TK.

The conclusion is: if you want to generate PWM with a satisfying resolution, you need additional hardware components which are doing this job. And these components can be easily controlled from the PIC. It's a similar approach you already know from Monome. It uses a MAX7221 to control a 8x8 LED matrix. This unloads the CPU, and it saves you from adding LED drivers. There are also chips which allow you to control LEDs with PWM instead. Sorry, that I cannot give you more details, because I haven't worked on such topics yet (and I'm not planning to do this) - but it would be great if some people could just colaborate together here. The implementation is very straightforward at the software side, and at the hardware side you only need to find chips which are available in low quantities (without ordering >= 1000 pieces) Best Regards, Thorsten.

End of setup_mbfm_v1.asm Best Regards, Thorsten. P.S.: README.txt of the current MBFM release is confusing, it contains some artifacts from the previous "flat" project structure - sorry about that! I already corrected this in the development version.

Why do I have this deja vu two times today? -> http://www.midibox.org/forum/index.php/topic,10911.0.html Best Regards, Thorsten.

Cool! Thank you! :) Best Regards, Thorsten.

See this answer, and the discussion below Best Regards, Thorsten.

PIC18F452 is only a subset, PIC18F4620 is compatible, but provides more flash, more RAM, and more bugs ;) Means: you are able to run programs compiled for PIC18F452 on a PIC18F4620 without changes. With the upcoming project setup (which is still not completely documented), it will be much easier to change the processor type. See apps/examples/ain64_din128_dout128_v2_0, how clear an application does look meanwhile :) So, in order to change the processor, you would only need to change the PROCESSOR variable, and pic18f452.o file in the Makefile - thats all. E.g., the register file of the processor is not included directly into main.c anymore, instead we are using <pic18fregs.h>, which internally selects the register file depending on the PROCESSOR argument, which has been predefined in the Makefile Also the Linker File will be selected automatically, so that you don't need to take care about this :) Best Regards, Thorsten.

I would say, it's impossible with the MBFM firmware for various reasons. But you could easily use a second microcontroller which handles the triggers and analog inputs, and which communicates with MBFM via MIDI. Maybe this will cost you 10 EUR more, but assumed that you are using a premade project like EDrum or Megadrum, you will get a perfectly working solution and don't need to develop new software. Best Regards, Thorsten.

You know, we have multiple sequencers, multiple engines, multiple possibilities to control the volume concurrently to velocity. Your question "what I've forgotten to setup" doesn't really make it easy for somebody to understand, what could be wrong ;) Please read the RIOFAQMARKER again: If you don't remember the details anymore, why not attaching the patch (in .syx format) to this posting, so that somebody could try it on his own hardware? Best Regards, Thorsten.

See MBSID firmware, USER_Init hook how to enable PWM (in this firmware it's used to provide the 1 MHz SID clock). For 38 KHz clock, you want to use the same configuration as shown in the Basic Code example. Transmitting informations at 1kbaud should be easy, because you could control the NAND from USER_SR_Service_Prepare or USER_SR_Service_Finish. These hooks are called each mS Using different baudrates is impossible, because all other timer sources are already allocated by the firmware. Best Regards, Thorsten.

Mode 5 should do this. You wrote, that you built MBSEQ V3 long time ago, but are you also using the latest firmware? Meanwhile the usage of mode 5 is more comfortable than before. Previously you had to configure the trigger layers correctly to get it working. Meanwhile you only need to push the GP button below "copy preset" in the event configuration page (where you already select the mode) :) Thats an unexpected usecase (I'm a big fan of unsynched changes), but no problem: just added this option. The change is already available on the SVN server (seq_song.inc), but unfortunately, the flow how to fetch such immediate versions isn't documented yet. So, if you are not familar with subversion, try to copy the cs_m_song.inc into the latest release If this doesn't work due to incompatibilities (I haven't tested it), open seq_song.inc of the original release, search for "call SEQ_CORE_ChangePattern_Song" and replace this line by: ;; in phrase mode (bit #7 set): allow synched pattern change BRA_IFCLR SEQ_SONG, 7, BANKED, SEQ_SONG_FetchPosLoop_Song SEQ_SONG_FetchPosLoop_Phrase call SEQ_CORE_ChangePatternSynched rgoto SEQ_SONG_FetchPosLoop_PhraseCont SEQ_SONG_FetchPosLoop_Song call SEQ_CORE_ChangePattern_Song ; change pattern SEQ_SONG_FetchPosLoop_PhraseCont [/code] Best Regards, Thorsten.

The MIDI channel is combined with the MIDI event type. We are using 0x90 for Note Events over MIDI channel #1 To change the channel, use: 0x91 for MIDI Channel #2 0x92 for MIDI Channel #3 0x93 for MIDI Channel #4 0x94 for MIDI Channel #5 0x95 for MIDI Channel #6 0x96 for MIDI Channel #7 0x97 for MIDI Channel #8 0x98 for MIDI Channel #9 0x99 for MIDI Channel #10 0x9a for MIDI Channel #11 0x9b for MIDI Channel #12 0x9c for MIDI Channel #13 0x9d for MIDI Channel #14 0x9e for MIDI Channel #15 0x9f for MIDI Channel #16 Best Regards, Thorsten.

Timing inaccuracies can be emulated on a MB808 by letting the analog input for swing pot unconnected (don't clamp it to ground! ;) Best Regards, Thorsten.

No, thats fine - the crystal is oscillating with high frequency, you won't be able to measure this with your multimeter, and probably only read the RMS value (or something similar) Best Regards, Thorsten.

Habe den Schaltplan nun geaendert, und hoffe, dass damit alle Klarheiten beseitigt sind: http://www.ucapps.de/midibox_sid/mbsid_v2_communication.pdf Gruss, Thorsten. P.S.: die CAN Rx/Tx Beschriftung habe ich mit Absicht entfernt, denn sie schien bei Newbies wohl auch nur unnoetige Fragen aufzuwerfen... ich sollte wohl mal eine Experten-Ansicht in uCApps einbauen ;)

it's fixed now... Best Regards, Thorsten.

The "official" J5_IO driver is now available at http://svnmios.midibox.org/trunk/modules/j5_io/ Note that the integration hints described in the README.txt are intended for the future programming environment. So long you are working with the old C wrapper approach, you need to copy j5_io.asm, j5_io.inc and j5_io.h into your project directory, and add j5_io.asm to your MAKEFILE.SPEC Best Regards, Thorsten.

Irgendetwas machst Du anders als die anderen... hat jeder PIC eine eigene MIOS Device ID? Gruss, Thorsten.

I received my kit, and was positively impressed about the PCB - very good work! :) Since I'm currently busy with MB808 and some other stuff (e.g. Seppoman's Stereo SSM2044 filter), I won't find the time to populate the board in the next weeks. But once this has been done, I will be able to provide a modular driver for MIOS based hardware quickly... I'm also planning to integrate support for Stribes into projects like MBSID V2 and MBSEQ V3 (for my own interest ;)) Creating an application for softpots will be extremely simple :) Best Regards, Thorsten.

Please note, that Wilba's .inc file is not the latest one (e.g. $Id: $ header is missing) Never overwrite files, only change them directly within the repository Best Regards, Thorsten.

It could also be a nice exercise for you to change the original file in the repository (special_characters.inc), because I like the changed layout as well :) Best Regards, Thorsten.

Works fine! Thank you for starting to improve the documentation! Best Regards, Thorsten.

A while ago I changed the behaviour of the ALL function based on user requests, so that it affects all steps and not only the selected, which is especially useful while editing CC tracks. Now people are starting to request the old behaviour again.... ;) ris8_allo_zen0's proposal is excellent! It would be the perfect solution to tell the UI exactly, which steps should be modified, thats much better than changing only the gated steps (which doesn't work on CC tracks), because it allows you to quickly specify a certain "pattern" (of steps within the layer) which should be set to a specific value. I will consider this in the next release. As you know, I'm currently busy with MB808, so it could take some weeks. But you can be sure that the idea won't get lost (and in addition, we will get new features which have been implemented during MB808 development :)) It would be interesting to me, if there are more use cases for a "focus" function, beside of using it in conjunction with the "All" button. Best Regards, Thorsten.

It's probably related to the MIDI In port of your MIDI interface (of PC or whatever platform you are using). If the middle pin isn't disconnected at the MIDI In side, there is a potential risk for a ground loop, which can result into a 50 (or 60) Hz buzz. In order to solve this, you could disconnect the ground connection to the middle pin of the MBHP_CORE MIDI OUT port. It doesn't really hurt Best Regards, Thorsten.

Haha, you really want to transfer a whole patch via NRPN? ;) Not only access to the mod matrix is missing, there are much more parameters only available via SysEx for good reasons. If you would like to change the whole patch, it would take ca. 5 seconds (!!!) via NRPN, while it takes ca. 300 mS via SysEx. There are also SysEx commands which allow you to modify individual locations (e.g. only mod matrix parameters) - therefore SysEx is provided as the only complete solution. If you want to provide a flawless user interface, I strongly recomment you to learn how to handle SysEx with your software. Best Regards, Thorsten.

Ich komme irgendwie mit der Fehlerbeschreibung nicht klar. Man kann bspw. waehrend der Laufzeit die SID Spannung wegnehmen, und wieder anlegen, der aktuelle Patch geht dadurch nicht verloren, denn die digitale Spannungs-Domaene wird ja mit 5V versorgt - und in dieser Domaene befinden sich die Konfigurationsdaten (mal ganz einfach formuliert) Ausserdem werden die SIDs erst 2 Sekunden nach Startup konfiguriert. Vorher wird auch nochmal die Reset Leitung aktiviert, was zu einem definierten Start fuehrt. Die Frage ist also, wie sich die SIDs genau verhalten, wenn sie mal "nicht anspringen". Hoerst Du bspw. ein Geraeusch, wenn Du vom CS aus den Patch wechselst? Gruss, Thorsten.