What is the maximum clock rate given the state of today's technology?

Hi:

What is the maximum physically-possible clock rate [measured in Hz] of
a 1-bit-per-cycle, single-core, purely-serial processor?


Thank a bunch,

Green Xenon

GreenXenon <glucegen1x@gmail.com> wrote:
 
> What is the maximum physically-possible clock rate [measured in Hz] of
> a 1-bit-per-cycle, single-core, purely-serial processor?

Using MOS transistors, it might be in the 10GHz range.

The high clock rates of current processors require on-chip
PLL frequency multipliers, as you don't get those frequencies
in and out of the chip.

For a serial processor, that would require some shift registers
to get the bits into and out of the one bit ALU.  That takes
a lot more logic than is in the ALU itself.

-- glen

On Mar 25, 12:58&#4294967295;pm, glen herrmannsfeldt <g...@ugcs.caltech.edu>
wrote:
> GreenXenon <glucege...@gmail.com> wrote:
> > What is the maximum physically-possible clock rate [measured in Hz] of
> > a 1-bit-per-cycle, single-core, purely-serial processor?
>
> Using MOS transistors, it might be in the 10GHz range.
>
> The high clock rates of current processors require on-chip
> PLL frequency multipliers, as you don't get those frequencies
> in and out of the chip.
>
> For a serial processor, that would require some shift registers
> to get the bits into and out of the one bit ALU. &#4294967295;That takes
> a lot more logic than is in the ALU itself.
>
> -- glen


Would an optical processor be able to run safely at a significantly
higher-frequency than an electronic processor?

Lets say the optical CPU uses 400-nm-wavelength lasers in place of
electronic signals, what is the max clock rate that can be performed
by this theoretical CPU without experiencing any physical damage?

Damage is not the issue: function is.

Jerry
-- 
Engineering is the art of making what you want from things you can get.

On Mar 25, 5:41&#4294967295;pm, Jerry Avins <j...@ieee.org> wrote:


> Damage is not the issue: function is.


Meaning?

Malfunction is the typical result of excessive clock speed, not damage.

Jerry
-- 
Engineering is the art of making what you want from things you can get.

On Mar 27, 12:47&#4294967295;am, Jerry Avins <j...@ieee.org> wrote:
> Malfunction is the typical result of excessive clock speed, not damage.
>
> Jerry
> --
> Engineering is the art of making what you want from things you can get.


Really? I thought the major risk factor in overclocking is thermal
damage.

On 2011/03/27 03:26, GreenXenon wrote:
> On Mar 27, 12:47 am, Jerry Avins<j...@ieee.org>  wrote:
>> Malfunction is the typical result of excessive clock speed, not damage.
>>
>> Jerry
>> --
>> Engineering is the art of making what you want from things you can get.
>
>
> Really? I thought the major risk factor in overclocking is thermal
> damage.

You're thinking of Intel chips.  They're the only ones that are 
manufactured with a fan affixed.  :-)

Seriously, though, not every chip is running up against thermal limits. 
  Many are dealing with much simpler problems like propagation delays. 
As Jerry says, in those cases the chip just malfunctions.  In other 
words, you get the wrong results if you ask the chip to go too fast, 
because it trips over itself and calculates a new, but incorrect output 
based on an out-of-date input.

Even more seriously, your question depends heavily upon what you're 
trying to do.  Propagation delays add up, so the more logic gates you 
have in a series for a single calculation, the slower you must run the 
clock in order to give time for the correct result to ripple to the 
output.  But, if you find that only a small percentage of your 
calculations take a long time, while most of the rest involve fewer gate 
delays, the you can just redesign to break the longer chains in half so 
that they run in two or more cycles ... then you can speed the clock up 
some more.

i.e. There is no single maximum clock rate.  "It depends" is the best 
answer you can get.  For a given instruction set, you might be able to 
determine a limit, but if the designer is free to change the 
instructions, change to ECL, or use any sort of technology, then I'm not 
sure what the limit would be.

Brian Willoughby
Sound Consulting

GreenXenon <glucegen1x@gmail.com> wrote:
> On Mar 27, 12:47&#4294967295;am, Jerry Avins <j...@ieee.org> wrote:
>> Malfunction is the typical result of excessive clock speed, not damage.
 
> Really? I thought the major risk factor in overclocking is thermal
> damage.

If you don't care at all about the results, but just want them fast,
then overclock all you want.  

Most processors now have thermal shutdown with a sensor to sense
that the chip is too hot.  (Usually because the vents are blocked.)
It will shutdown before thermal damage occurs.

Well, some increase Vdd, which speeds up the transistors, but 
increases the power even more.  Then add better heat sinks.
(Or even add active coolers.)

-- glen

On Mar 26, 8:58&#4294967295;am, glen herrmannsfeldt <g...@ugcs.caltech.edu> wrote:
> GreenXenon <glucege...@gmail.com> wrote:
> > What is the maximum physically-possible clock rate [measured in Hz] of
> > a 1-bit-per-cycle, single-core, purely-serial processor?
>
> Using MOS transistors, it might be in the 10GHz range.
>


Keep going...