[sdiy] Oberheim Matrix 6r ROM image needed

[rsdio at audiobanshee.com]

On Aug 20, 2014, at 6:22 AM, Rainer Buchty wrote:
> On Wed, 20 Aug 2014, Roman Sowa wrote:
>> Microchip claims 200 years data retention in their eproms.
> 
> But only when programmed with "authorized" programmers, which means something in the Data I/O range, not GALEP, and of course not some crude DIY programmer offering 50ms "slow but secure" and "program/read back until the bits stick" algorithms.
> 
> Rainer

There isn't much magic to the authorized programmers. I talked to Microchip about this when advising clients on how to manufacture with PIC firmware for boards that I designed and developed, and Microchip pointed out that it's important to check that the programmed values read back at all valid power supply voltages. So, if you have a PIC that will run at 5 V or 3.3 V or even lower, then your programmer needs to verify the data values at the full range of power voltages. It's really not much different from "program/read back until the bits stick," but with the added variable of supply voltage. I suppose you could design and build such a programmer yourself if you don't want to pay for Microchip's ~$300 programmer. The $30 PICkit devices just use whatever supply voltage is available on USB, and don't bother varying the supply because it's not really possible in such a bare-bones device.

For my vintage EEPROM programming, I first designed my own programmer on a hand-etched Apple II peripheral card back in the eighties. The programming voltage came from a raw transformer and regulator that I built, and the rest of the signals came from the Apple II bus. I remember that's when I learned to check that the power cord isn't plugged in to the wall when I'm soldering the leads on to the transformer! I refined the assembly driver code to take advantage of the "program until the bits stick" algorithm from the EEPROM data sheets as soon as I heard about it. Later, I got some company's properly manufactured EEPROM programmer for the Apple II, which was much easier to use than my own software and supported larger chips, but I noticed that their algorithm was slower than mine. Diving into the assembly language, I noticed that their printout of the programmed values was adding several minutes to the process. I hacked that to print the same amount of information, but in a way that only added a fraction of a second to the total time, and then continued to use their software with my mods.

Brian Willoughby
Sound Consulting