TK.

Sure, I could integrate this into the MB64 firmware, just another option... but I think that it's better to wait for MIDIbox NG where everybody will be able to create and distribute his own "MIDIbox Skin" (display, LED outputs, button functions programmable via PlugIn) :) Best Regards, Thorsten.

Hi Mujgy, this explains everything: you have a PIC16F874 with a memory range from 0000h-0FFFh, therefore addresses >= 1000h are not accessible. Just select the correct PIC derivative in IC-prog and the error message will vanish. Which firmware are you trying exactly? Best Regards, Thorsten.

Hallo Simolch, Vom Prinzip her alles richtig - mit den 3 Layern kannst Du mit einer 16-Poti-Reihe Tonhoehe/Accent und Slide steuern. Allerdings gibt es bei der PIC16F877 Version eine kleine Einschraenkung: im dritten Layer gibt es fuer die gesamte Reihe nur einen Wert. Du musst also vorher im Layer-Assign Menue selbst entscheiden, ob entweder die Tonhoehe, die Velocity oder die Gatelength immer gleich sein soll. Bei der 18F452-Version wird es diese Einschraenkung nicht mehr geben (ist mal wieder die leidige Memory-Geschichte). Andererseits: da Du ja nur 16 Potis verwendest, koennte ich eine spezielle Option in die Firmware einbauen, die dem dritten Layer das nicht verwendete RAM zuweist. LEDs: koennte ich ebenfalls als Option einbauen, macht ja Sinn. Die Sequenzer-Position ist trotzdem sichtbar sein, die jeweilige LED wird einfach invertiert. Gruss, Thorsten.

Hi Justin, unfortunately this is a known problem with MIDIbox Plus. The root cause is the 4-bit LCD interface without read signal. Since the MBPlus firmware is not able to ask the LCD if a character has been printed out or not, it has to wait for a certain time before continuing with the main program. If a lot of MIDI events are received continuously, the LCD delay routine will be interrupted very often and therefore takes much longer than expected, as a result the whole display update takes more than 300 uS and this can cause a MIDI buffer overflow on a back-to-back MIDI event stream. This issue has been fixed on all MBHP designs by accessing the LCD over a 8-bit interface and by polling the busy flag of the LCD. For MPlus two workarounds are possible: if you don't use a display, I could give you a special version which doesn't try to access the LCD. If you own a LCD, I've to include some code which is also used by the MB1664 variant. This variant would require a hardware modification (read-line has to be connected with the PIC) Best Regards, Thorsten.

Hi Mujgy, the PIC16F877 has an address range of 0000h-1FFFh, so it seems that the upper half cannot be programmed. Did you really try a 16F877 or a 16F874 (which contains only a 4k word flash) Check also the code protection setting at the right side of the IC-Prog mainwindow. It has to be disabled for both halfes of the memory (Code Protect: CP OFF) Best Regards, Thorsten.

Hi Steve, thanks for the positive feedback! :) The new MB64SEQ doesn't come with all planned features yet, but I think that I can release V1.00 in about 3 weeks (after I've finished all the other open issues). Best Regards, Thorsten.

Just a question to Ian and Christoffer: are you using an 1 MHz oszillator for the SID, or is the SID clock input supplied by the PIC? Because this could be the only difference between my own SIDs and yours, and my SIDs are running hours without any failure, and I neither use headsinks, nor a metal case... Best Regards, Thorsten.

You mean this kind of buttons? Yes, they are great! :) Best Regards, Thorsten.

Hi Oozi, yes - you can either use the MIDI learn function to assign a different MIDI event with a different channel to the row, or you can configure the appr. settings via MIDI dump. Yes, each row/track has three layers (functions): Note Value, Velocity, Gatetime. With the gatetime you are able to vary the note length; overlapping notes are not supported, but on overlaps first the new note will be played, thereafter the note off event of the last note will be sent. This is perfect for HiHat rhythms (and for TB303-like slides :)) What does the accent function exactly? Does it vary the velocity? The BankStick is not required, but still usefull if you want to save your sequences without sysex dump via PC. However, it can be left out... Suggested modules: Core, JDM, two AINX4, one DINX4, one DOUTX4 (two if you want to have a LED under each pot), maybe a LTC module for the MIDI In/Out LEDs. For 64 pots with knobs, 32 buttons, 32 LEDs, 1 LCD: about 50-100 EUR (the knobs and the case are the most expensive parts) Experienced people need about one week, the others maybe longer - it depends on your ambition :) Best Regards, Thorsten.

Hi Nasty, yes, you can leave them out without further modifcations - just take care for the 4*10k pull-up resistors, they must be connected, otherwise the MBPlus will send out random button values. See also: http://www.ucapps.de/midibox/midibox_plus_test_16.pdf Best Regards, Thorsten.

Hi Kristoffer, yes, you need 2 DINX4 modules. Regarding the DOUT: you can use one DOUTX4 and cut the SC, SO and RI track between the first three and the last shift register. On this way you've two chains which can be connected to J7 and J8 seperately. See also http://www.midibox.org/cgi-bin/yabb/YaBB.cgi?board=concepts;action=display;num=1037272152 Best Regards, Thorsten.

I would cut all tracks from the optical encoders and the buttons, so that the pins/connectors are free and can be directly controlled by the MIDIbox. The internal electronic of your mouse cannot be re-used, however the suggested solution is really uncomplicated. :) Note that optical encoders contain a LED which has to be powered - just take a 470 Ohm resistor in serial to the LED and connect it with Vss/Vdd (+5V) of the Core or DIN module. Best Regards, Thorsten.

This time from ThomasT - see his homepage for further details http://www.krachwerk.de/projekte.html total control! :)

You mean that one of the chip gets hot and has no functional cut-off effect? It seems that this chip is either damaged (after 20 years this could happen...) or a different chip-revision which requires another capacitor pair (C1/C2). You could try two 1 nF or 2.2 nF if available. Btw.: Since I built the MIDIbox SID I've tested 4 different SIDs and noticed that every chip has it's own sound ;-) --- especially the filters varying a lot. Best Regards, Thorsten.

I already answered via EMail, here a copy for the records: ------------------------------------------------------------------------------- just a quick but (hopefully :) helpful answer: it seems that you didn't solder a 1k pull-up on pin RA4 of the PIC. This resistor is absolutely required, otherwise the firmware cannot determine if a bankstick is connected or not. You will find the pull-up resistor as R2 here http://www.ucapps.de/mbhp/mbhp_core.pdf - the schematic will help you to locate the correct pin. Another important hint: the serial input pin of the last 74HC165 requires a 10k pull-up resistor. In http://www.ucapps.de/mbhp/mbhp_dinx4.pdf you will find this resistor as R36. Without this pull-up, the box will send random button events. Best Regards, Thorsten.

Hi Adam, 24AA256 should work, PIC16F877A also - it's just an (compatible) redesign Best Regards, Thorsten.

thanks for the reminder ;-) However, here a schematic of the touch sensor circuit which will be supported with the next MIDIbox MF version. In theory it could also work with other MIDIboxes as cheap button replacement, but this will not work until I've added the special charging pulse signal at Port J14 http://www.ucapps.de/mbhp/mbhp_din_touchsensors.pdf Best Regards, Thorsten.

no adaption is required. Just connect all open analog pot inputs to ground and don't forget the pull-up resistor on all digital inputs - otherwise the box will send random values. Best Regards, Thorsten.

Yes, the 24LC256 is ok; it's just the low-current version. Where did you find this chip? I would like to add it to the BankStick order list. Best Regards, Thorsten.

Alright - yes, for MB16E it's the same. You have to omit all optocouplers. They are only for "common" MIDI lines, which are working as opto-isolated "current loop". So long as you are wiring digital Rx/Tx signals, no additional components are required between MIDI IN and OUT. Best Regards, Thorsten.

Sure, this is realistic. You could use the MIDIO128 with just a single DINX1 or DINX2 module for 8 or 16 buttons. This would come to about 25 EUR (w/o case, buttons & JDM PIC programmer). Or - if ableton reacts on program changes - this cheaper circuit could be interesting for you: http://www.ucapps.de/floorboard.html Best Regards, Thorsten.

Optical encoders mostly send TTL pulses and can be connected to the DIN module on the same way like mechanical encoders - only difference: the pull-ups are not required. Best Regards, Thorsten.

For the records: my answer to kierans posting can be found here: http://www.midibox.org/cgi-bin/yabb/YaBB.cgi?board=concepts;action=display;num=1037621791

correct. And with the upcoming MIDIbox NG design you will be able to change the application PlugIns via MIDI on-the-fly :) Here the explanation of the MB64 tutorial: Snap Mode: When you adjust a knob in Snap Mode, the original setting determines the value until the point where the movement of the knob passes beyond this position. In other words, the parameter does not jump to the value as in Normal Mode. However, keep in mind that your adjustments have no effect on the parameter until the knob "snaps" on at the position equivalent to the original parameter. In SNAP mode, an right arrow symbol appears instead of the parameter bar. It indicates in which direction the knob must be turned to exceed the current parameter value. Arrow Left implies the value is lower. The parameter bar appears when the value is "snapped". And an example: your knob points to note C-1, you switch to another bank which restores the note F-3. When you now turn the knob clock-wise, it would send C#1, D-1, D#1 and not F#3, G-3, G#3 in normal mode. In snap mode no value will be changed until the knob passes the position for F-3 --- on this way parameter jumps are avoided. Best Regards, Thorsten.

The purpose of the LEDs: if the appr. push-button is in toggle mode, the LED shows if the button is active or not the LEDs can also be controlled via MIDI. If your MIDI software supports some kind of feedback, you can use the LEDs to display the status of something (the MIDI event which controls the LED is the same like the event assigned to the appr. button) the MB64SEQ uses the LEDs to display the sequencer position the LEDs can also be misapplied as christmas illumination, you need: a core module running with any MIDIbox firmware a DOUT module the appr. number of LEDs (different colours increase the effect) some christmas songs in MIDI file format a computer with MIDI player software a christmas tree hang the LEDs over the christmas tree, plug the MIDI OUT cable of your PC into the MIDI IN jack of the core module, assign the LEDs to some notes of different MIDI channels, start the MIDI file player and enjoy the show ;-) Best Regards, Thorsten.