MIDI LFOs and Env

cebec

This should be an easy one, and I've tried numerous things, but I just can't seem to get it. I'd like to have various waveforms at my disposal to send via ctlout (to an MIDI/CV converter): sine, tri, rectangle, saw, etc., and a bonus would be to be able to draw my own waveforms.
In addition, I've been trying to figure out how to use line with the function editor and send arbitrarily complex envelopes out of ctlout, as well.
My problem is: I can't figure out how to get basic waveshapes in Max without resorting to MSP, and when I do, there's the problem of 'translating' them to MIDI. I gather I need to scale any signal I intend to send out of ctlout to 0-127.
Any help or tips would be much appreciated,
Thanks!

Exit Only

You can use line in conjunction with a table to draw your own waveforms. Here is an example:

Max Patch

Copy patch and select New From Clipboard in Max.

Exit Only

Also, here is a patch someone posted on the forums a long time ago. I found this when trying to approach a similar idea. I don't know who made it, but its invaluable! It generates a number of different curves in tables, which you can cycle up and down with line to make waveforms.

Max Patch

Copy patch and select New From Clipboard in Max.

Also note that this number box in the upper right will adjust the curves. You can then just save the data into a text file for quick loading.

cebec

These are fantastic aids. Many thanks!

If anyone has any further suggestions re: the Function editor with line and creating basic waveforms with pre-existing objects, please chime in...

cebec

Is there a way to change the line's ramp time dynamically while it's in such a loop as that in the first patch?

Peter Castine

On 18-Jul-2006, at 22:51, matt wrote:
> If anyone has any further suggestions re: the Function editor with
> line and creating basic waveforms with pre-existing objects, please
> chime in...

I do this sort of stuff with lp.scampi.

Check out lp.scampi.help and the patcher newStuff.

URL below.

-------------- http://www.bek.no/~pcastine/Litter/ -------------
Peter Castine +--> Litter Power & Litter Bundle for Jitter
Universal Binaries on the way
iCE: Sequencing, Recording &
Interface Building for |home | chez nous|
Max/MSP Extremely cool |bei uns | i nostri|
http://www.dspaudio.com/ http://www.castine.de

cebec

Yes! And I own a license for Litter Pro, too... Sometimes the best solutions are right under our noses.

cebec

lp.scampi does not handle negative floats/ints, it seems, right? i still cannot, for e.g., use a low frequency cycle~ to control a VCO via MIDI/CV...

Peter Castine

On 19-Jul-2006, at 1:56, matt wrote:
> lp.scampi does not handle negative floats/ints, it seems, right? i
> still cannot, for e.g., use a low frequency cycle~ to control a VCO
> via MIDI/CV...

Scampi will handle any numbers you ask it to.

The default input range is 0..1, but you can write explicit arguments
in the object box for a different range, or send messages to the 2nd
& 3rd inlets.

The default output range is 0..127, but you can change that, too.

You can change everything else in scampi's behavior (symmetry, curve/
kink, handling of out-of-bounds input).

Hope this helps,
Peter

roger.carruthers

I've always been a big fan of Panaiotis' autocount
object for creating non-audio rate LFO's, because it's
an easy one-stop object for creating all the usual
LFO wave shapes but very flexible too.
And another worthy curve builder is Tristan Jehan's
mapper object.
I'm not sure if either ever got ported to Windows, but
you should be able find them via maxobjects.com,
cheers
Roger

--- matt wrote:

>
> These are fantastic aids. Many thanks!
>
> If anyone has any further suggestions re: the
> Function editor with line and creating basic
> waveforms with pre-existing objects, please chime
> in...
>
>

cebec

Thanks, everyone... No, those two objects haven't been ported... yet. That's ok, I think I can still make-do with what I have right now. I'm just surprised there isn't some REALLY easy way to do this. I thought a cycle~ to snapshot~ or number~ to lp.scampi or scale would've done the trick but it doesn't...

langsound

this may have been suggested already... but how about a for-loop, and a
formula. or maybe a for loop and a lookup table or lots of lookup
tables... i mean you could just build stuff that works the same way as
your favorite audio tools, but at lower and slower resolution...

if you know math and/or digital signal theory it should take you around
38 minutes to do what you want using javascript or java or rtcmix~ or
47 minutes using traditional max objects...

on/off or bang

for-loop or [counter]
formula or table
additional formula or table
...
output of LFO curves of your design at event rate...

> Thanks, everyone... No, those two objects haven't been ported...
> yet. That's ok, I think I can still make-do with what I have right
> now. I'm just surprised there isn't some REALLY easy way to do
> this. I thought a cycle~ to snapshot~ or number~ to lp.scampi or
> scale would've done the trick but it doesn't...

cebec

this is about as far as I've gotten. try a tri~, too... i haven't gotten it quite right...

you need lp.scampi

Max Patch

Copy patch and select New From Clipboard in Max.

Peter Castine

On 19-Jul-2006, at 22:23, matt wrote:
> this is about as far as I've gotten. try a tri~, too... i haven't
> gotten it quite right...

Make that an 'lp.scampi map -1 1'.

The map symbol puts scampi in map mode. In map mode the first two
numeric arguments set the input range. The default output range,
0-127, is fine so no need for any more arguments. As long as the
input values are in range, there's no need for the reflect/wrap/etc.
popup menu.

Somewhere along the line some of your patcher was HTML-ized, causing
some problems with "New from Clipboard." This should iron out those
issues:

Max Patch

Copy patch and select New From Clipboard in Max.

cebec

fantastic. that solidifies it for me, thanks.

Stefan Tiedje

matt wrote:
> Any help or tips would be much appreciated, Thanks!

The right inlets are for frequency in Herz, if you want to have the
timing more exact, you need to syncronize them instead of self triggering.
This should give you enough ideas to go on...

Max Patch

Copy patch and select New From Clipboard in Max.

--
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com

cebec

Thank you VERY much, Stefan. I found lines, too.

Stefan Tiedje

Peter Castine wrote:
> The default output range is 0..127, but you can change that, too.

unfortunately its still 0...128 (just checked again (with the hope that
it had changed in the mean time)), though I believe it would not break a
single patch if it would be changed to 0...127, as its pretty unlikely
that the default is ever used in a real patch...

Stefan

--
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com

Peter Castine

On 21-Jul-2006, at 11:40, Stefan Tiedje wrote:
> unfortunately its still 0...128 (just checked again (with the hope
> that it had changed in the mean time)), though I believe it would
> not break a single patch if it would be changed to 0...127, as its
> pretty unlikely that the default is ever used in a real patch...

You're quite right that lp.scampi's default output range is 0 to 128.
And, as Klaus Wowereit says, it's a good thing too. Let me explain.

The scampi/scampf pair were designed for mapping output from Litter
unit-range RNGs to MIDI and back. A careful look at the unit-range
random number generators reveals that they generate values in the
half-open range 0
multiply the input by 128 and truncate, giving integers in the range
0

Note also that if the input values have a flat distribution, so do
the output values. This would not be the case if scampi used the more
naive approach of multiplying by 127. You can test this yourself by
editing the lp.scampi object in lp.shhh.help to 'lp.scampi map 0 1 0
127'. Then bang Uzi a few times. How many 127s show up in the histogram?

As for breaking "real-world" patches: scampi was, as I said,
originally designed for scaling output of RNGs to the standard table
range, as exemplified in a dozen lp.*.help patches. Furthermore I've
used scampi with its default settings in many real-life projects.

Like right-to-left order, this is an example of where the most
obvious approach is not the best.

-- Peter

Max Patch

Copy patch and select New From Clipboard in Max.

Stefan Tiedje

Peter Castine wrote:
> The scampi/scampf pair were designed for mapping output from Litter
> unit-range RNGs to MIDI and back. A careful look at the unit-range
> random number generators reveals that they generate values in the
> half-open range 0
> multiply the input by 128 and truncate, giving integers in the range 0
>

Thanks for that clear and understandable explanation. I promise I'll
never rumble about that ever again. Just to avoid wrong thinking for
other users, just put that explanation patch into your help file...

In general a bit more basic explanation about all your wonderful
distributions would be helpful aswell. I know its possible to search for
such in the web, but if you are patching, a little explanation within
the helpfile is just more convenient than searching the web for "what
the hell does mu or sigma mean..."

I am sure you could write a book about all these distributions and their
occurances in the real world...

And many thanks for your great work...

Stefan

--
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com

Peter Castine

On 27-Jul-2006, at 7:50, Stefan Tiedje wrote:
> In general a bit more basic explanation about all your wonderful
> distributions would be helpful aswell. I know its possible to
> search for such in the web, but if you are patching, a little
> explanation within the helpfile is just more convenient than
> searching the web for "what the hell does mu or sigma mean..."

Up to the current release of Litter, I followed the documentation
paradigm given by Opcode and Cycling: quick overview in the .help
file, details in the Reference Manual.

My experience in coaching and user support is that real people very
rarely crack open the book, even when the book is a .PDF file. Also,
some of the reasons for keeping .help files small that were valid in
the late 80s no longer hold. So I am in the process of moving all
documentation into the .help files.

The process won't be completed in the next release, it's really a
massive task. But you should find a lot more information immediately
available. I'll try to include a note about mu and sigma.

> I am sure you could write a book about all these distributions and
> their occurances in the real world...

Yeah, that's the problem. Who wants to read a book?

(Hint: read my Skype Mood Message)

Stefan Tiedje

Peter Castine wrote:
> The process won't be completed in the next release, it's really a
> massive task. But you should find a lot more information immediately
> available. I'll try to include a note about mu and sigma.

Actually if you know of a possibility to use launchbrowser not only for
calling the manual, but also to jump to a specific chapter within the
PDF, let me know. (Preferably for Preview, as Acrobat Reader is way too
heavy for my taste.)
That might even do it as link within the help file...

Stefan

--
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com

Pamela Z

Oh yeah, of course! Sorry! Here's a quote from one of her emails:

At 4:39 PM -0700 10/5/06, Lauren Pratt wrote:
>My CalArts email address has changed: it is now lpratt@calarts.edu. Please
>note this in your address book.
>
>Lauren Pratt, Associate Producer for Music
>REDCAT (Roy and Edna Disney/CalArts Theater)
>California Institute of the Arts, School of Music
>24700 McBean Parkway
>Valencia, CA 91355-2397
>661-799-0226, 1-888-277-1190 toll-free
>Fax: 661-670-0151
>lpratt@calarts.edu
>
>http://www.redcat.org

I'll work on "stoned and in a dormroom". Maybe that's the answer to
all my problems of late...

Pamela Z
Composer/Performer

Contact info:

Telephone: 415.861.EARS (415.861.3277)
Mobile: 415.5PAMELA (415.572.6352)
FAX: 415.861.FAKS (415.861.3257)
(I forward my land line to my mobile phone when I'm travelling)

pamelaz@pamelaz.com
http://www.pamelaz.com

Skype: pamelazed AIM: pamelazdotcom

Snail Mail:
Pamela Z Productions
540 Alabama Street Studio 213
San Francisco, CA 94110, USA

shipping address (for packages larger than a 9" x 12" envelope):
Pamela Z 2440 Sixteenth Street PMB #171, San Francisco, CA 94103, USA

Pamela Z's CD "A Delay is Better" on the Starkland label is now
available at http://www.amazon.com, http://www.starkland.com, and in
stores near you.
......................................................................................................