On 2022-Apr-01, at 5:54 AM, Paul Koning via cctalk wrote: >> On Apr 1, 2022, at 2:56 AM, Mark Huffstutter via cctalk >> <cctalk@classiccmp.org> wrote: >> >> https://archive.org/details/TNM_Glass_computer_memories_-_Corning_Electronics_20171206_0185 > ... > That Corning document is also interesting because of its comparison of memory > technologies it shows. Tunnel diode memories? Hm. And cryogenic, in 1962? > Hm again.
Are you alluding to a question of actual use in a practical large-scale implementation? Tunnel diodes did work as storage (state-holding elements), at least on a small scale. I have a frequency counter that uses a 5-stage tunnel-diode counter for the first high-speed counting stage. That is the most tunnel diodes I've seen in use in one place. As two-terminal negative-resistance devices, I wonder if there were some design attempts to put them in a matrix, something along the lines of providing the matrix axes in whole with a 'holding current' to retain state, along with 2D addressed R/W. Cryotrons I only know of from reading the period 'laboratory announcements', don't know how far they got in any practical use.
