Jump to content

MIDIbox SEQ V4 Release + Feedback


TK.
 Share

Recommended Posts

Found it, fixed it!!! :)


MIDIboxSEQ V4.066
~~~~~~~~~~~~~~~~~

o fixed potential stack overflow in LPC17 build
[/code]

Background: on heavy USB traffic, the USB stack consumes more stack memory than expected (because it can be interrupted by high-prio handlers)

Actually this bug could happen independent from the configuration in the past... Lysander enabled some MIDI router connections, and this increased the probability of the occurrence.

The bug only existed in the LPC17 build due to the special RAM layout which requires some data section allocation tweaks, with the result that the stack area is closer to the critical application data than in the STM32 build.

I solved this by moving a huge block (mixer values) into a different RAM area.

A firmware update is strongly recommended for all MBSEQs which are using the MBHP_CORE_LPC17, it isn't required for MBSEQs which are using the MBHP_CORE_STM32!

Best Regards, Thorsten.

Link to comment
Share on other sites

  • 2 weeks later...

Hi there i noticed a bug (at least I guess it is a bug) in the "Track Event Configuration" page:

I have a new and empty track (no steps are set, sequencer is playing)

I change the value for a parameter layer from "note" to something else,

when i scroll throug "pitchbend" "controller" and "program change" the corresponding

events are send to midi out, which does some some unwanted changes to external gear under some

circumstances. (The events are only send the first time when scrolled throgh the parameters)

Here the record from midi Out:

[50897.312] b0 01 40 Chn# 1 CC# 1 = 64

[50905.079] e0 00 40 Chn# 1 Pitchbend 0

[50944.773] c0 00 Chn# 1 Program Change 0

I think these events should only be send when the track is filled with the corresponding data.

So that was my first remark, here the second:

My mixer maps are displayed as "unnamed" did I just miss where to give them a name??

And my final Proposal:

Wouldn't it be useful if one could also delete sessions or is this an unwanted feature and intensionally only possible

via usb connection?

Greetings

Gridracer

Link to comment
Share on other sites

Hi,

I think these events should only be send when the track is filled with the corresponding data.

confirmed!

The events are sent because they are not gated in Note tracks, only in CC tracks (that's intended)

The actual issue is, that they are sent while you are changing the configuration, and this isn't so nice.

A possible solution: confirming the new configuration by pressing the buttom below the item, similar to the CC value (that you requested some time ago).

I think that this will be the best solution.

My mixer maps are displayed as "unnamed" did I just miss where to give them a name??

No, it's still not possible to change the name of a mixer map from the control surface... ;)

Wouldn't it be useful if one could also delete sessions or is this an unwanted feature and intensionally only possible

via usb connection?

I'm unsure if this is really a wanted feature.

From my point of view, removing sessions is a very sensible topic.

Just consider that a kid is playing with your MBSEQ and removes all your sessions, because it's so easy and it's fun to clean up everything! ;)

Best Regards, Thorsten.

Link to comment
Share on other sites

HI,

Just consider that a kid is playing with your MBSEQ and removes all your sessions, because it's so easy and it's fun to clean up everything! ;)

I agree performing a factory reset is also lots of fun... :-)

Sorry, User failure :angel: everything working fine

But I noticed a new BUG with Bootloader 1009 and Firmware V067:

Record mode is no longer working, incoming notes from midi in are forwarded correctly but no longer recorded...

Best Regards

Gridracer

Edited by Gridracer
Link to comment
Share on other sites

From my point of view, removing sessions is a very sensible topic.

Just consider that a kid is playing with your MBSEQ and removes all your sessions, because it's so easy and it's fun to clean up everything! ;)

Possible solutions:

1. Lock mbseq in a room which is inaccessible to kid.

2. Remove memory card after each recording session and put in locked box or secret hiding spot.

3. Make the delete session process deliberate so that a kid can't just randomly push buttons and accidentally delete something. For example, the user would need to push a certain button sequence or spell out the word "yes" to confirm deletion

4. Require a password entry in order to delete sessions.

Link to comment
Share on other sites

I will reconsider my opinion ;)

We've now a new version which fixes the USB MIDI issues under Windows, and which also allows to use the brand new MIOS Filebrowser which is part of MIOS Studio 2.4.0! :)


MIDIboxSEQ V4.068
~~~~~~~~~~~~~~~~~

o USB device settings have been changed.
IMPORTANT: please install MIOS32 Bootloader V1.010 before uploading
the application! Otherwise code upload via USB won't work anymore!
(Backup: upload via MIDI IN1/OUT1)

o support for MIOS Filebrowser, which will be part of MIOS Studio 2.4

o support for GM5 driver under Windows.
MacOS and Linux users: just install this update, no additional changes required.
Windows users:
- upload the application with MIOS Studio
- close MIOS Studio
- install the GM5 driver which can be downloaded from http://www.ucapps.de/mios32_download.html
- start MIOS Studio -> 4 "GM5" USB ports should be visible
[/code]

Your other requests are not lost, I'm collecting them before doing the next update.

Best Regards, Thorsten.

Link to comment
Share on other sites

  • 4 weeks later...

A new version is available:

 

MIDIboxSEQ V4.069
~~~~~~~~~~~~~~~~~

   o USB MSD driver: changed endpoint to avoid conflict with MIDI driver.
     Note that it's still not possible to use MSD and MIDI in parallel,
     but Windows and MacOS should accept the connection (again).

   o Event configuration page: the selected layer type now has to be confirmed with 
     GP button below the appr. item.
     A warning message will pop up to notify about this unusual data entry method.
     In addition, the value will be marked with '!' as long as it doesn't match with
     currently active value.

    o it's now possible to delete a session from the main page

 

 

This update will allow you to clean up your SD Cards for the upcoming year! :smile:

 

Best Regards, Thorsten.

Link to comment
Share on other sites

HI!

 

Thanks for new version, but my MBSEQv4 (lpc core) show that last version is 4.068.

I think it is V4.69 (because there is session delete option in main menu), but just checking.

MIOS studio also showing V4.68.

 

Another question: Is there a way to change shortcut MENU+SYSEX to shortcut for Euclidean Rhythm Generator?(for dummies :)

 

  Thanks and Best regards, IGI

Link to comment
Share on other sites

You are right, the displayed version number isn't correct (will be fixed with the next release ;)

 

I don't see a simple way to replace MENU+SYSEX by another page.

Maybe a bookmark would be sufficient?

 

Best Regards, Thorsten.

Link to comment
Share on other sites

Currently this would require a code modification in seq_ui.c, SEQ_UI_LED_Handler_Periodic() function (search for "sequencer_running")

 

If more people are interested, I could provide this as an option.

If you are the only guy who is interested: you know how to build the application by yourself... ;)

 

Best Regards, Thorsten.

Link to comment
Share on other sites

  • 4 weeks later...

As I have a BLM 16x4 integrated into my SEQ I have decided to switch off the GP LEDs. This is easily done in the Config file:

# 0: no mapping of 8x8 LEDs
# 1: enable GP LED -> 8x8 matrix mapping for Wilba's MB-SEQ PCB
BLM8X8_DOUT_GP_MAPPING  0

However, also the GP buttons are not needed as trigger buttons any longer. Would it be possible to make that configurable, too, such that the GP buttons carry (by default) the functionality that they otherwise only have when holding down the MENU button?

 

At the same time, the GP LEDs could then also be used to indicate the current menu (i.e. MIXER, EVENT, MODE,...). That should of course be obvious from the LCDs, but would add a little bling and usefulness for those LEDs.

 

Best regards, ilmenator

Link to comment
Share on other sites

No problem, I can add this to the next version...

 

...if you do me a favor and test your LCDs with a MBHP_CORE_LPC17 core again - it seems that Doug was able to reproduce the same issue with his CLCDs, and we probably found a solution! (More about this topic + firmware update tomorrow)

 

Best Regards, Thorsten.

Link to comment
Share on other sites

Finally V4.070 is available with a some requested features:

 

 

 

MIDIboxSEQ V4.070
~~~~~~~~~~~~~~~~~
 
   o LPC17: robustness of LCD driver has been improved
 
   o the MIDI configuration page got a new subpage called "Ext.Ctrl"
     This subpage allows to assign CCs to functions like Morph Value,
     Scale, Song, Phrase, Patterns, Banks, All Note Off, etc.
     Some of these functions can also be assigned to a Program Change Event.
     The MIDI port and channel over which these functions are accessible is free assignable.
     The MIDI channel is disabled by default, which means that the Ext Ctrl. feature
     has to be explicitly enabled in the MIDI->Ext Ctrl. page!!
 
   o improved transpose handling if loopback track is running with sustain enabled,
     or with notes > step length (glide)
 
   o note stack now automatically cleared if transpose was controlled by a loopback
     track, and then manually from an external keyboard.
 
   o it's now possible to transpose with chords (or multiple notes) from a loopback
     track
 
   o the "sdcard_format" command allows to format the connected SD Card with FAT32
     from the MIOS Terminal
 
   o added BLM_GP_ALWAYS_SELECT_MENU_PAGE option to MBSEQ_V4.HW file
 

 

 

Best Regards, Thorsten.

Link to comment
Share on other sites

  • 2 weeks later...
   o added BLM_GP_ALWAYS_SELECT_MENU_PAGE option to MBSEQ_V4.HW file

 

Thanks!

 

There are a few things that do not work as expected with this mode set to "1": I guess there are a few dependencies that got lost along the way?

 

The simplest example: in EVENT mode, when I try to change from "Note 128..." to e.g. "Note 64..." I am asked to initialize the track. However, I can't do that by turning the last GP encoder, and pressing the GP button number 16 now obviously takes me to the SYS EX mode.

 

I am not sure I have just opened a can of worms here? If this is an "easy" fix and you are willing to keep this mode, can I also suggest to have the respective LED, e.g. the green one, being switched on while in one of the modes directly accessed by the GP buttons, and switched off again when back in EDIT mode?

 

Best, ilmenator

Link to comment
Share on other sites

The simplest example: in EVENT mode, when I try to change from "Note 128..." to e.g. "Note 64..." I am asked to initialize the track. However, I can't do that by turning the last GP encoder, and pressing the GP button number 16 now obviously takes me to the SYS EX mode.

 

The GP buttons are assigned to menu functions only, it isn't possible to set exceptions for certain special functions in certain pages.

Are you able to initialize the track with the 16th button of the upper BLM row?

 

 

 

I am not sure I have just opened a can of worms here?

 

Yes... everything which isn't used by myself (and especially which doesn't match with my own hardware) has a high potential for failures.

 

 

 

can I also suggest to have the respective LED, e.g. the green one, being switched on while in one of the modes directly accessed by the GP buttons, and switched off again when back in EDIT mode?

 

it should already work this way?

Could you please check, under which conditions a page mapped to a menu button isn't displayed by the appr. GP LED?

 

Best Regards, Thorsten.

Link to comment
Share on other sites

The GP buttons are assigned to menu functions only, it isn't possible to set exceptions for certain special functions in certain pages.

Are you able to initialize the track with the 16th button of the upper BLM row?

No, that simply switches the 16th "note" in G1T1 on or off, depending on the previous state. 

 

it should already work this way?

Could you please check, under which conditions a page mapped to a menu button isn't displayed by the appr. GP LED?

 

Ok, it seems that

# 0: no mapping of 8x8 LEDs
# 1: enable GP LED -> 8x8 matrix mapping for Wilba's MB-SEQ PCB
BLM8X8_DOUT_GP_MAPPING  0

simply switched the LEDs completely off - I have set that back to 1 and now they light up as expected :smile: , albeit pulsating. However, then I also get the red LEDs running across when the SEQ is started, something that I wanted to avoid :sad: .

Link to comment
Share on other sites

No, that simply switches the 16th "note" in G1T1 on or off, depending on the previous state. 

Aren't you using the "BLM_BUTTONS_NO_UI 0" function to emulate the GP buttons from your BLM?

Without GP button functions the usage of "BLM_GP_ALWAYS_SELECT_MENU_PAGE 1" doesn't make much sense - there are too many pages which rely on buttons (instead of encoders) - mostly for very good reasons.

 

 

 

However, then I also get the red LEDs running across when the SEQ is started, something that I wanted to avoid

 

:wacko:

Adding just another option really consumes some time at my side (enhancing HW cfg file, integration, testing, documentation, long term support for people who would like to know what all the options are doing, and why a certain conflicting option causes a problem somewhere else...).

Would it be ok for you to do the required code change by yourself when I tell you the places where seq_ui.c has to be modified?

Best Regards, Thorsten.

Link to comment
Share on other sites

Aren't you using the "BLM_BUTTONS_NO_UI 0" function to emulate the GP buttons from your BLM?

Without GP button functions the usage of "BLM_GP_ALWAYS_SELECT_MENU_PAGE 1" doesn't make much sense - there are too many pages which rely on buttons (instead of encoders) - mostly for very good reasons.

 

No, I am not using this - in other words, my setting is

# set this to 1 if these buttons should only control the "step triggers" (gate, and other assigned triggers) - and no UI functions
BLM_BUTTONS_NO_UI   1
because my understanding of workflow would be that the BLM buttons are exclusively meant for triggering, whereas the (former) GP buttons are exclusively meant for menu handling.

And even if I set this parameter to "0", what happens is that I then end up in SYS EX mode, the same as if I was pressing the GP16 button. That actually makes sense to me :smile: .

 

:wacko:

Adding just another option really consumes some time at my side (enhancing HW cfg file, integration, testing, documentation, long term support for people who would like to know what all the options are doing, and why a certain conflicting option causes a problem somewhere else...).

Would it be ok for you to do the required code change by yourself when I tell you the places where seq_ui.c has to be modified?

Yes, I totally understand that!
Link to comment
Share on other sites

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...
 Share

×
×
  • Create New...