Electrostatic chip for storing digital info?

Main Page | Report this Page

Science Forum Index » Physics - Electromagnetic Forum » Electrostatic chip for storing digital info?

Page 2 of 3 Goto page Previous 1, 2, 3 Next

Author

Message

dmm

Posted: Fri May 18, 2007 7:25 am

Guest

On 17 May 2007 11:31:25 -0700, Benj <bjacoby@iwaynet.net> wrote:

Quote:

Radium wrote:
Is it possible to design a chip that stores information in the form of
static electricity? If so, what would be the drawbacks of such a chip?

EEPROMs and Flash devices, (as well as microcontrollers that have these
to store program or data) use capacitors and FETs to store charge, which
define the logic state of each bit. They've been manufactured for years.

Quote:

Dear Radium the Troll.

Here is what your ignorant clueless brain can't figure out!
It is totally impossible to store information in a computer once the
power is off!

Ever heard of non-volatile memory? Probably not.
Non-volatile - memory that can retain data for later retrieval once power has been removed.
Volatile - memory that looses the data held in its cells once power has been removed.

Examples of non-volatile memory:
1. Electrically Eraseable Programmable Read Only Memory
2. Flash memory
3. Hard disk drive
4. Magnetic tape storage.

Examples of non-volatile memory that cannot be written to during run time.
1. Read Only Memory
2. Erasable Programmable Read Only Memory

Examples of volatile memory:
1. Random Access Memory
2. Dynamic Random Access Memory
3. DDR Ram.

Data can get written to any of these devices to be stored for later retrieval
if they are a component of the computer/controller system.

Have you ever saved a file on your PC? Know where it gets written to? Come on, have a guess...

Quote:

ALL computers simply "forget" everything they know once they are
turned off.

Nonsense. Computers don't "know" anything. They are programmed with
a series of instructions to perform a series of tasks. Currently they don't
have any cognitive reasoning. Once power is removed from the volatile
memory (ie RAM, dynamic ram (and all it's permutations)) data will be lost.
Data will be retained in non-volatile memory except if power is lost when
writing data to the device.

Quote:

Thus, the only option is to either NEVER turn off your computer or to
use the

After a reset (ie power up), ALL processors look for a series of instructions,
starting at a certain location. A PC is no different.

Quote:

very CLEVER method of "pulling a computer up by it's bootstraps" known
as
"booting" for short. Because of the total impossibility of
constructing a memory that can
store information in ANY form, let alone using electrostatic energy,

Where do you think the "bootstrap" program is stored?

BTW, memory is NOT constructed by the actions of turning a computer or microprocessor on.

Memory is a section of either a microcontroller or an external device to a processor
that can be either volatile or non-volatile, and can have data written to it (or overwritten) for
later retrieval. Memory can also be permanent (ie non-volatile), that stores a program, or data.
Memory is an integral hardware part of any processor/controller system. It does not get constructed
by the application of power.

Quote:

you will find that EVERY
computer today takes 5 to 15 minutes to "boot" every time you turn it
on.

Booting a PC up takes a long time principally because of the requirement to retrieve massive
amounts of data (typically the Windows OS) from the hard drive and place it into volatile
memory, from where it is run. The bottleneck is typically the hard drive, where the data transfer
rate is slowest. Anyway, you should be getting a cup of coffee, or something stronger during the
time it takes for the PC to boot up. 8)

In a PC, if you copy a program into a ramdisk ( a reserved section of volatile memory that has been
configured to look like a disk drive) and run it from there, you'll find that the program will
run significantly faster that if the program was run from the hard drive.

You can store Windows and any other program in solid state non-volatile memory that would enable the
PC to take less time to boot up than from a hard disk. Mind you, the cost of 0.5G of flash is fairly high,
but is dropping significantly with newer and larger flash memory.

Quote:

I hope this gives you a clue!

Keep trying.

Quote:

Benj
(Actually IBM once made a computer that didn't "boot", but everyone
hated it and it didn't sell)

Guest

Posted: Fri May 18, 2007 9:25 am

In sci.physics.electromag Benj <bjacoby@iwaynet.net> wrote:

Quote:

hhc314@yahoo.com wrote:
Excellent points. I guess it is not surprising that kids today do no
have any concept about how computers operate, or how they boot
themselves when they are turned on. Good grief, it wasn't always that
way. Some older readers will remember when you had to toggle in the
boot codes from front panel switches, simply to have a paper tape
reader read the tape which held the program or operating system
instructions.

And I'm surprised that geezers today who once had to toggle in bytes,
STILL think that computers need to "boot" Duh, like it's ALWAYS been
done that way!!!

It took about 10 years later, circa 1975, for self-booting systems to
arrive on the scene, and another 10 years later (circa 1985) for the
first personal computers to arrive. Networking came much later, but I
forgot in what year for the PCs. Unix systems had it much earlier,
with their networks mail systems over dial-up lines. That where the
Internet started, but originally only selected computers that were
allowed on the network.

Yeah, "self-booting" now THAT is a real "advance"! It just goes to
prove my previous statement that there is NO way for computers to
remember anything once they are turned off! Hey, if they could then
why would they need to "boot" at all? As for "hibernate" please allow
me to point out that I already covered that under "never turn it off"!

PS; Dynamic RAM does not remember when turned off.

But there are serveral types of memory that do.

There have been computers built that simply stop when the power goes
off and pick up where they left off when the power returns.

Since this costs more and is of no particular use in general
computing, you will only find such machines in niche applications.

--
Jim Pennino

Remove .spam.sux to reply.

Benj

Posted: Fri May 18, 2007 12:56 pm

Guest

dmm wrote:

<snip very nice lecture on computer memory and structure>

Quote:

Did anyone ever tell you guys that you are REALLY satire impaired? You
all really need to get out of the office more! Go see a nice movie
(NOT sci-fi!)

I'll draw you all a picture. Doesn't it occur to you all that after
this huge lecture on the existence of non-volatile memory, that THEN
"explaining" to me how it is really, really necessary for a computer
to "boot" is, well, just a tad dumb? WAKE UP! And just WHY should
windoze OS have "massive" amounts of data? WHY should it be such a
resource hog? WHY does it need to be saved on a hard disk? Why should
I have to transfer it into the volatile RAM each time? Why should it
take time to see if in the last 10 seconds I've installed a couple of
additional hard drives or slammed a disk into the CDROM drive? Why in
checking on what I've done should it wait a few minutes to make sure
I've not installed anything new when it already KNOWS in microseconds
if something has changed?

And FINALLY JUST WHY should I follow your advice and agree that
"booting" is something so necessary that it simply MUST be done and
*I* should go get a cup of coffee and put up with it and shut up!
Here. Do this calculation on you little computers. Multiply all the
computers in America (especially all those personal computers used in
business) by the amount of time wasted "booting" and "formating" and
"Installing". I won't even get into Windoze features like "constant
reinstall". Just HOW much wasted work-time and creative time is
wasted in America by this dogma you all believe in, that all these
time-wasting things MUST be done!

Here's my serious question to you. Why is it computer people are
incapable of asking the SIMPLEST questions, let alone answering them?

Quote:

Keep trying.

Yeah, but I really think it's a lost cause!

Martin Hogbin

Posted: Fri May 18, 2007 1:42 pm

Guest

"Radium" <glucegen1@gmail.com> wrote in message news:1179353731.425069.304520@y80g2000hsf.googlegroups.com...

Quote:

Hi:

Is it possible to design a chip that stores information in the form of
static electricity? If so, what would be the drawbacks of such a chip?

Already done. UV EPROMs and Flash store information as static
charges for years.

Quote:

Advantages I see are the ability to withstand high-voltages without
burning. Touching the chip with a static-charges conductor [such as my
hand] will not damage the chip but would probably corrupt the
information.

The voltages used in such chips are low.

Martin Hogbin

The Ghost In The Machine

Posted: Fri May 18, 2007 10:29 pm

Guest

In sci.physics.relativity, jimp@specsol.spam.sux.com
<jimp@specsol.spam.sux.com>
wrote
on Wed, 16 May 2007 23:45:03 GMT
<78hsh4-t82.ln1@mail.specsol.com>:

Quote:

In sci.physics.electromag Radium <glucegen1@gmail.com> wrote:
Hi:

Is it possible to design a chip that stores information in the form of
static electricity? If so, what would be the drawbacks of such a chip?

CCD camera, bubble memory.

Flash RAM, EEPROMs/EEAPROMS, a few other things.

IINM, "bubble memory" uses/used magnetic domains, though
there might be an electrostatic variant flitting about.

Quote:

snip clueless babbling

--
#191, ewill3@earthlink.net
Useless C++ Programming Idea #11823822:
signal(SIGKILL, catchkill);

--
Posted via a free Usenet account from http://www.teranews.com

The Ghost In The Machine

Posted: Fri May 18, 2007 10:30 pm

Guest

In sci.physics.relativity, jimp@specsol.spam.sux.com
<jimp@specsol.spam.sux.com>
wrote
on Thu, 17 May 2007 18:45:03 GMT
<dckuh4-rg.ln1@mail.specsol.com>:

Quote:

In sci.physics.electromag Benj <bjacoby@iwaynet.net> wrote:

Radium wrote:
Is it possible to design a chip that stores information in the form of
static electricity? If so, what would be the drawbacks of such a chip?

Dear Radium the Troll.

Here is what your ignorant clueless brain can't figure out!
It is totally impossible to store information in a computer once the
power is off!

Ever heard of core memory?

Ferrule core memory used magnetic domains, not electrostatic,
but since the same force (electroweak) drives both, one might
allow it. :-)

--
#191, ewill3@earthlink.net
Useless C++ Programming Idea #11823822:
signal(SIGKILL, catchkill);

--
Posted via a free Usenet account from http://www.teranews.com

Guest

Posted: Fri May 18, 2007 11:35 pm

In sci.physics.electromag The Ghost In The Machine <ewill@sirius.tg00suus7038.net> wrote:

Quote:

In sci.physics.relativity, jimp@specsol.spam.sux.com
jimp@specsol.spam.sux.com
wrote
on Thu, 17 May 2007 18:45:03 GMT
dckuh4-rg.ln1@mail.specsol.com>:
In sci.physics.electromag Benj <bjacoby@iwaynet.net> wrote:

Radium wrote:
Is it possible to design a chip that stores information in the form of
static electricity? If so, what would be the drawbacks of such a chip?

Dear Radium the Troll.

Here is what your ignorant clueless brain can't figure out!
It is totally impossible to store information in a computer once the
power is off!

Ever heard of core memory?

Ferrule core memory used magnetic domains, not electrostatic,
but since the same force (electroweak) drives both, one might
allow it. Smile

Errr, the response was to "It is totally impossible to store information
in a computer once the power is off!"

--
Jim Pennino

Remove .spam.sux to reply.

dmm

Posted: Sat May 19, 2007 8:20 am

Guest

On 18 May 2007 10:56:03 -0700, Benj <bjacoby@iwaynet.net> wrote:

Quote:

dmm wrote:

snip very nice lecture on computer memory and structure

Why, thank you Benj.

Quote:

Is there any other genre? ;-)

Quote:

I'll draw you all a picture. Doesn't it occur to you all that after
this huge lecture on the existence of non-volatile memory, that THEN
"explaining" to me how it is really, really necessary for a computer

Well, you made an VAST overgeneralisation, which was corrected.

Quote:

to "boot" is, well, just a tad dumb? WAKE UP! And just WHY should
windoze OS have "massive" amounts of data? WHY should it be such a resource hog?

Er...it's the end result of Microsoft screwing around?
<vast generalisation there>

Quote:

WHY does it need to be saved on a hard disk?

Currently its the most cost effective way to permanently store large amounts of
digital data. Especially media files.

Quote:

Why should I have to transfer it into the volatile RAM each time?

You don't do it, the PC does it for you. It's the way PCs work. Linux does the same thing,
'cos programs run faster in RAM.

Quote:

It's the nature of the beast. Or in other words, the way Windows works.
It's a multi-threaded and multi-tasking environment, which basically means
that there's a shiatload of stuff going on at once, each of which needs the
focussed, intimate attention of the processor driving it all. The only way to
guarantee that this happens is that the processor allocates a small amount
of time for each process. The end result is that everything is slowed down.

Quote:

And FINALLY JUST WHY should I follow your advice and agree that
"booting" is something so necessary that it simply MUST be done and
*I* should go get a cup of coffee and put up with it and shut up!

Nobody's saying you have to follow my advice. Would you rather spend
your time waiting for something to happen, or would you feel your time was
better spent doing something enjoyable?

Quote:

Here. Do this calculation on you little computers. Multiply all the
computers in America (especially all those personal computers used in
business) by the amount of time wasted "booting" and "formating" and
"Installing". I won't even get into Windoze features like "constant
reinstall". Just HOW much wasted work-time and creative time is
wasted in America by this dogma you all believe in, that all these
time-wasting things MUST be done!

Jeez Benj. Chill out dude. You're going to bust a nut.

Mosiad Technologies has a new version of flash (HLNAND) that will pretty
much snuff the waiting time. 800Megabytes per second reading and
writing time. You could bung all your OS and programs into the flash
and essentially not have to wait for anything to happen.

Quote:

Here's my serious question to you. Why is it computer people are
incapable of asking the SIMPLEST questions, let alone answering them?

Simple questions sometimes have complex answers. And it's not restricted
to computing. Try asking a quantum physicist what REALLY is mass?

Quote:

Keep trying.

Yeah, but I really think it's a lost cause!

Wow Benj. I reckon you need to relax a bit and get out and smell the roses sometime...

Regards
David

The Ghost In The Machine

Posted: Sat May 19, 2007 1:35 pm

Guest

In sci.physics.relativity, jimp@specsol.spam.sux.com
<jimp@specsol.spam.sux.com>
wrote
on Sat, 19 May 2007 04:35:02 GMT
<pva2i4-6ih.ln1@mail.specsol.com>:

Quote:

In sci.physics.electromag The Ghost In The Machine <ewill@sirius.tg00suus7038.net> wrote:
In sci.physics.relativity, jimp@specsol.spam.sux.com
jimp@specsol.spam.sux.com
wrote
on Thu, 17 May 2007 18:45:03 GMT
dckuh4-rg.ln1@mail.specsol.com>:
In sci.physics.electromag Benj <bjacoby@iwaynet.net> wrote:

Radium wrote:
Is it possible to design a chip that stores information in the form of
static electricity? If so, what would be the drawbacks of such a chip?

Dear Radium the Troll.

Here is what your ignorant clueless brain can't figure out!
It is totally impossible to store information in a computer once the
power is off!

Ever heard of core memory?

Ferrule core memory used magnetic domains, not electrostatic,
but since the same force (electroweak) drives both, one might
allow it. :-)

Errr, the response was to "It is totally impossible to store information
in a computer once the power is off!"

I'll definitely allow that. Smile

Of course, a small battery will take
care of SRAM in that case -- and flash doesn't even need that.

Perhaps Radium forgot his history...

--
#191, ewill3@earthlink.net
Useless C++ Programming Idea #10239993:
char * f(char *p) {char *q = malloc(strlen(p)); strcpy(q,p); return q; }

--
Posted via a free Usenet account from http://www.teranews.com

Guest

Posted: Sat May 19, 2007 3:55 pm

In sci.physics.electromag The Ghost In The Machine <ewill@sirius.tg00suus7038.net> wrote:

Quote:

In sci.physics.relativity, jimp@specsol.spam.sux.com
jimp@specsol.spam.sux.com
wrote
on Sat, 19 May 2007 04:35:02 GMT
pva2i4-6ih.ln1@mail.specsol.com>:
In sci.physics.electromag The Ghost In The Machine <ewill@sirius.tg00suus7038.net> wrote:
In sci.physics.relativity, jimp@specsol.spam.sux.com
jimp@specsol.spam.sux.com
wrote
on Thu, 17 May 2007 18:45:03 GMT
dckuh4-rg.ln1@mail.specsol.com>:
In sci.physics.electromag Benj <bjacoby@iwaynet.net> wrote:

Radium wrote:
Is it possible to design a chip that stores information in the form of
static electricity? If so, what would be the drawbacks of such a chip?

Dear Radium the Troll.

Here is what your ignorant clueless brain can't figure out!
It is totally impossible to store information in a computer once the
power is off!

Ever heard of core memory?

Ferrule core memory used magnetic domains, not electrostatic,
but since the same force (electroweak) drives both, one might
allow it. :-)

Errr, the response was to "It is totally impossible to store information
in a computer once the power is off!"

I'll definitely allow that. Smile

Of course, a small battery will take
care of SRAM in that case -- and flash doesn't even need that.

Perhaps Radium forgot his history...

Radium doesn't know any history.

The nly reason I mentioned core was because the technology is about
a half century old now.

--
Jim Pennino

Remove .spam.sux to reply.

Rich

Posted: Mon May 21, 2007 9:26 am

Guest

The Ghost In The Machine wrote:

Quote:

In sci.physics.relativity, jimp@specsol.spam.sux.com
jimp@specsol.spam.sux.com
wrote
on Sat, 19 May 2007 04:35:02 GMT
pva2i4-6ih.ln1@mail.specsol.com>:
In sci.physics.electromag The Ghost In The Machine <ewill@sirius.tg00suus7038.net> wrote:
In sci.physics.relativity, jimp@specsol.spam.sux.com
jimp@specsol.spam.sux.com
wrote
on Thu, 17 May 2007 18:45:03 GMT
dckuh4-rg.ln1@mail.specsol.com>:
In sci.physics.electromag Benj <bjacoby@iwaynet.net> wrote:

Radium wrote:
Is it possible to design a chip that stores information in the form of
static electricity? If so, what would be the drawbacks of such a chip?
Dear Radium the Troll.
Here is what your ignorant clueless brain can't figure out!
It is totally impossible to store information in a computer once the
power is off!
Ever heard of core memory?

Ferrule core memory used magnetic domains, not electrostatic,
but since the same force (electroweak) drives both, one might
allow it. Smile

Errr, the response was to "It is totally impossible to store information
in a computer once the power is off!"

I'll definitely allow that. Smile

Of course, a small battery will take
care of SRAM in that case -- and flash doesn't even need that.

Personally I use a hard disk, but that's just me. :^/

Rich

Quote:

Perhaps Radium forgot his history...

Szczepan Bia³ek

Posted: Mon May 21, 2007 2:00 pm

Guest

"Martin Hogbin" <

Quote:

"Radium" <glucegen1@gmail.com> wrote in message
news:1179353731.425069.304520@y80g2000hsf.googlegroups.com...
Hi:

Is it possible to design a chip that stores information in the form of
static electricity? If so, what would be the drawbacks of such a chip?

Already done. UV EPROMs and Flash store information as static
charges for years.

Advantages I see are the ability to withstand high-voltages without
burning. Touching the chip with a static-charges conductor [such as my
hand] will not damage the chip but would probably corrupt the
information.

The voltages used in such chips are low.

In such chips the electrons (static charges) are shoot inside an isolator.
Am I right?
S*

Benj

Posted: Tue May 22, 2007 2:16 am

Guest

dmm wrote:

Quote:

to "boot" is, well, just a tad dumb? WAKE UP! And just WHY should
windoze OS have "massive" amounts of data? WHY should it be such a resource hog?

Er...it's the end result of Microsoft screwing around?
vast generalisation there

So if Microsoft creates a dog because they are "screwing around" then
that's just fine with you, right? You got nothing better to do than
spent lots of time "screwing around" yourself because Microsoft told
you to, right? But I suppose all that wait time is useful for you as
it allows you to thumb through the pages of Wired looking at all the
over-priced stylish crap to spend your excessive salary on.

Quote:

WHY does it need to be saved on a hard disk?

Currently its the most cost effective way to permanently store large amounts of
digital data. Especially media files.

Is an OS a media file? Gosh my old Radio Shack had a VERY usable word
processor that ran in 32k! Is MS office really THAT much more
effective in helping me do my work? Divide 32k into X gig to get
answer.

Quote:

Why should I have to transfer it into the volatile RAM each time?

You don't do it, the PC does it for you. It's the way PCs work. Linux does the same thing,
'cos programs run faster in RAM.

Hey Bunkie, If I PAID for it, *I* have done it! I guess if Linux does
the same the same thing that makes it OK. We won't note that some of
the same people responsible for Linux are also responsible for that
Windoze trash! Dogma is dogma no matter who does it.

Quote:

Why should it take time to see if in the last 10 seconds I've installed a couple of
additional hard drives or slammed a disk into the CDROM drive? Why in
checking on what I've done should it wait a few minutes to make sure
I've not installed anything new when it already KNOWS in microseconds
if something has changed?

It's the nature of the beast. Or in other words, the way Windows works.
It's a multi-threaded and multi-tasking environment, which basically means
that there's a shiatload of stuff going on at once, each of which needs the
focussed, intimate attention of the processor driving it all. The only way to
guarantee that this happens is that the processor allocates a small amount
of time for each process. The end result is that everything is slowed down.

Oh, I'm sorry. I didn't understand this. Somehow my operating system
needs to have a SHITLOAD of stuff going on because, well, because
everyone agrees it's the nature of the beast! I understand
programmers are simply lazy-asses because it's the nature of the
beast! Ok. I agree! Nothing needs to change. Everything is just FINE
the way it is now!
You planning on running for president? You've GOT what it takes!

Quote:

And FINALLY JUST WHY should I follow your advice and agree that
"booting" is something so necessary that it simply MUST be done and
*I* should go get a cup of coffee and put up with it and shut up!

Nobody's saying you have to follow my advice. Would you rather spend
your time waiting for something to happen, or would you feel your time was
better spent doing something enjoyable?

What do you find enjoyable? Reading Ads in Wired? Surfing Internet on
company time? Making up bogus theories as to why it's impossible to
build an automobile that can start without a crank? My theory,
totally bogus as it is, happens to be that I'd rather SPEND MY TIME
DOING WORK ON MY COMPUTER RATHER THAN WAITING FOR IT TO GET AROUND TO
DOING SOMETHING!!! Do you have that through your thick skull yet?
I can see that YOU are a totally lost cause. My guess is that you
should report ME at once to the inquisition.

Quote:

Yeah, man. Can you loan me a few back issues of Wired to read? I just
got a ton of money for writing a totally shitty program and need to
spend it... Yeah, that'll be soothing...

Quote:

Mosiad Technologies has a new version of flash (HLNAND) that will pretty
much snuff the waiting time. 800Megabytes per second reading and
writing time. You could bung all your OS and programs into the flash
and essentially not have to wait for anything to happen.

And...And...and....?????? Oh, wait! This dude on the Internet told me
that my computer MUST boot. ALL computers MUST boot! "It's the nature
of the beast!" This almost sounds like I'm going to have to report YOU
to the inquisition for heresy!

Quote:

Here's my serious question to you. Why is it computer people are
incapable of asking the SIMPLEST questions, let alone answering them?

Simple questions sometimes have complex answers. And it's not restricted
to computing. Try asking a quantum physicist what REALLY is mass?

Or maybe try asking a programmer why his program totally SUCKS. He'll
probably answer just like you just did. The problem is that the
simple questions are the most important ones because they are most
fundamental. If you really don't understand mass, can you REALLY
understand quantum mechanics? No, you can't. Covering up that
ignorance with a bunch of complex math is like covering up programming
incompetence with a million lines of code. It's all fraud.

Quote:

Wow Benj. I reckon you need to relax a bit and get out and smell the roses sometime...

Yeah, I'm REALLY off base here. Anyway, that crank on your Lexus
helps keep you in shape, right? Yeah, my computer is reinstalling
it's OS again, so I've got lots of time now to wander outside....

Obviously anyone who does not blindly accept everything you say as
ultimate truth has a problem...right? How about this. Think of me as
one of your colleagues rather than as one of your students. Ok?

Benj

Benj

Posted: Wed May 23, 2007 1:50 am

Guest

Benj wrote:
<snip my previous rant>

Since this is a science cross-post rather than a computer engineering
discussion, let me just add a few thoughts about computer theory. And
by that I mean fundamental philosophy rather than binary mathematics
of calculations and the like.

A computer is an interesting device. It is a way to build a "machine"
by doing operations sequentially. And even better, this new device is
actually capable of emulating a great many different hyper-complex
machines by simply changing instructions as to what it does. Think
about this. In the mechanical age, you got some metal, a machinist
turned out parts and you put the gears etc. together and you ended up
with a "machine" that did something. One machine and "one" thing it
did, even though that "one thing" could at times be a plurality of
separate functions.

Suppose now you wanted to build a machine that took a million gears
and cams etc. to build. That is expensive and complex. But now we
invent another type of machine that uses just ONE gear and ONE cam but
has the ability to store what these two parts did so that the
operations of the 1,000,000 gears can be done sequentially on the TWO
parts! The second machine now perfectly emulates the first with MUCH
greater simplicity and a fractional cost, but the price you pay is
that it is much slower because the original did things in parallel
while the emulator does them in series. Today, if the gears are
electronic, doing a couple million operations a second is child's
play. Just imagine the mechanical monstrosity you are duplicating in
a fraction of a second!

So expanding that thought to a more modern computer system, we make
the observation that we have two choices. One is to sequentially
perform operations to construct a mathematical machine of enormous
complexity (if say it were built out of gates) OR we can say, hey,
look how clever I am! I figured out how to build a timer-counter
using software. The problem with the later is that our serial action
model is very ill-suited to these kinds of operations and emulating
such a device on your computer simply eats up all the available cycles
for no useful purpose. It is nothing more than a misapplication of a
very powerful principle, to an area where it is no longer powerful! In
other words you just used a multi-million gate circuit to emulate a
couple of integrated circuit chips!

Now lets think about making a computer as a truly useful machine
emulator. One fact you notice right off is that there is a human
psychological factor. If any operations takes less than say a half
second to complete they are viewed by the operator as instantaneous.
Irritation factor is nil. When time expands to several seconds, the
operator notes the device is "sluggish", when time expands to minutes,
irritation sets in and when it expands to hours, use gets restricted
ONLY to operations deemed so important as to be worth the wait.

Therefore, we can draw a conclusion here. A PROPERLY designed
computer system designed to emulate a variety of business machines,
should have NO waits to completion longer than a half second or maybe
a second if you push it. The idea is that the machine emulations are
ALL waiting on the operator, NOT the other way round. When it is the
other way round, valuable work-time is being wasted and operator
irritation is being created, none of which are good for productivity.
You may say, hey, it's only a few minutes! But as the song goes in the
show "The Pajama Game" give it to me every hour, of every day, of
every week, of every month, of every year and... well it adds up!

The bottom line is that such an efficient design is NOT beyond any
current technology. The problem is that people are determined NOT to
think about these things. They are determined to emulate things in
software that the sequential nature of the machine is ill-suited for.
And worst of all, once ONE person designs something wrong, it's monkey-
see monkey-do time and the story goes round that somehow it always
MUST be done the way the first guy did it.

And of course as time goes on the situation just gets worse and
worse. There is always the programming question of should I write my
own code from scratch or should I just adapt this already working code
to do my job. Even though virtually everyone who programs knows that
either answer to both these questions almost always gives virtually
the same amount of work to final product, but the universal choice
(alas, even for me) is to adapt existing code because you see all that
work already done and are convinced that an adaptation can't be THAT
difficult, can it? Yes...it CAN! Plus it insures that YOUR version
will include all the bad design and mistakes of the original.

So what we need is for someone to say look at say Linux and say, hey,
that thing is hosed. It takes nearly a half hour to boot, it has all
manner of carryover from early days when people didn't know better,
and really needs to be started again from scratch, this time based on
sound philosophy and what we know now. But we look at all that code,
and we sigh and decide the job is just too big. I have decided the
job is too big. And what's worse, as time goes on the job only gets
BIGGER! Pretty soon the job will become so imposing that NOBODY will
have the balls to undertake starting over.

Thus, not just ME but the entire world is saddled with computer
systems wasting our precious time, hogging precious resources and
generally being some small fraction as effective as they might be with
the same identical (or close to it) hardware and a better design
philosophy. Anybody out there ready to change the world for the
better? Yeah, I thought not.

<end rant>

Goodbye.

Michael A. Terrell

Posted: Fri Jun 01, 2007 2:08 pm

Guest

Benj wrote:

Quote:

(Actually IBM once made a computer that didn't "boot", but everyone
hated it and it didn't sell)

You've never heard of an analog computer?

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

Page 2 of 3 Goto page Previous 1, 2, 3 Next All times are GMT - 5 Hours The time now is Sat Jul 05, 2008 6:30 pm Science Forum Index