TK.

Hast Du ein Link zum Datasheet des LCDs parat? Evtl. muss man da ein wenig tricksen. Bootloader: der Command Handler ist nur in der "mios32_bootloader" application implementiert. Lade sie einfach via MIOS Studio auf, danach sollten die Kommandos funktionieren. Gruss, Thorsten.

As promised I tested the bipolar (RS422-like) MC3487/MC3486 based line driver today, and it works! :smile: At least with a 2m cable (ribbon, not twisted pair), but 5m should be feasible as well (has to be confirmed by somebody else) With a scope I can clearly see the effect of the long transmission line, how the bipolar transmission and especially the termination resistors help, etc. - I've a good feeling! ;-) Schematic: http://www.ucapps.de/mbhp/line_drivers.pdf This way you can transmit the signals for the AOUT* module and DOUT (gate/trigger/switches) over long distances. Best Regards, Thorsten.

Hast Du schonmal versucht, den SysEx string direkt von MIOS Studio aus zu senden (mit dem SysEx Tool)? Damit kannst Du sicherstellen, dass er richtig verarbeitet wird. Falls das klappt, weisst Du, dass Cubase den String offenbar nicht sendet... Gruss, Thorsten.

This conflict only exists for the MIDIbox CV V2 application which accesses the AOUT module from an interrupt service routine, and not from a task where Mutex based synchronisation would be possible. If the application would access the AINSER module as well, it would need to do this from the ISR, and this dramatically reduces the CV update rate. I guess that everybody prefers high update rates instead of the possibility to connect analog inputs to an AINSER module. The good news are, that the analog inputs of STM32F4 are more stable than the ones of STM32F1 and LPC17. Again +1 for using a MBHP_CORE_STM32F4 module! Best Regards, Thorsten.

The AINSER8 or 64 board will neither be supported by LPC17, nor STM32F4 since it conflicts with the AOUT module at J19 which will be accessed by a fast timer interrupt. Therefore: LPC17: up to 6 analog inputs at J5A and B (J5B.A6 and A7 are used for MIDI3) STM32F4: up to 8 analog inputs at J5A and B Best Regards, Thorsten.

Der Update dauert manchmal ein paar Stunden - je nach RSS Client Gruss, Thorsten.

STM32F4 is typically ca. 50% faster than a LPC17 core. Also the DMA implementation is better (no shared interrupt for all channels) which results into faster SRIO and SD Card accesses. Best Regards, Thorsten.

Nice that you find the error! :) Here a preliminary version which provides more event pool memory for the STM32F4: http://www.ucapps.de/mios32/midibox_ng_v1_030_pre1.zip Best Regards, Thorsten.

Dafuer gibt es noch keine Moeglichkeit. Wahrscheinlich werde ich das als Meta Event bereit stellen, die Du dann mit den Tastern ausfuehren kannst. Ich bin mir noch unschluessig, wie ich die Anzeige der aktuellen Oktave ermoeglichen soll (ich denke, zumindest via LEDs aber auch via LCD), doch die waere fuer Dich erstmal uninteressant, richtig? Gruss, Thorsten.

Das 4x20 LCD ist vergleichbar mit zwei 2x20 LCDs, es hat sicherlich auch zwei E (enable) Eingaenge, richtig? In diesem Fall musst Du den zweiten E Eingang an J15B:E anschliessen, sowie "lcd_height 2" und "lcd_num_y 2" einstellen. Falsch - sie Posting weiter unten. Zum Script: koenntest Du mal alle relevanten Kommandos aus dem NGC und NGR File posten, so dass ich das Problem exakt bei mir reproduzieren kann? Gruss, Thorsten.

These are good news! :) Best Regards, Thorsten.

This is the typical result of a stack overflow, you've to reserve more memory for the task which accesses the SD Card. Examples, how to do this, can be found in the mios32_config.h and task.c resp. app.c of applications like MIDIbox SEQ, MIDIbox NG, MIDIO128. MIDIO128 has probably the most simple solution in app.c, it just reserves twice the memory of standard tasks for the 1mS_LP and 1mS_SD task: xTaskCreate(TASK_Period_1mS, (signed portCHAR *)"1mS", configMINIMAL_STACK_SIZE, NULL, PRIORITY_TASK_PERIOD_1mS, NULL); xTaskCreate(TASK_Period_1mS_LP, (signed portCHAR *)"1mS_LP", 2*configMINIMAL_STACK_SIZE, NULL, PRIORITY_TASK_PERIOD_1mS_LP, NULL); xTaskCreate(TASK_Period_1mS_SD, (signed portCHAR *)"1mS_SD", 2*configMINIMAL_STACK_SIZE, NULL, PRIORITY_TASK_PERIOD_1mS_SD, NULL); and it reserves more heap memory: configMINIMAL_STACK_SIZE is declared in programming_models/traditional/FreeRTOSConfig.h If you need more memory for all tasks, then don't change this define, but MIOS32_MINIMAL_STACK_SIZE in your mios32_config.h file. The standard settings are: How to find out the best stack size value? Call: umm_info( NULL, 1 ); e.g. from a terminal command after you application was running (and especially executed the most memory hungry functions) It will dump all memory blocks. Memory areas which haven't been overwritten during runtime are marked with the value A5 Best Regards, Thorsten.

SmashTV ordered another prototype some days ago which allows to mount an alternative SD Card socket which is better available (e.g. at Mouser) We also fixed two imperfections and want to ensure that this board is 100% right before more people are using it. I fixed this in the schematic. The connections are like Sneakthief said Best Regards, Thorsten.

Es fehlt noch der SysEx string fuer die MTC digits: # this generic receiver (without controller assignment) receives the MTC digits # and notifies all events which select syxdump_pos=10:<pos-in-dump> EVENT_RECEIVER id=10 type=SysEx stream="0xf0 0x00 0x00 0x66 ^dev 0x10 ^dump" Gruss, Thorsten.

yes no, the long leg (anode) has to be connected to the 5V line towards the bottom. Best Regards, Thorsten.

yes, e.g. with PitchBender or NRPNs :) @Ilmenator: thanks for the nice explanation! I added it to the MBHP_AOUT page. Best Regards, Thorsten.

Wow! 2 releases at once, because the preliminary ones were never officially published: V1.028: MIDIbox NG V1.028 ~~~~~~~~~~~~~~~~~ o DOUT_MATRIX n=8 properly supported o now up to 16 DOUT_MATRIX and DIN_MATRIX can be configured V1.029: MIDIbox NG V1.029 ~~~~~~~~~~~~~~~~~ o added USB MIDI workaround for Windows o if a EVENT_SENDER is part of a radio group, and got a new value, it will only send the MIDI event if the new value is matching with the specified range. Usage example: cfg/templates/pg300.ngc o if EVENT_RECEIVER forwarded to EVENT_AIN, EVENT_AINSER or EVENT_BUTTON, send also MIDI event. Usage example: cfg/templates/pg300.ngc o support for MBHP_CORE_STM32F4 module + many bugfixes and minor improvements reported/done in the last weeks. Best Regards, Thorsten.

Wow! 3 releases at once, because the preliminary ones were never officially published: V3.017: MIDIO128 V3.017 ~~~~~~~~~~~~~~~ o the MIDI router forwards SysEx messages regardless if channels are filtered, or "All" is used. It's ensured that messages are only sent once if multiple router nodes forward to the same destination port. o corrected MIDI clock output V3.018: MIDIO128 V3.018 ~~~~~~~~~~~~~~~ o added USB MIDI workaround for Windows V3.019: MIDIO128 V3.019 ~~~~~~~~~~~~~~~ o added SysEx support for MIDI file player and recorder o MIDI file recorder now sets "end of track" marker at the end of a 4/4 bar to comply with the MIDI file spec, and to allow proper looping o support for MBHP_CORE_STM32F4 Best Regards, Thorsten.

V1.015 is available under http://www.ucapps.de/mios32_download.html workaround for USB MIDI under Windows when more than 1 USB MIDI port is enabled various LCD/GLCD driver updates for STM32F4 Best Regards, Thorsten.

V1.4h is available: MIDIboxFM V1.4h ~~~~~~~~~~~~~~~ o Wavetables can now be changed via SysEx when they are stored in RAM Best Regards, Thorsten.

this is now changed in the official V1.4h release strange! All 32 steps should be stored iN RAM Best Regards, Thorsten.

Cool - finally somebody who confirms that the "new attempt" for a workaround solves the USB MIDI issues under MIDI! :-) See also Best Regards, Thorsten.

Yes, you are right - it's much more work than expected, it's not a simple change! :-/ I started to adapt mbng_matrix.c, so that it should also allow MIOS32_SRIO_NUM_DOUT_PAGES == 64, but then I found much more places which actually need changes. Also mbng_dout.c as you already noticed. The table has to be enhanced by more granular patterns, and also the variable width (e.g. u64 for 64 dim levels), and finally MBNG_DOUT_NotifyReceivedValue which transfers the pattern into the pages. So, most of the code is constrained for 32 pages and 16 dim levels. Adding more flexibility would result into too much work at my side - I don't want to do this. Best Regards, Thorsten.

Well done!!! Great case, great buttons! :) Could you please create a new Wiki page which explains the MCP6004 approach? I would like to link it to the MBHP_OPL3 page to inform other builders about this option. Best Regards, Thorsten.

Ok, then this preliminary version should be sufficient for the first tests: http://www.ucapps.de/mios32/midibox_kb_v1_013_pre1.zip It depends on the number of scanned rows; with the "optimized" flag enabled, ca. 130 uS for a typical keyboard Best Regards, Thorsten.