Analogue-sequencer

sorry a bit OT but did anyone here see or use the Genoqs Octopus? and
verdict ?
one thing is sure, it looks great... like a starship controlboard or
something. button and knobs make it look a bit like the Notron, but without
being shaped like a toilet seat ...

BTW I ordered a P3 last friday. Can't wait to play it , the three weeks to
come are going to be a long wait... Colin can you hear me ;-)
But there is definitely a good thing about this production lead time, it is
that I'm already learning the machine by the manual. Call me a geek but I'm
reading a few pages every night and it really helps and I hope it'll help
when the P3 arrives. I remember reading a previous version of the manual
last year and the one that is online now is much much easier to read and
understand. Well done Colin !

O.


[Non-text portions of this message have been removed]

On 6/6/06, Olivier Tison <otison@gmail.com> wrote:
>
>   I remember reading a previous version of the manual
> last year and the one that is online now is much much easier to read and
> understand. Well done Colin !
>



>


Agreed.   The new manual is very good.  I'm still a little fuzzy about the
accumulator masks after reading that section but it is a somewhat complex
feature.

-- 
miles


[Non-text portions of this message have been removed]

> Agreed.   The new manual is very good.  I'm still a little 
> fuzzy about the
> accumulator masks after reading that section but it is a 
> somewhat complex
> feature.

I could probably explain masking a little better...
It's one of those things that is really quite simple once you understand it.

The mask events are all based on the logical operation AND.
Logical operations work on values that are either TRUE or FALSE.
The AND operation takes two values, and the result is only TRUE if *both*
the values are TRUE.

It works the same way as AND in English...
If you say "The grass is green and the sky is blue", the whole statement is
true.
If you say "The grass is purple and the sky is blue", the whole statement is
false.
Unless you live somewhere with purple grass.

In the case of a mask event "Mask gate,dAcc>n", P3 evaluates this
expression:

GATE = GATE AND (D-accumulator > n)

The value 'n' is the aux value for the step in the pattern.

If GATE is OFF (i.e. FALSE), then the mask event will never turn it on
because GATE is the first value in the AND operation.
Both values must be TRUE for the result to be TRUE.
GATE is being set to the result of the AND operation.

If GATE is ON (i.e. TRUE), then P3 evaluates whether the current value of
the Aux D accumulator is greater than the aux value set on the pattern step.
If the accumulator value is greater, then the expression "D-accumulator > n"
is TRUE, so the result of the AND is TRUE.
GATE is set to the result of the AND, so the GATE stays on.

If the accumulator value were not greater than the step aux value,
"D-accumulator > n" is FALSE, the result of the AND is FALSE, and GATE is
turned off.

Does that makes sense ?
All other mask events work the same way.
You have a status value from the pattern which is ANDed with some comparison
of the aux step value and either the aux D accumulator or a knob position.
In the case or random masks, the status value is ANDed with a TRUE/FALSE
value generated at random based on the probability set by the step value.

Best regards,
Colin Fraser
Sequentix Music Systems Ltd
http://www.sequentix.com

--- In analogue-sequencer@yahoogroups.com, "Colin Fraser" <colin@...>
wrote:

> I could probably explain masking a little better...
> It's one of those things that is really quite simple once you
understand it.

Thanks! That helps!

Another related stupid question:

Are there only accumulators for note, velocity and Aux D?

-Jim

On 6/6/06, Colin Fraser <colin@sequentix.com> wrote:
>  If GATE is ON (i.e. TRUE), then P3 evaluates whether the current value of
>  the Aux D accumulator is greater than the aux value set on the pattern step.
>  If the accumulator value is greater, then the expression "D-accumulator > n"
>  is TRUE, so the result of the AND is TRUE.
>  GATE is set to the result of the AND, so the GATE stays on.

That's definitely clearer, thanks.  So auxD = the D accumulator,
whatever accumulator is active in the D slot?  It's just a raw integer
value under the hood, regardless of what it's accumulating?

Also, if I haven't explicitly activated an accumulator in D, is it
just treated as a zero value?

Thanks for helping to explain this part.  It seems like a powerful feature.

Which sort of leads me to my next question, once I've got a
complicated sequence running with a bunch of accumulators, masks,
randomize events etc, it can get pretty tricky to unroll what's going
on, particularly if I come back to the P3 after a while.  Any tips for
keeping things straight or is it just a question of keeping focused
and paying close attention?  There's no way to temporarily mute some
or all aux events or anything like that, right?

-- 
miles

>  So auxD = the D accumulator,
> whatever accumulator is active in the D slot?  It's just a raw integer
> value under the hood, regardless of what it's accumulating?

Aux D is the fourth aux after A, B anc C.
The patterns hold a value for each aux on every step.

The Aux D accumulator is a persistent value, which is added to the step
value for aux D on every step.
All three accumulators have a range of -128 to +127.
There is an option to 'disconnect' the Aux D accumulator, so you can use aux
D for un-accumulated events or CCs, while still using aux D masking.
 
> Also, if I haven't explicitly activated an accumulator in D, is it
> just treated as a zero value?

The accumulators are always active, but they are reset to zero whenever P3
is started or a new pattern selected, so unless you explicitly set up an
event to alter the accumulator value, it won't have any effect.

> Which sort of leads me to my next question, once I've got a
> complicated sequence running with a bunch of accumulators, masks,
> randomize events etc, it can get pretty tricky to unroll what's going
> on, particularly if I come back to the P3 after a while.  Any tips for
> keeping things straight or is it just a question of keeping focused
> and paying close attention?  There's no way to temporarily mute some
> or all aux events or anything like that, right?

Once you get the hang of what the events do, you can pretty quickly work out
what's going on by stepping through the four auxes in a pattern you haven't
looked at for a while.
Sometimes it can be good not to be entirely sure what's going on ;-)

Best regards,
Colin Fraser
Sequentix Music Systems Ltd
http://www.sequentix.com

> Another related stupid question:
> 
> Are there only accumulators for note, velocity and Aux D?

There are only note, velocity and aux D accumulators for each track. So 24
in total.
The aux D accumulator can be applied to some other values, using "add dAcc
to length" or "add dAcc to delay", or any "set ..." events, which pretty
much covers everything.

Best regards,
Colin Fraser
Sequentix Music Systems Ltd
http://www.sequentix.com

"Sometimes it can be good not to be entirely sure what's going on ;-)"

aha, the secret philosophy of the P3 is revealed at last!

Richard

 

[Non-text portions of this message have been removed]

>>sorry a bit OT but did anyone here see or use the Genoqs Octopus?
and verdict ?
one thing is sure, it looks great... like a starship controlboard or
something. button and knobs make it look a bit like the Notron, but
without being shaped like a toilet seat ...<<

nor is it built from dustbin lids, like the notron mk1 was. 

mine arrived this week. it's a bit early for me to deliver a full
review, but the build quality is superb. it has about half a tree's
worth of what looks like mahogany holding the whole shebang together,
& a huge enamel steel control surface with genuine ball bearings for
buttons (i.e. not the fake ball bearings that the notron has). heavy,
it is. 
the buttons roll around under your hand. it feels fantastic. it just
fits on the lid of my mellotron. which is nice.
the knobs are all rotary encoders & shaped like you could operate them
with sockets or spanners instead of fingers. this is all very exciting
& makes you want to spend hours driving it. under the hood, it's not
as flexible as the p3- not anywhere near- but then (for me, at least)
this is akin to comparing a rickenbacker 12 string with a
stratocaster. I shall use them both, & the maq, & the notron (when
it's mended).
(& the zeit, when it turns up!)

duncan/r.m.i.

Hi

when the mixer is enabled why does it use the row of knobs about "part" instead of "track"? Seems counter intuitive to me - I would much prefer to have the knobs be above the actual tracks that are being controlled. It would use less of brain to know immediately which knob was going where.

Richard

[Non-text portions of this message have been removed]

Richard wrote:
> Hi
>
> when the mixer is enabled why does it use the row of knobs about "part" instead of "track"? Seems counter intuitive to me - I would much prefer to have the knobs be above the actual tracks that are being controlled. It would use less of brain to know immediately which knob was going where.
>   
When you assign knobs for a variety of functions (knob grab, knob mask 
etc.), I think it makes sense for those assignments to be directly above 
the tracks they relate to.
I like to know that the mixer is always in the same place and, as I use 
it a lot, I find I can handle it without thinking, making the 1-8 
translation in my head (I also moved the knobs around so their colours 
are in a different layout on the right hand side).

Generally, you might only assign a few knobs here and there, and maybe 
not in every bank, so if you use the mixer as you suggest, being able to 
mentally link the knob assignments for each bank with a few selected 
knobs from the right hand side could be mind-blowingly hard.

Dunno for sure but that might be it

Paul

---
www.softroom.co.uk / www.JointIntelligenceCommittee.com

Hi

What I would really like is to be able to set a high and low note limit for the random aux settings, so I could set velocity and pitch not to stray beyond a certain limit - I use the random features for all my work on the P3 but at the moment too many things can jump out in an out of control way. I could also for example set a cello part to be played on a sampler so it didn't stretch beyond the natural range of cello... I could compose a playable string quartet on the P3!!

I made various suggestions a few months ago about controlling randomness - most of which were judged too CPU or memory intensive  - Surely this abilty to limit randomness just using two numbers would not be too expensive and it could be useful addition to other users than myself too? If so - do speak up

I also wonder of RNPNs are still in the pipeline?

Richard

[Non-text portions of this message have been removed]

Richard wrote:
> Hi
>
> What I would really like is to be able to set a high and low note limit for the random aux settings, so I could set velocity and pitch not to stray beyond a certain limit - I use the random features for all my work on the P3 but at the moment too many things can jump out in an out of control way. I could also for example set a cello part to be played on a sampler so it didn't stretch beyond the natural range of cello... I could compose a playable string quartet on the P3!!
>   
You can set a high and low note for the randomness - randomisation is 
always positive so the lowest output value is the actual 
note/velocity/whatever. The highest value depends on the random amount 
you set for each step using the knob. One of the things I did always 
thing would be cool was a means of "compressing" certain output values 
post-aux event processing (then you could do ducking of values based on 
other tracks etc.) but there are probably more fun things to cram into 
the remaining OS space.

Anyway, time to go pick some magic mushies... ;)

Paul

---
www.softroom.co.uk / www.JointIntelligenceCommittee.com

Another cool thing you can do with randomness is vary the random values 
over time. To do this, increment the aux D accumulator then set the 
random aux you require on row D. Any steps set this way will change as 
the D accumulator counts up. You can do groovy things using the 
accumulator behaviours - giving a whole new dimension from randomness. 
Or use knob control of the randomness amount so you can start to 
increase or decrease for specific steps whenever you want.

All good clean fun!

Paul

> What I would really like is to be able to set a high and low 
> note limit for the random aux settings, so I could set 
> velocity and pitch not to stray beyond a certain limit - I 
> use the random features for all my work on the P3 but at the 
> moment too many things can jump out in an out of control way. 
> I could also for example set a cello part to be played on a 
> sampler so it didn't stretch beyond the natural range of 
> cello... I could compose a playable string quartet on the P3!!
> 
> I made various suggestions a few months ago about controlling 
> randomness - most of which were judged too CPU or memory 
> intensive  - Surely this abilty to limit randomness just 
> using two numbers would not be too expensive and it could be 
> useful addition to other users than myself too? If so - do speak up

Thinking about this, there is a relatively simple solution for note limiting
that wouldn't take too much code space.
The 'note rng' settings allow you to set the base note and range covered by
the NOTE knobs when you are editing a pattern.
These are stored separately for each track.
How about a user config option that allows the note range for input to also
be enforced as a limit for randomised and accumulated notes ?
It would probably need a little finer control over the span than at present,
but that wouldn't be too hard.
 
> I also wonder of RNPNs are still in the pipeline?

It's on the possible TBI list, but there are other things before it that are
higher priority use of code space.

Best regards,
Colin Fraser
Sequentix Music Systems Ltd
http://www.sequentix.com

_"You can set a high and low note for the randomness - randomisation is 
always positive so the lowest output value is the actual 
note/velocity/whatever. The highest value depends on the random amount 
you set for each step using the knob." 

ah, mm, this machine is still confusing me after a year: not having the latest OS didn't help!

thanks paul!

Hmm, a year later and i am still confused about this machine! What you say means the randomness is always adding a value and never subtracting it so I still think the abilty to define upper and lower limits would be more flexi
"You can set a high and low note for the randomness - randomisation is 
always positive so the lowest output value is the actual 
note/velocity/whatever. The highest value depends on the random amount 
you set for each step using the knob." 


Hmm, a year later and i am still confused about this machine! What you say means the randomness is always adding a value and never subtracting it so I still think the abilty to define upper and lower limits would be more flexible.

_,_._,___ 

[Non-text portions of this message have been removed]

Richard wrote:
> Hmm, a year later and i am still confused about this machine! What you say means the randomness is always adding a value and never subtracting it so I still think the abilty to define upper and lower limits would be more flexible.
>   
I see what you mean: you'd have a random value generated that was 
between x and y values rather than 0 and x? Me, I wanted bi-polar random 
values.
Another good way to get random values taken from a specific list of 
values is to grab them from another track. You could set the other track 
to a random order but with 16 values you've set to be useful in your 
receiving pattern. Then use the grab events to get their value. You 
would grab any one of 16 values at random, more if you use multiple 
patterns in a playlist.

I should probably knock up a load of pattern examples as sysex so you 
can grab them and try them out. I use a lot of random functions, usually 
randomly masking ties and set note or set velocity events but sometimes 
randomly masking note randomise events, repeats, direction changes and 
so on. It's a whole wacky random world out there... ;)
Paul

---
www.softroom.co.uk / www.JointIntelligenceCommittee.com

> _"You can set a high and low note for the randomness - 
> randomisation is 
> always positive so the lowest output value is the actual 
> note/velocity/whatever. The highest value depends on the 
> random amount 
> you set for each step using the knob." 
> 
> ah, mm, this machine is still confusing me after a year: not 
> having the latest OS didn't help!

What Paul means is you can set a high note and low note for randomness on a
step-by-step basis - if you set the note to C4 and the randomisation amount
to 12, then the result will be in the range C4 to C5, but that's a lot more
tricky than being able to set the high and low limit for the track, and then
not having to do too many sums...

Best regards,
Colin Fraser
Sequentix Music Systems Ltd
http://www.sequentix.com

Folks,

FAO those of you waiting on delayed wooden end cheeks...
I've been in to the factory this morning to approve a sample from a revised
CNC program.
It looked very nice.
Hopefully they'll get the rest of them done next week, then take a few days
for finishing.
So they could be ready to ship out the week after...

Best regards,
Colin Fraser
Sequentix Music Systems Ltd
http://www.sequentix.com

On 15 Sep 2006, at 03:52, Paul Nagle wrote:

> Another good way to get random values taken from a specific list of
> values is to grab them from another track.

This is also a useful technique for building long sequences: the  
Loremarie sequence (appearing at the Different Skies webcast sometime  
soon) is only four notes, except that the first note is a grab from  
another track with a timebase four times longer; the result is a  
cycle of 128 notes.

(This is when I realised that the GBar countdown on the status LED's  
is actually 8 bits; d'oh.)

	-- N.


   nick rothwell -- composition, systems, performance -- http:// 
www.cassiel.com

Groovy

--- In analogue-sequencer@yahoogroups.com, "Colin Fraser" <colin@...>
wrote:
>
> Folks,
> 
> FAO those of you waiting on delayed wooden end cheeks...
> I've been in to the factory this morning to approve a sample from a
revised
> CNC program.
> It looked very nice.
> Hopefully they'll get the rest of them done next week, then take a
few days

> for finishing.
> So they could be ready to ship out the week after...
> 
> Best regards,
> Colin Fraser
> Sequentix Music Systems Ltd
> http://www.sequentix.com
>

"The 'note rng' settings allow you to set the base note and range covered by
the NOTE knobs when you are editing a pattern.
These are stored separately for each track.
How about a user config option that allows the note range for input to also
be enforced as a limit for randomised and accumulated notes ?"

could be a good work around - it certainly makes sense to do this on a track by track basis - of course it may be confusing if I forget it and wonder why the accumaltors aren't doing what I expect

Richard

  ----- Original Message ----- 
  From: Colin Fraser 
  To: analogue-sequencer@yahoogroups.com 
  Sent: Friday, September 15, 2006 10:14 AM
  Subject: RE: [analogue-sequencer] random limiters



  > What I would really like is to be able to set a high and low 
  > note limit for the random aux settings, so I could set 
  > velocity and pitch not to stray beyond a certain limit - I 
  > use the random features for all my work on the P3 but at the 
  > moment too many things can jump out in an out of control way. 
  > I could also for example set a cello part to be played on a 
  > sampler so it didn't stretch beyond the natural range of 
  > cello... I could compose a playable string quartet on the P3!!
  > 
  > I made various suggestions a few months ago about controlling 
  > randomness - most of which were judged too CPU or memory 
  > intensive - Surely this abilty to limit randomness just 
  > using two numbers would not be too expensive and it could be 
  > useful addition to other users than myself too? If so - do speak up

  Thinking about this, there is a relatively simple solution for note limiting
  that wouldn't take too much code space.
  The 'note rng' settings allow you to set the base note and range covered by
  the NOTE knobs when you are editing a pattern.
  These are stored separately for each track.
  How about a user config option that allows the note range for input to also
  be enforced as a limit for randomised and accumulated notes ?
  It would probably need a little finer control over the span than at present,
  but that wouldn't be too hard.

  > I also wonder of RNPNs are still in the pipeline?

  It's on the possible TBI list, but there are other things before it that are
  higher priority use of code space.

  Best regards,
  Colin Fraser
  Sequentix Music Systems Ltd
  http://www.sequentix.com



   

[Non-text portions of this message have been removed]

Hi

I'd really appreciate finding a way to send simple program changes at the beginning of each track without using up an aux each time (my lovely old Akai SGO1v won't remember multis)

Is this possible? Could it be? - its a shame to use up a whole aux just to send one value each time

Also, is it possible to add a 16 bank config with 2 patterns per travck - this would cover 95% of what i do.

Richard

[Non-text portions of this message have been removed]

On 16 Sep 2006, at 00:56, Richard wrote:

> Also, is it possible to add a 16 bank config with 2 patterns per  
> travck

Wouldn't that be 16 x 3?


   nick rothwell -- composition, systems, performance -- http:// 
www.cassiel.com

"Wouldn't that be 16 x 3?"



  well 16 x 3 would be nice,  but only 3 x 16, 6 x 8, or  12 x 
  4 are availlable under the current OS



  Richard 

  Recent Activity
    a..  3New Members
    b..  3New Files
  Visit Your Group 
  SPONSORED LINKS
    a.. Sampler 
    b.. Voice synthesizer 
    c.. Text to speech synthesizer 
    d.. Electronic music 
    e.. Electronic music production 
  Yahoo! News
  Sexual Health

  Get important

  sex health news

  Y! GeoCities
  Share Your Passion

  Join the web's lar- gest community.

  New business?
  Get new customers.

  List your web site

  in Yahoo! Search.
  . 
   

[Non-text portions of this message have been removed]

On 16 Sep 2006, at 12:43, Richard wrote:

>   well 16 x 3 would be nice,  but only 3 x 16, 6 x 8, or  12 x
>   4 are availlable under the current OS

What I meant was: why would you want 16 x 2 if there's (presumably)  
room for 16 x 3?


   nick rothwell -- composition, systems, performance -- http:// 
www.cassiel.com

>   "Wouldn't that be 16 x 3?"

>   well 16 x 3 would be nice,  but only 3 x 16, 6 x 8, or  12 x 
>   4 are availlable under the current OS

The way patterns are divided up is based on taking bits from the pattern
number and giving them to the bank number.
So it only works easily with binary increments - 4, 8, 16...
2 patterns per bank would be possible in theory, but there isn't enough
memory left to hold the extra data for bank settings, parts and playlists.

Best regards,
Colin Fraser
Sequentix Music Systems Ltd
http://www.sequentix.com

>>Thinking about this, there is a relatively simple solution for note
limiting that wouldn't take too much code space.
The 'note rng' settings allow you to set the base note and range
covered by the NOTE knobs when you are editing a pattern.
These are stored separately for each track. 
How about a user config option that allows the note range for input to
also be enforced as a limit for randomised and accumulated notes ?
It would probably need a little finer control over the span than at
present, but that wouldn't be too hard.<<

both elegant & intuitive. 
to be honest, I thought it worked like that anyway, with the base-note
& range functioning "after the fact", so to speak, rather like the FTS
calculation.
I too wish to create string quartets using my h/w sequencers, with the
p3's lined up to perform all the monophonic parts & my octopus playing
piano & drums. I know, not really a quartet... chamber orchestra then.
with drums. :-)
paul, is it that time of year again already? opinion on y'r newest JIC
follows under seperate cover. 

duncan.