@csepp very carefully? presumably engineers and mathematicians could figure it out. as for axtual pruduction, most of these look like standard photolithography printing processes, and perhaps a rivet machine.

@csepp before computers and desk calculators, this is what a room full of “computers” looked like

http://ece.umich.edu/bicentennial/images/timeline/human-computers.jpg

@csepp but maybe i don’t understand your questoon

@zens I was thinking more about that "figure it out" part. Must be a lot of ancient forbidden trigonometry and stuff behind it. 😮

@csepp well, i don’t really know the answer to that, and i really wish i did because i wanna make my own.

the only intuition i can offer is the function must be transformed to turn non-linear relationships into linesr relationships by remapping the input or output density- e.g. you force the function to be linear by changing the distance between marks or renumbering them.

@csepp so the simplest example of this is the standard plain slide rule, that just uses a log scale for the marks, which changes the quadratic relationship between two multiples and a result- into a linear relationship.

@csepp a large number of the computers in the gallery are in fact, just simple multipliers with some semantic task specific labelling.

for more complicated formulae, i imagine the linearisation process would guide the design