From bw@ashbysolutions.com Wed Dec 13 15:18:29 2017
Subject:Re: Digital Korg CX3 (no sound) using HX3 as controller

Hi Daniel,

I stand corrected.  I was never able to get it to work on one of my XP
machines.  The Korg updater was a very 'fussy' piece of software with
certain hardware combinations.  It also took a very long time to update
over MIDI, on the order of 15-20 minutes.  If anything happened during
the process -- power glitch, accidental disconnect of a cable, etc. --
it didn't have much in the way of recovery.

To be fair, it was never intended to be a DIY process.  The updater was
designed to be used by Korg Service as an in-house tool, and for
manufacturing.  Korg Canada posted it on their website, and later
removed it.  Korg USA convinced Korg Japan to release it as a goodwill
gesture to customers, but came to regret it.  In the early days, if one
took all 'reasonable' precautions and ended up with a bricked mainboard,
Korg was very gracious about replacing it FOR FREE.  In later years,
though, it meant an expensive repair. Today?  I don't even know if Korg
still has mainboards; the product was discontinued in 2007 or 2008.

I have no insight into the software internals, only experience from
personally performing 20-25 updates.  Half of those were on my personal
CX-3, as part of the beta test team; the others were for Ashby Solutions
customers and close friends.  With my personal PC system and these 10-15
CX-3 units, I had only one hiccup -- and this, fortunately, was after
the Sub processor was programmed.  But I saw a number of things that
made me concerned:

* The updater did not work with a number of popular USB-to-MIDI adapters.
* It did not work on *every* PC running the supported operating
systems.  We could not determine why.
* Even if the MIDI Test passed, it was possible to drop packets during
programming.  (The reliability increased if the test passed multiple
times.)
* In most cases, when the updater dropped a packet, it did NOT stop
the update, but continued to send data, which was ignored.
* There did not seem to be a method of retrying failed packets, etc.
built into the software.
* Even if the Sub processor was already current, the updater
re-flashed it on each new try.

Thus, I would have to categorize the updater is being designed for
limited use under controlled conditions.  There's certainly no reason
why one could not play around with the updater to try to obtain success
-- as long as it is recognized that there are a number of ways to fail,
and a common result is a 'bricked' CX-3.  It sounds like the OP would
like a working CX-3, but wouldn't necessarily give up the MIDI
controller aspects to get it.

Regards,

-BW

Bruce Wahler
Ashby Solutions.com™
bw@ashbysolutions.com
http://music.ashbysolutions.com
http://halfmoon-switch.com
http://ashboys.com
978.597.7008