Sequencing patches ?
I know asking for patches is not best practice around here, and i'm sorry about that.
But are there any, rather advanced, sequencing patches that you'd recommend ?
Sequencing is basically the reason i'm getting into MAX, but i'd like to get ideas from other peoples' concepts and designs.
cheers
Quote: jeanfrancois.charles wrote on Fri, 08 June 2007 08:04
----------------------------------------------------
> > Sequencing is basically the reason i'm getting into MAX.
>
> Well, that's maybe not the best reason. Actually, sequencing is the worst
> thing in Max for the moment.
thats a radical statement.
i like radical statements.
but.
max at least lets you program your sequencer like you want, making your own gives you freedom.
try probability in cubase or changing signatures in protools.
so, mister "disturb", start with [metro] connected to a [counter]. then try out what results you can get.
dont plan too much. start making it.
Roman Thilenius skrev:
> Quote: jeanfrancois.charles wrote on Fri, 08 June 2007 08:04
> ----------------------------------------------------
>
>>> Sequencing is basically the reason i'm getting into MAX.
>>>
>> Well, that's maybe not the best reason. Actually, sequencing is the worst
>> thing in Max for the moment.
>>
>
>
> thats a radical statement.
>
> i like radical statements.
>
So do I.
That's why I say: SO TOTALLY WRONG! (shouting is a bit of net-Tourettes,
sorry).
Andreas (almost solely doing sequencing in Max.)
what are your grudges with sequencing actually ?
is it timeline-related ?
On Jun 8, 2007, at 6:12 AM, disturb wrote:
> But are there any, rather advanced, sequencing patches that you'd
> recommend ?
>
> Sequencing is basically the reason i'm getting into MAX, but i'd
> like to get ideas from other peoples' concepts and designs.
You can check out the Monolake sequencer on Cycling's share pages:
It's eccentric, but you can learn lots by taking it apart.
Vlad
Vlad Spears
Urbi et orbi
Quote: disturb wrote on Fri, 08 June 2007 11:28
----------------------------------------------------
> what are your grudges with sequencing actually ?
> is it timeline-related ?
yes what jeanfrancois had in mind was the "timing issues" stuff which is m uch discussed around here.
it is well possible in max to program a tight sequener app, just not every day, not for everyone, on every computer.
i believe it also depends on the weather.
at least for beginners it is difficult, and it is also difficult to build a tight sequencer for LOTSA EVENTS on a slower computer.
if you want to play a concerto grande partiture with max, you would have to add a short "timing errors compensation delay" at the end of your data path.
which is what seqencer programs like cubase do - with the difference that they are also able to calculate and play the data preemtive. (=earlier than your eyes see reaching that timepoint)
-110
At 7:12 AM -0600 6/8/07, disturb wrote:
>I know asking for patches is not best practice around here, and i'm sorry about that.
>
>But are there any, rather advanced, sequencing patches that you'd recommend ?
>
>Sequencing is basically the reason i'm getting into MAX, but i'd like to get ideas from other peoples' concepts and designs.
I'm using a patch that gives me step-sequencer-like functions as well as a bunch of other note generation methods, that I call Gyre. You may be able to grab some ideas from Gyre, although I've not released the patch itself. Info on this patch can be found here: http://www.xfade.com/Gyre
I put up a couple very simple examples that illustrate some very basic sequencing techniques here though:
http://www.xfade.com/max/examples/
-C
--
Chris Muir | "There are many futures and only one status quo.
cbm@well.com | This is why conservatives mostly agree,
http://www.xfade.com | and radicals always argue." - Brian Eno
My experience also is that it is not easy to achieve an always accurate sequencing machine from MAX. But atomic clock precision has nothing to do with music in my opinion. A lot of people use MAX because it is so convenient to interface it with various human situations (gesture, ...). That way you can do human music, that ultraprecised sequencers cannot.
Domi
So the problem seems to be some jitter in the timing ?
Sounds kinda bad...
I couldn't checkout that monolake patch thopugh, i'm on PC.
Quote: disturb wrote on Fri, 08 June 2007 11:30
----------------------------------------------------
> So the problem seems to be some jitter in the timing ?
> Sounds kinda bad...
Well, I've never found it to be as bad as some people. For example this is four voices of Buchla 200e sequenced in time:
http://www.xfade.com/Buchla/#TableThing
There is a _little_ timing variation, but it's also not too bad, IMO, as long as Overdrive is enabled. I find Max pretty usable for sequencing stuff, at least in the way I've been using it.
-C
hey chris, funny i just stumbled over your webpage through matrixsynth this morning.
I must say i checked every square inch of the page looking for a download link ;)
Your patch looks very interesting, hope you release it someday...
forgot to ask if the ppl having a hard time with timimng inconsistancy, experience the same problems when the patch is pluggo-ed and loaded in a vst host ?
Yeah, no Gyre download link, sorry. As far as releasing it, I'm still thinking about how, and if, I should make it available.
Pros:
- Other people will have good ideas for it, and implement them
Cons:
- It's tied to my specific setup. Work to generalize it.
- Have to create at least _some_ sort of documentation.
- I'm still whacking at it fairly regularly, and trying to figure out how to integrate my changes with a release process is a PITA.
- No really good way for multiple contributers, that I know of, in Max development. One of the drawbacks of Max being a graphic environment.
So far, it's seeming like releasing it will slow me down more than it will help me, sorry to say.
I've tried to release parts of it, like the gesture system, at any rate.
On Jun 8, 2007, at 11:30 AM, disturb wrote:
> I couldn't checkout that monolake patch thopugh, i'm on PC.
You should still be able to open it. I just checked and the only
externals are:
label, which is a Lobject and available for Windows
iGate, which was an optimized gate for ints in OS9. You can
substitute a regular gate object for this, if I remember correctly.
If nothing else, you can still check out the components of the patch
to get ideas for ways to accomplish various tasks. Prior to Live, I
remember hearing this patch was used for sequencing early Monolake
tracks, so there's one example of Max/MSP being used to sequence with
acceptable timing.
Vlad
Vlad Spears
Urbi et orbi
Quote: spears@2secondfuse.com wrote on Fri, 08 June 2007 13:47
----------------------------------------------------
> On Jun 8, 2007, at 11:30 AM, disturb wrote:
>
> > I couldn't checkout that monolake patch thopugh, i'm on PC.
>
> You should still be able to open it. I just checked and the only
> externals are:
>
> label, which is a Lobject and available for Windows
>
> iGate, which was an optimized gate for ints in OS9. You can
> substitute a regular gate object for this, if I remember correctly.
>
> If nothing else, you can still check out the components of the patch
> to get ideas for ways to accomplish various tasks. Prior to Live, I
> remember hearing this patch was used for sequencing early Monolake
> tracks, so there's one example of Max/MSP being used to sequence with
> acceptable timing.
>
> Vlad
>
>
> Vlad Spears
> Urbi et orbi
>
> http://www.daevlmakr.com
> http://www.2secondfuse.com
>
>
>
>
>
----------------------------------------------------
thanks it seems to have loaded fine (didn't signal any missing external and whatnot).
I'm having a hard time figuring it out though, and can't read the manual with a regular text editor, there is no file extension either ; could anyone tell me what's the file format ?
cheers everyone
Quote: disturb wrote on Fri, 08 June 2007 12:55
----------------------------------------------------
> forgot to ask if the ppl having a hard time with timimng inconsistancy, experience the same problems when the patch is pluggo-ed and loaded in a vst host ?
>
----------------------------------------------------
thats a good question dude. in fact some people are using their sequencers as pluggos because of a "better timing" (without going into technical details now).
i can only recommend it, especially to steinberg- and DP-users which would like to sequence 100 controllers at once, which is almost impossible in a maxmsp-patch (though inside max it would still be more accurate than when sent to coremidi, lol, sorry for that.)
but seriously, it is definetly possible to build an 8-row sequencer playing 1/64 notes with maxmsp. the inaccuracy of data inside max is still much less than data send from a mac to a hardware midi interface - you would be surprised how "inaccurate" a cubase->midiinterface->norlead system in fact is.
with 32 rows it might be a different situation ...
On Jun 8, 2007, at 2:01 PM, disturb wrote:
> thanks it seems to have loaded fine (didn't signal any missing
> external and whatnot).
Definitely check the Max window. There are some abstractions in a
separate directory which it should be complaining about, and the two
missing externals which are included in the archive, but are only for
OS9. There is a third external in there, as well, [modifers], but
it's part of the regular Max/MSP distribution now.
>
> I'm having a hard time figuring it out though, and can't read the
> manual with a regular text editor, there is no file extension
> either ; could anyone tell me what's the file format ?
It's just a one page Max patch. ;)
Vlad
Vlad Spears
Urbi et orbi
Vlad Spears schrieb:
> iGate, which was an optimized gate for ints in OS9. You can substitute
> a regular gate object for this, if I remember correctly.
This is correct, either open the patch as text and replace every igate
with gate, or save the patch below as igate into your search path, then
it will substitute automatically...
(if you need more than 32 outlets, open it with shift-cmd to prevent the
deletion of outlets and edit to your liking...)
You can also save it as bgate, fgate, sgate, lgate... to substitute the
other ancient optimized gates with unoptimized ones...
The speed of computers has been raised by a factor of 100 or something
like that in the mean time (means unoptimized is fine...)
--
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com
On 11 juin 07, at 11:23, disturb wrote:
> • error: table: doesn't understand "signal"
I didn't follow the thread but this one means that there's a signal
object connected to a table.
ej
Well, if i were looking for traditionnal sequencing, i wouldn't be wasting my time on MAX...
Trying PX18 i experienced the timing problems ppl mentionned.
MidiYoking it back into LIVE, MAX, would sometimes freeze for a little bit when switching program windows, which would fuck up my manual sync.
I guess this has to do with the gui, i should look into how to properly sync those two, i'm not sure what are the options available though, as far as syncing goes (rewire, midi clock ?), can this be improved ?
cheers
Now seems like a good time to share the sequencer I've been working on with jsui. The forums told me to be skeptical of getting accurate timing, but it seems to work okay.
I was more interested in writing it to come up with a convenient way to carry around all the numbers I wanted to send to my synth than because I'm planning on writing lots of loopy beats. What's the more conventional solution to this problem?
The code and an example patch (and a save file for the sequencer and the patch for the synth it's driving) are at www.dandroid.org/max/ddseq/ .
There are probably still some bugs to be worked out, but it's pretty stable. Let me know if you think you might find it useful.
cheers,
Dan
I agree. If your going to do normal linear sequencing,
use MIDIYoke/Rewire into Ableton live. Assuming you have it.
No sense in reinventing the wheel.
Anthony
----- Original Message -----
From: disturb
Date: Monday, June 11, 2007 3:49 pm
Subject: [maxmsp] Re: Sequencing patches ?
>
> Well, if i were looking for traditionnal sequencing, i wouldn't be
> wasting my time on MAX...
>
> Trying PX18 i experienced the timing problems ppl mentionned.
> MidiYoking it back into LIVE, MAX, would sometimes freeze for a
> little bit when switching program windows, which would fuck up my
> manual sync.
> I guess this has to do with the gui, i should look into how to
> properly sync those two, i'm not sure what are the options
> available though, as far as syncing goes (rewire, midi clock ?),
> can this be improved ?
>
> cheers
>
Quote: dgood@alum.mit.edu wrote on Mon, 11 June 2007 15:26
----------------------------------------------------
> Now seems like a good time to share the sequencer I've been working on with jsui. The forums told me to be skeptical of getting accurate timing, but it seems to work okay.
>
> I was more interested in writing it to come up with a convenient way to carry around all the numbers I wanted to send to my synth than because I'm planning on writing lots of loopy beats. What's the more conventional solution to this problem?
>
> The code and an example patch (and a save file for the sequencer and the patch for the synth it's driving) are at www.dandroid.org/max/ddseq/ .
>
> There are probably still some bugs to be worked out, but it's pretty stable. Let me know if you think you might find it useful.
>
> cheers,
> Dan
----------------------------------------------------
Dan, this is VERY interesting stuff !
I was just wondering how i could get a gui module to provide that type of drawing feature...
I only tried the demo patch though, but i can report that i often got small chunks of notes left after deleting with right click.
I'm not sure thefunctions on the right are acting properly too, maybe that's just me.
And last but not least, this was very cpu taxing, i don't know how much of it should be accounted for the synth or the seq, but that was pretty worrying (reached around 75 % on my pentium4 3.5ghz).
Please keep us informed of your progress !
Quote: Roald Baudoux wrote on Mon, 11 June 2007 06:54
----------------------------------------------------
> Timing is not the main issue with Max.
>
> The problem comes from the graphical user interface.
now that is another interesting radical statement.
you cant make sequencers in max because of the lack of fancy 3D knobs, is that what you wanted to say? :)
> I only tried the demo patch though, but i can report that i often got small chunks of notes left after deleting with right click.
I'm not sure what you mean deleting with the right click. The delete button works fine from here. I don't have a windows computer to try it out with, but if you want to email me exactly what you're doing I'll see if I can figure it out.
Likewise with the ui objects on the right side. I did my best to integrate the max objects with the jsui, but I'm not sure if I did it the best way possible.
> And last but not least, this was very cpu taxing, i don't know how much of it should be accounted for the synth or the seq, but that was pretty worrying (reached around 75 % on my pentium4 3.5ghz).
Yeah, that seems to be the case, it's all in the sequencer. I'll see if I can simplify what it's putting in the high-priority thread.
> And last but not least, this was very cpu taxing, i don't know how much of it should be accounted for the synth or the seq, but that was pretty worrying (reached around 75 % on my pentium4 3.5ghz).
Actually it turns out that the processor hosage comes from all the redrawing, not from the data manipulation.
The sequencer sends out data when it gets an int and redraws when it gets a bang. That's what the bangbang at the top is for. So you can turn off redrawing completely by deleting the left wire out of the bangbang. On my computer processor usage drops from 90% to 25% when I do that. Or you can redraw every Nth tick using % and sel. Drawing every 4 - 10 ticks or so works nicely.
Roald Baudoux schrieb:
> It could be considered as a kind of merging between timeline, coll, lcd
> and some multidimensional functions both for drawing and editing.
Something a bit like technoui.js? Did anybody dive into the code? I
think it could be a good starting point for this kind of stuff...
Stefan
--
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com
Tried downloading this. I get a .js file that is readable but the other
three files show up with blank icons that Max 4.6 OSX 10.4.9 doesn't
recognize.
On 6/11/07 5:26 PM, "Dan Good" wrote:
>
> Now seems like a good time to share the sequencer I've been working on with
> jsui. The forums told me to be skeptical of getting accurate timing, but it
> seems to work okay.
>
> I was more interested in writing it to come up with a convenient way to carry
> around all the numbers I wanted to send to my synth than because I'm planning
> on writing lots of loopy beats. What's the more conventional solution to this
> problem?
>
> The code and an example patch (and a save file for the sequencer and the patch
> for the synth it's driving) are at www.dandroid.org/max/ddseq/ .
>
> There are probably still some bugs to be worked out, but it's pretty stable.
> Let me know if you think you might find it useful.
>
> cheers,
> Dan
Cheers
Gary Lee Nelson
Oberlin College
www.timara.oberlin.edu/GaryLeeNelson
could be useful.
>
> It could be considered as a kind of merging between timeline, coll,
> lcd and some multidimensional functions both for drawing and editing.
>
> Roald Baudoux
ah so you want a cubase object for mamsp, i see. :)
i didnt notice you were talking about the timeline object before hence my dumb answer.
if i ever need a "timeline" thing in max i will use a signal buffer at 22 khz and put signal numbers into it.
> Tried downloading this. I get a .js file that is readable but the other three files show up with blank icons that Max 4.6 OSX 10.4.9 doesn't recognize.
Seems Mac OS really wants them to be text files since they don't have extensions. I downloaded the files onto my computer (Max 4.6 OSX 10.4.9) and saw what you say. I was able to fix it on the two patches by going into get info, deleting the .txt extension and changing open with to Max/MSP. That ought to be enough to get it to work.
if you zip them and throw them all on a server it will preserve the
metadata for OS X. But probably best to name them for PC users anyway.
On Jun 12, 2007, at 3:13 PM, Dan Good wrote:
>
> Seems Mac OS really wants them to be text files since they don't
> have extensions. I downloaded the files onto my computer (Max 4.6
> OSX 10.4.9) and saw what you say. I was able to fix it on the two
> patches by going into get info, deleting the .txt extension and
> changing open with to Max/MSP. That ought to be enough to get it
> to work.
v a d e //
www.vade.info
abstrakt.vade.info
Ive done a few sequencer patches (www.axiom-crux.net) that I will be making available to the public soon. They took a long time to engineer and are very useful for making abstract electronic rhythms and chordal motions.
I realized that max isn't perfect at timing BUT msp is because its always signal rate. so now I use phasor metros for everything which is way more stable, not to mention you can do cool things with it that you cant with regular bang based sequencing (IE use a sine wave to drive the sequence or sah)
The ones I downloaded do not have .txt extensions.
On 6/12/07 3:13 PM, "Dan Good" wrote:
>
>> Tried downloading this. I get a .js file that is readable but the other three
>> files show up with blank icons that Max 4.6 OSX 10.4.9 doesn't recognize.
>
>
> Seems Mac OS really wants them to be text files since they don't have
> extensions. I downloaded the files onto my computer (Max 4.6 OSX 10.4.9) and
> saw what you say. I was able to fix it on the two patches by going into get
> info, deleting the .txt extension and changing open with to Max/MSP. That
> ought to be enough to get it to work.
Cheers
Gary Lee Nelson
Oberlin College
www.timara.oberlin.edu/GaryLeeNelson
> The ones I downloaded do not have .txt extensions.
Hmm, I'm not sure what the problem is in that case, sorry about that.
I took the directory and zipped it so you should be able to download it in the right format now. Hopefully that will work.
While I was at it I tweaked the patch so it's a little easier on the cpu. It now consumes about 45% of my G4 1.25 GHz.
Gary Lee Nelson schrieb:
> The ones I downloaded do not have .txt extensions.
I just added .pat extensions and they worked, but in general its better
to just zip all together, one file no problems...
Stefan
--
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com
Quote: f.e wrote on Wed, 13 June 2007 01:26
> It looks for sequence05.js, which is not in the folder. So i've replaced
> it with ddseq.js.
Wow, I apologize for that. Keeping versions of things straight is trickier than I expected.
> Then hitting the toggle freezes max to death.
I'm not so sure what's happening there, I can't get it to freeze my system. As mentioned it's a bit of a cpu hog, but it shouldn't be that bad. Does it change anything if you delete the wire from the select banging into the inlet? That should cut down on the cpu usage.
I don't know if anything strange might have happened with the setup when it was trying to load the wrong js file. If you want to try a fresh copy I fixed the patch online. I'm sure it works over here.
Sorry for the trouble, I'm new at sharing my code with the rest of the world..
Dan
The zipped version loaded fine. I had to make that replacement in text
because the outlets were lost with sequence05.js.
On 6/13/07 5:24 AM, "Dan Good" wrote:
>
> Quote: f.e wrote on Wed, 13 June 2007 01:26
>> It looks for sequence05.js, which is not in the folder. So i've replaced
>> it with ddseq.js.
>
> Wow, I apologize for that. Keeping versions of things straight is trickier
> than I expected.
>
>
>> Then hitting the toggle freezes max to death.
>
> I'm not so sure what's happening there, I can't get it to freeze my system.
> As mentioned it's a bit of a cpu hog, but it shouldn't be that bad. Does it
> change anything if you delete the wire from the select banging into the inlet?
> That should cut down on the cpu usage.
>
> I don't know if anything strange might have happened with the setup when it
> was trying to load the wrong js file. If you want to try a fresh copy I fixed
> the patch online. I'm sure it works over here.
>
> Sorry for the trouble, I'm new at sharing my code with the rest of the world..
>
> Dan
Cheers
Gary Lee Nelson
Oberlin College
www.timara.oberlin.edu/GaryLeeNelson
The ones I download first must have had hidden .pat extensions because they
opened in Final Cut. But all is well with the zipped archive.
On 6/13/07 2:08 AM, "Stefan Tiedje" wrote:
> Gary Lee Nelson schrieb:
>> The ones I downloaded do not have .txt extensions.
>
> I just added .pat extensions and they worked, but in general its better
> to just zip all together, one file no problems...
>
> Stefan
Cheers
Gary Lee Nelson
Oberlin College
www.timara.oberlin.edu/GaryLeeNelson