TK.

Hi Jouni, thanks for this great input! :) I just wrote the appr. SysEx parser, but it seems that your re-engineered spec is not complete. The order of <data> is not clearly defined. Example: at the beginning, SIDPLAY sends following SysEx command: F0 2D 4E 7F 7F 7F 0F 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0F 00 00 00 F7 I guess that this command should set all 25 registers to zero - except for the volume register (address 0x18 ) which should be set to 0x0f. The "featured" double-writes to the waveform registers 0x04, 0x0b, 0x12 (<mask4>, bit 7-5) are flagged to zero, which seems also to be ok. Assumed that the data is sent in ascending order (like your spec could be interpreted), the volume register (0x18 ) will be set to 0 Assumed that the data in this example is sent in following order of address offsets: 6, 5, 4, 3, 2, 1, 0, 13, 12, 11, 10, 9, 8, 7, ... it will be set to 0x0f, so this order seems to be correct - or not? WIth the first conversion rule the SID plays mostly nothing, with the second one I mostly hear some rhythmic sounds, but they are still incorrect (wrong parameters) So, I would like to know about which parts of the document you are 100% sure, and which parts are assumptions. Best Regards, Thorsten.

And again: no further discussion before somebody tells me how to transfer the old articles to a new forum... ;-) Best Regards, Thorsten.

See MIDIbox64 Tutorial -> Snap Mode (Soft-Overtake function) Best Regards, Thorsten.

Just use the MIDIO128 application, it provides up to 128 LEDs Best Regards, Thorsten.

You made a good analysis! :) Best Regards, Thorsten.

Hi Ryan, crystals with parallel cut are optimized for parallel resonance with the oscillator. If you would use a crystal with serial cut (which can mostly be found in radio modules), the PIC won't run with the correct frequency. It must run at 40 MHz (external: 10 MHz), otherwise the MIDI baudrate will be wrong. Unfortunately every distributor has it's own way on identicating the crystal type. I can only say: if the type is not declared, it's a parallel cut crystal with a probability of 99% Best Regards, Thorsten. FAQMARKER

Maybe the 2 remaining pins? You should definitely search for a datasheet before destroying the LCD. Best Regards, Thorsten.

Hi Adam, here a more general answer (I answered this some months ago, but I will do it again just for the records) the MB64SEQ provided by me will only support 4 tracks. This is enough for my private needs (my groovebox is running in parallel for the drums and synced via MIDI clock). The firmware has been especially designed for an analog style sequencer (and btw: the MIOS version will also support 8 AOUTs (CV outputs) for analog gear) The firmware has to be adapted by somebody else if another usemodel is desired (e.g. for a TB303 or 808/909 like sequencer or for a 16 channel sequencer with minimal hardware requirements (some buttons + datawheel) or whatever) - this won't be my job. The sequencer core routines donn't take care for the number of tracks, but everything else (user interface, etc.) is dedicated to my hardware and mostly useless for other variations. To your questions: 32 knobs: will work, the minimum requirement will be: 8 pots or 8 rotary encoders or 8 motorfaders. Maximum: 64 pots or 48 rotary encoders or 8 motorfaders LEDs: my firmware will support 4 tracks only, so you only need 4 mute buttons/LEDs and 16 mute buttons/LEDs for every single seq. position (the same buttons/LEDs are assigned to different functions depending on the selected menu). The general purpose LEDs also display the sequencer position of a track. (and to prevent any confusion: yes, you could also add a seq position LED to every pot which is usefull when the tracks are running with different settings) Please check also the given wiring diagrams of my own box: http://www.ucapps.de/mbhp/mbhp_dinx4_mb64seq.pdf http://www.ucapps.de/mbhp/mbhp_doutx4_mb64seq.pdf Best Regards, Thorsten. FAQMARKER

Ooops! Sorry, I was totally wrong. I mixed Michael with Christian (I delete this from my old posting to avoid unnecessary confusion if somebody uses the search function from the forum). So, Michael aka Mike is an absolutely reliable person, he must be in holidays. Normaly he delivers within 2-3 days Best Regards, Thorsten.

40 is a hexadecimal number. So: "176 7 65" for increments. Or better write "0xb0 0x07 0x41", which is the same, but looks better. You never wrote what you plan to do with this knowledge...? ;-) Best Regards, Thorsten.

Ok, here a description how it could work: I used an IR sensor with integrated amplifier of an old TV (such sensors can be found in most consumer products) and the IR LED of an old remote control. The IR LED is switched with a constant frequency. And here the results from the scope: (Channel 1: IR Input, Channel 2: IR Output) Than shorter the distance between my hand and the IR sensor/LED pair, than more pulses are received. So, the only thing which has to be done by the PIC is to count the pulses in order to measure the distance. This could be done by a seperate - small - PIC device (Hi Duggle, are you reading this? This would be a nice IIC gadget) or by MIOS projects which still have a free counter input ( Pin C.1 or C.2) Best Regards, Thorsten.

Pilo is now also a moderator of this group (hope this is ok, Pilo? I haven't asked you ;-)) Best Regards, Thorsten.

-> PM Gruss, Thorsten.

Das Posting stammt nicht von mir, sondern von Benni, aber nun gut: ich habe die GLCDs erhalten und muss mir nun noch ein paar PCBs aetzen lassen, um damit irgendetwas anfangen zu koennen - der Folienleiter ist so duenn, dass man ihn einem normalen Flachbandkabel nicht verloeten kann. Also nix fuer Gelegenheitsloeter. Nunja, bis ich mal wieder dazu komme, ein Layout anzufertigen, kann es noch dauern. Ich habe mir uebrigens auch das Schaltnetzteil fuer 2,95 EUR bestellt - es liefert 5V/5A und 12V/3A. Die meisten haben mit Schaltnetzteilen ja bisher schlechte Erfahrungen gemacht, aber dieses scheint fuer MIDI-Anwendungen (ausser MIDIbox SID!!!) ganz gut geeignet zu sein. Ich fuehre gerade einen Langzeittest an meiner MIDIbox NG durch. Gruss, Thorsten.

Hallo Seppoman, am besten schliesst Du es direkt an das 64er Netzteil an. Wie Du schon richtig vermutet hast, wuerde es den T1 ueberlasten. Er steuert den Strom. Eine Helligkeitsreglung via PWM kann ich nicht empfehlen, dadurch entsteht Jitter an den analogen Eingaengen (und speziell beim SID: extremer Audio-Noise) VFD Einstreuungen: keine Ahnung, habe solch ein Display noch nie betrieben. Vielleicht weiss jemand anderes mehr? Bestueckung: 7805 und Gleichrichter weglassen, C6 ist ohne den 7805 nicht notwendig, tut aber auch nicht weh. C5 drauflassen und mit der 5V-Leitung verbinden. Gruss, Thorsten.

bass woofer

Hi, <old text deleted to hide my own confusion ;-)> Best Regards, Thorsten.

Hi Nat, yes, your assumptions are correct. Most software can also handle with the "+/- <speed>" setting which should be more prefered, it "feels" more natural (hard to describe if you never tweaked a MIDIbox64E ;-)) The last mode, which is based on the MIDI spec, doesn't include a speed value, therefore most software manufacturers prefer the other (non-standard) modes. Best Regards, Thorsten.

Hi, you wrote that your synth has a 5V trigger output, so thats fine, you can connect it to a IO pin of the PIC directly. The purpose of the 10k resistor is to have some kind of default logic-level when nothing is connected to the input. This ensures plug-and-play (or better to say: "unplug and no random MIDI triggers") Your suggested transistor circuit is not required. But btw.: it would also work in this way (but it's better to use a 1k resistor at the base input) Best Regards, Thorsten.

If you've luck (and a good power supply) 50k could work without jittering values. 10k is the maximum suggested impedance for the analog inputs - this value can be found in the PIC18F452 datasheet (AC/DC characteristics). Alternatively you could maybe add a small (transistor based) amplifier in order to lower the output impedance of the joystick. Best Regards, Thorsten.

Hi, if this is a graphical display with an unknown controller it will be very very hard to get it working. You possibly need to program your own MIOS LCD driver, and this requires a good datasheet which contains at least the pinning and the commands that are necessary for the initialization. Ok, and it requires some programming skills (you will find some driver examples in the MIOS download section). If this is a 16x100 dot display, it would be sufficient for a common MIOS application. The character size is 6x8 pixels by default, this means that 16x2 characters can be displayed w/o problems. Best Regards, Thorsten.

Hi Ryan, this circuit is not perfect for adapting it to a PIC, you need a second power supply, and the output voltage range is from 0 to 12V. In fact a digitally controlled D-Beam could be realized on a very easy way, it could run as background task in a MIOS application. I will evaluate this in the next days (a D-Beam would be a nice feature for my MIDIbox SID :-) ) Best Regards, Thorsten.

Hi Nat, there are different formats for sending relative values, they are documented under http://www.ucapps.de/midibox64e_tutorial_en.html (search for "encoder mode"). You need a coding example? Then tell me for which type of message (or for which host application)... Best Regards, Thorsten.

Das stimmt nicht ganz, der 8580 ist eigentlich viel besser als sein Ruf. Das Filter zeigt eine andere Charakteristik, die sich jedoch softwaremaessig an die des 6581 anpassen laesst (deshalb gibt es in der MIDIbox SID Firmware den 8580 Switch) - in den aelteren C64 Spielen wurde auf diese Charakteristik keine Ruecksicht genommen, und deshalb kommen die Sounds nicht so aus dem 8580 raus, wie es eigentlich beabsichtigt war. Bei der Gelegenheit ein Preview: Jess D. Skov-Nielsen hat mir in den letzten Tagen ein paar Tips gegeben, wie man noch etwas mehr aus dem SID herausholen kann. In der neusten (noch nicht veroeffentlichten) Firmware gibt es einen "Curve" Parameter, mit dem sich die beiden Huellkurven aus der Modulations-Matrix "verbiegen" lassen, so dass sie nicht mehr linear verlaufen, sondern exponentiell - so wie man es von teuren Analog-Synthesizern her kennt. Und hier die Ergebnisse: (*) http://www.kubarth.de/sid/files/8580FLT.mp3 http://www.kubarth.de/sid/files/8580PHS.mp3 http://www.kubarth.de/sid/files/8580LOG.mp3 ...wer braucht da noch einen Virus? ;-) Dieser Sound laesst sich nur mit dem 8580 erzeugen - das 6581-Filter klingt wesentlich "dumpfer". Gruss, Thorsten. (*) Diese Beispiele stammen nicht von der MIDIbox SID, sondern von Jess' selbstgebauten SID-Synthesizer - seine Algorithmen gehen nun in die MBSID Firmware ein. :)

folk music