c0ntr0l update

[toxaco]

I've almost finished building my first x0xb0x and have been trying to run the patched c0ntr0l

I couldn't get it to work on WinXp

Installed python25, wxpython, pyserial but when trying to run c0ntr0l got lot's of errors:

'Traceback (most recent call last):
File "C:\c0ntr0l\c0ntr0l.py", line 40, in <module>
import model
File "C:\c0ntr0l\model.py", line 3 in <module>
import serial
File "C:\Python25|Lib\site-packages\serial\__init__.py", line 18, in <module>
from serialwin32 import *
File "C:\Python25\Lib\site-packages\serial\serialwin32.py", line 9, in <module>
import win32file # The base COM port and file IO functions.
ImportError: No module named win32file

Got fed up trying so installed the same on OSX 5.6 and it worked !?!

Am i missing something obvious on XP ?

[eil]

Looks like your pyserial isn't working correctly. Google the last line of the traceback, there are some relevant-looking results.

[toxaco]

thanks eil,

just had to install pywin32 and it worked fine

i know nothing about python cheers

[alex_dubinsky]

I used py2exe to bundle this up into an executable. I think it'd be nice if links to the old c0ntr0l were removed and this became official c0ntr0l 1.1. I couldn't get this to attach to my post, so download it from here:

http://x0xsource.com/c0ntr0l_v1.0Eil.r10_Win32.zip

[alex_dubinsky]

It'd be great if someone could use py2app to create an OS X executable. Unfortunately I can't do it from Windows... If someone tries this and has trouble (dependency errors, etc.), PM me. I'd be glad to do this together.

[E:pp:ik]

I'm afraid I cant help out on that one but you, sir, are awesome for posting that. I just successfully backed up my patterns and uploaded Sokkos on COM7 after all this time.

THANK YOU!!!

[phono]

the exe package is great, tested and worked, however the octave down/up and double up arent saved in an eprom backup or copy&paste, if anyone could fix this then i would agree this should be the official next version

[alex_dubinsky]

Is the octave problem something that worked in 1.0 and got broken? I haven't studied this source code at all and in fact I'm not a python programmer, but I could try taking a look. If it was one of the patch changes that broke it, it might be easier to find. Eil.. sokkan... where are you guys!

Btw, in the big picture, I'm not a fan of how this whole computer-controlled thing got implemented. If I were to do it over, I'd throw out all the logic on the firmware side (saving 5KB) except for an interface to do the LD and ST instructions, and write all the logic host-side in C# (or python). This is actually what I did for the tester software I use. It'd be easier, more flexible, and you wouldn't have all these weird bugs (and the performance is actually decent). Storing and loading the EEPROM would be a very simple data transfer. In fact, I might just go and do that, although I'm not sure if I would take the time to re-implement the note-editing GUI.

Also, there is a deeper issue. When uploading firmware and manually re-checking it (see the firmware burning thread) I often discover bit-errors. Where are they coming from? I suspect the problem is in the RS-232 side, since the USB interface should have error-checking. I didn't really notice anything on my scope as to its hardware cause, but anyway. This should be dealt with in software. Ie, RS-232 parity bits have to be enabled and checked and retries done on failures. Firmwares, patterns, etc are getting screwed up right now as a result of this.

[phono]

Is the octave problem something that worked in 1.0 and got broken? I haven't studied this source code at all and in fact I'm not a python programmer, but I could try taking a look. If it was one of the patch changes that broke it, it might be easier to find. Eil.. sokkan... where are you guys!

afaik it never worked, it kind of makes eprom dumps almost useless since you loose all your octave shifts

[LivePsy]

Is the octave problem something that worked in 1.0 and got broken? I haven't studied this source code at all and in fact I'm not a python programmer, but I could try taking a look. If it was one of the patch changes that broke it, it might be easier to find. Eil.. sokkan... where are you guys!

afaik it never worked, it kind of makes eprom dumps almost useless since you loose all your octave shifts

Hasn't anyone noticed that eeprom dumps are useless in Windows because c0ntr0l reads and writes ascii files instead of binary? I thought I had lost 90 patterns after a backup and erase. Fortunately I figgered it out and could fix the backup file. Surely every windows user experiences this? Just for the record, edit the patternfile.py file and change the 'w' to 'wb' and 'r' to 'rb' in each open statement. And if you have a useless backup file, you should be able to find and delete '0D' if it is immediately in front of an '0A' - the entire 10 bank has the extra 0D eight times

As for the octaves, it seems fine to me. They are not dedicated bits like accent or slide so I can't see how it could restore the root note but not the octave. I got a complete restore over an erased bank so I am sure it all works.

Cheers,
B

[LapaPatrolBoat]

I'm having a bit of trouble getting c0ntr0l to run. whenever I open the program the window for it minimizes and when i try to open it i get a message saying "can't load image from 'resources/sixteenth.gif':file does not exist." what I really dont understand is the file "sixteenth.gif" is right there :S anyone else have this issue??

[LivePsy]

Does it work if you copy the sixtheenths.gif to up C0ntr0l (windows) folder?

[LapaPatrolBoat]

unfortunately no.

[buddax]

removed

[antto]

i hope this thread isn't dead
got an issue here.. i've never been able to backup the whole EEPROM to a file
at first i was using the official c0ntr0l app (found in the ladyada webpage)
recently i found this from the x0xsource website, got the "c0ntr0l_v1.0Eil.r10_Win32.zip" installed and i'm now using it all the time
it greatly improved the firmware update speed and it looks better than the stock app
but it still seems to have this issue, i know the EEPROM is 4KB and mine is filled with patterns (as i'm with my custom firmware which uses 2688 bytes now (it'll use probably the whole 4KB when finished) but the most i can get out of the "Backup EEPROM" function is up to ~2.3KB .. while most of the times it's 1.85KB (it varies for an unknown reason)

do any of you experience this problem?
i assume that the c0ntr0l app tries to extract the whole 4KB of data from the x0x, right?