Re: Rewriting SSA. Is This A Chance For GNU/Linux?

On 4/8/25 7:59 AM, Farley Flud wrote:

On Mon, 07 Apr 2025 17:59:45 -0400, c186282 wrote:
 

>
    Indeed ! However ... probably COULD be done, it's
    a bunch of shifting values - input to some accts,
    calx ops, shift to other accts ....... lots and
    lots of rheostats ........
>

 How is conditional branching (e.g. an if-then-else statement)
to be implemented with analog circuits?  It cannot be
done.

   Ummmmmm ... a lightly-latching op amp maybe, "IF (v1) > (v2)
   THEN amp = ON" ?

Analog computers are good for modelling systems that are
described by differential equations.  Adders, differentiators,
and integrators can all be easily implemented with electronic
circuits.  But beyond differential equation sytems analog
computers are useless.
 The Norden bomb site of WWII wan an electro-mechanical
computer.  It's job was to calculate the trajectory of
a bomb released by an aircraft and the trajectory is described
by a differential equation.

   Note those kinds of systems require the use of
   something called 'dithering' ... essentially a
   small mechanical vibrating device. This would
   help overcome the natural 'stickiness' of parts
   on parts, making finer actions possible by
   adding just a little chaos. Some pure-electronics
   systems use a form of 'dithering' too.
https://www.planetanalog.com/can-adding-noise-actually-improve-system-performance/

One of my professors told a story about a common "analog"
practice among engineers of the past.  To calculate an integral,
which can be described as the area under a curve, they would plot
the curve on well made paper and then cut out (with scissors)
the plotted area and weigh it (on a lab balance).  The ratio
of the cut-out area with a unit area of paper would be the
value of the integral.  (Multi-dimensional integrals would
require carving blocks of balsa wood or a similar material.)

   That's impressively clever !

Of course it worked but today integration is easy to perform
to unlimited accuracy using digital means.

   Most things are ... but at the cost of great complexity
   and often power consumption. IF you can find ways to make
   use of 'natural calculations' they may be worth using,
   or at least including in the mostly-digital machine. If
   Ohm's law can do near-instant floating-point calx it
   MAY be easier to have a tiny circuit of a few resistors
   and then read it with an A/D converter than to do all
   the calx step by dozens/hundreds/thousands of digital
   instruction steps.