TK.

Ok, in this case we've to store the set of brightness levels in each EVENT item. This will increase the memory consumption, however I will add this tomorrow... Best Regards, Thorsten.

Ok, I see... The solution in V1.019 only allows to change the hue individually for each LED. But in V1.020 a new "unicolour" mode will be available which allows to set the levels for all LEDs of a matrix via Meta events. You can already try out the changes by compiling the latest version in the repository. The "unicolr.ngc" file shows, how you can control the levels with CC#16, #17 and #18 by using the SetColour[RGB]FromValue event: http://svnmios.midibox.org/filedetails.php?repname=svn.mios32&path=%2Ftrunk%2Fapps%2Fcontrollers%2Fmidibox_ng_v1%2Fcfg%2Ftests%2Funicolr.ngc It's nice for finding out the best matching levels. And the "unicolr2.ngc" file demonstrates your usecase: http://svnmios.midibox.org/filedetails.php?repname=svn.mios32&path=%2Ftrunk%2Fapps%2Fcontrollers%2Fmidibox_ng_v1%2Fcfg%2Ftests%2Funicolr2.ngc Each bank change will call the meta events SetColour[RGB] to set the brightness levels Note that with this solution, it's possible to use different hues for the different matrices! :smile: Best Regards, Thorsten.

You are right - I fixed this in the repository (/edit: and now also in the prebuilt v1.019 binary) Note that EVENT_LED_MATRIX automatically provides access to each pin with cc=16, 17, 18, ..., not only a single pin. Accordingly, the colours should be assigned to the same cc (set cc=16 for all three EVENT_LED_MATRIX commands) If still individual control is desired, use different channels for each colour - or enhance the maps. Best Regards, Thorsten.

Sounds like a loose contact or bad solder joint between PIC and the MIDI OUT. Best Regards, Thorsten.

If you are using the original core, then there is a high probability (to be exact: 50%) to swap the MIDI OUT pins! Could it be that MBSID is going to format the banksticks? You've to wait until this initialisation procedure has been finished, it can take several minutes (and if your SIDs are already working: you will hear some bleeps... - and if the LCD is already working: it will show messages). Another question: do you get an upload request message (displayed in MIOS Studio) after power-on? Third question: did you already measure voltages to ensure that there is no power issue with the additional components? Best Regards, Thorsten.

Hm! In this case the EEPROM definitely won't help... Best Regards, Thorsten.

You've to measure the voltage on the displayed pin after the message has been print. If you measure the wrong voltage, you know that there is an interconnection error (e.g. broken cable, bad solder joint, etc...) It's the best test to ensure, that the SID is controlled correctly from the PIC. Best Regards, Thorsten.

Everything implemented in V1.019 :) Best Regards, Thorsten.

Hi Lars, my assumption was, that by using some clever min/max values it would be possible to shift the middle position into the 7bit based deadband which sends pitchbend=0 so that no additional measure is required. It sounds like the variance is much higher than expected (I don't have such a variance with my own Pitchwheel)! Before implementing the wrong counter measure (if any required), could you please enable debug mode ("set debug on"), let the pitchwheel spring to the middle position and write down the resulting 12bit value. Please do this test at least 10 times (movement from both sides) - I would like to know all measurement results. This test doesn't make sense before the min/max values have been specified with the AINSER command, because they will change the scaling. Sidenote: this test requires an USB connection to display the debug messages! V1.019 is available: MIDIbox NG V1.019 ~~~~~~~~~~~~~~~~~ o added "Autoload" function: whenever a .NGC file has been uploaded via the MIOS Filebrowser, it will be automatically loaded, so that "load <file>" doesn't need to be entered in the MIOS Terminal anymore. o EVENT_LED_MATRIX: now expects the 'colour' parameter to address the red (colour=0), green (colour=1) or blue (colour=2) LED. Note that it's still possible to control the LEDs from a single incoming MIDI message, e.g. by assigning the 3 EVENT_LED_MATRIX entries to the same CC number. But now it's also possible to address the LEDs independent from each other. o LEDs can now be dimmed with 16 brightness levels over the value range. This feature has to be enabled with dimmed=1 in the EVENT_* definition. Configuration example: cfg/tests/dimled.ngc o LEDs in a matrix configuration can be dimmed as well. The dim range is the same (0..15), but the effective dim level is limited by the number of scanned rows: - 4 rows: only 8 levels (0..1, 2..3, 4..5, 6..7, 8..9, 10..11, 12..13, 14..15) - 8 rows: only 4 levels (0..3, 4..7, 8..11, 12..15) - 16 rows: only 2 levels (0..7, 8..15) Configuration example: cfg/tests/dimled_m.ngc o the new MIOS Terminal command "lcd" allows to directly output a (formatted) string to the LCD(s). E.g. try "lcd @(1:1:1)Hello World!" Note that terminal strings can also be sent via SysEx from a DAW: F0 00 00 7E 32 00 0D 00 <ascii-text> 0A F7 Accordingly, this SysEx string will print "Hello World!" as well: F0 00 00 7E 32 00 0D 00 6C 63 64 20 40 28 31 3A 31 3A 31 29 48 65 6C 6C 6F 20 57 6F 72 6C 64 21 0A F7 Best Regards, Thorsten.

Congratulations for the successful upgrade! :) Best Regards, Thorsten.

Since you are using a customized layout: could it be that the two pins of the MIDI OUT socket are swapped? Best Regards, Thorsten.

As far as I remember, the GM5 driver will only work with VID 0x16c0 PID 1022 (MIDIbox) or with Ploytec VID (0x0a4a) and any PID Best Regards, Thorsten.

I must say that I never tried more than 1 GM5x5x5 on a Windows PC... If there is a driver issue, only Ploytec can help. Or somebody, who tried more than 2 GM5x5x5 as well and just can confirm, that it works at his side (resp. what he did to get this working) Best Regards, Thorsten.

The most important step has been done: I added MIOS32_SRIO_NUM_DOUT_PAGES to MIOS32_SRIO. And I overworked the MBNG DIN/DOUT matrix handler, so that it stores the selection and output patterns in these pages. As a result, the RAM consumption is a bit lower (no "double-buffering" anymore), but more important: the CPU doesn't need to copy the next patterns into the DOUT registers with each update cycle anymore. So: a big performance benefit + dimmed LEDs for free! ;-) Best Regards, Thorsten.

I will work on this tomorrow. :smile: But to give you already some informations: - after the KB driver has been integrated into MBNG, the scan rate is ca. 4 kHz (250 uS), and not only 1 kHz (1 mS) anymore - I gave up the idea to support a second SRIO for various reasons - MBNG won't support a matrix where RGB LEDs are connected this way! Each color needs a dedicated SR (BLM_SCALAR design). Alternative scramblings would make the configuration interface too complicated. I also want to prevent, that somebody starts a PCB based on such a wiring and than forces me to support it for other MIDIbox firmwares as well, although I can't test it -> reduce complexity! - I will start with 16 brightness levels for LEDs directly connected to DOUT pins (w/o matrix) based on the MIOS32_SRIO enhancement that you proposed some time ago (actually the paging can also be used to offload the matrix drivers by storing the patterns directly into the pages) - next step would be to support this also for matrices with 3 levels only Best Regards, Thorsten.

Hi, the link in the Wiki pointed to an older version which doesn't support the PIC18F4685 properly. I corrected the link: http://www.midibox.org/dokuwiki/windows_toolchain_quickstart Please try it again with this version. Best Regards, Thorsten.

I would add a label statement for each button which prints the bank number. Please let me know if this doesn't work in conjunction with radiogroups. Best Regards, Thorsten.

Hi, since the LCD outputs nothing else (even no black bar in the first line?) it's probably related to the contrast: move the contrast trimpot into the Vs direction Sound: there are two troubleshooting applications which allow to debug this: -> http://www.ucapps.de/mios_download.html mbsid_interconnection_test_v2.zip Testprogram for the interconnections between Core and SID module of MIDIbox SID (read the README.txt file for the usage) mbsid_testtone_v3c.zip Just plays a 1kHz triangle wave on the SID - nice for testing the audio output without the need for a MIDI sequencer or keyboard. With the new v2 version, both MBHP_SID modules connected to a core are initialised. In addition, a 1kHz square wave will be generated on the CS pin (#8) Best Regards, Thorsten.

Welcome to the forum! :smile: If the keyboard already has a scan controller which can be re-used, there is no really need for going the hard route. You can connect the MIDI OUT of the keyboard to one of the MBNG based MIDI INs, and route it to USB with a ROUTER node. Best Regards, Thorsten.

Just to be clear, the connections to J19 are: J1:1 -> J19:Vs J1:2 -> J19:Vd J1:3 -> J19:RC1 J1:4 -> J19:SO J1:5 -> J19:SC In addition, the AOUT_NG interface has to be selected in the CV Configuration menu (by default, the AOUT module is configured) Best Regards, Thorsten.

J8 open: Ploytec VID selected J8 closed: MIDIbox VID selected All other jumpers untouched. Can you confirm this? Best Regards, Thorsten.

Ok, I see. Forget this option, it won't work properly without a lot of additional work at different places (not only seq_ui.c)... :-/ Best Regards, Thorsten.

With V1.018 the "SCS lcd_pos=<device>:<x>:<y>" command is available. With: SCS lcd_pos=1:1:7 the SCS page should be displayed at the two lower lines. The menu isn't always visible, only after pushing the MENU button. The reason is, that in the so called "main page" you could use this space for other meaningful output, such as the selected bank number. Best Regards, Thorsten.

V1.018 is available: MIDIbox NG V1.018 ~~~~~~~~~~~~~~~~~ o added EVENT_KB. See cfg/tests/kb_*.ngc for usage examples o new LCD formats: %N prints the note name based on the specified MIDI event, and %n based on the event value o new meta event: SwapValues - it swaps the primary and secondary value of an event. E.g. on a note event, velocity can be swapped with the key value. A usage example can be found in cfg/tests/kb_4.ngc, where the key and velocity value should be output by separate CV channels o AIN/AINSER configuration: added pinrange parameter. It allows to specify individual ranges for the connected pots to calibrate the physical min/max limits. o SCS configuration: added SCS num_items and lcd_pos parameters Please note: after the installation of this update some WARNINGs about non-available KEYBOARD parameters could be print. These parameters (ain_* and midi_*) are obsolete, please remove them from your .NGC file To users who are already testing the keyboard functions: this is a new configuration example: http://svnmios.midibox.org/filedetails.php?repname=svn.mios32&path=%2Ftrunk%2Fapps%2Fcontrollers%2Fmidibox_ng_v1%2Fcfg%2Ftests%2Fkb_1.ngc Best Regards, Thorsten.

Beware: without the USB connection you won't get debugging messages in MIOS Terminal, it could be very difficult troubleshoot a non-working configuration. This issue could be related to the electrical installations in your flat/house. E.g. a problem with earth. Here for example some postings from somebody who has a similar issue: http://www.eevblog.com/forum/beginners/connecting-system-ground-to-mains-earth/ And just to confirm: I tested this with my own Mac PSU today and haven't measured a voltage between earth and USB ground of my MBP. I would still go for a PSU with integrated mains connector - such a PSU can be built into your metal case without the need for connecting the metal of the case to ground. Btw.: you will face the same issue once you are connecting a MIDI cable from your Mac to the core module, because the cable shield is connected to ground (at both sides) as well. Best Regards, Thorsten.