Here is a video about simulating the air flow around an airfoil
by using a Telefunken RA 770 analog computer from 1966.
Is there any use of these machines today? (I mean the idea of analog computing)
https://www.youtube.com/watch?v=nP84Dv01y4A
http://www.analogmuseum.org/english/introduction/
Thomas
Analog computing
Moderator: oliveroxtoby
Forum rules
and Helpful information for the FEM forum
and Helpful information for the FEM forum
-
- Veteran
- Posts: 3129
- Joined: Sat May 20, 2017 12:06 pm
- Location: Germany
Re: Analog computing
Analog computers have the advantage of continuous representation of real numbers, and for some applications like integration the idea is intriguing to do it continuously. However, due to temperature or other shifts the precision is limited.
With digital computers it is possible to have arbitrary precision (though not unlimited) and the results are absolutely repeatable. So they can do far better in any practical application.
With digital computers it is possible to have arbitrary precision (though not unlimited) and the results are absolutely repeatable. So they can do far better in any practical application.
A Sketcher Lecture with in-depth information is available in English, auf Deutsch, en français, en español.
Re: Analog computing
Of course
Re: Analog computing
See PID controllers.
A long time ago in a distant galaxy, i used to build for power generation and control systems (coal/gas/nucl/desalinization) stringing together primitive math and logic functions with canned op-amps and so fourth.
That airfoil is quite impressive; the other examples seem to look like actively-biased lissajous pattern/deflections.
See https://ifac.papercept.net/conferences/ ... Web_3.html
As for "So they can do far better in any practical application"; I am 99.46467546754765476546457% sure this is not so!
- Analog lightspeed response
- Digital DAC Quantisation Error
- Form Factor: Analog size vs Digital's computing on the head of a pin and on and on...
However if given a choice, of course a digital computer solution is typically simpler to design, smaller, and cost effective!
A long time ago in a distant galaxy, i used to build for power generation and control systems (coal/gas/nucl/desalinization) stringing together primitive math and logic functions with canned op-amps and so fourth.
That airfoil is quite impressive; the other examples seem to look like actively-biased lissajous pattern/deflections.
See https://ifac.papercept.net/conferences/ ... Web_3.html
As for "So they can do far better in any practical application"; I am 99.46467546754765476546457% sure this is not so!
- Analog lightspeed response
- Digital DAC Quantisation Error
- Form Factor: Analog size vs Digital's computing on the head of a pin and on and on...
However if given a choice, of course a digital computer solution is typically simpler to design, smaller, and cost effective!
Re: Analog computing
You are right with these applications, although it is open to discussion if some electronic device can be called a computer or not, there is probably a large grey zone.bill wrote: ↑Sun Jun 24, 2018 11:24 am As for "So they can do far better in any practical application"; I am 99.46467546754765476546457% sure this is not so!
- Analog lightspeed response
- Digital DAC Quantisation Error
- Form Factor: Analog size vs Digital's computing on the head of a pin and on and on...
A Sketcher Lecture with in-depth information is available in English, auf Deutsch, en français, en español.