Discussion in 'Electronic Basics' started by Radium, May 2, 2007.

Hi:

Below is an example of "parallel Hz"

http://img56.imageshack.us/img56/2427/clocksignalexample8is.gif

If each clock signal is 1 Hz, and you have a billion of them,
staggered such that every 1ns part of the CPU can start, and finish,
an instruction - making the effective 'clock rate' 1 GHz.

The benefit of using a billion 1 Hz clock signals to make a clock rate
of 1Ghz is that such a system would not get as hot as system running
one 1 GHz clock signal . While the overall amount of heat generated by
both systems maybe around the same, the system running a billion 1 Hz
clock signals will have less heat per area than the system running one
1 billion Hz clock signal. Hence, the former system is far less
vulnerable to thermal damage than the latter.

Let's say two CPUs of different frequencies have been running at the
same voltages and amperages and for the same amount of time. The CPU
with a higher-frequency will be hotter than the CPU with a lower-
frequency.

In a "parallel Hz" device the bits maybe completely in serial and the
algorithms and tasks maybe totally non-parallelizable. However, the
frequency is still parallel.

The device I am proposing is completely serial except for the clock
rate.

My proposed device is completely serial except for the frequency. It
uses "parallel Hz" but in terms of everything other than frequency, it
is totally serial and non-parallel. Only the clock rate is parallel.

Parallel Hz = a method using N number of 1 Hz clock signals to gain a
clock rate of N Hz.

My design has a clock rate of 4 GHz that is obtained by using 4
billion 1 Hz clock signals. But otherwise, it is completely serial.

This design would go great for any application that cannot be
efficiently parallelized [in terms of bits]. Examples of such are
arithmetics and Boolean logic. Parallel Hz would work for serial-only
problems because the bits are still in serial. Parallel Hz does not
require that the bits be parallel.

There is a significant difference between "parallel Hz" and "parallel
bits".

A parallel printer is an example of a device that uses "parallel
bits". This has nothing to do with "parallel Hz" because both serial
and parallel devices can use parallel Hz.

Thanks,

2. ### Michael A. TerrellGuest

Bye:

--
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

3. ### Robert BaerGuest

If each processor drew a microwatt, a billion would draw (and
dissipate) a kilowatt.

4. ### David L. JonesGuest

That's hilarious.

Dave.

Hey idiot.

Graham

6. ### Ken HaganGuest

I note that actually producing a compiler that can exploit an ILP
of 10**9 is left as an exercise for the reader. That would be a 51
on the Knuth scale of difficulty, no?

8. ### Gary SevenGuest

: Hi:
:
: Below is an example of "parallel Hz"
:
<snip rant>

"Sir, there is a multi-legged creature crawling on your shoulder." [Spock,
c.1967]

9. ### DaveMGuest

LMAO!!!!!!!!!! And just how are you going to make a CPU with a billion separate
clock lines going into it, not to mention all the supporting inter-CPU
communication circuits? All those CPU "parts" need to be managed somehow... the
microcode to do that would require gigabytes of memory on the CPU. Heat would
be the least of your worries.

The OP needs to redirect his/her energy to earthbound reality and stop watching
so many StarTrek reruns.

--
Dave M
MasonDG44 at comcast dot net (Just substitute the appropriate characters in the

Life is like a roll of toilet paper; the closer to the end, the faster it goes.

10. ### Anthony FremontGuest

x-posting reduced

Hmmm....... not seeing Radium over here on Supernews. Don't see the flood
messages on Supernews either....... double hmmmmm........ Gee, I wonder if
there is a correlation?

12. ### Guest

"My design has a clock rate of 4 GHz that is obtained by using 4
billion 1 Hz clock signals. But otherwise, it is completely serial. "

Design? Design? All I see is a MSPAINT squiggle. Get a LED to flash
with a PIC first. It has a 4 level parallel Hz processor inside. This
will save you money, you just need to buy 250 million of them instead
of a billion. Makes the PCB a bit more manageable too.

No microcode necessary. My design is hard-wired.

quotes from http://en.wikipedia.org/wiki/Microcode :

"Each machine instruction (add, shift, move) was implemented directly
with circuitry. This provided fast performance, but as instruction
sets grew more complex, hard-wired instruction sets became more
difficult to design and debug."

I still prefer the "hard-wired instruction sets"

"a bug could often be fixed by replacing a portion of the microprogram
rather than by changes being made to hardware logic and wiring."

But I still prefer the "hardware logic and wiring".

In addition, I like my device to be set to the lowest gear -- i.e. a
max of 1-bit per cycle.

Here are some analogies of gears, CPU clock rate, and RPM :

http://sound-on-sound2.infopop.net/2/OpenTopic?a=tpc&s=215094572&f=351097254&m=101104492
quotes :

"The Mhz of a CPU is like the max rpm of a car engine. It's not the
same thing as power, because a big engine running at moderate rpm can
produce the same power as a smaller engine running at higher rpm."

"Under this analogy, AMD CPUs are about 50% larger than Intel CPUs to
compensate for their lower clock speeds. It's a question of design -
neither approach is intrinsically right or wrong."

"The Athlon XP number is a power (or performance or speed) rating -
it's measuring the rate at which the CPU will execute a program."

"The P4 Ghz number measures only the clock rate of the CPU - i.e just
the max rpm in my analogy. If you compare two otherwise identical CPUs
the power will increase as the GHz increases. For a long time all CPUs
were similar enough to mean that this meant Ghz was also a valid power
rating."

"That's no longer true. A P4 with a 800Mhz FSB and large cache will be
much faster than amother P4 with a slower FSB or smaller cache running
at the same CPU clock speed. An Intel Pentium-M mobile CPU gives about
50% more processor power than a P4, Mhz for Mhz, just like an AMD XP
or Athlon64."

"The advantage of CPU clock cycle is that it's easy measure and easy
to sell - just one number and bigger means better. The disadvantage is
that it's not actually correct."

"The advantage of Performance Ratings is that it does actually tell
you what you need to know; but it's hard to measure and a bit
subjective because actual CPU performance depends on a large numbr of
factors and can vary quite a lot depending on what benchmarks you
happen to use."

http://forums.techguy.org/hardware/542050-uncertainty-principal.html
quotes:

"A better race car analogy would be thinking of GHz as RPM."

"The RPM of an engine is only enough information to give you a
relative idea of it's power output. Engine "X" running at 3,000RPM is
more powerful than the same engine running at 2,000 RPM. Unless you
know lots of other info (number of cylinders, gear ratios, etc) you'll
have no idea if it is as powerful as engine "Y" at 4,000 RPM."

"Back to your CPU - a Core 2 Duo or Athlon 64 does more work per clock
cycle than a Pentium 4, so a Core 2 Duo at 2 GHz may be substantially
more powerful than a Pentium 4 at 3 GHz. This could be related to a 6
cyl car vs a 8 cyl."

Okay, I'll have to admit, I have am a bit obsessed with theoretical
PCs that are as much hardware, chip-based, massively-serial, use
"parallel Hz" [to the highest extent that is mathematically-possible]
and are set to the lowest gear -- and use as little of software and
memory -- as mathematically-possible for a PC to run efficiently.

14. ### Eugene MiyaGuest

So, you were this guy who sent me this post to comp.parallel 3 years back?
And let me guess, that you want it to run Windows?

Return-Path: <>
Path: not-for-mail
From: (Curious)
Newsgroups: comp.parallel
Subject: Parallel Quartz Clock
Date: 21 Jun 2004 17:14:29 -0700
Message-ID: <>

Is it possible to have a processor with 1 billion 1 Hz clocks to make
1 GHz frequency?

This was and is a test case for Spam Assassin in my mail box.

Let us know when YOU reach the stage of using nitric acid.

--comp.parallel moderator
--

15. ### ChrisQuayleGuest

Sounds a great idea, but to clarify, I guess the instruction stream,
which of course wouldn't originate from parallel programming techniques,
languages, or compilers, just utilises simple hardware serial to
parallel converters to dish out intructions to all the cpu's, radially ?.

Amazing that no one has thought of this before...

Chris

There is no serial-to-parallel [or visa versa] conversion. Why do you
think there would be such a conversion?
(much like a serial printer)] but always in parallel Hz. All parts of
the device use parallel Hz but are otherwise completely serial.

Serial bits. Parallel Hz.

Serial bits. Parallel Hz.

Serial bits. Parallel Hz.

Serial bits. Parallel Hz.

Serial bits. Parallel Hz.

Serial bits. Parallel Hz.

Serial bits. Parallel Hz.

Serial bits. Parallel Hz.

Serial bits. Parallel Hz.

Serial bits. Parallel Hz.

18. ### Rich GriseGuest

Which is called a "pipeline", and they've been doing it for decades.

Cheers!
Rich

Okay. According to my research [on google] pipelining doesn't have
much to do with "parallel Hz".

In addition, pipelining uses buffers and has significant latency. Not
something I am found of.

My dream PC does not have any buffers or latency.

My dream PC uses RAM chips -- instead of magnetic discs -- in to store
information. It is entirely chip-based.

This PC is built in such a way that it freshly generates the correct
electric signals ["on the fly"] instead of playing them back from its
ROM chips.

There are sets of instructions stored in ROMs. In the case of most PC,
these instructions load before the CPU "knows" it has a hard drive or
other peripheral devices. However, in my dream PC, those instructions
be generated in real-time instead of storing them.

I am aware that EEPROM is reliable, low power, customizable, reprogram-
able, cheap and proven. But just out of personal preference, my dream
PC is hard-wired in such a way that it does not need any ROM.

Other specs are below. The stuff below also do not need any ROM memory
because they are physically-built to generate the signals which
correspond to the following.

OS: Windows 98SE
Browser: Mozilla Suite 1.8b

No fans, no discs, no moving parts, no ROM [except for the CD/DVD
recording/playing and re-writing].

IOW, my dream PC would work perfectly but would not need any moving
parts, discs, or fans. The "HDD" would consist of silicon RAM chips in
place of disc-platters and electric parts in place of magnetic parts.
No moving parts, no noise, no fans, no magnets, no hazardous heat.

To put it simply, what I am describing is a PC that does not need to
store any information because all of the signal codings for the info
is generated in real-time.

PC reading info from memory = sample playback synth playing back its
samples of sounds of an FM synth.

PC generating its signals in real-time = an *actual* FM synth freshly-
generating its tones "on the fly".

Yes, I know, the above is a poor analogy but I couldn't think of
anything better.

Most importantly, though, my dream PC uses parallel-Hz and is
massively-serial!!

20. ### John LGuest

If each clock signal is 1 Hz, and you have a billion of them,
I hear that if you have nine women working in parallel, you can get a
baby in one month, too.

R's,
John