Run Your Vehicles On Water (HHO)

Look on the web for, Geniac Computers

I think they date back to the 1950s.
cuhulin

Don Stauffer in Minnesota > wrote:
>On Mar 28, 9:55 am, "HLS" > wrote:
>
>> Analog computers were relatively simpler, I believe, and had some unique
>> applications.
>> Never used one in my branch of technology, although some were available on a
>> rental,
>> batch, or timeshare basis.
>
>Ah! I did get to play with analog computers at one time- late
>sixties. After a couple of years they were replaced as digital models
>got faster and larger. Our analog was a pretty big cabinet full of
>stuff, as big as some of the smaller digitals around then. But it was
>one of the more powerful units. This was at McDonnell Douglas. We
>were simulating advanced radar systems. Slow- took forever to
>simulate a few seconds of real time operation. I forget what the time
>scaling was, but as I say we only ran a few pulses at a time. I
>really liked using those things, though. Nobody talks about them
>anymore. They sure had a short lifetime!

Analogue computing systems lasted well into the early eighties in various
disguises, mostly for control systems rather than for doing batch
calculations. EAI was making hybrid digitally-controlled-analogue
computers much longer than you'd ever expect.

But if you want to see a work of art, open up the fuel injection control
unit in a 1972 Volvo. It really is a special-purpose analogue computer,
with gain blocks, differential and integration stages, and pulse forming
networks all made with op-amps and discrete components. There must be
a hundred potentiometer adjustments inside there too. Some guy in Sweden
probably knows what they all do; everyone else is afraid to touch them.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

On Fri, 28 Mar 2008 14:55:33 GMT, "HLS" > wrote:

>Mechanical "computers", like Enigma, of course were there,

Enigma was hardly a "computer" and it was electrical, rather than
mechanical. Colossus, part of the attempts to break those codes, was the
first truly electronic programable digital computer.

If you want sophisticated digital computers built out of mechanical
logic, look at Victorian railway signalling interlock.

>The first "electronic digital computers" I remember being touted were in the
>mid 40's or early 50's which used racks of relays and vacuum tubes, toroids,
>etc. Eniac was the name if I remember correctly.

Early, electronic, but non-programmable (in the sense we use the term
now). It was more like a fast calculator than a decision-maker..

"Andy Dingley" > wrote in message
...
> On Fri, 28 Mar 2008 14:55:33 GMT, "HLS" > wrote:
>
>>Mechanical "computers", like Enigma, of course were there,
>
> Enigma was hardly a "computer" and it was electrical, rather than
> mechanical. Colossus, part of the attempts to break those codes, was the
> first truly electronic programable digital computer.

I have only seen pictures of Enigma.. Thought they were mechanical.
It was, if I understand correctly, just a coding/decoding apparatus, but
I considered it a simple single purpose computer.

Andy Dingley > wrote:
>On Fri, 28 Mar 2008 14:55:33 GMT, "HLS" > wrote:
>
>>Mechanical "computers", like Enigma, of course were there,
>
>Enigma was hardly a "computer" and it was electrical, rather than
>mechanical. Colossus, part of the attempts to break those codes, was the
>first truly electronic programable digital computer.

Enigma was strictly a combinational device. Colossus was an iterative
device, but not really programmable.

The bombes that the US Navy built using the Colossus model are worth
looking at; the NSA museum in Maryland has one of them and it uses
gas-filled thyratrons for actual logic devices. They're very slow by
Eniac standards, but they're faster than the mechanical stuff that fed
them.

Today when we say "computer" we mean a finite automaton (ie. a Turing
machine with a fixed length tape). It loads an instruction from memory,
decodes the instruction, and does something to memory given the contents
of the instruction, which includes the ability to change the program counter
(that indicates where the next instruction is) conditionally.

I think the first "computer" by that definition was the Harvard Mark I
relay computer in the late 1930s. However, a modern programmer would have
a hard time with it since it used different storage for program and data.

(On the other hand, the 8048 and 8051 microcontrollers that you will see
in ECUs of the eighties also were Harvard-class machines with seperate
memories.)

>If you want sophisticated digital computers built out of mechanical
>logic, look at Victorian railway signalling interlock.
>
>>The first "electronic digital computers" I remember being touted were in the
>>mid 40's or early 50's which used racks of relays and vacuum tubes, toroids,
>>etc. Eniac was the name if I remember correctly.
>
>Early, electronic, but non-programmable (in the sense we use the term
>now). It was more like a fast calculator than a decision-maker..

Eniac was a full finite automaton. It could fetch, decode, and execute.
It was goofy in a lot of ways and it also had split instruction and data
storage, but it was the real thing and it didn't use any mechanical
components, just tubes. This made it MUCH faster.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

HLS > wrote:
>"Andy Dingley" > wrote in message
.. .
>> On Fri, 28 Mar 2008 14:55:33 GMT, "HLS" > wrote:
>>
>>>Mechanical "computers", like Enigma, of course were there,
>>
>> Enigma was hardly a "computer" and it was electrical, rather than
>> mechanical. Colossus, part of the attempts to break those codes, was the
>> first truly electronic programable digital computer.
>
>I have only seen pictures of Enigma.. Thought they were mechanical.
>It was, if I understand correctly, just a coding/decoding apparatus, but
>I considered it a simple single purpose computer.

It has mechanical and electrical components. And it is just a simple
transform. The Unix "crypt" command is basically modelled after the Enigma
encryption and the manual page for it is well worth reading for a good
idea of how it works inside.
--scott

Oh yes, and don't use "crypt" for anything you care about because it turns
out not be too hard to break...
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Andy Dingley wrote:
> Enigma was hardly a "computer" and it was electrical, rather than
> mechanical.

As far as I can tell, the only electrical thing about Enigma was the wire
paths that sent power to the light bulbs. Correct me if I'm wrong.

--
One meter, to within 0.0125% accuracy (off by just under .005 inches):
Three feet
Three inches
Three eights of an inch

clifto > wrote:
>Andy Dingley wrote:
>> Enigma was hardly a "computer" and it was electrical, rather than
>> mechanical.
>
>As far as I can tell, the only electrical thing about Enigma was the wire
>paths that sent power to the light bulbs. Correct me if I'm wrong.

That's true, but that doesn't make the electrical part any less important.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Scott Dorsey wrote:
> clifto > wrote:
>>Andy Dingley wrote:
>>> Enigma was hardly a "computer" and it was electrical, rather than
>>> mechanical.
>>
>>As far as I can tell, the only electrical thing about Enigma was the wire
>>paths that sent power to the light bulbs. Correct me if I'm wrong.
>
> That's true, but that doesn't make the electrical part any less important.

Dingley said it was "electrical, rather than mechanical". I had to protest
that it was mainly mechanical.

It could have been made without any electrical parts, presumably, but the
readout would have been massively complicated or really user-unfriendly.

--
One meter, to within 0.0125% accuracy (off by just under .005 inches):
Three feet
Three inches
Three eights of an inch

"clifto" > wrote in message
...
> Andy Dingley wrote:
>> Enigma was hardly a "computer" and it was electrical, rather than
>> mechanical.
>
> As far as I can tell, the only electrical thing about Enigma was the wire
> paths that sent power to the light bulbs. Correct me if I'm wrong.
>

That is what I thought too, but am amicable to being corrected in this case.