Yahoo Groups archive

Analogue-sequencer

Index last updated: 2026-04-28 23:15 UTC

Thread

Something else...

Something else...

2006-02-11 by colinfraser_com

I forgot to mention, the new beta adds another record option.
nDub has been replaced with a 'Dub' option, with settings 
of 'off', 'note' and 'gaps'.
Note is the same as nDub, gaps is like normal record but new notes 
will only be added into gaps in the existing pattern.

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

Re: [analogue-sequencer] Something else...

2006-02-12 by Martin Naef

Hi Colin

colinfraser_com wrote:
> I forgot to mention, the new beta adds another record option.
> nDub has been replaced with a 'Dub' option, with settings 
> of 'off', 'note' and 'gaps'.
> Note is the same as nDub, gaps is like normal record but new notes 
> will only be added into gaps in the existing pattern.

Very nice idea. In a similar vein: Would it be possible to have sculpt 
only affect active steps? That would enable quick entry of patterns e.g. 
consisting of a constant base note and the syncope playing some pattern, 
somewhat arpeggiator-like. I guess the same would be useful for 
randomization.

On a similar issue: The Midibox sequencer has a special button "All", 
which, when pressed, changes the value of all active steps. E.g., if I 
change the note, all active notes are changed. Admittedly, this can be 
achieved with the sculpt function, but for patterns with timebase 1, 
this can be awfully slow...

BTW: The direct selection of the menus is *very* nice. As I'm writing 
this e-mail, I'm preparing a small paper template so I know always what 
key goes to what menu. I think once the feature set is more stable, it 
might be a good idea to print something like that onto the front panel.

Bye
Martin

Re: [analogue-sequencer] Something else...

2006-02-12 by Martin Naef

Hi Colin

> BTW: The direct selection of the menus is *very* nice. As I'm writing 
> this e-mail, I'm preparing a small paper template so I know always what 
> key goes to what menu. I think once the feature set is more stable, it 
> might be a good idea to print something like that onto the front panel.

... while I'm at it: How about using the same selection mechanism for 
the AUX events? Page+StepXX select the category, data then chooses the 
event within the category.

Bye
Martin

RE: [analogue-sequencer] Something else...

2006-02-12 by Colin f

> ... while I'm at it: How about using the same selection mechanism for 
> the AUX events? Page+StepXX select the category, data then 
> chooses the 
> event within the category.

Since the step keys are not active during aux config, it doesn't even need
PAGE+step.
Step keys could select groups directly.
This may be easier to use than the current toggling between groups and
events, which I don't like much.
I'll do a test build this way and see how it feels.

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

RE: [analogue-sequencer] Something else...

2006-02-12 by Colin f

> Yup - they could also select the CC mode. EG, step 1 means CC, the 
> others select the group.

That's neater still, so I did it that way.

Test beta 41 is in Yahoo files now.

In aux config, you can still press PAGE to toggle between events and CCs.
But FUNC is no longer. To select an Event group, press one of the step keys.
The group name will be shown briefly.
Step key 1 = CC select.

It also adds PAGE+step access to all play mode pages.
One oddity with this is that step key 4 gives you the same page as step 1,
and 8 gives you the same page as 5.
This is because pages 4 and 8 are unimplemented, and just reset the page
index to 1 or 5 respectively.

I'm tempted to drop the incremental PAGE only action altogether.
I'm find the direct access much quicker, and I'll probably stop using the
old method myself.
Comments welcome.

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

Re: [analogue-sequencer] Something else...

2006-02-12 by Martin Naef

Hi Colin

Colin f wrote:
>> ... while I'm at it: How about using the same selection mechanism for 
>> the AUX events? Page+StepXX select the category, data then 
>> chooses the 
>> event within the category.
> 
> Since the step keys are not active during aux config, it doesn't even need
> PAGE+step.

Good point - reduces the number of buttons pressed. It might still be 
worth keeping the page button for the sake of coherency between the 
different modes though. Not a trivial decision.

> Step keys could select groups directly.

Yup - they could also select the CC mode. EG, step 1 means CC, the 
others select the group.

> This may be easier to use than the current toggling between groups and
> events, which I don't like much.

It's better than the previous method, but it could use some 
enhancements... Using the step keys should be a very workable solution, 
especially if I enhance my additional piece of paper that's now attached 
to the P3... ;-)

> I'll do a test build this way and see how it feels.

Looking forward!

Bye
Martin

Re: Something else...

2006-02-13 by Gary Chang

"Colin f" <colin@...> wrote:
>
> 
> I'm tempted to drop the incremental PAGE only action altogether.
> I'm find the direct access much quicker, and I'll probably stop
using the
> old method myself.
> Comments welcome.
> 

Colin, What you are doing is revolutionizing the P3's work flow, and I
must express my personal gratitude for your patient to hear our
comments and the persevering to think this through and to come up with
practical and meaningful new improvements.

Although we haven't abolished the enter/exiting routine, (and perhaps
that isn't practical), I think that the utilization of the previously
unused buttons to select pages simplifies our P3 - now that all of the
page access doesn't all fall into the page-page-page syndrome....

Thanks for your ingeniuty, Colin!

gc

RE: [analogue-sequencer] Something else...

2006-02-13 by Colin f

> Add a shiny
> "groove-box-style" printed panel (sticky transfers anyone?), and I
> think this will utterly rock!

I could print that on the silver laser film I use for the serial numbers...

> There's something rather wonderful about watching evolution in action.

I think this is more a case of intelligent design ;-)

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

Re: [analogue-sequencer] Something else...

2006-02-13 by Paul Nagle

On Sun, 12 Feb 2006 23:14:58 -0000, "Colin f" <colin@colinfraser.com>
wrote:

>I'm tempted to drop the incremental PAGE only action altogether.
>I'm find the direct access much quicker, and I'll probably stop using the
>old method myself.
>Comments welcome.

Woooo - now I'm faster than *ever*. Superb change dude. Add a shiny
"groove-box-style" printed panel (sticky transfers anyone?), and I
think this will utterly rock!

There's something rather wonderful about watching evolution in action.

Thanks man,

Paul
---
Paul Nagle - Joint Intelligence Committee - www.JointIntelligenceCommittee.com
                           SoftRoom Music - www.softroom.co.uk

RE: [analogue-sequencer] Something else...

2006-02-13 by Colin f

> > One oddity with this is that step key 4 gives you the same 
> page as step 1,
> > and 8 gives you the same page as 5.
> 
> I understand the implementation side of it, but it feels 
> rather strange. 
> The little math required to fix this should be simple enough.

Yeah, I'm just lazy.

> I wouldn't be surprised 
> if some people therefore still prefer the old method.

Nor would I.
The PAGE key should become sticky if that option is enabled, along with
FUNC, PLAY/EDIT and REC.
That would need to break the original functionality.
 
> The new method, however, clearly has the advantage when you add more 
> menu pages. There's no reason not to do that from a usability 
> point of 
> view, whereas previously you'd introduce a penalty due to the 
> increased 
> number of page-presses to reach any specific point.

I'll need to think of a couple more pages for play mode, to save me having
to fix the blanks.

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

Re: [analogue-sequencer] Something else...

2006-02-13 by Martin Naef

Hi Colin

Colin f wrote:
>> Yup - they could also select the CC mode. EG, step 1 means CC, the 
>> others select the group.
> 
> That's neater still, so I did it that way.
> Test beta 41 is in Yahoo files now.

Very neat - works beautifully here. Having a small overlay printed helps 
a lot to keep overview, and access is *a lot* faster.

> It also adds PAGE+step access to all play mode pages.
> One oddity with this is that step key 4 gives you the same page as step 1,
> and 8 gives you the same page as 5.

I understand the implementation side of it, but it feels rather strange. 
The little math required to fix this should be simple enough.

> I'm tempted to drop the incremental PAGE only action altogether.
> I'm find the direct access much quicker, and I'll probably stop using the
> old method myself.
> Comments welcome.

 From a usability point of view, the old page mode has the advantage 
that you always keep your eyes on the screen, whereas with PAGE+STEP you 
tend to look over to the step keys to choose the page and then need to 
refocus on the display (especially when you work with text labels below 
the step keys to find what you're looking for). I wouldn't be surprised 
if some people therefore still prefer the old method.

The new method, however, clearly has the advantage when you add more 
menu pages. There's no reason not to do that from a usability point of 
view, whereas previously you'd introduce a penalty due to the increased 
number of page-presses to reach any specific point.

So unless there's a good reason, I wouldn't drop the old method yet. But 
the new method is clearly a *very* good thing.

Bye
Martin

Re: [analogue-sequencer] Something else...

2006-02-13 by Martin Naef

Hi Colin

Colin f wrote:
>> I wouldn't be surprised 
>> if some people therefore still prefer the old method.
> 
> Nor would I.
> The PAGE key should become sticky if that option is enabled, along with
> FUNC, PLAY/EDIT and REC.
> That would need to break the original functionality.

Good point - that won't really work with the current selection method.

> I'll need to think of a couple more pages for play mode, to save me having
> to fix the blanks.

Let me think:
- MIDI reroute (all incoming MIDI is rerouted to channel X)
- Copy Pattern (all tracks)
- Mute all/none/stored
- Solo track/off
- Mixer mode toggle (not that I ever needed that one, but hey...)

... and of course in play mode (as I already mentioned)
Instead of
- Sculpt (note/upper/tap)
have:
- Sculpt Note (all, active, tap)
- Sculpt Upper (all, active, tap)
  plus an option to invert, e.g. FUNC-F2 means sculpt inactive steps

Randomize should have a similar option, e.g. to only randomize active 
steps. Maybe FUNC-F1? (but then you'd probably want to use the same 
mechanics for Sculpt, but have F2 invert the step selection)

Additional issues independent from the above:

Sculpt: Why don't I see the sculpt value immediately? Especially for 
slow patterns it's hit and miss when I try to set a specified value 
(e.g., I only see the current data value once I hit the next step)

Signed value entry: As an engineer, I understand why the current 
implementation starts at 0 - 63 and then wraps-around to -64, yet I find 
it counter-intuitive from a user's point of view. How about having 0 on 
the top/center with a small dead-zone to enable quick selection of 0?

Enough for tonight - I'm sure I'll get some more ideas... ;-)

Bye
Martin

RE: [analogue-sequencer] Something else...

2006-02-13 by Colin f

> Additional issues independent from the above:
> 
> Sculpt: Why don't I see the sculpt value immediately? Especially for 
> slow patterns it's hit and miss when I try to set a specified value 
> (e.g., I only see the current data value once I hit the next step)

Sculpting is captured at the start of each beat.
If you pressed the sculpt key just before the start of a beat, then the old
way, you would sculpt two steps.
The ideal implementation is probably to have a window for sculpt from half a
beat before, to half a beat after, so you can easily pick out one step to
sculpt.
 
> Signed value entry: As an engineer, I understand why the current 
> implementation starts at 0 - 63 and then wraps-around to -64, 
> yet I find 
> it counter-intuitive from a user's point of view. How about 
> having 0 on 
> the top/center with a small dead-zone to enable quick selection of 0?

I didn't start out doing it that way. The transpose knobs in playlist edit
still have zero in the middle.
But there isn't enough resolution to add a dead zone and still have the full
range of values.
I got fed up hunting for a mid-point zero on bi-polar knobs, and decided
just to make zero fully anticlockwise.
I think once you accept that's the way it works, it can make life easier.

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

Re: [analogue-sequencer] Something else...

2006-02-14 by Richard Scott

"How about displaying the actual, continuously updated data value in
brackets somewhere on the screen?"

I agree with Martin that something like that would be very helfpul - its
often impossible to get back to where you were on the P3

Richard

----- Original Message -----
From: "Martin Naef" <mnaef@acm.org>
To: <analogue-sequencer@yahoogroups.com>
Sent: Tuesday, February 14, 2006 9:35 AM
Subject: Re: [analogue-sequencer] Something else...


> Hi Colin
>
> Colin f wrote:
> >> Sculpt: Why don't I see the sculpt value immediately? Especially for
> >> slow patterns it's hit and miss when I try to set a specified value
> >> (e.g., I only see the current data value once I hit the next step)
> >
> > Sculpting is captured at the start of each beat.
> > If you pressed the sculpt key just before the start of a beat, then the
old
> > way, you would sculpt two steps.
> > The ideal implementation is probably to have a window for sculpt from
half a
> > beat before, to half a beat after, so you can easily pick out one step
to
> > sculpt.
>
> The window method is certainly a good idea. However, my issue is really
> about *visual* feedback. I need to see immediately where my data pot is,
> otherwise it's extremely difficult to hit a certain value. Try the
> following:
>
> I recently tried to add a set of AUX rel notes one octave below the main
> note. The track was running at timebase 4 (global Tempo 66), so rather
> slow. I only knew where my knob was at the start of each step, so it
> took more than one full track length just to adjust the data knob to
> -12. Now I had to keep pressing the sculpt key for another eternity
> until the track actually run through all steps. Of course, I slightly
> moved my left hand, so at some time I started writing -13. The result: I
> had to go through all steps again... In retrospective, it would have
> been faster to use the individual pots instead of sculpt.
>
> If the screen represented the data value at all times, it would have
> taken me one second to get to -12, and there's a good chance I would
> have spotted the change to -13 before it was written.
>
> How about displaying the actual, continuously updated data value in
> brackets somewhere on the screen?
>
> > I didn't start out doing it that way. The transpose knobs in playlist
edit
> > still have zero in the middle.
> > But there isn't enough resolution to add a dead zone and still have the
full
> > range of values.
> > I got fed up hunting for a mid-point zero on bi-polar knobs, and decided
> > just to make zero fully anticlockwise.
> > I think once you accept that's the way it works, it can make life
easier.
Show quoted textHide quoted text
>
> Agreed. If the dead-zone can't be implemented, having the 0 at full left
> makes sense.
>
> Bye
> Martin
>
>
>
>
> Yahoo! Groups Links
>
>
>
>
>
>
>

RE: [analogue-sequencer] Something else...

2006-02-14 by Colin f

> How about displaying the actual, continuously updated data value in 
> brackets somewhere on the screen?

I had planned to do this at one point, but put it off because it's more
complex to do than it might seem.
The display would need to show the value that would be sculpted for both the
note and upper row.
The note is easy, but for the upper row, it would need to check the upper
mode and scale the displayed value for velocity, length, delay, or whichever
aux was active.
Auxes would be the tricky bit, since there are a variety of different value
formats that can be displayed.
I'll look into it...

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

Re: [analogue-sequencer] Something else...

2006-02-14 by Martin Naef

Hi Colin

Colin f wrote:
>> Sculpt: Why don't I see the sculpt value immediately? Especially for 
>> slow patterns it's hit and miss when I try to set a specified value 
>> (e.g., I only see the current data value once I hit the next step)
> 
> Sculpting is captured at the start of each beat.
> If you pressed the sculpt key just before the start of a beat, then the old
> way, you would sculpt two steps.
> The ideal implementation is probably to have a window for sculpt from half a
> beat before, to half a beat after, so you can easily pick out one step to
> sculpt.

The window method is certainly a good idea. However, my issue is really 
about *visual* feedback. I need to see immediately where my data pot is, 
otherwise it's extremely difficult to hit a certain value. Try the 
following:

I recently tried to add a set of AUX rel notes one octave below the main 
note. The track was running at timebase 4 (global Tempo 66), so rather 
slow. I only knew where my knob was at the start of each step, so it 
took more than one full track length just to adjust the data knob to 
-12. Now I had to keep pressing the sculpt key for another eternity 
until the track actually run through all steps. Of course, I slightly 
moved my left hand, so at some time I started writing -13. The result: I 
had to go through all steps again... In retrospective, it would have 
been faster to use the individual pots instead of sculpt.

If the screen represented the data value at all times, it would have 
taken me one second to get to -12, and there's a good chance I would 
have spotted the change to -13 before it was written.

How about displaying the actual, continuously updated data value in 
brackets somewhere on the screen?

> I didn't start out doing it that way. The transpose knobs in playlist edit
> still have zero in the middle.
> But there isn't enough resolution to add a dead zone and still have the full
> range of values.
> I got fed up hunting for a mid-point zero on bi-polar knobs, and decided
> just to make zero fully anticlockwise.
> I think once you accept that's the way it works, it can make life easier.

Agreed. If the dead-zone can't be implemented, having the 0 at full left 
makes sense.

Bye
Martin

Re: [analogue-sequencer] Something else...

2006-02-14 by Martin Naef

Hi Paul

Paul Nagle wrote:
>> I recently tried to add a set of AUX rel notes one octave below the main 
>> note. The track was running at timebase 4 (global Tempo 66), so rather 
>> slow. I only knew where my knob was at the start of each step, so it 
>> took more than one full track length just to adjust the data knob to 
>> -12. Now I had to keep pressing the sculpt key for another eternity 
>> until the track actually run through all steps. Of course, I slightly 
> 
> Martin, the absolute best way to get round this is to change the tbase
> to 16, maybe even up the tempo, then make your changes and switch back
> to tbase  4. At least that's what I do in such situations. 
> With the new direct access to menu pages this is pretty darn fast and
> saves all that waiting around.

That's certainly a possible work-around (not one I would recommend live 
unless you have the track muted...), but but I'd prefer to fix the issue 
at the core.

Bye
Martin

Re: [analogue-sequencer] Something else...

2006-02-14 by Paul Nagle

On Tue, 14 Feb 2006 09:35:04 +0000, Martin Naef <mnaef@acm.org> wrote:

>I recently tried to add a set of AUX rel notes one octave below the main 
>note. The track was running at timebase 4 (global Tempo 66), so rather 
>slow. I only knew where my knob was at the start of each step, so it 
>took more than one full track length just to adjust the data knob to 
>-12. Now I had to keep pressing the sculpt key for another eternity 
>until the track actually run through all steps. Of course, I slightly 

Martin, the absolute best way to get round this is to change the tbase
to 16, maybe even up the tempo, then make your changes and switch back
to tbase  4. At least that's what I do in such situations. 
With the new direct access to menu pages this is pretty darn fast and
saves all that waiting around.

Paul

---
Paul Nagle - Joint Intelligence Committee - www.JointIntelligenceCommittee.com
                           SoftRoom Music - www.softroom.co.uk

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.