max # of eeprom read cycles

[Brigman, Corley]

>So the EPROM's and EEPROM's are getting old...Sounds like the thing to
>do is remove the EPROM or EEPROM  (if socketed)  while it's still
>retaining  data, read the data on a PROM programmer, and store the data
>on a floppy disc for re-programming the memory chip when it gets flaky.

My question: How many synths will this really affect? Per the original
poster, 
it's "eeproms" that are suspect, not "eproms" (and definitely not "proms"), 
right? No synths I have use eeproms for the OS, they all use eproms. The
program
memory is usually battery-backed SRAM, from what i've seen? In fact, the
only 
synth I personally own (mostly hybrids so they're good candidates) is the 
Ensoniq ESQ-M, whose memory cartridges (not the synth itself!) are eeproms. 

every synth i've ever used had a way to store patches, either to tape or
over 
midi, so i doubt a prom reader is necessary.

A point of note: If actual software was run from eeprom, and this actually
was a
problem (since it can't be backed up from tape/midi), then consider this:
The 
average processor most of these have runs at 1Mhz. These processors had no 
cache, so fetched every instruction separately. Figuring 3
cycles/instruction 
(about average for the 8086 of the time, at least, I think), that's .3M 
instructions/second, or 300,000 read cycles/second. Someone threw out 10
million
read cycles as a limit; you'd hit this in less than 60 seconds...

corley brigman
intel corp.
corley.brigman at Intel.com

speaking for me, not for intel.