Analog CPU ??????????????
Fuboco
I was just handed a photocopy of an article about an analog CPU
replacement for the Pentium II's? Is this real or is this just a joke.
The article came from 'Electronics Now April 99' and claimed that an
analog computer CPU existed (Ecraf Techn.) and is pin compatible with
the PII's and that it was a 4000 time improvement over the PII with a
ridiculus 900000 MHz clock rate. Using a standard ASUS motherboard and
analog RAM (4G from Diamondback electronics) and standard 4G Harddrive
(with the analog system it was increased to 1600G). So does this exist ?
I did a prelimanry web search and found no info on either Ecraf Tech or
Diamondback elec. So is this a big April fool's joke ala Syd Finch in SI
(couple of years ago).
Thanks
charles copeland
Brian Jonathan Lee
so that it could be sent over modem using an Apple II in Radio Electronics
(who eventually became Electronics Now). All an April fool's joke :)
In article <charles1F...@netcom.com>,
Tim Weiler
news:19990329171942...@ng68.aol.com...
> Hi
>
> I was just handed a photocopy of an article about an analog CPU
> replacement for the Pentium II's? Is this real or is this just a joke.
> The article came from 'Electronics Now April 99' and claimed that an
> analog computer CPU existed (Ecraf Techn.) and is pin compatible with
> the PII's and that it was a 4000 time improvement over the PII with a
> ridiculus 900000 MHz clock rate. Using a standard ASUS motherboard and
> analog RAM (4G from Diamondback electronics) and standard 4G Harddrive
> (with the analog system it was increased to 1600G). So does this exist ?
> I did a prelimanry web search and found no info on either Ecraf Tech or
> Diamondback elec. So is this a big April fool's joke ala Syd Finch in SI
> (couple of years ago).
>
> Thanks
>
I read the same article last week, and it all sounded pretty real and
exciting, however, if it is just an April fool's day joke, I will be
severely disappointed because 'Electronics Now' published the story, and
secondly because this kind of technology seems promising.
David Feustel
so I scanned the article looking for a clue.
The last sentence in the article states that
the chip is scheduled for delivery on April 1st.
So it's an April Fools joke as far as I'm concerned.
--
David Feustel
Fort Wayne, Indiana
219-483-1857
Ian St John
David Feustel <feu...@ix.netcom.com> wrote in message
news:37000ED9...@ix.netcom.com...
> This article had me going when I read it earlier today.
> so I scanned the article looking for a clue.
> The last sentence in the article states that
> the chip is scheduled for delivery on April 1st.
> So it's an April Fools joke as far as I'm concerned.
Didn't it clue you in when the junctions would run for weeks on aa
batteries, while the photon radiation was so great that it blinded
people through opaque packaging? Heard of the conservation of energy?
Curt McDowell
The picture showed a CRT strapped face-down to a copier machine :-)
Curt
Brian Jonathan Lee wrote in message
<1999Mar29.1...@jarvis.cs.toronto.edu>...
Lewin A.R.W. Edwards
Ecraf = Farce.
--
-- Lewin A.R.W. Edwards <http://www.zws.com/>
Realtime/Embedded Programmer & Embedded HW Eng
You'll catch more programmers with donuts than with bagels.
Julian Gardner
faster the hotter it got and the faster it ran the hotter it got ad
infinitum. The article was by a Lirpa Loof ( april fool)
joolz
Fuboco wrote:
> Hi
>
> I was just handed a photocopy of an article about an analog CPU
> replacement for the Pentium II's? Is this real or is this just a joke.
> The article came from 'Electronics Now April 99' and claimed that an
> analog computer CPU existed (Ecraf Techn.) and is pin compatible with
Michael A. Covington
---------------------------------------------------------------------------
Thank you for your correspondence. The article to which you refer was
our April Fool's Day joke, as indicated by the date of April 1 appearing
in it. There is no such thing as an analog microprocessor.
Occasional spoofs are a tradition at Electronics Now, going back to our
founder, Hugo Gernsback, who published facetious items under the name
"M. U. Fips."
You can contact the magazine's editorial office at cla...@gernsback.com.
Yours sincerely,
Michael A. Covington
Q&A Columnist, Electronics Now Magazine
---------------------------------------------------------------------------
A lot of people sent really angry letters. On the other hand, we had one
fellow ready to invest $50,000... Note that ECRAF spelled backward is FARCE,
and that the device claims physically impossible things such as enormous
light output.
I was not involved in publishing the article. However, as "Q&A" columnist I
had to field most of the inquiries.
Mark Rogers
of humor. How can you get through life, or even just a marriage without it?
How can you do debug an embedded design without it (especially in the
dreaded software hardware integration phase).
A few years ago, at my company, all the icons for a custom file type in our
desktop publishing software turned into fish on April 1st. People were
freaking out pretty bad, but it wasn't long before our IS department tracked
down the cause and issued a company wide email. It turns out that the
custom application was written and maintained by our parent company in
France, and fish are a big part of there Aril fool's day tradition. A few
people with no sense of humor caused them to change the application so that
it would not do that in the US. I was really embarrassed and ashamed. I
could only imagine what horrible stick in the mud's the French must think we
are. They probably talk worse about us then they do the English now.
Mark
Michael A. Covington <covi...@mindspring.com> wrote in message
news:7dqo0h$uo7$1...@camel19.mindspring.com...
D. Henkel-Wallace
<curt.m...@worldnet.att.net> wrote:
A few April 4th's ago they had a project for the "Poor Man's Laser
Printer."
The picture showed a CRT strapped face-down to a copier machine :-)
Back in the CP/M days (late '70s) you could buy "printers" that were a
pad of actuators that you laid on the keyboard of a selectric
typewriter!
As Dave Barry would say: "I am not making this up." Check out any old
issues of Byte, Creative Computing, etc.
Stephen Maudsley
>
>I was just handed a photocopy of an article about an analog CPU
>replacement for the Pentium II's? Is this real or is this just a joke.
>The article came from 'Electronics Now April 99' and claimed that an
>analog computer CPU existed (Ecraf Techn.) and is pin compatible with
>the PII's and that it was a 4000 time improvement over the PII with a
>ridiculus 900000 MHz clock rate. Using a standard ASUS motherboard and
>analog RAM (4G from Diamondback electronics) and standard 4G Harddrive
>(with the analog system it was increased to 1600G). So does this exist
?
>I did a prelimanry web search and found no info on either Ecraf Tech or
>Diamondback elec. So is this a big April fool's joke ala Syd Finch in
SI
>(couple of years ago).
If you read the developers guide for WindowsAnalogue, a potentiometer
array for the low level APIs is being shipped so it must be true.
--
Stephen Maudsley mailto:Stephen....@esgem.com
Esgem Limited: embedded system design http://www.esgem.com
Tel: +44-1453-521626 Mobile: +44-370-810991
Personal pages: http://www.esgem.com/people/Stephen.Maudsley
Kenneth A. McIsaac
Jack Crenshaw had an article in a recent _Embedded Systems Programming_
(one of his wistful remember-the-old-days? pieces) that describes an
early microcomputer (for large values of "micro"--the word here means
single user, I think) that printed text by back-driving the keyboard.
Apparently they spooked the cleaning staff from time to time when the
"ghost" machine would start typing on itself in the middle of the night.
Spehro Pefhany
> Back in the CP/M days (late '70s) you could buy "printers" that were a
> pad of actuators that you laid on the keyboard of a selectric
> typewriter!
Yes, I bought one. It cost a lot of money for the time, and used a 6502 to
control the solenoids. Worked ok with Wordstar. The whole thing failed in
spectacular fashion burning up, IIRC the solenoids were only capable of
taking short pulses before they overheated.
> As Dave Barry would say: "I am not making this up." Check out any old
> issues of Byte, Creative Computing, etc.
Now try to explain what a "Selectric(tm)" was ;-)
--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany "The Journey is the reward"
sp...@interlog.com
Fax:(905) 271-9838 (small micro system devt hw/sw + mfg)
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Mike Silva
Spehro Pefhany wrote in message <7du7qu$lpg$2...@news.interlog.com>...
>the renowned D. Henkel-Wallace <gu...@zembu.spam_dissuader.com> wrote:
>
>> Back in the CP/M days (late '70s) you could buy "printers" that were a
>> pad of actuators that you laid on the keyboard of a selectric
>> typewriter!
>
>Yes, I bought one. It cost a lot of money for the time, and used a 6502 to
When I was still in school I worked part time for a guy building up an S-100
system (hands up if you remember). He wanted a printer but didn't want to
spend much so he bought a bare dot-matrix print mechanism. I wired the
print head and linefeed solenoid up to a power supply and some driver
transistors, and connected up the other bits (two end-of-travel sensors and
a perforated wheel / phototransistor setup for clocking the printhead
columns, IIRC), entered in a dot matrix font by hand, and by gum if I didn't
eventually get that thing working. While I was developing that software I
went through a *lot* of printhead pins as my code fired solenoids and for
whatever reason didn't immediately release them so the printhead could move.
I also spent a lot of time sanding the gouges out of the platten, if that's
the right word for the behind-the-paper part. Great fun and great embedded
experience!
Mike
Ray Irvine
>
> the renowned D. Henkel-Wallace <gu...@zembu.spam_dissuader.com> wrote:
>
> > Back in the CP/M days (late '70s) you could buy "printers" that were a
> > pad of actuators that you laid on the keyboard of a selectric
> > typewriter!
>
> Yes, I bought one. It cost a lot of money for the time, and used a 6502 to
> spectacular fashion burning up, IIRC the solenoids were only capable of
> taking short pulses before they overheated.
>
> > As Dave Barry would say: "I am not making this up." Check out any old
> > issues of Byte, Creative Computing, etc.
>
> Now try to explain what a "Selectric(tm)" was ;-)
>
Ah yes, the good old days. Around 1980/1981, the I/O version of the
Selectric was becoming available on the surplus market. These units
already had solenoids and had been used in printer/terminal
applications. Several of us went together and bought a bunch of these
to convert to a centronics type interface for our own use.
There were some interfaces sold for these type printers but they cost
more than we wanted to spend. We already had developed a Z80 based SBC
for another project, so I built the rest of the necessary hardware and
wrote the software. It seemed to work at first but after a few lines the
first printer self- destructed.
Those old Selectrics were slow and definately would not be rushed!
Stephen Maudsley
Mike Silva <mjs...@jps.net> wrote in message
news:7ducrc$1eo$1...@its.hooked.net...
> I was developing that software I
>went through a *lot* of printhead pins as my code fired solenoids and
for
>whatever reason didn't immediately release them so the printhead could
move.
>I also spent a lot of time sanding the gouges out of the platten, if
that's
>the right word for the behind-the-paper part.
So that's what embedded means.....
Øystein Brandt Kjelsen
> Now try to explain what a "Selectric(tm)" was ;-)
>
The IBM Selectric typewriter, or rather *the* typewriter everything else
was measured against.
It is (was) an electric ball-head type typewriter. Widely used, had nice
"fonts", a nice keyboard and was not too noisy.
However it was like a BMW, attractive, but too expensive.
So devices were made to enable printouts done with the most prestiogous
of fonts, and one did not have to adopt them to several different
keyboards..
Remember, this was in the age of 8-pin needle printers, and drum- and
chain type printers.
-Øystein
Øystein Brandt Kjelsen
> our April Fool's Day joke, as indicated by the date of April 1
> There is no such thing as an analog microprocessor.
>
> Yours sincerely,
> Michael A. Covington
Nearly nothing is impossible! Analog computers do exist!
I have made several applications where I have programmed
LM358 to add, subtract, integrate, derivates,sort signals,
and even do trig functions.
Øystein Brandt Kjelsen
Michael A. Covington
from.) But analog *microprocessors* (with memory, etc.) don't.
--
Michael A. Covington / AI Center / The University of Georgia
http://www.ai.uga.edu/~mc http://www.mindspring.com/~covington <><
Kevin D. Quitt
<covi...@mindspring.com> wrote:
>Yes, yes, we know analog computers do exist. (We know where op-amps come
>from.) But analog *microprocessors* (with memory, etc.) don't.
'Micro' is meaningless in terms of analog computers. I know whereof I speak -
my first years in computing were spent with analog computers. At the time, the
general consensus was that digital computers would have a very small utility
compared to analog computers. At the time, that was correct.
But analog computers do have memory, just nothing like RAM.
--
#include <standard.disclaimer>
_
Kevin D Quitt USA 91351-4454 96.37% of all statistics are made up
Per the FCA, this email address may not be added to any commercial mail list
EWMCo
But analog computers do have memory, just nothing like RAM.>>
Isn't the idea of an analog computer somewhat similar to fuzzy logic? Or some
of the biologically-based computers that have at least been thought of if not
actually built recently? More than one technological breakthrough has started
as a joke or a flight of fancy.
-Scott Burgess
Ra
Michael A. Covington wrote:
Don't some old naval battleships still use analog computers for their main
'cannons'?
I was a dog, not a squid, so I have no real clue what I'm talking about
here. ;)
Rufus V. Smith
Fuboco wrote in message <19990329171942...@ng68.aol.com>...
>Hi
>
>I was just handed a photocopy of an article about an analog CPU
>replacement for the Pentium II's? Is this real or is this just a joke.
>The article came from 'Electronics Now April 99' and claimed that an
>analog computer CPU existed (Ecraf Techn.) and is pin compatible with
>the PII's and that it was a 4000 time improvement over the PII with a
>ridiculus 900000 MHz clock rate. Using a standard ASUS motherboard and
>analog RAM (4G from Diamondback electronics) and standard 4G Harddrive
>(with the analog system it was increased to 1600G). So does this exist ?
>I did a prelimanry web search and found no info on either Ecraf Tech or
>Diamondback elec. So is this a big April fool's joke ala Syd Finch in SI
>(couple of years ago).
>
>Thanks
>
erikc
>|Yes, yes, we know analog computers do exist. (We know where op-amps come
>|from.) But analog *microprocessors* (with memory, etc.) don't.
Ever heard of a sample/hold circuit?
Erikc
_ _
.nnn0 / ) ( \ 0nnn.
( ) / ( ) \ ( )
\ ( ( ) ( ) ) /
\_) .uuu0 0uuu. (_/
Hard at work, but hardly working.
Christian Fundamentalism: The doctrine that there is
an absolutely powerful, infinitely knowledgeable,
universe spanning entity that is deeply obsessed with
my sex life.
Iain McCracken
> Selectric was becoming available on the surplus market. These units
> already had solenoids and had been used in printer/terminal
> applications. Several of us went together and bought a bunch of these
> to convert to a centronics type interface for our own use.
>
> There were some interfaces sold for these type printers but they cost
> more than we wanted to spend. We already had developed a Z80 based SBC
> for another project, so I built the rest of the necessary hardware and
> wrote the software. It seemed to work at first but after a few lines the
> first printer self- destructed.
>
> Those old Selectrics were slow and definately would not be rushed!
Hey, don't blame your interface too much. Those old I/O versions of the
Seletric were a little scary. I had one hooked up to a TRS-80 (Model I
with the expansion interface -- and we had the expansion interface only
for the printer -- <sigh> I would have died for 80K of storage on a
floppy). It had this interface box about 12-18 inches cubed which
converted the centronics-style interface to the Selectric interface. It
also had a standard silver toggle switch on the front which told it to
swap upper and lower case letters for the TRS-80, which had them
backwards, for some reason I once knew...
One day, I was printing a long file and the Selectric decided that it
had done enough printing, and proceeded to spit out all those little 3mm
steel ribbons at high speed. I remember being terrified my dad would
skin me alive for breaking his machine, but he'd had a secretary blow up
a regular Selectric by typing too fast for too long... The repair guy
just shrugged and said that "today's computers" just drive the thing too
fast -- they apparently couldn't handle more than about 60CPS for long
without causing the drive mechanism to fall off the rails. I wrote a
little BASIC ROM-extender to time the output, but the Selectric never
was fixed -- the repairs would have been more than the cost of the
typewriter. We got an Epson MX-80 and moved into the modern age. Much
quieter too -- we could actually carry on a conversation in the next
room with the MX-80 going :)
Iain McCracken
>
> Yes, yes, we know analog computers do exist. (We know where op-amps come
> from.) But analog *microprocessors* (with memory, etc.) don't.
>
> Michael A. Covington / AI Center / The University of Georgia
> http://www.ai.uga.edu/~mc http://www.mindspring.com/~covington <><
For all that are insisting that analog computers do exist, let's all
remember that yes they do exist, but they did not develop much after the
digital computer took over the world. Now, everything you can buy at
Egghead or Computer City is digital. To those who say "micro" is
irrelevant, let me substitute VLSI for that. Analogue computers are
built from oscillators, amplifiers, filters, etc, all of which require
capacitors and inductors. Large-value capacitors and inductors can't be
done in VLSI, so baseband (unmodulated) computing is impossible at that
scale.
But don't lose hope, all you analog fans! Tiny (ie
pico-farad/pico-henry and smaller) capacitors and inductors are possible
in VLSI (ie a pie-shaped trace is a cap, a spiral is an inductor, and a
bow-tie shaped trace can be used as a band-pass filter). The only thing
is that you need to drive this thing AC at frequencies up into the
microwave region.
No problem -- just modulate your data (FM, FSK, whatever) onto a carrier
in that range, and presto, you can do all sorts of things with it at the
VLSI scale! Especially if you use narrow-band filters to select out
information, and build your circuits from demodulators and so on...
Then, an analogue microprocessor would exist.
Analog ram would probably be constructed from resonant ring circuits...
Hmmm. I wonder if there's an RF electronics engineer and a CMOS VLSI
specialist out there who could team up and do this?
--Iain.
Øystein Brandt Kjelsen
But wow! 60CPS on a half mecanical typewriter ? !!!!
Øystein
I
d...@cutout.gci-net.com
>> There were some interfaces sold for these type printers but they cost
>> more than we wanted to spend. We already had developed a Z80 based SBC
>> for another project, so I built the rest of the necessary hardware and
>> wrote the software. It seemed to work at first but after a few lines the
>> first printer self- destructed.
>>
>> Those old Selectrics were slow and definately would not be rushed!
The I/O had sensors which would allow a *smart* interface to
anticipate "end of cycle" so the clutch would never have to disengage.
> Hey, don't blame your interface too much. Those old I/O versions of the
> Seletric were a little scary. I had one hooked up to a TRS-80 (Model I
> with the expansion interface -- and we had the expansion interface only
> for the printer -- <sigh> I would have died for 80K of storage on a
> floppy). It had this interface box about 12-18 inches cubed which
> converted the centronics-style interface to the Selectric interface. It
> also had a standard silver toggle switch on the front which told it to
> swap upper and lower case letters for the TRS-80, which had them
> backwards, for some reason I once knew...
> One day, I was printing a long file and the Selectric decided that it
> had done enough printing, and proceeded to spit out all those little 3mm
> steel ribbons at high speed. I remember being terrified my dad would
> skin me alive for breaking his machine, but he'd had a secretary blow up
> a regular Selectric by typing too fast for too long... The repair guy
> just shrugged and said that "today's computers" just drive the thing too
> fast -- they apparently couldn't handle more than about 60CPS for long
Actually, I think the Trendata machine (to which, I assume you
are refering) ran about *13* characters per second at the high end
(I can try to dig up my service manuals for the IBM "I/O" on which
it was based).
EPSCO had modified a conventional Selectric mechanism before IBM
introduced the "I/O"s. It was part of their Edityper product
(an early "word processor")
I keep an Edityper and a Trendata 11200 (?) just for the sake
of nostalgia... :-/
> without causing the drive mechanism to fall off the rails. I wrote a
> little BASIC ROM-extender to time the output, but the Selectric never
> was fixed -- the repairs would have been more than the cost of the
> typewriter. We got an Epson MX-80 and moved into the modern age. Much
> quieter too -- we could actually carry on a conversation in the next
> room with the MX-80 going :)
--don
Iain McCracken
>
> Iain McCracken wrote a very nice piece.
>
> But wow! 60CPS on a half mecanical typewriter ? !!!!
>
>
> I
I think I may have goofed there... Don mentions that his Trendata only
handles 13 CPS -- That's probably right. I recently got a copy of a
program called MESS which emulates the TRS-80 Model I very well. Along
with re-discovering all the old games from Leo Christopherson, Scott
Adams (not the Dilbert guy), and Big 5, I tried to LLIST a BASIC
program. It came out of the emulator at about 60-80 CPS, so my
nostalgia probably skipped a track there... :)
As for the type of machine it was, I can't remember. It had a silver
sticker where the Selectric logo goes -- the left half was silver, and
had the Selectric name on it, and the right half was black and had some
other name on it, so it maybe wasn't the IBM one. I do remember that
the actuator pads started getting heavy advertizing in Kilobaud
Microcomputing and 80 Microcomputing at the time we got it. (Where is
Wayne Green now?) The interface box had the same logo tag on it, and
was covered in the same IBM powder blue stucco as the typewriter.
After the typewriter broke, I took off the brown metal cover from the
bottom which housed the IO board. The board sprouted a rat's nest of
heavy wires from a morass of 50?volt resistors. The rocker arms from
the keys seemed unmodified -- a back panel that they fed into was where
the wires went. The keys jiggled a little as the thing tap danced across
the desk, but I don't know whether that was from the actuators or the
seismic activity...
Thanks for the memories, guys. Now I have to get back to that
newfangled 68360 thingy...
--Iain.
Ken Keys
Well Scott, stand by for a flood of replies from old farts like myself.
I had the joy of playing with analog computers in college. They were
collections of op-amps connected to a plugboard that could be
"programmed" with banana plugs and jumper wires. For certain classes of
problems involving the solution of differential equations, they were
much faster than digital computers. In fact, that is the origin of the
present use of the term "analog"
in referring to continuous linear circuits. The computer could solve
mathematical problems in a number of fields such as vibrations or
thermodynamics by creating an electrical circuit which was analogous to,
or was a mathematical analog of, the problem in question.
Sometimes specialized analog circuitry was referred to as a computer
especially in the area of missile and spacecraft guidance. That use of
the term was not unreasonable since they were about the same as a
general purpose analog computer with the plugboard replaced by fixed
interconnect wiring.
Back in the '70s a guy by the name of Von Daniken (sp?) become briefly
famous by writing a book called _Chariots of the Gods_. His thesis was
that the ancient Egyptians, Incas and Mayans were too stupid to have
accomplished everything that they did and that they must have had help
from extra-terrestrials. I wonder if some future archeologist will
discover analog computers and the roll that they played in putting man
on the moon and come to much the same conclusion.
KLK
Mike Silva
Ken Keys wrote in message <3708E745...@west.raytheon.com>...
>EWMCo wrote:
>
>Well Scott, stand by for a flood of replies from old farts like myself.
>I had the joy of playing with analog computers in college. They were
>collections of op-amps connected to a plugboard that could be
Reaching -way- back, as a kid I would get Heathkit catalogs and on my list
of kits to buy "someday" was their analog computer. All the meters and
knobs and connectors made it simply irresistable. Didn't know what I'd do
with it, other than that it was a "computer".
Mike
vi...@sonydnse.com
Iain McCracken <iain.mc...@sympatico.ca> wrote:
> It
> also had a standard silver toggle switch on the front which told it to
> swap upper and lower case letters for the TRS-80, which had them
> backwards, for some reason I once knew...
IIRC, the TRS-80 Model 1s were shipped with a bit missing from the video RAM
so that there was only one case, even though most of the character generators
suppoted both. I believe that bit line was pulled high (my working knowledge
of the TRS has become somewhat shaky over the years...), which would leave
you with either a case swap, or everything in lower case. Most users added
the extra bit & enabled lower case soon after purchase.
> We got an Epson MX-80 and moved into the modern age. Much
> quieter too -- we could actually carry on a conversation in the next
> room with the MX-80 going :)
Wimps! I had a KSR-33 in my bedroom hammering out 10cps and a fine
impersonation of the Russian Army on manoeuvres!
Vic.
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Ulf Samuelsson
Iain McCracken skrev i meddelandet <37071408...@sympatico.ca>...
>Michael A. Covington wrote:
>>
>> Yes, yes, we know analog computers do exist. (We know where op-amps come
>> from.) But analog *microprocessors* (with memory, etc.) don't.
>>
>> --
>> Michael A. Covington / AI Center / The University of Georgia
>> http://www.ai.uga.edu/~mc http://www.mindspring.com/~covington <><
>
If I am not highly mistaken Intel has(had) a little known microprocessor
for neural networks which works with data represented as charges
over capacitors or in another analog form. Don't remember..
Saw this 3-5 years ago.
It is extremely expensive and probably in use by the military only.
--
This is a personal view which may or may not be shared
by my employer Atmel Sweden
Ulf Samuelsson ulf 'a't atmel 'd'o't com
rickp...@gmail.com
I remember reading the original article back then. Brings back memories.
john.l...@gmail.com
> Holy Shit!
> I remember reading the original article back then. Brings back memories.
c4urs11
Be prepared :-)
Cheers
George Neuner
>Holy Shit!
>I remember reading the original article back then. Brings back memories.
Regardless, it would be nice if you gave a reference to what you are
commenting on ... not everyone who reads C.A.E uses Google groups.
George
Tonton Th
--
------- http://weblog.mixart-myrys.org/?post/2016/02/THSF-V7 -------
I'm <tth> on freenode. Film at 11, take your popcorn.