TK.

Here my teaching words: successfull scientists are not the guys who only have a detailed knowledge about everything they are doing, but the guys with good social skills. If you would communicate with people in a friendly and respectful way, you could get more useful informations than just demanding for details in a rude way which don't really bring you forward. Some of us have already noticed this, and therefore didn't spent much effort to help you (thanks Moogah and all others for your patience!). If you wouldn't focus too much on details, but if you would start a chatter about a certain topic in a normal way, like you would talk with reallife people, then you would propably already have experienced, that perhaps Traktor doesn't support resolutions higher than 7 bit (this was at least the state in R4), and that this has to be evaluated first with a newer version (R5 or higher?). You would propably also already know, how to determine 14bit capabilities, and where to find ready made programming examples for 14bit MIDI messages, etc... these are the small but important details which cannot be read in any documentation, but which can be easily found out by asking the right questions to the right people at the right time with the right words. Keep up learning! Best Regards, Thorsten.

This is not correct, and I must say: I'm also out of the game; it doesn't make much fun to "discuss" with somebody who ignores most of the previously written stuff. Hope that you at least know, that MIOS is copyrighted and especially not allowed to be sold via Ebay... My last hint: you can easily find out the real behaviour by doing some experiments on a MIDIbox. Best Regards, Thorsten.

I've uploaded the files to this location: http://www.ucapps.de/tmp/mbseqxc.zip Best Regards, Thorsten.,

In this case the distance between a value change (-> USER_AIN_NotifyPin() is called) is 3, 5, 6, 7; accordingly the hook might be called more often then required. Calculation example: a common 7bit CC should be sent, the 10bit conversion result is rightshifted 3 times to get the 7bit value (the same as if the conversion value would be divided by 2^3 = 8 ) lets move a pot from 0 to 25. With a deadband of 6 following values would be propagated to USER_AIN_NotifyChange: 0, 7, 14, 21 The converted 7bit values are: 0, 0, 1, 2 With a deadband of 7 (which is the recommented value for 7bit transformations), USER_AIN_NotifyChange will only be called three times: 0, 8, 16 - and the converted 7bit values are: 0, 1, 2 Best Regards, Thorsten.

Normaly you are using a deadband of 7 for 7bit, 3 for 8 bit, 1 for 9 bit, 0 for 10 bit resolution. There is no need for providing values higher than 255, therefore it's only an unsigned char (this saves memory) did you write this question before or after you've read audiocommander's quotes? Best Regards, Thorsten.

Hallo, den 78xx hast Du richtig angeschlossen; kann es vielleicht sein, dass es ein Problem mit der Massefuehrung gibt? Gegen welchen Masse-Pin hast Du die Spannung denn gemessen? Zu 2) ja, das liegt am grossen Elko, wird sich aber schnell wieder geben, sobald bspw. die Hintergrundbeleuchtung des LCD angeschlossen ist. Gruss, Thorsten.

A promising project! :) I don't use the schematic tool of eagle, because the results are looking ugly and not always understandable for electronic newbies. Instead I'm using xcircuit (a derivative of xfig) - not because it's the best, but because I'm familiar with this program since ca. 10 years and just love it :) Best Regards, Thorsten.

Just to be sure: did you already test the encoder and DINX4 module connections? Details can be found here: http://www.midibox.org/dokuwiki/doku.php?id=din_module Best Regards, Thorsten.

Yes, thats it! Please: add this description to the Wiki (http://www.midibox.org/dokuwiki/doku.php?id=mios_details_ain_deadband) so that it won't get lost, it's really good and could be enhanced later if required. Best Regards, Thorsten.

Controlling CutOff and Resonance with analog pots is not natively supported by the firmware, has anybody written a GOOD documented extension in the meantime? I don't think that an AIN module is required, so long not more than 8 pots are connected to the core. 300 mA sounds a little bit weak if the LCD has a backlight, a 500 mA PSU would do better. Best Regards, Thorsten.

This is the intention of the Wiki - instead of buring the informations in the forum, they should be collected there. But this is a long process, and it especially relies on the contribution of the users. The target is, to have ucapps.de as some kind of "reference manual", which describes things in a compact and exact form, and the Wiki as "user manual", which describes things more detailed for non technical experts. Therefore I asked you for help (please read my posting again) You are now going to understand the Deadband feature, so once you know how it is working, why not writing down a report at http://www.midibox.org/dokuwiki/doku.php?id=mios_details_ain_deadband, so that next time it will be easier for other people to get into this topic? Best Regards, Thorsten.

You have replaced the 1.2k by a 100 Ohm resistor? This cannot work, and this is propably the reason for the failure... if the MIDIbox is directly connected to a gameport, you can also left out 1.2k, or (better) you can replace it by a 10k or higher in order to avoid that the MIDIbox will receive random (and invalid) MIDI events when it is not connected to your PC Best Regards, Thorsten.

You can only mute a group of 3 voices Drum mode is a real tinkering solution, my target was to bring this into the firmware with as less code changes as possible (I don't use it by myself, it was requested by hallucinogen), and with no panel changes... Best Regards, Thorsten.

Short hint: if you want to control more gates and low resolution CV outs, all you need to change is the MIDI_EVNT_Send_Fx function in midi_evnt.inc Best Regards, Thorsten.

The upload request already notifies a reset. If multiple requests are sent, then due to a watchdog timer reset - this reset happens on a timeout (e.g. no SysEx code received, or MIOS not uploaded). It doesn't make much sense to add more variations here, it wouldn't give you any useful information, it would only increase the complexity. So far I know, the "Received unexpected Upload Request" message can only happen in two cases: 1) the core doesn't receive a SysEx block with command 02 (Write), therefore a watchdog reset will happen 2) you are sending the code with a different device ID number - your core has device ID 00 (therefore I asked for the MIDI In dump, it shows me the number), is the same number selected in the upload window? I don't know other cases for such a behaviour Best Regards, Thorsten.

SysEx messages can be sent with MIDIbox64 and MIDIbox64E by using "Meta Events", but this requires that you are doing (simple) enhancements in the application. An programming example for SysEx can be found in mb64_meta.inc/mb64e_meta.inc within the source code packages In addition it makes sense to run the applications in patch mode, some infos can be found here http://www.midibox.org/dokuwiki/doku.php?id=midibox64_e_patchmode Best Regards, Thorsten.

yes, you are right, MBSEQ neither supports the AOUT_LC module, nor more than 4 triggers when the normal AOUT module is used. I just have added this request to the MBSEQ V3 wishlist Supporting more than 4 gates will require some programming effort at my side, since gates are currently handled channel based, and not note based. If somebody wants to implement it by himself: the changes have to be made in midi_evnt.inc, MIDI_EVNT_Send_Fx function Best Regards, Thorsten.

With this new build: http://www.ucapps.de/mios/midibox_sid_v1_7303b_rc6.zip the constant time glide feature (E2P flag) can be used in combination with the arpeggiator Note that since only one envelope is used for this feature (in MBSID V1, will be improved in MBSID V2), only the arpeggiator of voice 1 resets the envelope in order to avoid multiple triggers Best Regards, Thorsten.

What is the output of the MIDI In window? Best Regards, Thorsten.

this doesn't make much sense once the function uses subfunctions. Why not using the search function over the source code? (Windows: "search" button, Unix: "grep") Best Regards, Thorsten.

Just don't ignore the hints of the others, and spend some time to read the relevant postings in the forum. I've explained the purpose of a deadband several times in the forum, I've also written some more details at the http://www.ucapps.de/mbhp_mf.html page (search for deadband), I drew diagrams, I demonstrated the behaviour with pseudo code, and not at least I released the source code which contains some helpful comments. This is my distribution, now it's your term - you could be the man who takes care that other people don't have to spend the same work in searching for all these scattered informations. I've created a Wiki page where you can start explaining the AIN deadband function with your own words: http://www.midibox.org/dokuwiki/doku.php?id=mios_details_ain_deadband Hope that you understand this motivation. Best Regards, Thorsten.

Would you help me to write a GOOD documentation? Best Regards, Thorsten.

Hi Jaicen, if you have programming skills: yes of course, it's just an alternative user interface Without programming skills the only possibility is to ask somebody who did this before. Have you already seen Alkex's MIDIbox SID? -> http://www.midibox.org/forum/index.php?topic=6553.0 it doesn't fit with all your requirements, but maybe it will give you a good starting point. Best Regards, Thorsten.

You could improve the readability of a sequence by using constants (#define) for each note value. Unfortunately # (for major) is not allowed in a macro name, but "m" should do the same: #define NOTE_____ 0 #define NOTE_Cm0 1 #define NOTE_D_0 2 #define NOTE_Dm0 3 ... #define NOTE_C_4 48 #define NOTE_Cm4 49 #define NOTE_D_4 50 ... [/code] Now you can program sequences in this way: [code] const char pattern[] = { NOTE_C_4, NOTE_C_5, NOTE_C_4, NOTE_C_5, ... }; As you can see, I've already prepared a NOTE____ for an empty step. This is a simple solution, because it's very unlikely that somebody wants to play C_0 Alternatively, you could define an empty note as 0x80 ... in fact, 0x80 and 0xff can be easily used for extensions, such as chords, arpeggio definitions (entry number in the sorted list of held keys), or other control words Back to the implementation: the #define's should be put into a seperate include file You could also think about the integration of your sequencer routines into a seperate .c file - than earlier, than better. Just do it in a similar way like it has been done for the MCLOCK_* routines. Create a seq.c and seq.h file, and announce the new .c file in MAKEFILE.spec Best Regards, Thorsten.

MIOS V1.9 aufladen, und die Speicherbegrenzung in main.asm entfernen (Zeile 611, wie im Error File nachzulesen ist) Gruss, Thorsten.