From bruce@ashbysolutions.com Fri Nov 18 05:40:50 2005
Subject:Re: CX-3 version upgrade danger???

Hi Lou,

>I would differ with you just a bit. If the firmware can get messed up to the point a component has to be replaced, there IS a deficiency in the design (of the CX-3). I write firmware for a living and I know full well the problems you describe. Most (probably almost ALL) of them in the case of re-programming the CX-3 are the fault of something on the PC side of things, or at least external to the CX-3. But the CX-3 should be able to protect itself.
>
> The process the re-programming uses is a form of boot-loading. A small program called the bootstrap loader (a.k.a. boot-loader) is used to load instructions into a different area of memory where the normal program runs. If the firmware is correctly written, the boot-loader program is hardware protected and can NEVER be damaged. If that is true, which it apparently is not in the case of the CX-3, the bootloading process is permanently protected and can never be damaged; the unit can always be re-programmed. In the CX-3, I suspect the boot-loader program area is not protected and can be over-written. If that happens, you can no longer boot-load and the unit is toast.

I understand what you are saying. (I, too, write firmware for my living.) However: The CX-3 was NOT designed as a user-upgraded system. No promise was made in the CX-3 sales pitch about upgrade capability. I don't know whether the boot loader can't be separated on the ICs that Korg used, or the firmware wasn't written that way, but the bottom line is that once the Sub Processor starts erasing, the boot loader is gone.

There were a lot of improvements between the CX-3 (v1) and the BX-3/CX-3 v2, and Korg USA lobbied the parent company to be allowed to release the upgrade to CX-3 v1 owners. In retrospect, I'll bet that they wish they didn't -- no good deed goes unpunished, right? -- but clonewheel owners as a whole benefited from the release.

It's a situation where the process seemed to work flawlessly during testing, but after release some problems were seen. Because of the lack of a permanent boot loader, the risk was serious. The number of failed updates was very small, but if you were one of those unlucky few, I'm sure it was no comfort to know this. Check the archives from about three years back, and you can read all about the problems.

Regards,

-BW

--
Bruce Wahler
Design Consultant
Ashby Solutions™ http://consult.ashbysolutions.com
978.386.7389 voice/fax
bruce@ashbysolutions.com

[Non-text portions of this message have been removed]