MIDIbox SEQ V4 Release + Feedback

[TK.]

Interesting: I've a similar issue with my menu button!

It could make sense to add a debouncing mechanism here...

Best Regards, Thorsten.

[Echopraxia]

How can you make a program change when switching patterns? I have some patterns that sound good on one patch and other patterns that sound good on another patch and want to flip in between the two patterns with two different patches.

[stuartm]

I don't think there is a quick qay you can do this by just changing the pattern.

Program Changes can be done via the Mixer.

I use the mixer maps in song markers for that to switch between whole tracks.

I.e:

Song 1 - Marker A1: Dump Mixer Map 1 -> Marker A2: Change to Patterns ...

Song 2 - Marker A2: Dump Mixer Map 2 etc.

[Echopraxia]

I don't think there is a quick qay you can do this by just changing the pattern.

Program Changes can be done via the Mixer.

I use the mixer maps in song markers for that to switch between whole tracks.

I.e:

Song 1 - Marker A1: Dump Mixer Map 1 -> Marker A2: Change to Patterns ...

Song 2 - Marker A2: Dump Mixer Map 2 etc.

Thanks for the tip. I have not even gotten in to song mode yet!

On another subject I was trying to change a tracks event type from 256 step to only 64 to get the extra parameter layers And was only somewhat successful importing the track. Here is what I was trying to do.

1: copy pattern

2: on new pattern, change event type to 64 notes, and recreate layer cc settings from first pattern (no-preset made, didn't bother because the previous pattern would save it as 256 steps)

3: save new pattern

4: go back to first pattern

5: Try to copy parameter settings of cc layers from old track and paste to the new 64 step pattern.

When I copy the cc, roll2 etc, layers in to the new pattern they get pasted with a random value even though I made precise adjustments to each step in each layer on the pattern I was copying from. Am I making sense?

I am stating to get into the whole layer concept adn would like to know of how to copy, for example, parameter layer d cc#3 and its values for each step into a completely new pattern even though the new pattern will be a different length and type.

Regards

[stuartm]

Did you notice the right LCD which says "Please initialize the track" and the bliking "INIT" above GP16 ?

Thats what I get when I change the length.

[Echopraxia]

Did you notice the right LCD which says "Please initialize the track" and the bliking "INIT" above GP16 ?

Thats what I get when I change the length.

Right that's normal. When you do this it erases everything on the track and changes the length and parameter allocating but does not save any pattern and layers within it unless you use save it as a preset but presets don't matter in this case because if you load a preset it loads the 256 note length or whatever it was set to .

It is difficult to get a 256 length pattern down to 128,64 etc and keep all the settings within the original pattern.

[gomiboy99]

Hi All,

I've been playing with beta35, I have noticed that the scroll function corrupts the tracks that I scroll through when there are >=17 steps. I find that instead of just moving to the next 16 steps, all steps are moved by 2 or 3 steps. Also scrolling only goes in one direction and all 256 steps must be scrolled through before getting back to step 1. I am not sure if this is expected behaviour, or a hardware problem (or worse a lack of understanding on my part). UPDATE, I have tried beta33 and the scroll function does the same thing. I have also checked the hardware thoroughly and found no problems with the construction.

Is there a simple answer? I cannot find a solution in this thread or the user manual. Any help would be appreciated.

OK, dumb question, RTFM is the answer, as always it was hidden in plain sight.

After going round in circles try to get the SD card to work reliably, I found that Sandisk cards work fine 100% of the time, maybe this will help others avoid much frustration.

Cheers

Tim.

Edited March 13, 2011 by gomiboy99

[TK.]

Right that's normal. When you do this it erases everything on the track and changes the length and parameter allocating but does not save any pattern and layers within it unless you use save it as a preset but presets don't matter in this case because if you load a preset it loads the 256 note length or whatever it was set to .

It is difficult to get a 256 length pattern down to 128,64 etc and keep all the settings within the original pattern.

Also software-wise it's difficult to do this, some clever copy routines are required to change the partitioning properly under all circumstances.

It's the same issue like with the copy function that you reported before.

There is currently no solution.

OK, dumb question, RTFM is the answer, as always it was hidden in plain sight.

After going round in circles try to get the SD card to work reliably, I found that Sandisk cards work fine 100% of the time, maybe this will help others avoid much frustration.

Great that you found a solution!

I will highlight in the documentation that Sandisk cards should be preferred.

Best Regards, Thorsten.

[gomiboy99]

Great that you found a solution!

I will highlight in the documentation that Sandisk cards should be preferred.

Best Regards, Thorsten.

I tried various different makes before the Sandisk, not sure why the others were not reliable.

Would it be possible to modify the operation of the metronome to make it independent of G1T1?

It would be nice to specify time signature based solely on the BPM, e.g. 3/4 4/4 etc. As there is a dedicated button for the metronome already, it would make it easier to turn on and off rather than making a track into a metronome and having to find the mute when it's not needed.

Thanks

Tim

[TK.]

I tried various different makes before the Sandisk, not sure why the others were not reliable.

I tried different brands as well and never noticed unreliable operation, only different performance factors.

Would it be possible to send me the non-working SD Cards for further analysis?

Would it be possible to modify the operation of the metronome to make it independent of G1T1?

It would be nice to specify time signature based solely on the BPM, e.g. 3/4 4/4 etc. As there is a dedicated button for the metronome already, it would make it easier to turn on and off rather than making a track into a metronome and having to find the mute when it's not needed.

The metronome already works independent from any track, if you noticed a dependency on G1T1 you found an unexpected bug!

The signature is defined as "steps-per-measure", which has to be set in the options page (UTILS->OPT) - it even allows you to define unequal steps (free definable from 1-256)

For 3/4 just select 12 (1/16) steps.

Best Regards, Thorsten.

[gomiboy99]

TK,

I would love to send you the SD cards but I have reused them elsewhere, the worst offenders were 2 Fuji brand cards, these would work once and then fail from then on, they work fine in my camera.

The others were unbranded and I have given them away, AFAIK they are working in their new application, sorry.

The metronome problem is interesting as I only noticed this behaviour since Beta33. I wonder if something has gone wrong in the update process. I will try reloading the firmware. I will also try a fresh session in case that is the problem.

Thanks for your help.

Tim.

Edited March 14, 2011 by gomiboy99

[TK.]

Beta36 is available now:

MIDIboxSEQ V4.0beta36

~~~~~~~~~~~~~~~~~~~~~


   o The metronome function considers the "steps per measure" correctly now


   o received MIDI song position command sets the track positions correctly

     now - this is important when MBSEQ is controlled by a clock master.

     E.g. Logic Studio sends startpoint-1 when the sequencer continues

     in a loop

[/code]



The metronome problem is interesting as I only noticed this behaviour since Beta33. I wonder if something has gone wrong in the update process. I will try reloading the firmware. I will also try a fresh session in case that is the problem.



The changes in beta36 are not really related to your observations (I haven't found any dependency on G1T1), but while checking the code I found that steps per measure wasn't used by the metronome as intended, especially on "unequal steps".


Also the song position handling of previous versions could lead to a misaligned step counting if MBSEQ is clocked by an external MIDI clock master.



Init/Copy issues reported by Echopraxia: nothing done yet, still needs some more thinking about all possible cases.

A fix for the copy function is difficult, but feasible.

The INIT function can't be realized without clearing all track values (e.g. otherwise layers would play wrong events if their mode is changing, e.g. from CC to Note - very bad while playing live!)

Maybe I will implement a new repartitioning function, I would find it useful as well


Best Regards, Thorsten.

[gomiboy99]

Thanks TK, this has solved the issue for me.

Regards

Tim.

[Echopraxia]

Just messing with the loop feature :)

Do you think we could get one, two or several additional loop points so we can trigger between them live? I found so many useful loops in my tracks that this would make MB life even sweeter!

Example

loop point 1, offset 4 steps 8

loop point 2, offset 8 steps 4

loop point 3, offset 24 steps 12

and so on....

I am still trying to figure out what is the best way to enable them because as of now we use off/on for loop 1.

Maybe use loop encoder to switch from "off-1-2-3-4-5-6-7-8-9-10" :) and still have the loop button still turn loop on/off. That way you can play your full track, then turn loop on and already be at loop point 8. Then be able to turn off loop to go back to the full track then turn loop encoder to go to loop point 3 and then turn on etc. Or set your loop points and start turning that encoder for crazy live predetermined mashups.

To go even further with the idea, If someone wants to install 16 individual track select buttons like I mentioned previously and set loop to "selected tracks" you could mash up for example only the drum tracks and do all this in record time.

Wada ya think?

Regards

PS can I get my transpose to default to "octave" first instead of "semitone"? I know you changed it in the past but wanted everybody's opinion as I am finding that I use the octave setting a LOT more even though I was the one to suggest semitones be the default :whistle: which was to move drum kits up and down precisely. Are people using semitones this way or find that they use full octave changes more?

Edited March 17, 2011 by Echopraxia

[grizz]

Me and my bandmate are going through our songs and trying to reorganize tracks and patterns on each of our V4 sequencers, as we did things kinda willy-nilly while writing.

I'm wondering if there is a way to load the session(s) we're reorganizing onto our laptops, reorganize them and then reload them onto the sequencers? If there isn't already a way, does anyone have any thoughts on the viability of this idea?

thanks,

grizz

[TK.]

Me and my bandmate are going through our songs and trying to reorganize tracks and patterns on each of our V4 sequencers, as we did things kinda willy-nilly while writing.

I'm wondering if there is a way to load the session(s) we're reorganizing onto our laptops, reorganize them and then reload them onto the sequencers? If there isn't already a way, does anyone have any thoughts on the viability of this idea?

It's viable, but creating a useful GUI will be a huge task.

You could already re-organize your sessions by using the session import/export function under Util->Disk

It allows to copy patterns from other sessions to create a new session (or vice versa: copy patterns of the current session into another session)

Once this has been done, just copy the appr. session directory (/SESSIONS/<session-name>) to the SD Card of the other MBSEQ V4

Best Regards, Thorsten.

[TK.]

Just messing with the loop feature :)

Do you think we could get one, two or several additional loop points so we can trigger between them live? I found so many useful loops in my tracks that this would make MB life even sweeter!

Using multiple looppoints is the wrong way, as you would have to define them again and again whenever you created a new pattern.

They would also consume some (expensive) pattern memory... although they are never used at the same moment.

Therefore I added a new "loop preset" feature into Beta37 (see ChangeLog below)

PS can I get my transpose to default to "octave" first instead of "semitone"? I know you changed it in the past but wanted everybody's opinion as I am finding that I use the octave setting a LOT more even though I was the one to suggest semitones be the default :whistle: which was to move drum kits up and down precisely. Are people using semitones this way or find that they use full octave changes more?

Well, as I mentioned earlier, I'm using octave changes more often as well.

In order to move drum kit instrument, I set the semitone individually for each instrument under Track->Event

So: I changed it back to octave on page entry in Beta37

Best Regards, Thorsten.

[TK.]

Beta37 is available - from the ChangeLog:

MIDIboxSEQ V4.0beta37

~~~~~~~~~~~~~~~~~~~~~


   o Length Page: quick selection now also available for loops.


   o Length Page: length/loop presets are now stored in MBSEQ_C.V4 file (located

     in session directory) under the name "QuickSelLength", "QuickSelLoopLength" and

     "QuickSelLoopLoop".

     They could be edited with a text editor if you would like to have different values.    


   o track transpose page now always starts with "octave" item instead

     of "semitones" item


   o it's now possible to define 16 individual "direct track selection" buttons

     in MBSEQ_HW.V4 (search for BUTTON_DIRECT_TRACK*)


   o it's now possible to define 2 shift registers for outputing the current track

     selection in MBSEQ_HW.V4 (search for TRACKS_DOUT_L_SR and TRACKS_DOUT_R_SR)


   o copy function works correctly now if a track is pasted into another track

     with different partitioning in Paste/Clr = Steps mode (selectable in Utils->Options page)

[/code]


Best Regards, Thorsten.

[Gridracer]

Hi there, i did not notice anything unexpected since quite a while seems it's getting closer to "final" release ;-)

But yesterday i recognized that something little is wrong with the "move" function:

As long as tracklength is </=16 everything is ok, but when you move a step from one stepview to another something funny happens.

If you move the step forward a new step is created on the first step of the new step view e.g. 17 and the "cursor" position jumps to the eighth

step position of the new step view. If you move the step backward a new step is created on the last step of the new step view e.g. 16, the "cursor" position does not jump.

Hope i could explain it properly.

Best regards

Gridracer

[TK.]

Thanks for feedback, this bug was obvious!

Fixed in Beta38 - as a bonus, the performance has been improved. :)

MIDIboxSEQ V4.0beta38

~~~~~~~~~~~~~~~~~~~~~


   o CPU load reduced by more than 20%, this especially results into faster SD Card operations!


   o fixed move step function for moves which select the next step view

[/code]


Best Regards, Thorsten.

[vcfool]

Just upgraded from Beta35 to 38.

I am using Wilba's CS and found that BUTTON_PAR_LAYER_A is not working for me.

I can confirm that it is not a hardware problem: MIOS Studio detects that the button has been pressed/depressed but says that it's not mapped.

I went back to Beta35 and the button works fine again.

Thanks!

:)

[nuke]

i have exactly the same issue...

BUTTON_PAR_LAYER_A is assigned but not working...MiosStudio is telling pressed/depressed...

[gomiboy99]

Beta38 works fine with my set-up. The control surface is my own PCB with separate DIN DOUT modules to replicate Wilba's 8x8 matrix set-up. Just to be sure I tried Beta35 and there is no difference.

I have had to modify the Wilba MBSEQ_HW.V4 file to reassign buttons and LEDS to suit the control surface design (for easier routing on a single sided board).

Is there an error in the HW.V4 file that comes with Beta38?

Tim.

[TK.]

:ike:

MIDIboxSEQ V4.0beta39

~~~~~~~~~~~~~~~~~~~~~


   o BUTTON_PAR_LAYER_A working again (was overwritten by BUTTON_DIRECT_TRACK16)

[/code]


next bug, please! ;)


Best Regards, Thorsten.

[vcfool]

Great! Thanks!

:)