Forums > MaxMSP

rhythm experiment…

September 4, 2008 | 4:40 pm

I have been doing some experimenting with the global transport
and the new timing features of phasor. I want to be able to
construct rhythms by randomly selecting note divisions (1n, 2n,
4n, etc.) each time there is a new phasor cycle. But for some
reason the chain gets broken when the note division is is
set to 64nt. Any ideas why it might be doing that?

Thanks,
Anthony

– Pasted Max Patch, click to expand. –

September 4, 2008 | 6:14 pm

I didn’t test your patch at all, but just from looking at it- it could be that when phasor~ is running that fast it never goes above 0.995 before it jumps back down to 0. Maybe instead of using >~, try using delta~

– Pasted Max Patch, click to expand. –

September 4, 2008 | 6:41 pm

Instead of having it change when the phasor is about to get to the top of its ramp, you should have it change when it restarts at the bottom. In other words, change your >=~ 0.995 to < ~ 0.5. It's much more reliable, and you don't have to worry about deciding on some arbitrary value close to one (e.g. 0.995).


September 4, 2008 | 6:42 pm

Hi there,
Just a thought….
Looking through the table of ticks and meter values in

file:///Applications/Max5/Cycling%20’74/docs/vignettes/core/maxtime_syntax.maxvig.xml

I noticed a while ago that 64nt is omitted. Also, trying to type in 64nt into the inspector for a metro objects’ interval returns to 0. when entered.
I assume that 64nt is not included in permissible divisions, though why not does make me curious. I hazard a guess that it would be 20 ticks, so ‘omission due to fractional amount of ticks’ isn;t it!
If I’m wrong on that some one please correct me..

Kind Regards
Leigh

Quote: Anthony Palomba wrote on Thu, 04 September 2008 17:40
—————————————————-
> I have been doing some experimenting with the global transport
> and the new timing features of phasor. I want to be able to
> construct rhythms by randomly selecting note divisions (1n, 2n,
> 4n, etc.) each time there is a new phasor cycle. But for some
> reason the chain gets broken when the note division is is
> set to 64nt. Any ideas why it might be doing that?
>
>
> Thanks,
> Anthony
>
>
>
>

– Pasted Max Patch, click to expand. –

>
—————————————————-


September 4, 2008 | 6:45 pm

>
> I noticed a while ago that 64nt is omitted. Also, trying to type in 64nt into the inspector for a metro objects’ interval returns to 0. when entered.

Same goes for phasor object btw,


September 4, 2008 | 6:49 pm

Thanks Nick, your way is much more accurate. But every now and
then it will stop, it tends to stop on either 64nt or
128nt. Could this be a bug? Here is the corrected patcher…

– Pasted Max Patch, click to expand. –

September 4, 2008 | 7:35 pm

I had been frustrated with the transport object until i discovered the ticks output. I was frustrated because part of a metrical timing system should include more interesting rhythms such as "5 against 4". By attaching a modulo op on the ticks output, you can get pretty much any normal/abnormal metrical division you’d like. Please check this patch out and turn off the randomizer to hear the 5:4 (5 quarter notes in the place of 4) entry at the end of the menu which I’ve added to your patch. In addition, the 64nt and 128nt divisions also work properly in this patch. It just might be a good workaround or even a useful trick to get more than just the "n, nd, nt" limited offerings of the transport system as it’s been defined so far, hope it helps:

– Pasted Max Patch, click to expand. –

September 4, 2008 | 8:09 pm

also, I like listening to this patch which has an added constant-quarter-for-comparison beat:

– Pasted Max Patch, click to expand. –

September 4, 2008 | 9:16 pm

Symbols such as 4n are just a convenient shorthand for numerical tick values, which I would suggest are a more general and flexible way to specify tempo-relative time. Note that numerical tick values can also be fractional. The set of symbols built-in into the system was never supposed to be anything other than a basic set. I feel a little uncomfortable about these symbols since they are not "computational" and thus represent a bit of an exception to a general Max design principle which is to try to specify as much as possible using numerical spaces. However, given that they are just a shorthand for numbers, you need not use them and you shouldn’t complain if your favorite metrical grouping is not represented.

There never was a value assigned to the symbol 64nt, and if this is listed in the documentation or in a help file, it is an error and we should remove it.

David Z.


September 5, 2008 | 2:00 am

David Z. wrote:
"I feel a little uncomfortable about these symbols since they are not ‘computational’ and thus represent a bit of an exception to a general Max design principle which is to try to specify as much as possible using numerical spaces. However, given that they are just a shorthand for numbers, you need not use them and you shouldn’t complain if your favorite metrical grouping is not represented."

I think this is what I initially had a problem with is that the numerical shorthand was tempting me to be limited in rhythmic space. But then I discovered that using the transport’s tick output was liberating and, hopefully, my posts explain how the transport is actually a great new tool, NOT a complaining point. I would hate to be misunderstood.

To further the promotion of how flexible you can be using transport, one way to use the shorthand of ITM to communicate fully (sans connection; sans ticks-output) as the ITM is designed to do, is by using phasors set to "1n" which are locked to the transport and then using rate~ to get any metrical division you’d like.

It’s nicely extendable.


September 5, 2008 | 1:54 pm

Will it appears that I can not manually set a phasor to
64nt or 128nt. I am not sure where I got that original ubumenu,
I think from a previous example.

RabidRaj, I like your solution to this problem. Using the raw
tick count is a good alternate way.

Now that we have solved this problem, I was wondering if anyone
had any interesting ideas as to concatenating rhythmic intervals.
Prob is the first thing that comes to mind. But I was hoping
someone could spark my interest with some other ideas.
I am also very interested in representing the note divisions in
some non linear way, like say for example a matrix. Where the x-axis is length and y-axis is swing. Or maybe convert that to
polar coordinates and gave some Lissajous function pass through it.


Viewing 11 posts - 1 through 11 (of 11 total)