TK.

Servus, das interne PIC EEPROM ist viel zu klein um mehrere Baenke zu speichern, du benoetigst also einen BankStick. Um zwischen den Baenken umzuschalten muesstest Du den Tastern eine sog. "Special Function" zuweisen. Die verfuegbaren Funktionen sind hier aufgelistet: http://svnmios.midibox.org/filedetails.php?repname=svn.mios&path=%2Ftrunk%2Fapps%2Fcontrollers%2Fmidibox64e%2Fdoc%2Fmidibox64e_sfb_table.txt AIN (analog input): kann es sein, dass Du DIN (digital input) meinst? Gruss, Thorsten.

Copy fails because the FIL structure is created locally (accordingly it will allocate stack memory) Stack is reduced to 1024 bytes, but each FIL allocates more than 512 bytes, therefore the function doesn't work deterministic anymore (or just crashes) Solution: declare the FIL structures outside the function ("static FIL fsrc, fdst:") Best Regards, Thorsten.

I moved your posting to the right location, because in the MBSEQ V4 thread nobody can read about the history... FatFs reports error status 13, which stands for invalid file system. So, neither FatFs nor DosFs are able to read this SD Card, therefore no hope for this strange format. Phil prepared a new application which allows to format SD Cards, you could beta test this one. Best Regards, Thorsten.

Should work now: we reached beta18c ;) Best Regards, Thorsten.

I uploaded another update (midibox_seq_v4_0beta18b) which fixes sporadic "FatFs" errors, e.g. after a preset pattern has been stored and a new pattern is selected. Best Regards, Thorsten.

Beta18 is now available: o the current patterns/mixer map/song/groove templates will be cleared before a new session is created. o last patterns now correctly selected after startup [/code] Best Regards, Thorsten.

I can reproduce this and will create an update soon Best Regards, Thorsten.

Beta17 is now available for download: o implemented "Sessions" concept. Now you can change between different setups stored on a single SD Card o Sessions can be opened/saved/created in the main menu (press EXIT button) o your old session isn't lost, it's just at the wrong place. To take over this session: - enter MSD mode (UTILITY->DISK) - create a backup of all *.V4 files on your PC (recommented for Beta17!!!) - create a /SESSIONS directory if it doesn't exist yet - create a session subdirectory (8 characters maximum) if you name it "DEFAULT", it will be loaded after startup! - copy following files: MBSEQ_B1.V4, MBSEQ_B2.V4, MBSEQ_B3.V4, MBSEQ_B4.V4, MBSEQ_S.V4, MBSEQ_G.V4, MBSEQ_C.V4 - create a second session subdirectory - copy following files: MBSEQ_B5.V4, MBSEQ_B6.V4, MBSEQ_B7.V4, MBSEQ_B8.V4, MBSEQ_S.V4, MBSEQ_G.V4, MBSEQ_C.V4 - rename these files to MBSEQ_B1.V4 to MBSEQ_B4.V4 - don't copy MBSEQ_HW.V4 (always read from root directory) - remove the copied *.V4 files from root directory, or move them to another directory for backup - unmount the SD Card on your computer - wait until MSD mode is disabled! - press EXIT, thereafter OPEN the session by pressing GP button 9 o changed default track configuration to 256 steps/4 parameter layers since it's more useful in most cases [/code] To give you a full overview I created a dedicated manual section for this topic: http://www.ucapps.de/midibox_seq_manual_sessions.html Other parts of the manual have to be overworked to cover the changes. Note that I haven't "stress tested" the code yet - means in other words: there could be bugs. I will do this in the next days (by creating some nice Sessions :)) Please help beta testing - if the sequencer should crash or behave strangely, please describe the scenario in a way which allows me to reproduce the issue. Best Regards, Thorsten.

I also assumed that this could be a software bug and quickly tried to reproduce this on my MB-6582 after the initial posting of jbartee, but without success (there are no artifacts with with the MB-6582 matrix configuration). The matrix control handling is (a bit) different on the original MBSID design, but I haven't found the time to test this with the latest firmware yet. It could take some days... Therefore I haven't replied to this posting yet. It would be helpful if somebody could confirm that he noticed exactly the same issue on his MBSID before I'm spending hours in searching for a bug which probably is related to something we don't know yet (e.g. failed firmware upload or invalid configuration - e.g. DOUT pins mapped to LED Matrix pins)? Best Regards, Thorsten. P.S.: and who buys me a beer? ;)

and me too... ;) The session concept is implemented, and I can't believe it myself: the sequencer is running pretty stable while a new session is created or stored into a new directory. Magic! :sorcerer: This means, that the sequencer doesn't has to be stopped, e.g. you can play a nice bridge while the copy operations are in progress. :) And changing between existing sessions doesn't lead to any hick-ups of course! :) Beta17 will be released tomorrow. Best Regards, Thorsten.

Yes, each bank stores 64 patterns, and I'm normaly using a separate bank for each group (e.g. one bank with drum patterns, one with bassline patterns and synth lines, one for filling the song, another for loopback sequences to transpose/arpeggiate other patterns) It's half finished, but especially since a mkdir() is available now, there is no blocking point anymore already available, just create /SYSEX/<mysynthfolder> Do you really want such a deep hierarchie? Just create three folders for your synth Have you already tried to navigate through the SysEx menu? It's really fast - I don't really see a need for subcategories! Not implemented yet Not implemented yet, but prepared (you are just quoting the concept that I wrote some times ago in a different thread ;)) Thats the way how it will work. It's already possible to send dumps by pushing a button, have you already tried this? Is it working at your side? How would you exactly like to edit such a list? Would a .txt file with references to the dumps be sufficient? I'm not sure if I will find the time and motivation to integrate such an editor into the firmware (no usecase for myself) Again: how exactly would you like to edit this? I don't see a clear vision how the UI should be realized. It would be interesting how many people would really work this way, so that I know if it is really worth the effort to integrate additional editors so that absolutely no computer is required. To repeat Strophlex' statement: > There are so many concepts in the SEQ V4 so I get a bit confused... and this is what many people are saying. Best Regards, Thorsten.

Please don't reformat the SD Card, it would be interesting if it is working with beta17. Debug messages: thats a bug, I will fix this in beta17 as well. Best Regards, Thorsten.

I guess that the SD Card access is working, but: FAT12 is unusual, I never tested this format with DOSFS. However, since MBSEQ will use a new FAT driver (FatFs), there is a certain chance that the card can be successfully read with beta17 Best Regards, Thorsten.

The migration to FatFs was successful, directories can be created and especially copy operations are working much faster now! :) Questions to the users: do you really need 8 banks, or would it be sufficient to work with 4 banks in a session? This would speed up the creation of a new session, resp. the copy operation of an existing session would work faster as well. Note that it will be possible to switch between sessions in zero-time! :) But creating/cloning a session will consume some time (e.g. copy: ca. 30 seconds, with 4 banks instead of 8 maybe 20 seconds), and the sequencer will run unstable while files are copied. Best Regards, Thorsten.

No, this isn't normal. What are the results of the (overworked) MIDI troubleshooting guide? -> http://www.ucapps.de/howto_debug_midi.html It would be especially interesting if the loopback is working on TEST MIDI2, and *not* working if the Rx/Tx loopback is removed. The results of the other tests would be interesting as well to get a full picture. Another hint: use MIOS Studio 2 - it works more reliable than the Java based version. Best Regards, Thorsten.

Looks very interesting! It's worth a try - not at least to compare the performance (the integrated caching mechanism will make things easier) Best Regards, Thorsten.

Great results! :) the MBSEQ application uses the status LED for a different purpose, but since it's even possible to read the directory, the communication between core and SD Card seems to be ok. Could you please try usb_mass_storage_device_v1_1.zip? This version uses exactly the same SD Card driver like the MBSEQ release. The latest changes have been added by Phil to support MMC and HCSD - although your card seems to be a common type, it seems that the changes solve previous access problems. If the MSD application still crashes, it will be very difficult to debug this remotely :-/ (even with a TCP/IP connection to the core ;) Best Regards, Thorsten.

Hi David, it would be interesting if the MBSEQ V4 application is able to read the SD Card, or if the core will be rebooted as well. In the most recent release (beta16) I integrated a command interface to access debugging functions from the MIOS32 terminal, which is part of MIOS Studio 2: Could you please try it? Just enter "help" to test that the communication is working, and "sdcard" to retrieve informations from the SD Card Best Regards, Thorsten.

A new release is available: MIDIboxSEQ V4.0beta16 ~~~~~~~~~~~~~~~~~~~~~ o MIDI->Section Control: Octaves which are not taken for section selection can now be optionally forwarded to a configurable MIDI port. This saves you from adding a keyboard splitter to your MIDI keyboard, if you want to play a MIDI instrument from an unused keyboard zone. o fixed bug in live recording function (notelength sometimes not recorded correctly) o MIOS terminal functions have been added. Just enter "help" in the terminal which has been integrated into MIOS Studio 2 o copy function now works correctly together with the section concept o Follow function now available as dedicated button and with key combination MENU+EXIT o the Follow Song option under UTILITY->Options has been removed. The selected follow mode won't be stored on SD Card anymore. Follow is deactivated after power-on and has to be manually enabled (e.g. by pressing MENU+EXIT) [/code] Plans for the next version: it will allow to store/restore the complete setup (banks, songs, configuration, grooves, etc...) into "session" directories. This will lead to a new way how to organize patterns. E.g., it won't be required anymore to remember the banks where you stored your favourite sequences from last week, or the week before, or 3 years ago - just give the session a name, and store your work into a separate directory. Switch between the directories (load different sessions) to restore the complete setup. There is currently only one blocking point to realize this concept: the FAT32 driver I'm using to access the SD Card (DOSFS) doesn't support "mkdir" - I will have to implement this by myself. Best Regards, Thorsten.

Yes, because only 64 patterns per bank are available. I will change this in the documentation I'm still unsure if I will take over this function into V4, since the handling is inconsistent. It's easier to select (and change) a parameter with the encoder There wasn't so much progress on this project lately. I will probably create a prototype on veroboard first, but before I'm starting with this, I have to be 100% sure about the additional buttons/LEDs. fixed in beta16 Btw.: I noticed a bug in the live recording function - it's fixed in beta16 as well Best Regards, Thorsten.

Hi *, MIOS Studio 2 is now available for download :) Thanks to all beta testers and especially to Phil who helped with the Windows adaption! Best Regards, Thorsten.

Hi David, It shouldn't flash - do you mean the same dimming effect which will be performed by the bootloader after power-on? This indicates that something resets the core. E.g. a short circuit once the card has been connected. After reset, and after the application has been restarted, the IO pin drivers will be enabled, and if this causes a short (-> power drop), the core will boot again. Period. Since you already checked the adapter, the issue could be located at the core side. E.g., check that J24 is connected to 3.3V position (not 5V!), check that R30 resistor arrays *not* mounted, check with a multimeter that there is no short between SI/SO/SC/RC pins Best Regards, Thorsten.

you got a link to the most recent MIOS Studio version, which isn't based on Java anymore and solves all Mac issues. Best Regards, Thorsten.

An EEPROM application based on this application note is already available under modules/eeprom Best Regards, Thorsten. P.S.: I gave you access to the programmers section of this forum.

Wow! Not only the construction, also your demonstration is really really good! I hope to see more - and that you share your patches, which are great as well :) Blogged Best Regards, Thorsten.