Updating MegaDrum firmware via bootloader with MIDI-OX

[dmitri]

Ken, does it bother you that 4 times out of 3 it ends up understood only by you what you've tried to say or explain?

[Ken Forgettable]

You mega644 settings *are* wrong - check.

[dmitri]

You mega644 settings *are* wrong - check.

Good! After all you *can* tell things so that others understand it.

[Ken Forgettable]

XYZZY

[jman 31]

Ken, does it bother you that 4 times out of 3 it ends up understood only by you what you've tried to say or explain?

Glad it's not just me. I was starting to think that there was stuff about this megadrum that was just not meant for me to understand!

[Ken Forgettable]

stuff about this megadrum that was just not meant for me to understand!

How crap ATMELs tools and documentation is; that it should take three months and 21 countries to make sense of it. Microchip are way ahead here but the guys who implemented that architecture have only just retired leaving the PIC to emerge from mists of time (they'd be dead if they hadn't got the credit card market). We'd all be much better off and have much more fun if we got into mCUs with some true pedigree.
"... Hic".

[elrules]

Could it be possible,dmitri, as you asked in another thread, that sending the "erase" sysex and the firmware inmediately (or 100ms, as in MCT) after, does not let enough time to MD to erase it?

[dmitri]

Could it be possible,dmitri, as you asked in another thread, that sending the "erase" sysex and the firmware inmediately (or 100ms, as in MCT) after, does not let enough time to MD to erase it?

MegaDrum bootloader has nothing to do with it. When "secure" version of the bootloader is selected in MCT it fails to send a firmware Sysex to any MIDI port. Even to virtual MIDI ports. Try it.

[Ken Forgettable]

This is from the bootloader source at the beginning of this thread.

Code: Select all

#define UART_CTRL2_DATA   ((1<<UMSEL01) | (1<<UCSZ01) | (1<<UCSZ00))

Shouldn't we be using asynchronous (8, n, 1 @ 31250baud) comm's
If so the code should read;

Code: Select all

#define UART_CTRL2_DATA   ((1<<UCSZ01) | (1<<UCSZ00))

[dmitri]

This is from the bootloader source at the beginning of this thread.

Code: Select all

#define UART_CTRL2_DATA   ((1<<UMSEL01) | (1<<UCSZ01) | (1<<UCSZ00))

Shouldn't we be using asynchronous (8, n, 1 @ 31250baud) comm's
If so the code should read;

Code: Select all

#define UART_CTRL2_DATA   ((1<<UCSZ01) | (1<<UCSZ00))

Yes, it should.