TK.

And please use smart mode! Even when MIOS has been installed, it could happen that code blocks will be missed or refused due to a bad MIDI connection or a strange working PC Best Regards, Thorsten.

Yes, when I read your description today, it sounds more like a configuration issue. It makes sense to have a proper conditions when debugging this after so many months - means: reinstalling MIOS and the application. I guess that you are using a very old MIOS version, uploading the current MB64 release would crash the core. Therefore just upload the pic18f452/midi/update_with_old_mios.hex of the MIOS update package first, and thereafter the most recent release of MIDIbox64 Thereafter the internal bank should be properly initialized again (btw.: the S in "Snternal Bank" has maybe be changed manually, this can be done in one of the configuration menus...) As doc mentioned, it really makes sense to test all button functions with the ain64_din128_dout128_v1_3 application. Buttons should send notes independent on how the application handles the button events. Best Regards, Thorsten.

The easiest way is to add the code which handles the delay into an existing timer hook. Normaly I'm using the SRIO hooks (USER_SR_Service_Prepare or USER_SR_Service_Finish) for this, as they are called each mS independent from the other timers. Even the names indicate, that they are used for SR operations, you can also use them for other tasks w/o conflicts. Best Regards, Thorsten.

Please read in my first posting, how the encoder handling should be changed Having the alternative to change ensembles with encoder is very useful, I don't know, why you don't like it (however, it doesn't matter - so long you don't press the menu button within the main page, you won't notice it anyhow ;-) Best Regards, Thorsten.

Using PAGE_UP/DOWN buttons in addition is a great idea - it ensures, that users will find an uncomplicated way w/o searching for details in the documentation. Consequently, PAGE_UP/DOWN should select the patch within the "normal" patch menu. Ensemble Init: could be put into the SHIFT page when you are in one of the ENS menu pages as replacement for "Patch Init" Best Regards, Thorsten.

Hi Julien, no problem - the only difference between 6581 and 8580/6582 is the filter calibration range - you can edit this manually within the ensemble for each SID seperately. Recommented range for 6581: 000-FFF, for 8580/6582: 000-600 An auto detection is not possible Best Regards, Thorsten.

Thanks for the hints! and it could add even more confusion if the quality of the select button is so bad, that it double-triggers sometimes (I know that some people are using very cheap miniature buttons with limited debouncing) I like this option, but unfortunately it makes the SHIFT button handling inconsistent, as the SHIFT function speeds up all encoders, not only the datawheel :-/ Best Regards, Thorsten.

Currently the ensemble has to be selected within the ENS menu by moving the datawheel. Normaly the datawheel is used to scroll the menu items The special behaviour of the dataweel within the ENS menu leads to problem, that the 5th menu item of the ENS menu (Save) cannot be selected when only a 2x16 display is connected to MBSID V2. In addition, it doesn't allow me to add more menu items to the ENS menu in future, since this would lead to the same problem for 2x20 displays My proposal is to change the ensemble selection, so that it can be done within the main menu (where ensemble number, patch name, etc... is visible). Within the main screen, the datawheel changes the patch; by pressing SHIFT button in addition patches will be incremented faster. The ensemble could be changed by pressing the MENU button while moving the datawheel. Since this isn't really an intuitive usage, I would like to know, if somebody has an alternative idea. Best Regards, Thorsten.

Yes, thats an error in MBSID V1, which won't be fixed anymore (as this is an untypical corner case) Best Regards, Thorsten.

it's your personal desition. I'm using 2x8580, 4x6582 and 2x6580 Disadvantage: the filter of 8580 and 6582 sounds equal, whereas the 6580 filters sound different (also between the two chips). Accordingly, I can use the super-polyphony feature only for the 6 8580/6582 SIDs Best Regards, Thorsten.

Ok, a variation for CS_MENU_DISPLAYED_ITEMS == 4 has to be added to cs_menu_enc_table.inc hm! sh* happened - for this I don't know a solution yet /edit: discussion here: http://www.midibox.org/forum/index.php?topic=10039.0 they are pre-defined to 0..fff in setup_6581.asm I guess that you made an error by uploading setup_8580.asm instead of setup_6581.asm? In this case, a "virgin ensemble bankstick" will be completely initialized with the default values for 8580 So, you need to change the correct range manually now just increase the depth of the modulation path - it's 64 by default, max value is 127 Best Regards, Thorsten.

It definitely is strange, as nobody has reported this issue before. There are only 4 lines of additional code I inserted into RC6, and it's only related to voice allocation. Today I checked typical cases which could lead to such effects, e.g. uploading setup_mb6582.hex instead of setup_tk.hex, reviewing the code at the 64k boundary, uploading rc5/rc6 and back to rc5 - but I wasn't able to reproduce this. Please: could anybody who uses a MB6582 please confirm, that cloning is still working, and that the slave cores are running? Narwhal: it could be helpful to know the MBNet and application state - just create a SRAM dump of 0x000 (range 0x80) and 0x600 (range 0x80) with MIOS Studio as shown in this picture: and copy&paste the hex output into your posting. Note that I've selected device ID 1 at the left upper corner. The MIDI output is jumpered accordingly. It would also be interesting to know, why you are waiting for the MIOS boot screen instead of the MBSID version number screen as suggested. This screen follows after the MIOS boot screen. Could it be, that it is not shown on your setup? Best Regards, Thorsten.

Ensure that you are using the latest MIOS Studio version, older versions can only upload code to the 32k flash of a PIC18F452 Best Regards, Thorsten.

Since you've posted this question also in the french forum, I'm not sure if the issue is already solved. I don't know, why TL's patch manager shouldn't be able to access Bank A... maybe some pins of the EEPROM are not connected? Best Regards, Thorsten.

I haven't changed something on the cloning mechanism in any rc* release Could you please try the following: press the MENU button once the MBSID version number is displayed (don't press is during the boot process). This ensures that all cores are ready for receiving the update request. Best Regards, Thorsten.

This topic has been moved to MIDIbox of the Week. [iurl]http://www.midibox.org/forum/index.php?topic=9996.0[/iurl]

Hallo Stuart, danke fuer den Erfahrungsbericht - ich habe den Artikel gleich mal in die MIDIbox of the Week Sektion verschoben. :) English speakers: please use babelfish Das Problem mit den LEDs stimmt mich langsam bedenklich; Du bist wohl nicht der einzige, der davon berichtet, andererseits gibt es MIDIboxen, die mit wesentlich mehr LEDs bestueckt sind, und die trotzdem stabil laufen. Falls Du hier nicht weiterkommst, schreibe mir doch bitte eine PM, damit wir das gemeinsam debuggen koennen. Gruss, Thorsten.

It's normal that there is some crackling noise in background, especially when slow envelopes are used. Best Regards, Thorsten.

Fine! :) I guess you mean a 24LC512 - yes, it's supported, but it doesn't make a difference if you connect a 24LC256 or 24LC512, only 32k will be used. Best Regards, Thorsten.

MBSID V2.0 has already reached its final state; currently the release candidates will be published with the last bugfixes of corner cases (which most people wouldn't notice anyhow). Just today I found an imperfection in the drum voice allocation when Mono mode is activated. It's fixed in rc6 Best Regards, Thorsten.

If the menu button is directly connected to the DIN (no usage of a button matrix), the button has to be pressed while the MBSID version number is displayed. Just press it multiple times to ensure that the event will be notified... Best Regards, Thorsten.

No, you cannot slow down the track with the scrub function. It's better to use the BPM clock divider for this - this one has been made for live usage Best Regards, Thorsten.

The panel looks really good, especially with the flat buttons! :) What are you planning to insert into the lower right corner? LED behaviour: individual assignments are not provided by the firmware, but you can do this for each shift register (group of 8 pins). E.g., if the LEDs connected to the first SR should receive MIDI events which correspond with the first 8 buttons, and the second SR should control the LEDs according to the state of the next 8 buttons, then use following assignments: [tt] LED_SR1 = 16 # (MIDI Status of Button ID #1-#8) LED_SR2 = 2 # (Button ID #9-#16) [/tt] all variations can be found in the midibox64.ini file of the mk_syx.zip package Best Regards, Thorsten.

Yes, I tested it with a 808 some time ago - it works. Best Regards, Thorsten.

It allows you to FFW (fast-forward) the sequencer lines while notes are played. But I must say that I don't really use it - it's only nice when you want to demonstrate some nice features to friends ;-) Best Regards, Thorsten.