Jump to content

MIDIbox SEQ V4 Release + Feedback


Recommended Posts

Posted

...i think i´ve found a bug too... :whistle:

when a track is set to retrigger ( restart "on" ) and a note is received, it seems that it doubletriggers the first step...so when i.e. cubase is driving the transposition of a track via notes and restart is "on" , it´s allways offsync/shifted...

i think its not really doubletriggering but looses one step...just looks like this when syncing via midiclock and retriggering via notes...

Posted

Beta42 is available now (with less features than planned, but I thought that it would be better to release the bugfixes with higher priority)


MIDIboxSEQ V4.0beta42
~~~~~~~~~~~~~~~~~~~~~

o Handling of Start/Stop/Pause button changed for Slave and Auto Mode to allow
proper synchronisation to the measure when multiple sequencers are playing
together.

Whenever a MIDI clock is received:
- the STOP button won't stop the sequencer, but just disables the output of
all tracks (like a Mute). The step progression continues to run.
- the START button won't start the sequencer immediately.
Instead MBSEQ will wait for the next measure and then enable track output again.
- the PAUSE button will enable/disable track output, but the track progression
continues to run.

In order to notify that track outputs are disabled, the START/STOP/PAUSE
LED flashes to the beat.

o Added Live page.
The page can be entered by pressing UTILITY -> Live (GP Button #11)

Inside this page notes (and drum instruments) can be played with the GP buttons.

In addition the 4 MIDI IN Busses for Transposer/Arpeggiator/Live Playing can be
configured like in the MIDI page.

Following Live Play parameters are available:
- track selection
- track mute/unmute
- Octavewise transpose
- Velocity (only used for GP button keyboard)
- FTS (Force-To-Scale)
- FX on (enables Humanizer, LFO, Limiter and Echo)
Note: the Echo effect will only work when the sequencer is running!


o parameter layer mutes are now also working for
velocity/length/probability/delay/roll/roll2

o Echo Fx now also works with "zero delay". In conjunction with the Note incrementer
and (optionally) force to scale this results into a "poor man's chord" mechanism. ;-)

o Move function works beyond step 16 as intended again
[/code]

and on the Copy function the selection, of the lowest step to copy, by use of the leftmost encoder does not work.

Instead the leftmost encoder affetcts the highest step to copy what the rightmost encoder should do, and still does.

Sorry, I wasn't able to reproduce this

when a track is set to retrigger ( restart "on" ) and a note is received, it seems that it doubletriggers the first step...so when i.e. cubase is driving the transposition of a track via notes and restart is "on" , it´s allways offsync/shifted...

i think its not really doubletriggering but looses one step...just looks like this when syncing via midiclock and retriggering via notes...

I tried to improve this a bit, but without much luck.

The problem: Cubase sends the note event after the MIDI clock event, accordingly the transpose function will restart the track after MBSEQ processed the first step.

In Logic I'm able to apply a negative delay to the events which are sent to MBSEQ, this solves the issues at my side.

This is the recommended approach anyhow, you would do the same when controlling (for example) the arpeggiator of an external synth which is synchronized to the MIDI clock.

Best Regards, Thorsten.

Posted

Just gut to go updating to Beta 42 :)

But first i will have a closer look on the copy function.

View PostGridracer, on 30 May 2011 - 23:53, said:

and on the Copy function the selection, of the lowest step to copy, by use of the leftmost encoder does not work.

Instead the leftmost encoder affetcts the highest step to copy what the rightmost encoder should do, and still does.

Sorry, I wasn't able to reproduce this

After having a closer look on the misbehaviour of the copy function,

I noticed that the failure only occures under certain circumstances:

(Condition A, B and C have to be fulfilled)

A. Track length has to be longer than 16. (other track configuration parameters seem to have no influence)

B. The position indicators of the highest and lowest step to copy have to be

on different step views eg.: <13>-<24>

(if both indicators are on the same step view everything works fine eg.: <17>-<24>)

C. Finally but most important: The highest step to copy has to be changed once before the selection routine of the

lowest step starts to mess up.

(so if you always set up the lowest step to copy first before you ever change the

highest setp to copy, the effect will never show up)

So her a cenario to provoke the bug:

1. Have a track with Length=32

2. Call copy function (steps to copy should be initially 1-32)

3. Change the highest step to copy from 32 to a value higher than 16: 24eg.

4. Try to change the lowest step to copy-> the leftmost encoder will

change the highest step to copy instead of the lowest.

Hope you can reproduce it now.

Best Regards

Gridracer

Posted

Thanks, got it!

It makes sense that the leftmost encoder always changes the left boundary instead of doing strange calculations ;)

Next release in ca. 2 weeks

Best Regards, Thorsten.

  • 3 weeks later...
Posted

Hi, I have been using the random generator over the weekend and found something a bit strange:

I thought on MBSeq3 that when you used the encoder to change the LayA / B / C parameter that you still had to press Generate to modify the layer. It seems on MBSeq4 that as soon as I move the encoder under LayA that my layer gets randomised. Is that normal?

David

Posted

I thought on MBSeq3 that when you used the encoder to change the LayA / B / C parameter that you still had to press Generate to modify the layer. It seems on MBSeq4 that as soon as I move the encoder under LayA that my layer gets randomised. Is that normal?

yes, thats intended - I prefer immediate response on value changes.

Best Regards, Thorsten.

Posted (edited)

yes, thats intended - I prefer immediate response on value changes.

Best Regards, Thorsten.

Hi TK, thanks for clarifying. Would it be possible to request a switch to control this behaviour?

It also raises another question from me: what is the purpose of the generate button in this case?

Cheers

David

Edited by monokinetic
Posted

I've to think about a generic solution for providing "poweruser flags" without much overhead first ;)

The generate button allows to randomize again with exactly the same parameter settings.

Best Regards, Thorsten.

Posted

I've to think about a generic solution for providing "poweruser flags" without much overhead first ;)

The generate button allows to randomize again with exactly the same parameter settings.

Best Regards, Thorsten.

how would it be, by pressing "select" in random-page-mode, to be able to change the layers?...for now we have LayA-C and TrgA-C.

pressing and holding select, change the Layer&Trg from A-to H when available with the underlying encoders...?

  • 3 weeks later...
Posted

Beta43 is available now:


IDIboxSEQ V4.0beta43
~~~~~~~~~~~~~~~~~~~~~

o delayed mute/unmute which is synched to the measure.
It can be enabled individually in the global options menu (UTILITY->OPT)

o trigger assignment page: trigger layer of the current assignment automatically
selected (alternative way to select a trigger layer)

o random page: random values can now be generated for all parameter and trigger
layers (resp. drums: parameter and drum instruments).
The selection is scrollable with a GP encoder.

o random page: special behaviour for drum tracks: parameter layers and each
individual drum instrument can be randomized (no triggers in this mode
to keep the code simple)

o random page: new values will only be generated when "Generate" button is
pressed, or for a single layer if the GP button under the appr. layer is
pressed.
[/code]

Additional informations:

This package contains two precompiled binaries: one for MBHP_CORE_STM32, another for the MBHP_CORE_LPC17 module

Gridracer: the reported copy issue should be fixed

Best Regards, Thorsten.

Posted

I'm using beta42 and noticed a UI issue related to the transpose feature.

Select Menu -> Transpose.

When the >Octave:+0< is selected I can rotate the data wheel clockwise to increase the octave value but I cannot rotate the wheel counter-clockwise below 0. Also, if I use the data wheel to increase the value past +7 the value jumps to a -octave value. The same behavior happens when using the up and down arrow buttons.

I noticed similar strange behavior when modifying the semitone values. In this case, I couldn't use the data wheel or arrow keys to select a -1 semitone value. If I selected the -1 semitone value using the general button and then turned the data wheel clockwise the semitone value would change from -1 to -2... odd.

Posted

I'm using beta42 and noticed a UI issue related to the transpose feature.

thanks for reporting this! It always worked this (wrong) way, and it seems that nobody noticed this before. ;)

Fixed - the change will be available with beta44

Best Regards, Thorsten.

Posted

Hi there,

I had to recognize that i am not able to open sessions anymore with BETA43:

If I try to open a session I always get the message "Changed to session xxx"

xxx is the session that was opend last under BETA42 regardles what session i try to open.

So I am stuck with the Session i last worked on with BETA42.

When I change back to Beta 42 everything works well again.

Best Regards

Gridracer

Posted

Thank you!

This mistake happened during code changes that I made to share parts of the MBSEQ firmware with future MIOS32 applications.


MIDIboxSEQ V4.0beta44
~~~~~~~~~~~~~~~~~~~~~

o session can be selected again
[/code]

Best Regards, Thorsten.

Posted

Hi, I noticed some minor issues:

1.

The LOOP function seems to mess up the PROGRESSION:

FORWARD -> everything OK

BACKWARD -> only OK if tracklength <17, otherwise only step 1-16 will be played when LOOP is activated regardless of Tracklength (direction OK it plays backwards)

PINGPONG+PENDULUM -> does not work at all with LOOP only step 1-16 will be played when LOOP is activated regardless of Tracklengt (even direction is wrong, it plays forward)

RANDOM directions -> only OK if tracklength <17, otherwise only step 1-16 will be played when LOOP is activated regardless of Tracklength (direction OK it plays randomly)

2.

Regarding the EDIT screen on a CHORD Track:

The PARAMETER LAYER "CHORD" displays the Chords as expected, but the PARAMETER LAYER "VELOCITY" displays a note Value instead:

Example: Layer CHORD shows "G/3" on a step, switching to VELOCITY "C-3" is displayed

Not sure if this is Intension? Are those the root notes of the chords?

3.

The SCRUB function causes the stopped sequencer to start, this makes it impossible to use this Function to scroll through ones own track to find this damn one wrong note for example ;-)

Best regards

Gridracer

Posted (edited)

Hi, I noticed some minor issues:

1.

The LOOP function seems to mess up the PROGRESSION:

FORWARD -> everything OK

BACKWARD -> only OK if tracklength <17, otherwise only step 1-16 will be played when LOOP is activated regardless of Tracklength (direction OK it plays backwards)

PINGPONG+PENDULUM -> does not work at all with LOOP only step 1-16 will be played when LOOP is activated regardless of Tracklengt (even direction is wrong, it plays forward)

RANDOM directions -> only OK if tracklength <17, otherwise only step 1-16 will be played when LOOP is activated regardless of Tracklength (direction OK it plays randomly)

sounds like "sync to measure" is activated...i still have v43 installed, so can´t proove...

edit: or maybe "global loop" is on?

Edited by nuke
Posted

Thanks for the hint nuke,

sounds like "sync to measure" is activated...i still have v43 installed, so can´t proove...

but sync is not activated.

edit: or maybe "global loop" is on?

I am talking about the global loop function which is not working propperly with other progressions than forward.

I use the mode "Selected Track/Step View"

Best regards

Gridracer

Posted

1.

The LOOP function seems to mess up the PROGRESSION:

FORWARD -> everything OK

BACKWARD -> only OK if tracklength <17, otherwise only step 1-16 will be played when LOOP is activated regardless of Tracklength (direction OK it plays backwards)

PINGPONG+PENDULUM -> does not work at all with LOOP only step 1-16 will be played when LOOP is activated regardless of Tracklengt (even direction is wrong, it plays forward)

RANDOM directions -> only OK if tracklength <17, otherwise only step 1-16 will be played when LOOP is activated regardless of Tracklength (direction OK it plays randomly)

I already see a lot of exceptions in the code to get this properly working %-)

2.

Regarding the EDIT screen on a CHORD Track:

The PARAMETER LAYER "CHORD" displays the Chords as expected, but the PARAMETER LAYER "VELOCITY" displays a note Value instead:

Example: Layer CHORD shows "G/3" on a step, switching to VELOCITY "C-3" is displayed

Not sure if this is Intension? Are those the root notes of the chords?

will be fixed

3.

The SCRUB function causes the stopped sequencer to start, this makes it impossible to use this Function to scroll through ones own track to find this damn one wrong note for example ;-)

will be changed (starting the sequencer was my intention, but your argument not to do this is stronger)

Short note from my side: the progress is currently not displayed if a new session is created, or an existing session is copied.

It looks like the sequencer hangs up - but in fact only the LCD won't be serviced until the operation is finished.

This should be fixed in the next release.

Best Regards, Thorsten.

Posted

Beta 45 is available:


MIDIboxSEQ V4.0beta45
~~~~~~~~~~~~~~~~~~~~~

o USB update for MacOS 10.7 Lion

o Chords displayed correctly when a velocity layer is edited
[/code]

The scrub enhancement will need some work, as you won't notice any effect on the display if the sequencer is not running.

The loop improvements for different track modes needs even more work.

And session copy/create progress still not displayed

Best Regards, Thorsten.

Posted (edited)

Very nice!

Just as another feature idea - what do you think about a "complete track clone" feature?

It could be activated by pressing any available hotkey + "Copy" - it will copy all of the active tracks´ notes, length, mode and associated settings to a buffer (e.g. on SD for main memory conservation?)

This buffer, including full track information could than be pasted on to any other tracks, when the same hotkey + "Paste" is pressed. It would be nice, if the paste would also work on a group of active tracks, e.g. tracks 5-8 in group 2.

(I am aware, that the note copy/paste features are available, but cloning would affect "everything").

Personally, I would use it a lot, as I find myself quite often adjusting multiple tracks to use the same midi-out and channel, have force-to-scale on, use the same pattern type and length, fx, etc.

Also, having a prepared fallback "on disk" buffer would allow for longer and more risky track editing, involving many steps not only on the notes, but also on track parameters - if the editing fails miserably, you can "paste" and try again.

Greets and thanks!

Peter

Edited by Hawkeye
Posted

Finally an useful feature which was easy to implement. :)


MIDIboxSEQ V4.0beta46
~~~~~~~~~~~~~~~~~~~~~

o added multi copy function: press MENU+COPY to export all selected tracks
into the /PRESETS folder on SD Card. The resulting files are named
COPY[track].V4T (the [track] numbers are independent from the actual track
position)

o added multi paste function: press MENU+PASTE to import "COPY[track].V4T"
files from the /PRESETS folder into all selected tracks, starting from
COPY1.V4T to (up to) COPY16.V4T

o use these new functions to quickly copy&paste tracks between patterns
and sessions, to create temporary backups, to duplicate tracks, or whatever.

Note that you can also import these COPY[track].V4T files from the
MENU->Event->PRESETS menu, e.g. to exclude certain parts of the preset

o buttons of Wilba's Frontpanel are now debounced
[/code]

Best Regards, Thorsten.

/edit: re-released at 21:15 since only the visible track was exported into the COPY*.V4T files

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...