Analogue-sequencer

Right,

Following on from my previous post, i decided to test out dumping an old firmware file (that i did have) on to my p3, in order to
see if the problems id experienced previously were still in effect...

Symptoms:

Dumping to the p3 is pretty much not possible.  BAD DATA all the time, even at 500ms between blocks.

Now on boot im just getting CHK FAIL = BED5 (or whatever) and i then have to wait a about a minte for "bo" to appear first, followed
shortly after by "ot", and then a long time (like 5 mins or so) for "flash" to appear next to it.


Colin, you suggested at some point in the past (ive just checked the archive) checking that the electrolytic just to the left of the
cpu was of the correct type and orientation, and, yes, i have a 10uf 16v there orientated correctly.  The crystal is AEL18.000.  I
have also resoldered both parts, to ensure against dry(ish) joints.

Because i had the firmware sorted before in the end (just through trying and trying again, so far as i remember it) i kinda forgot
about the issues (hey, it was working, so...) but now we come to the whole bootloader issue, where i only get one chance, hence my
interest in getting it right before i start.

Now i dont know if this is a related problem, but just occasionally, in normal use, the p3 takes a similarly long time to boot.  To
the extent that just on one occasion i thought something was really wrong, and then looked again a bit later, and, yes, it had
booted.

Hmm... Looking at the device now, ive been waiting for "flash" to appear, whilst writing this mail, and, as of yet, it hasn't done
so :/  Maybe ill go make a cup of tea, and see what its saying when i get back?  Oh no, just as i said that, the "f" has appeared,
so it looks like it may be on its way (???)

Suggestions?

Julian

(long mail, but trying to be as descriptive as possible)

Julian,

I'd try the bootloader update on your machine.
If it screws up, I'll swap out the CPU for you.
If it works, it may well fix the problem with your current bootloader,
assuming the problem is related to corrupt code in the bootloader rather
than some hardware issue.
The only explanation I can think of for the slow rate of text appearing on
your display (other than a problem in the bootloader code) would be a
problem with the 'ready' bit that is read back from the display module by
the CPU. The routine that writes a string to the display writes one byte,
then has to wait for the display to signal it's ready for another. If
there's a problem with the connection for that bit, or a problem in the
display module itself, it could cause slow writing to the display. That
wouldn't explain why the checksum displays OK, then slows down at the
"bo..."

Cheers,
Colin f

Came back this evening, read your email, Colin, crossed fingers, turned on the p3, and ran that bootloader dump.

It went fine.  I loaded the new firmware.  That went fine too.

I then started messing about trying to get P3tools to work, to no avail.  Then tried messing about with midi ox, doing sysex dumps
and stuff, but, i must admit, im fairly lost here anyway.  I do know that p3 tools manages to both request and transmit a few units
(?) but then expires as the retries increase.

Then, just in the course of things, i turn on the p3, whilst holding func and page, and, sure enough, i get "Checksum = 20DF", and
little else.  Been waiting a while now, but it should show something else, like boot and flash, right?

Now i wont pretend i know what im talking about, but, could the length of time that i had the unit off for have made a
difference?  Some cap somewhere slowly discharging or something?

Somewhat confused here.  Never tried to dump off any sysex before from the p3, so i have nothing to go on, but things certainly
don't seem to be working as id expect them to, esp. considering i now seem unable to enter the firmware update screen...

Confused, i am...

Julian

Julian,

> Then, just in the course of things, i turn on the p3, whilst 
> holding func and page, and, sure enough, i get "Checksum = 20DF", and
> little else.  Been waiting a while now, but it should show 
> something else, like boot and flash, right?

If you've done the bootloader update, and it's still doing the same thing,
I'd suspect a problem in the PLED module, or the wiring between the CPU and
the module. It could be a CPU problem, but that doesn't seem likely since
the updated bootloader is in a different location to the previous one. But I
can swap out the CPU to 'eliminate it from your enquiries' if you like -
drop me a mail privately to arrange.

Cheers,
Colin f