VentureBeat: The death of disk? HDDs still have an important role to play

[jim bell]

 On Wednesday, September 4, 2019, 10:15:00 PM PDT, \0xDynamite <dreamingforward at gmail.com> wrote:
 
 
 >  The industry developed UV-erasable EPROM as a substitute, which allowed
> only the erasure of the entire memory chip, , and some early EEPROM.
> (Electrically erasable programmable Read-Only-Memory).Eventually
> "flash-EPROM" was developed.  Flash memory

| 
| 
| 
|  |  |

 |

 |
| 
|  | 
Flash memory

Flash memory is an electronic (solid-state) non-volatile computer storage medium that can be electrically erased...
 |

 |

 |





>Ah, I hadn't made the connection between FLASH drives and Flash
EEPROMS.  However,  this pushes the question back:  how would an
EEPROM do it?  Old PROMs used to burn fuses to maintain state, but how
to restore state, eh?

>I believe that one could make flash drives that start at all-1s and
then burn the 0s into the memory.  A small battery-powered device
could maintain the address of the next writable word, otherwise once
the device reached the end of memory, it would no longer be useful
for. writing.

These devices injected a tiny electric charge into a "floating gate", an otherwise-insulated conductor surrounded by an extremely good insulator.  https://en.wikipedia.org/wiki/Floating-gate_MOSFET

The level of insulation required was phenomenal:  In order to store a bit of data for many decades, an effective resistance of over 10*20 ohms would have been necessary.   If you assume a capacitance of one picofarad (10E(-12) farads) in parallel with 10**20 ohms, that's a time constant of 10**8 seconds, or 3 years.  
Erasing an EPROM with UV light activated electrons into 'conduction bands' that would not ordinarily be populated, and thus erase the entire device.  
The first EPROM I ever heard of was the Intel 1702 2 kilobit  (256 by 8 bits) device, of which I used one:  As the program store of the "Dyna-Micro" single-board computer that I built in 1977.  It was extremely difficult to program, requiring very high voltages.  The next EPROM I used was a 2708 device, which was far easier to program.
          Jim Bell

  
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/html
Size: 8158 bytes
Desc: not available
URL: <https://lists.cpunks.org/pipermail/cypherpunks/attachments/20190905/4fc0b6b9/attachment.txt>