TK.

Some years ago I had a lot of different sound cards in my PC at the same time (e.g. a Gravis Ultrasound, AWE32, SB Live and XG compatible), not at least to get as much MIDI ports as possible ;-) But in the meantime I also switched to RME Hammerfall DSP (+Multiface) - no driver & latency issues anymore + 8 analog IOs for realtime processing + support for Linux. :) For me this card + the PC software are a replacement for the even more expensive Fx gear Best Regards, Thorsten.

Oh man, it's time to dedicate an own gallery page for your designs ;-) Well made! :) Best Regards, Thorsten.

Hi Icam, the PIC is running fine, the LCD will not be initialized. So it must be a wiring error between the PIC and the LCD. In which way did you attach the LCD to the core module? Are you using an 2-row PCB connector, or did you solder the cables directly to J15? Best Regards, Thorsten.

Hi uclaros, Maybe this time out is caused by a feedback loop. If the "automatically attach Inputs to Outputs" option is enabled in MIDI-Ox, such feedback loops will be created without informing the user. This option should be disabled like shown here: Thereafter deselect all output devices and select them again, or open the port mapping and delete all input ports Best Regards, Thorsten.

Hi Nick, you should see the device ID in the "Download request" SysEx message which is sent by the bootstrap loader immediately after power-on. If the PIC doesn't send any MIDI data, the reason must be somewhere else. I'm currently not able to check the IDE (cannot be started under Linux...) but I guess that the predefined settings in the .hex file will be programmed correctly into the configuration sector although they are not displayed. With the wrong baudrate you should receive invalid MIDI events (some Pitch Bender and controllers). Maybe it's better to start with the LED test first to check if the bootloader has been started: Best Regards, Thorsten. P.S.: is your board stuffed with a 20 MHz or 10 MHz crystal?

Great! :) Best Regards, Thorsten.

Hi Dan, yes, the meters will be realized like the LED rings. The number of LEDs for each meter can be defined by the user (also the patterns), up to 16 LEDs will be supported. and yes: the MTC display will be similar like shown here http://www.ucapps.de/mbhp/mbhp_doutx4_mtc.pdf but with 12 digits instead of 8 Best Regards, Thorsten.

Yes Ian, thanks for your help! :) Best Regards, Thorsten.

Hi Cjlargear, the intention was to access the Yamaha OPL3 soundchip without desoldering it from a soundcard (as this is a SMD chip, there is a high danger that it will be destroyed in this way). Using the wave audio capabilities of a soundcard is not planned, because it would require a lot of additional hardware (external memories, etc.) and would increase the effort a lot, especially because of all the quirks which are necesary for different cards - see http://www.alsa-project.org to get informed about the complexity for such a driver software. The PIC is no Mini-PC ;-) From the MIOS intro page: Best Regards, Thorsten.

Hi DriftZ, the possibility to select the track directly is a very good idea! I would prefer a toggle function (1->2->3->4->1) which works faster in this case :) It should replace the stop function. Best Regards, Thorsten.

Thanks for the applause! :D I feel that the new system gives me the possibility for advancing to new sound dimensions ;-) Yesterday I made some experiments with a virtual vocoder (which comes with Reaktor). I used two SIDs for the carrier input, and the other two SIDs for the modulation input. As a result I get mostly dissonant, but sometimes also very expressive, organic sounds - by tuning the pitch, pulsewidth and LFO parameters the combination of 4 patches can change from very ill to extreme impressive (I'm still exploring the coherences ;-)) Dan, if you would like to have some fun, start Reaktor, connect the vocoder Fx between Audio In and Out and try following patches: Carrier: "P15 Sync Pad" and "P7 Simple Pulse" Modulation: "P18 Arpeggio" and "P49 Neutron" You can either assign all SIDs to one MIDI channel, or play them on 4 different channels (which makes it even more interesting) Try different octaves and chords (-> arpeggiator) Take care for your bass speakers! ;D For extreme percussive sounds, try "P31 Metal" and one of the "WT" drums as modulation source However, If I would have the time, I would begin with Step D immediately - nothing better than direct access to all parameters without those button-click-click-click-origies ;-) Best Regards, Thorsten.

Hi Steven, you have to press "MBLC_HOLD_LAYER" button in order to switch to the second layer. So long as this button is pressed, you can select the display page with one of the first four buttons ("MBLC_DISPLAY_PAGEx"). The appr. LEDs will indicate the selected page. When you release the layer button, the default button and LED functions of the first layer will be available again. The button numbers, which are equivalent to the table position, can be found in the lc_io_tables.inc file. In this file you can also assign these MBLC specific functions to different buttons, and you are able to select your favourite button behaviour for every MBLC_* entry (toggle, hold or switch function) Best Regards, Thorsten.

For all who can't wait to hear the monster sound of 4 SIDs, turn on the soldering iron, build some SID and core modules and try the new release of MIOS based MIDIbox SID V1.4 Step B :-) But also for people who can life with a single SID this update makes fun: patch upload issue solved - no inconsistend bank data anymore when you update to MIOS V1.2 (required for this SID version) Pitch range can be changed (was missing in the last version) Patch name, MIDI channel and device ID can be edited local mode allows to send CC data via MIDI in order to automatize parameters changes (now you can record filter sweeps with your MIDI sequencer program) And here the feature list for the master/slave solution: up to 3 additional core/sid pairs can be accessed from the control surface message protocol is compatible with the PIC16F877 version of the SID (V1.4) One BankStick for all cores to simplify the patch exchange spectacular stereo effects are now possible :-) Sources: MIOS V1.2 (if you haven't updated yet) http://www.ucapps.de/mios/mios_v1_2.zip SID V1.4/18F Step B http://www.ucapps.de/mios/midibox_sid_v1_4_18f_stepB.zip Simplified connection diagram for master/slave modules http://www.ucapps.de/midibox_sid_cs/sid_csB_connections.gif Example for stereo configuration: http://www.ucapps.de/midibox_sid_cs/mbhp_sid_joined_audio.gif Required preparation steps (only once after firmware upload): every slave SID needs an unique device ID - see the SID page, how this ID can be set via MIDI Slave 1 (SID#2): ID 1 Slave 2 (SID#3): ID 2 Slave 3 (SID#4): ID 3 The master SID should get ID 0 Thereafter connect the MIDI lines of the modules like shown here: http://www.ucapps.de/midibox_sid_cs/sid_csB_connections.gif After power-on you can select the SID in the mainscreen with the 4 first select buttons, the 5th select button enters the local mode (-> no MIDI data will be forwarded to the slaves, but master sends CC for automatization) More informations hopefully soon. In the next days I will especially take care for the the missing docs of MIDIbox LC, MB64 sequencer and SID step B ;-) Best Regards, Thorsten.

Hi Icam, thats really strange! How about the MIOS upload, did you receive any 0E xx F7 there? Maybe the OS is not complete Best Regards, Thorsten.

Hi js, did you really exchange the "call MIOS_AIN_UnMuxed" in sid_init.inc by "call MIOS_AIN_Muxed". Because the effect you are describing here is exactly that what would happen with an un-multiplexed device ;-) Maybe you should also check the multiplexers with the ain64_din128_dout128_v1_1 application first Best Regards, Thorsten.

Hi pay, unfortunately only bad news yet: thats an imperfection of the firmware, it stores the MIDI clock and SMPTE data into the same registers, the values are coded in BCD format, therefore numbers greater than 99 cannot be stored. For SMPTE this is ok, as the numbers will be received in BCD anyhow, but the MIDIclock handler normaly would require a totally different handling. In fact I would have to write a new routine especially for MIDIclock --- I will do this for the PIC18F port of the MIDImon (because with MIOS the programming makes really fun), once the routine is up and running, I will maybe integrate the new implementation also into the PIC16F firmware. Release date: unknown I cannot reproduce this problem. Yes, the end of the bargraph is not visible on screen like in the "numbers mode", but the routine which prints out the bar takes care for this and suppresses the last character. I'm not sure, maybe your display behaves different from mine, it's really hard to search for chars which are printed out but not visible on screen. However, I could send you the source code if you would like to help me ;-) yep (unfortunately, but the PIC hasn't much more memory). The PIC18F implementation will be able to store more data. yes, here I used a special shadow effect, the text and the -X value are alternated with a high frequency, so that you have the impression that the value overlays the text below Best Regards, Thorsten.

Hi Icam, it seems that the upload of the TC application failed for some reasons. Which "acknowledge" messages are sent during the upload? All messages should end with ...0F xx F7 (xx = checksum, ignore this number). On receive errors, you will receive ...0E xx F7 Maybe you should also try another application, e.g. the SID which provides the same control interface. Best Regards, Thorsten.

good idea! :) Best Regards, Thorsten.

Hi Frank, if the application has been written on a proper way, so that it doesn't rely on previous initializations which are made by the bootstrap loader, you could bypass the bootloader by overwriting address 0x0000-0x0002 with "NOP" instructions. Do you have access to the source code? If not, you could also patch the hex file Best Regards, Thorsten.

Hi Frank, if the application has been written on a proper way, so that it doesn't rely on previous initializations which are made by the bootstrap loader, you could bypass the bootloader by overwriting address 0x0000-0x0002 with "NOP" instructions. Do you have access to the source code? If not, you could also patch the hex file Best Regards, Thorsten.

Hi Amando, PIC18F and PIC16F are not binary compatible, there is now chance to run an old firmware on the PIC18F452 The most current version of IC-Prog also provides the PIC18F452 Best Regards, Thorsten.

Hi Dendy, I've no answer to 1) because it must be a strange electrical problem, but to number 2) it could be the reason that you are using an old bootstrap loader version? Versions below V1.1b didn't start when they received any MIDI message during the startup phase. This has been fixed in V1.1b - this version only waits for SysEx messages which begin with F0 00 00 7E 40, all other events will be ignored. Best Regards, Thorsten.

Hi Richard, the firmware changes the "encoder speed" depending on the value range (max value). Some values with a short range require more detents than others, in order to allow a precise control. So, this effect is ok. There are also some parameters like the oscillator ADSR or the filter cutoff/resonance, which are only 4 bit wide (limited by the SID chip itself). Thats the reason why these values can only be incremented/decremented by 8 But regarding the Rectangle waveform I'm unsure ("Pul" for Pulse should be displayed). If some inc/dec pulses from the encoder are missing (see also http://www.ucapps.de/mios/mios_encoder_modes.gif), you either have to change the connections (see the schematics under Concepts->MBHP->DIN) or the encoder mode in the application setup (-> mios_tables.inc) Hope this helps - have fun with the SID! :) Best Regards, Thorsten.

Hi Richard, the firmware changes the "encoder speed" depending on the value range (max value). Some values with a short range require more detents than others, in order to allow a precise control. So, this effect is ok. There are also some parameters like the oscillator ADSR or the filter cutoff/resonance, which are only 4 bit wide (limited by the SID chip itself). Thats the reason why these values can only be incremented/decremented by 8 But regarding the Rectangle waveform I'm unsure ("Pul" for Pulse should be displayed). If some inc/dec pulses from the encoder are missing (see also http://www.ucapps.de/mios/mios_encoder_modes.gif), you either have to change the connections (see the schematics under Concepts->MBHP->DIN) or the encoder mode in the application setup (-> mios_tables.inc) Hope this helps - have fun with the SID! :) Best Regards, Thorsten.

Hi Dan, in theory: yes in practice: I don't want to touch the MB64 firmware anymore, it's very hard to add even more features without relocating code (which mostly causes trouble and new bugs) Looking forward for the MIOS implementation :) Best Regards, Thorsten.