feritova pamat

[Richard Balogh]

Este som cosi vygooglil k teme:

> Viete niekto ako sa to vyrabalo? Tie zeny co to vysivali museli mat 
> sakramensky dobre ruky...

Robili to napr. v IBM zeny rucne pod stereomikroskopmi
(Edwin D. Reilly: Milestones in Computer Science and Information Technology).

Takto:

http://vt100.net/docs/misc/core/matrix.jpg
http://vt100.net/docs/misc/core/threading.jpg
http://vt100.net/docs/misc/core/testing.jpg

Vyroba vraj vysla 1 USD / 1 bit. Dnes stoji 1USD / 1MB.

Jadierka samotne vyrabali najprv podomacky na upravenom lise na aspirin,
potom sa vyrabali takto:
http://ed-thelen.org/comp-hist/TheCompMusRep/TCMR-V07.html#Core

Zaujimave je, ze ferity s pravouhlou hystereznou sluckou objavili
Nemci cez druhu svetovu vojnu. Vzorky materialu sa dostali do USA
a tam niekoho napadlo, ze by sa to dalo pouzit v pocitacoch
(A History of Modern Computing  By Paul E. Ceruzzi).

Prve pocitace, kde sa pouzili boli znamy Eniac a Whirlwind.
Potom sa este chvilu pouzivali ako "cache" pre diernostitkove pamate :)

Zopar perliciek na zaver:

Stare obrazky
http://ed-thelen.org/comp-hist/TheCompMusRep/TCMR-V13-P13.jpg

137 strankovy manual k rozsirujucemu modulu feritovej pamate k pocitacu PDP-6:
http://www.bitsavers.org/pdf/dec/pdp6/DEC-06-H1BA_164core_Dec67.pdf

... forty thousand bit capacity would be sufficient, provided there was a
sufficiently large serial mechanical memory to back it, i.e., tape, drums
and later discs... (Von Neumann)

:)

Richard Balogh