical scales using a [coll] object but I'm having difficulties translating the patch into pd (the tutorial is Max 6 based).

I'm using pd-extended and using certain objects from the cyclone library to imitate Max objects (namely [coll] and [gate]).

Perhaps there are some minor differences between the way these objects behave in pd compared to Max?

I've downloaded the demo of Max 6 and I've been able to recreate the the tutorial as far as 8mins:54secs. (The interesting bit begins at around 6:30).

But the same patch in pd won't work.

Any suggestions?

http://www.pdpatchrepo.info/hurleur/coll_of_duty_tutorial_pd_and_max_patches.zip

A few things:

• [length( should be a message box, not an object.

• [coll] does not appear to output the bang from the third outlet when it loads. It only does when you send it the [read( message, so you'll have to do that to get the length this way.

• [t ] works a bit differently in Pd than Max. In Pd, the arguments ONLY refer to types. They do not send specific messages. In Max, if you have [t 1] and send it a bang, it will output 1. In Pd, however, it will output 0. The reason is because Pd just considers a float argument to just mean "float" (i.e., it's the same thing as [t f]). So in Pd, you have to do this to get the same functionality:

[t b]
|
[1(

http://www.pdpatchrepo.info/hurleur/coll-fixed.pd

Thanks very much for your response.

I've made what I think are the correct changes but I seem to be missing something still.

The ultimate goal with this is that I'll be able to select a type of scale from a umenu or something similar (popup I think? but I haven't looked into it yet)
and then have that scale send out the corresponding midinotes. It's meant to be used with a kinect so the scale will be played linearly from left to right. I'm using a vslider and a rvslider in another example for testing purposes.

Am I going about this scale business the right way or are there any alternatives I should be aware of?

Thanks again for the help.

http://www.pdpatchrepo.info/hurleur/coll_pd_patch_2.pd

What's missing here is the bang coming from [coll]'s third inlet that only occurs in Pd when the [read <filename>( message is sent. The patch I attached above says 0 downloads, so I'm guessing you missed it, but it has the fixes to make this work.

If you are only sticking with 7-note scales (i.e., modes of the major scale, not symmetrical ones like whole tone), then it would be significantly easier to do this with a

Here's the example. You actually don't need [coll] at all--no matter what length the scale is--if you store your scale mappings as messages in the patch, as shown here.

I know this actually looks like more patching, but the reason why I think it's simpler is because: a) the loading of the scales and the mapping-lookup are separated, you don't have to have a bunch of separate files for each scale (they are stored in the patch itself), c) typing in scales in [coll]'s text editor doesn't necessarily save time in patching, and d) I think it just might be more efficient to look up an array here.

[coll] is great for storing lists or symbols associated with an index, but for simple one-to-one mappings where the index is an integer and the element is a float, it's overkill. A http://www.pdpatchrepo.info/hurleur/scale_mapper.pd

I only got a chance to look at this now, thanks very much, I think it's exactly what I need! It's definitely a better way to go about it.