Velocity Reviews - Computer Hardware Reviews

Velocity Reviews > Newsgroups > Computing > NZ Computing > Moores Law will hold true for a while yet

Reply
Thread Tools

Moores Law will hold true for a while yet

 
 
Shane
Guest
Posts: n/a
 
 
Reply With Quote
 
 
 
 
news.xtra.co.nz
Guest
Posts: n/a
 
      02-26-2006

"Shane" <(E-Mail Removed)-a-geek.net> wrote in message
news:dtrqrd$1it$(E-Mail Removed)...
> http://www.newscientist.com/article....nfo-tech_rss20
>


but, has moores law held true for the last 3 years?

3 years ago (early 2003), I bought a pc with a 3ghz intel processor.
Looking around today, I should be able to find a pc with the equivalent
power of a 12ghz processor.

I bet there aren't any pc's that could encode video 3 times faster than my
machine.

Multi core technology doesn't count either. It should be the output from a
single processor that counts, (to compare apples with apples)


 
Reply With Quote
 
 
 
 
Shane
Guest
Posts: n/a
 
      02-26-2006
news.xtra.co.nz wrote:

>
> "Shane" <(E-Mail Removed)-a-geek.net> wrote in message
> news:dtrqrd$1it$(E-Mail Removed)...
>>

http://www.newscientist.com/article....nfo-tech_rss20
>>

>
> but, has moores law held true for the last 3 years?
>
> 3 years ago (early 2003), I bought a pc with a 3ghz intel processor.
> Looking around today, I should be able to find a pc with the equivalent
> power of a 12ghz processor.
>
> I bet there aren't any pc's that could encode video 3 times faster than my
> machine.
>
> Multi core technology doesn't count either. It should be the output from a
> single processor that counts, (to compare apples with apples)



Multicore tech does count, moores observation was concerned with the number
of transistors on a chip, its only a consequence of those transistors that
has led to clock speeds.

A possible anology
A one horse cart with the horse getting progressively
bigger/stronger/faster, followed by a two horse cart, with two string
horses, speeds the same but capable of carrying heavier loads

AMD has shown that clock speed alone is a poor measurement of a chip, and
like I always say, its not the size that counts, but its what you can do
with it that matters
 
Reply With Quote
 
news.xtra.co.nz
Guest
Posts: n/a
 
      02-26-2006

"Shane" <(E-Mail Removed)-a-geek.net> wrote in message
news:dtt2hg$b33$(E-Mail Removed)...
> news.xtra.co.nz wrote:
>
>>
>> "Shane" <(E-Mail Removed)-a-geek.net> wrote in message
>> news:dtrqrd$1it$(E-Mail Removed)...
>>>

> http://www.newscientist.com/article....nfo-tech_rss20
>>>

>>
>> but, has moores law held true for the last 3 years?
>>
>> 3 years ago (early 2003), I bought a pc with a 3ghz intel processor.
>> Looking around today, I should be able to find a pc with the equivalent
>> power of a 12ghz processor.
>>
>> I bet there aren't any pc's that could encode video 3 times faster than
>> my
>> machine.
>>
>> Multi core technology doesn't count either. It should be the output from
>> a
>> single processor that counts, (to compare apples with apples)

>
>
> Multicore tech does count, moores observation was concerned with the
> number
> of transistors on a chip, its only a consequence of those transistors that
> has led to clock speeds.
>
> A possible anology
> A one horse cart with the horse getting progressively
> bigger/stronger/faster, followed by a two horse cart, with two string
> horses, speeds the same but capable of carrying heavier loads
>
> AMD has shown that clock speed alone is a poor measurement of a chip, and
> like I always say, its not the size that counts, but its what you can do
> with it that matters



Basically, they are not speeding up the processing for a 'single thread'
applications. They are improving chips multitasking abilities, rather than
making it absolutely faster.

My analogy is of a racing car. The 'moores' law applies to the absolute
top speed of the racing car. Then, each year, the racing car should be twice
as fast. ie, this year, the top speed is 100kph. In 18 months, they will
build one that can go 200kph, and so on.

In my opinion. the 'single thread' execution speed is what counts the most.
Actually, I am not exactly sure the definition of moores law but I thought
it related to speed, rather than power.



 
Reply With Quote
 
Andrew
Guest
Posts: n/a
 
      02-26-2006
news.xtra.co.nz wrote:
> "Shane" <(E-Mail Removed)-a-geek.net> wrote in message
> news:dtt2hg$b33$(E-Mail Removed)...
>> news.xtra.co.nz wrote:
>>
>>> "Shane" <(E-Mail Removed)-a-geek.net> wrote in message
>>> news:dtrqrd$1it$(E-Mail Removed)...

>> http://www.newscientist.com/article....nfo-tech_rss20
>>> but, has moores law held true for the last 3 years?
>>>
>>> 3 years ago (early 2003), I bought a pc with a 3ghz intel processor.
>>> Looking around today, I should be able to find a pc with the equivalent
>>> power of a 12ghz processor.
>>>
>>> I bet there aren't any pc's that could encode video 3 times faster than
>>> my
>>> machine.
>>>
>>> Multi core technology doesn't count either. It should be the output from
>>> a
>>> single processor that counts, (to compare apples with apples)

>>
>> Multicore tech does count, moores observation was concerned with the
>> number
>> of transistors on a chip, its only a consequence of those transistors that
>> has led to clock speeds.
>>
>> A possible anology
>> A one horse cart with the horse getting progressively
>> bigger/stronger/faster, followed by a two horse cart, with two string
>> horses, speeds the same but capable of carrying heavier loads
>>
>> AMD has shown that clock speed alone is a poor measurement of a chip, and
>> like I always say, its not the size that counts, but its what you can do
>> with it that matters

>
>
> Basically, they are not speeding up the processing for a 'single thread'
> applications. They are improving chips multitasking abilities, rather than
> making it absolutely faster.
>
> My analogy is of a racing car. The 'moores' law applies to the absolute
> top speed of the racing car. Then, each year, the racing car should be twice
> as fast. ie, this year, the top speed is 100kph. In 18 months, they will
> build one that can go 200kph, and so on.
>
> In my opinion. the 'single thread' execution speed is what counts the most.
> Actually, I am not exactly sure the definition of moores law but I thought
> it related to speed, rather than power.
>
>
>

Multi threaded in terma of the racing car analogy... that car that can
travel 200kph could probably only travel at 100kph up a hill, So if it
was multithreaded it would have more power and be able to travel up the
hill at 200kph
 
Reply With Quote
 
Shane
Guest
Posts: n/a
 
      02-26-2006
news.xtra.co.nz wrote:

>
> "Shane" <(E-Mail Removed)-a-geek.net> wrote in message
> news:dtt2hg$b33$(E-Mail Removed)...
>> news.xtra.co.nz wrote:
>>
>>>
>>> "Shane" <(E-Mail Removed)-a-geek.net> wrote in message
>>> news:dtrqrd$1it$(E-Mail Removed)...
>>>>

>>

http://www.newscientist.com/article....nfo-tech_rss20
>>>>
>>>
>>> but, has moores law held true for the last 3 years?
>>>
>>> 3 years ago (early 2003), I bought a pc with a 3ghz intel processor.
>>> Looking around today, I should be able to find a pc with the equivalent
>>> power of a 12ghz processor.
>>>
>>> I bet there aren't any pc's that could encode video 3 times faster than
>>> my
>>> machine.
>>>
>>> Multi core technology doesn't count either. It should be the output from
>>> a
>>> single processor that counts, (to compare apples with apples)

>>
>>
>> Multicore tech does count, moores observation was concerned with the
>> number
>> of transistors on a chip, its only a consequence of those transistors
>> that has led to clock speeds.
>>
>> A possible anology
>> A one horse cart with the horse getting progressively
>> bigger/stronger/faster, followed by a two horse cart, with two string
>> horses, speeds the same but capable of carrying heavier loads
>>
>> AMD has shown that clock speed alone is a poor measurement of a chip, and
>> like I always say, its not the size that counts, but its what you can do
>> with it that matters

>
>
> Basically, they are not speeding up the processing for a 'single thread'
> applications. They are improving chips multitasking abilities, rather
> than making it absolutely faster.
>


This is correct, unfortunately this means its up to the software designers
to take full advantage of that power, so older programs will not have any
noticeable change.

> My analogy is of a racing car. The 'moores' law applies to the absolute
> top speed of the racing car. Then, each year, the racing car should be
> twice
> as fast. ie, this year, the top speed is 100kph. In 18 months, they will
> build one that can go 200kph, and so on.
>
> In my opinion. the 'single thread' execution speed is what counts the
> most. Actually, I am not exactly sure the definition of moores law but I
> thought it related to speed, rather than power.


From the ubiquitous wikipedia
http://en.wikipedia.org/wiki/Moores_Law
Moores original statement on the phenomenon (try and spell that without a
spellcheka)
The complexity for minimum component costs has increased at a rate of
roughly a factor of two per year ... Certainly over the short term this
rate can be expected to continue, if not to increase. Over the longer term,
the rate of increase is a bit more uncertain, although there is no reason
to believe it will not remain nearly constant for at least 10 years. That
means by 1975, the number of components per integrated circuit for minimum
cost will be 65,000. I believe that such a large circuit can be built on a
single wafer.


 
Reply With Quote
 
news.xtra.co.nz
Guest
Posts: n/a
 
      02-26-2006

"Shane" <(E-Mail Removed)-a-geek.net> wrote in message
news:dtt86k$b33$(E-Mail Removed)...
> news.xtra.co.nz wrote:
>
>>
>> "Shane" <(E-Mail Removed)-a-geek.net> wrote in message
>> news:dtt2hg$b33$(E-Mail Removed)...
>>> news.xtra.co.nz wrote:
>>>
>>>>
>>>> "Shane" <(E-Mail Removed)-a-geek.net> wrote in message
>>>> news:dtrqrd$1it$(E-Mail Removed)...
>>>>>
>>>

> http://www.newscientist.com/article....nfo-tech_rss20
>>>>>
>>>>
>>>> but, has moores law held true for the last 3 years?
>>>>
>>>> 3 years ago (early 2003), I bought a pc with a 3ghz intel processor.
>>>> Looking around today, I should be able to find a pc with the equivalent
>>>> power of a 12ghz processor.
>>>>
>>>> I bet there aren't any pc's that could encode video 3 times faster than
>>>> my
>>>> machine.
>>>>
>>>> Multi core technology doesn't count either. It should be the output
>>>> from
>>>> a
>>>> single processor that counts, (to compare apples with apples)
>>>
>>>
>>> Multicore tech does count, moores observation was concerned with the
>>> number
>>> of transistors on a chip, its only a consequence of those transistors
>>> that has led to clock speeds.
>>>
>>> A possible anology
>>> A one horse cart with the horse getting progressively
>>> bigger/stronger/faster, followed by a two horse cart, with two string
>>> horses, speeds the same but capable of carrying heavier loads
>>>
>>> AMD has shown that clock speed alone is a poor measurement of a chip,
>>> and
>>> like I always say, its not the size that counts, but its what you can do
>>> with it that matters

>>
>>
>> Basically, they are not speeding up the processing for a 'single thread'
>> applications. They are improving chips multitasking abilities, rather
>> than making it absolutely faster.
>>

>
> This is correct, unfortunately this means its up to the software designers
> to take full advantage of that power, so older programs will not have any
> noticeable change.
>
>> My analogy is of a racing car. The 'moores' law applies to the absolute
>> top speed of the racing car. Then, each year, the racing car should be
>> twice
>> as fast. ie, this year, the top speed is 100kph. In 18 months, they will
>> build one that can go 200kph, and so on.
>>
>> In my opinion. the 'single thread' execution speed is what counts the
>> most. Actually, I am not exactly sure the definition of moores law but I
>> thought it related to speed, rather than power.

>
> From the ubiquitous wikipedia
> http://en.wikipedia.org/wiki/Moores_Law
> Moores original statement on the phenomenon (try and spell that without a
> spellcheka)
> The complexity for minimum component costs has increased at a rate of
> roughly a factor of two per year ... Certainly over the short term this
> rate can be expected to continue, if not to increase. Over the longer
> term,
> the rate of increase is a bit more uncertain, although there is no reason
> to believe it will not remain nearly constant for at least 10 years. That
> means by 1975, the number of components per integrated circuit for minimum
> cost will be 65,000. I believe that such a large circuit can be built on a
> single wafer.
>
>


yes, so, it is the number of components on a 'single' wafer. If they can
continue putting extra cores onto a wafer then it seems moores law is
upheld.

I just like the idea, of being able to encode an mp3 in 1/2 the time in
18months time. But, it seems under the existing definition this is not
predicted.



 
Reply With Quote
 
-=rjh=-
Guest
Posts: n/a
 
      02-26-2006
news.xtra.co.nz wrote:

> In my opinion. the 'single thread' execution speed is what counts the most.
> Actually, I am not exactly sure the definition of moores law but I thought
> it related to speed, rather than power.


Neither, it related to the number of components that could be built into
an integrated circuit for the same basic cost - Moore didn't specify the
function of the circuit. Bear in mind that he was thinking that
transistor counts of up to 65,000 should be possible.

At the time that Moore first made his observation, (which others later
called Moore's Law) the microprocessor didn't exist, and even when Moore
was working at Intel later, Intel was focusing on memory production, not
microprocessors.

Fortunately Japan thought memory was more important at that time and
made a huge government backed push to control that industry, while Intel
practically stumbled into processor production; the results of this
split in production are still evident today.

"Inside Intel" is probably at your local library, it is a very
interesting read.

http://www.amazon.com/gp/product/0452276438/
 
Reply With Quote
 
Have A Nice Cup of Tea
Guest
Posts: n/a
 
      02-26-2006
On Mon, 27 Feb 2006 10:25:44 +1300, news.xtra.co.nz wrote:

> In my opinion. the 'single thread' execution speed is what counts the
> most. Actually, I am not exactly sure the definition of moores law but I
> thought it related to speed, rather than power.


Moore's law relates to the density of transistors located on an
intergrated circut.


Have A Nice Cup of Tea

--
Buffer-overflow vulnerabilities are simply programming errors; they occur when
coders fail to deploy proper memory-management techniques.

 
Reply With Quote
 
Jerry
Guest
Posts: n/a
 
      02-27-2006
news.xtra.co.nz wrote:
> "Shane" <(E-Mail Removed)-a-geek.net> wrote in message
> news:dtrqrd$1it$(E-Mail Removed)...
>
>>http://www.newscientist.com/article....nfo-tech_rss20
>>

>
>
> but, has moores law held true for the last 3 years?
>
> 3 years ago (early 2003), I bought a pc with a 3ghz intel processor.
> Looking around today, I should be able to find a pc with the equivalent
> power of a 12ghz processor.
>
> I bet there aren't any pc's that could encode video 3 times faster than my
> machine.
>
> Multi core technology doesn't count either. It should be the output from a
> single processor that counts, (to compare apples with apples)
>
>

Moore's law has to do with the number of transisters per given area of a
chip.

at http://www.webopedia.com/TERM/M/Moores_Law.html it gives a
definition, but it appears the page is a few years old. Multi core
would count for sure, it increases the number of transistors on a given
sized chip.

(mrz l) (n.) The observation made in 1965 by Gordon Moore,
co-founder of Intel, that the number of transistors per square inch on
integrated circuits had doubled every year since the integrated circuit
was invented. Moore predicted that this trend would continue for the
foreseeable future. In subsequent years, the pace slowed down a bit, but
data density has doubled approximately every 18 months, and this is the
current definition of Moore's Law, which Moore himself has blessed. Most
experts, including Moore himself, expect Moore's Law to hold for at
least another two decades.
 
Reply With Quote
 
 
 
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Re: while True or while 1 BartC Python 8 12-28-2010 03:00 PM
Re: while True or while 1 Krister Svanlund Python 6 12-14-2010 04:52 PM
[False,True] and [True,True] --> [True, True]????? bdb112 Python 45 04-29-2009 02:35 AM
does the saying "don't optimise" hold true for GUI? - Java 3 03-23-2005 06:16 AM
while 1 vs while True Timothy Fitz Python 16 12-15-2004 09:27 PM



Advertisments