Sign in with
Sign up | Sign in
Your question

Laptops, History and comparrison to desktop counterparts.

Last response: in Laptop General Discussion
Share
Anonymous
a b D Laptop
June 24, 2005 12:35:55 AM

Archived from groups: comp.sys.laptops (More info?)

This is a post relating to a school project i am currently doing on
Laptops compared to Desktops.

http://www.mochima.com/net/notebooks

The above link is a link to our paper, and concentrates, as indicated
in the subject, mainly on laptop history and desktop comparison. There
is a lot of dispute over what was the first true laptop, what is your
input? As well as what's truly better a laptop or a desktop?
Seriously pros and cons here people!

The first pro to using a laptop are of course portablity, i mean that's
why they were invented in the first place right? But is the price
really worth the ability to hvae a computer wherever you go?

And of course with a desktop we get more power and upgradability... and
of course we don't have to worry about heat nearly as much as with the
notebook systems.

Give us your input and experiences, we'd love to hear them.

-Cameron Ward
-Norman Nadeau
Anonymous
a b D Laptop
June 24, 2005 11:38:40 AM

Archived from groups: comp.sys.laptops (More info?)

"Cameron Ward" <cameron_ward19@yahoo.ca> wrote:
>really worth the ability to hvae a computer wherever you go?

Well, isn't that kind of a #define? For some people it is, and the
sales of laptops would seem to indicate that quite a number of people
prize portability over other features...

In fact, laptops are a compromise between cost, size, battery life,
and horsepower. Any desktop of equivalent vintage will have far more
horsepower, a better display, and lower cost than a laptop, all other
things being equal.

Note, however, that horsepower is becoming less important, as even
low-end laptops can handle the usual tasks (Email, WWWeb browsing,
etc) that most peolle want a computer for, so for a lot of folks it's
a tradeoff between portability and cost, and with costs potentially
under $1K, the laptop frequently wins.
Anonymous
a b D Laptop
June 24, 2005 7:17:32 PM

Archived from groups: comp.sys.laptops (More info?)

William P. N. Smith wrote:

> Note, however, that horsepower is becoming less important, as even
> low-end laptops can handle the usual tasks (Email, WWWeb browsing,
> etc) that most peolle want a computer for, so for a lot of folks it's
> a tradeoff between portability and cost, and with costs potentially
> under $1K, the laptop frequently wins.

Perhaps one issue that is often overlooked when talking about horsepower
is responsiveness.

Notebooks have had good speed for quite a while, but still a bit low
responsiveness (you run two applications at a time, and the things
slows down to a crawl).

Do we already have dual-core mobile CPUs? (I haven't checked lately)

If not, I guess we'll soon have them, and from that moment on, no
notebook will ever have insufficient horsepower (well, until people
decide that notebooks are now useful for number-crunching, local
weather prediction, etc. :-))

Carlos
--
Related resources
Anonymous
a b D Laptop
June 24, 2005 9:51:27 PM

Archived from groups: comp.sys.laptops (More info?)

Okay, but even without Dual Cores there are some laptops that are
viciously fast. And for people who only use their computer for
internet, office and email... well an old pIII 700 laptop would be more
than addequate. I think the major issue here is heat, and wouldn't
dual core chips produce ridiculous amounts of heat? As for laptop
sales, like i said, to do email and web browsing you dont' need
anything for a p3, so why now are laptop sales begining to surpass
desktop sales? why not years ago?

-Cameron
June 24, 2005 10:20:20 PM

Archived from groups: comp.sys.laptops (More info?)

Carlos Moreno <moreno_at_mochima_dot_com@xx.xxx> wrote in
news:WZYue.28473$rz.1042282@wagner.videotron.net:


> Do we already have dual-core mobile CPUs? (I haven't checked lately)
>
> If not, I guess we'll soon have them, and from that moment on, no
> notebook will ever have insufficient horsepower (well, until people
> decide that notebooks are now useful for number-crunching, local
> weather prediction, etc. :-))
>
> Carlos
> --
>

There aren't any dual-core mobile cCPUs out yet, but it's just a matter
of time. Google turned up the following... and it look's like Intel is
having to play catch-up. Should be interesting..

"Intel shed a little more light on its forthcoming dual-core mobile chip
Thursday in a briefing with reporters, confirming that the Yonah
processor's two cores will share a single bank of cache memory."

"Intel plans to aggressively introduce dual-core processors over the next
year. The company was forced to accelerate its dual-core processors after
realizing that it could no longer wring any more performance out of
single-core Pentium 4 chips without melting motherboards. Processor
analysts and chip enthusiasts have turned their noses up at Intel's
initial Pentium D design, labeling it inelegant compared to AMD's dual-
core chips."

Norm


----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
Anonymous
a b D Laptop
June 25, 2005 12:46:42 AM

Archived from groups: comp.sys.laptops (More info?)

Carlos Moreno <moreno_at_mochima_dot_com@xx.xxx> wrote:
>Notebooks have had good speed for quite a while, but still a bit low
>responsiveness (you run two applications at a time, and the things
>slows down to a crawl).

You are confusing raw processor speed with computer performance, just
as all the computer companies have hoped you would. Other factors
(FSB speed, memory word size, disk {speed, interface, cache size})
have a much larger effect than most folks realize, and are typically
much lower in a laptop than in a desktop.

There's nothing magical about laptops as compared to desktops, they
are just optimized for different uses.
June 25, 2005 1:19:03 AM

Archived from groups: comp.sys.laptops (More info?)

"Cameron Ward" <cameron_ward19@yahoo.ca> wrote in
news:1119660687.572284.135700@g14g2000cwa.googlegroups.com:

> I think the major issue here is heat, and wouldn't
> dual core chips produce ridiculous amounts of heat?

Intel new dual-core chips have reduced power consumption, reducing heat and
increasing use per battery charge.

As for laptop sales vs desktop sales, it got to be portability. After all,
it's their reason for being.

Norm

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
Anonymous
a b D Laptop
June 25, 2005 2:09:23 AM

Archived from groups: comp.sys.laptops (More info?)

William P. N. Smith wrote:
> Carlos Moreno <moreno_at_mochima_dot_com@xx.xxx> wrote:
>
>>Notebooks have had good speed for quite a while, but still a bit low
>>responsiveness (you run two applications at a time, and the things
>>slows down to a crawl).
>
> You are confusing raw processor speed with computer performance

I'm very curious as for how do you draw that conclusion from what I
said... I referred to the speed *of notebooks* (which would correspond
to what you refer to as computer performance), and not the speed of the
notebook's processor.

That performance (or lack thereof) is what makes them less responsive,
compared to desktops/workstations (which have higher performance, and
thus it takes more to make them slow down to a crawl).

Dual-CPU (or dual-core CPUs) will certainly alleviate that problem
almost to the point of making it disappear completely (assuming a
"normal" performance in the rest of the components).

> There's nothing magical about laptops as compared to desktops, they
> are just optimized for different uses.

Agreed. And as usual, the Laws of Physics are cruel enough to guarantee
that every optimisation always leads to trade-off and compromising
certain aspects in favor of another.

Carlos
--
Anonymous
a b D Laptop
June 25, 2005 2:52:46 AM

Archived from groups: comp.sys.laptops (More info?)

Norm wrote:

> "Cameron Ward" <cameron_ward19@yahoo.ca> wrote in
> news:1119660687.572284.135700@g14g2000cwa.googlegroups.com:
>
>> I think the major issue here is heat, and wouldn't
>> dual core chips produce ridiculous amounts of heat?
>
> Intel new dual-core chips have reduced power consumption, reducing heat
> and increasing use per battery charge.

Or so says Intel. Believe it when you see it.

> As for laptop sales vs desktop sales, it got to be portability. After all,
> it's their reason for being.

Not necessarily. Size without regard to portability is also an issue--in a
typical dorm room for example a laptop takes up a lot less of the tiny
amount of desk space available than does a desktop, even though the
pilferability is a major negative.

> Norm
>
> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
> News==---- http://www.newsfeeds.com The #1 Newsgroup Service in the World!
> 120,000+ Newsgroups ----= East and West-Coast Server Farms - Total Privacy
> via Encryption =----

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)
June 25, 2005 3:08:41 AM

Archived from groups: comp.sys.laptops (More info?)

"J. Clarke" <jclarke.usenet@snet.net.invalid> wrote in
news:D 9ihbq01i0r@news3.newsguy.com:

>
> Or so says Intel. Believe it when you see it.
>

Agreed..


----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
Anonymous
a b D Laptop
June 25, 2005 1:13:58 PM

Archived from groups: comp.sys.laptops (More info?)

Norm wrote:
> "Cameron Ward" <cameron_ward19@yahoo.ca> wrote in
> news:1119660687.572284.135700@g14g2000cwa.googlegroups.com:
>
>
>>I think the major issue here is heat, and wouldn't
>>dual core chips produce ridiculous amounts of heat?

Even if they did, reducing clock speed (which would reduce heat) but
having two CPUs would offer a *much* higher overall performance.

Doing only one thing at a time twice as fast is not nearly as
good as being able to do two things at a time.

Carlos
--
Anonymous
a b D Laptop
June 25, 2005 2:52:40 PM

Archived from groups: comp.sys.laptops (More info?)

Of course dual processors would work... sure there might be a
bottleneck with memory bandwidth but even if you were running with 64
MB of ram, dual processors would still make things run faster. It
won't be twice the speed mind you, but you'd deffinatly make things go
smoother.

how much memory bandwidth does a standard laptop have anyways?

-Cameron
Anonymous
a b D Laptop
June 25, 2005 4:43:32 PM

Archived from groups: comp.sys.laptops (More info?)

Carlos Moreno <moreno_at_mochima_dot_com@xx.xxx> wrote:
>William P. N. Smith wrote:
>> Carlos Moreno <moreno_at_mochima_dot_com@xx.xxx> wrote:
>>
>>>Notebooks have had good speed for quite a while, but still a bit low
>>>responsiveness (you run two applications at a time, and the things
>>>slows down to a crawl).
>>
>> You are confusing raw processor speed with computer performance

>That performance (or lack thereof) is what makes them less responsive,
>compared to desktops/workstations (which have higher performance, and
>thus it takes more to make them slow down to a crawl).

Where do you think performance comes from? [Maybe I'm asking how you
define performance.]

>Dual-CPU (or dual-core CPUs) will certainly alleviate that problem
>almost to the point of making it disappear completely (assuming a
>"normal" performance in the rest of the components).

Why would you assume 'normal' (desktop) performance from the rest of
the components in a laptop? As I pointed out in my previous note,
that's where one of the tradeoffs happens.

How will dual-core processors help if the memory bandwidth is only 1/4
that of a modern desktop?
Anonymous
a b D Laptop
June 25, 2005 4:49:41 PM

Archived from groups: comp.sys.laptops (More info?)

well of course it wouldn't be cheap, nothing in computers ever is at
first. But seriously now, for what people use them for, laptops are
more then effecient, even without dual processors or fast ram.
Desktops will always be better, and i think people who want to push
their computers to the limit will never look at a notebook as a viable
option... they will always look to a desktop.

-Cameron
Anonymous
a b D Laptop
June 25, 2005 6:09:46 PM

Archived from groups: comp.sys.laptops (More info?)

Cameron Ward wrote:

> Of course dual processors would work... sure there might be a
> bottleneck with memory bandwidth but even if you were running with 64
> MB of ram, dual processors would still make things run faster. It
> won't be twice the speed mind you, but you'd deffinatly make things go
> smoother.

Depends on what you're doing. Some tasks are amenable to multiprocessing,
others aren't. The main benefit of duals IMO is the ability to handle two
interrupts at the same time. For general use there's no real difference in
performance, but sometimes there is a difference in responsiveness.

> how much memory bandwidth does a standard laptop have anyways?
>
> -Cameron

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)
June 25, 2005 6:17:57 PM

Archived from groups: comp.sys.laptops (More info?)

"J. Clarke" <jclarke.usenet@snet.net.invalid> wrote in
news:D 9k7rf1uks@news4.newsguy.com:

> Cameron Ward wrote:
>
>> Of course dual processors would work... sure there might be a
>> bottleneck with memory bandwidth but even if you were running with 64
>> MB of ram, dual processors would still make things run faster. It
>> won't be twice the speed mind you, but you'd deffinatly make things
>> go smoother.


And it wouldn't be cheap, at first anyways.

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
June 25, 2005 9:48:04 PM

Archived from groups: comp.sys.laptops (More info?)

"Cameron Ward" <cameron_ward19@yahoo.ca> wrote in
news:1119728981.487559.187420@f14g2000cwb.googlegroups.com:

> well of course it wouldn't be cheap, nothing in computers ever is at
> first. But seriously now, for what people use them for, laptops are
> more then effecient, even without dual processors or fast ram.
> Desktops will always be better, and i think people who want to push
> their computers to the limit will never look at a notebook as a viable
> option... they will always look to a desktop.
>
> -Cameron
>

Well, performance to a business man is how long his battery will last and
how fast he can download the latest proposal from the head office. He
really doesn't care what's inside as lond as it does the job, and he's
willing to pay the extra.

As for pushing the limit, I agree that desktop's are the way to go.

Norm

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
Anonymous
a b D Laptop
June 25, 2005 10:26:56 PM

Archived from groups: comp.sys.laptops (More info?)

"Cameron Ward" <cameron_ward19@yahoo.ca> wrote:
>Of course dual processors would work.

You really have bought into the hype, haven't you? IIRC I saw a
benchmark recently that dual-processor systems gain you 10%
improvement.

>how much memory bandwidth does a standard laptop have anyways?

Well, mine is 1/2 the speed (400MHz) and not dual-channel like my
desktop, so that's why I threw out 1/4 the bandwidth in my previous
missive.

Again, there's essentially no difference between a 2GHz processor and
a 3GHz processor if the FSB is only running at 400MHz.
June 25, 2005 11:46:29 PM

Archived from groups: comp.sys.laptops (More info?)

Carlos Moreno <moreno_at_mochima_dot_com@xx.xxx> wrote in news:3Lcve.31002
$rz.1335043@wagner.videotron.net:

> Norm wrote:
>> "Cameron Ward" <cameron_ward19@yahoo.ca> wrote in
>> news:1119660687.572284.135700@g14g2000cwa.googlegroups.com:
>>
>>
>>>I think the major issue here is heat, and wouldn't
>>>dual core chips produce ridiculous amounts of heat?
>

Reduce power consumption and you reduse the head, not to mention extending
the battery life.

N

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
Anonymous
a b D Laptop
June 26, 2005 11:36:15 AM

Archived from groups: comp.sys.laptops (More info?)

There is no hype to dual processors!!!!! sure for running one
application you don`t see that much improvment, but the whole point is
that you can have two processors doing two completely different things
with half the lag to your pc.... dual processors aren`t speed boosts
they are performance boosts. There is a difference between raw speed
and performance, and last time i checked a 3.0ghz machine was more, way
way way more than capable of running any application.

-Cameron
Anonymous
a b D Laptop
June 26, 2005 3:21:07 PM

Archived from groups: comp.sys.laptops (More info?)

Thanks Joseph, your response gave some fresh perspective to the thread.
I deffinatly see how a laptop could be an advantage in your industry,
and i'll use your example in my paper.

-Cameron
Anonymous
a b D Laptop
June 26, 2005 3:31:01 PM

Archived from groups: comp.sys.laptops (More info?)

Cameron Ward wrote:

> There is no hype to dual processors!!!!! sure for running one
> application you don`t see that much improvment, but the whole point is
> that you can have two processors doing two completely different things
> with half the lag to your pc.... dual processors aren`t speed boosts
> they are performance boosts.

IF you have two CPU-bound tasks to be performed. Most users rarely have
_one_ CPU-bound task.

As for "half the lag to your PC", if your PC is showing "lag" that's usually
a configuration problem or a software bug or just plain bad program design.

> There is a difference between raw speed
> and performance, and last time i checked a 3.0ghz machine was more, way
> way way more than capable of running any application.

Running, yes. Running at adequate speed, not always. Look into video
production and CGI animation for a couple of areas where even the fastest
machines fall short. Some games are also CPU-bound.

> -Cameron

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)
June 26, 2005 8:57:02 PM

Archived from groups: comp.sys.laptops (More info?)

"C.Joseph Drayton" <kalek1@mindspring.com> wrote in news:42bf004f$1_7
@Output.100ProofNews.com:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi Cameron,
>
> Please feel free to do so. If you would like some additional real
> world examples, let me know.
>


Joseph

Thanks for the support.
We really didn't know what to expect by posting the article on the usenet.
So for it's been a positive experience, and we've learned a few things
about laptops along the way.

Norm

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
Anonymous
a b D Laptop
June 27, 2005 2:07:43 PM

Archived from groups: comp.sys.laptops (More info?)

In article <1119584155.069922.254630@o13g2000cwo.googlegroups.com>,
Cameron Ward <cameron_ward19@yahoo.ca> wrote:
...
>The above link is a link to our paper, and concentrates, as indicated
>in the subject, mainly on laptop history and desktop comparison. There
>is a lot of dispute over what was the first true laptop, what is your
>input?

You can't really talk to the history of laptops without mentioning the
Dynabook, conceived at the Xerox PARC. This was 35 years ago,
and it was conceived as primarily an educational device.

http://en.wikipedia.org/wiki/Dynabook
http://www.squeakland.org/school/HTML/essays/dynabook_r...
http://thinkubator.ccsp.sfu.ca/Dynabook

The research that stemmed from the Dynabook project led to a lot of things
we take for granted: Object ORiented Programming, the Graphical interface,
laser printers, e-mail, and the modern laptop computer (among other
things).

I'd also suggest mentioning the Compaq Portable. The Osborn 1 is well and
good, but the Compaq was the first PC compatible portable. It also
basically launched the non-IBM DOS-compatible PC business.

>As well as what's truly better a laptop or a desktop?

They both hit different design points. Both are valid. That's why both are
still sold.

>The first pro to using a laptop are of course portablity, i mean that's
>why they were invented in the first place right?

Portability is part of it, but I'd say that it's what portability enables
that's more significant. With a portable machine that can always be
there, it opens up the door to new applications. Just as an example, a
GPS driven map/navigation application makes more sense for a portable
machine than a fixed desktop. The same thing is true for pen-based
applications and media players. The software (and hardware) for both has
existed for years on desktops, but it makes the most sense and gets the
most use on portables.

The other thing that's worth mentioning is that this line "The same
Pentium technology, but packaging and features for laptops. " is actually
quite wrong. The Pentium M shares a lot more in common with the Pentium 3
than the Pentium 4. It's been a source of some controversy, since the
mobile chip can actually perform _better_ than the Pentium 4, despite the
fact it consumes much less power. There's an article at Tom's hardware
guide that speaks to that point.

-Mike
--
http://www.mschaef.com
Anonymous
a b D Laptop
June 27, 2005 2:13:37 PM

Archived from groups: comp.sys.laptops (More info?)

In article <d9mjeo022up@news2.newsguy.com>,
J. Clarke <jclarke.usenet@snet.net.invalid> wrote:
>Cameron Ward wrote:
>
>> There is no hype to dual processors!!!!! sure for running one
>> application you don`t see that much improvment, but the whole point is
>> that you can have two processors doing two completely different things
>> with half the lag to your pc.... dual processors aren`t speed boosts
>> they are performance boosts.
>
>IF you have two CPU-bound tasks to be performed. Most users rarely have
>_one_ CPU-bound task.

The software engineering community is starting to acknowledge the gap.
With single CPU performance doubling every 18 months, there was relatively
little need to think about writing individual applications that are
multi-threaded. These days, with single CPU performance growth slowing
and chip vendors going multi-core, people are starting to say things like
this:

http://www.gotw.ca/publications/concurrency-ddj.htm

30 years from now, single CPU machines might seem as quaint and niche as
mainframes and discrete transistors do today.

-Mike
--
http://www.mschaef.com
Anonymous
a b D Laptop
June 27, 2005 4:01:07 PM

Archived from groups: comp.sys.laptops (More info?)

MSCHAEF.COM wrote:

> In article <d9mjeo022up@news2.newsguy.com>,
> J. Clarke <jclarke.usenet@snet.net.invalid> wrote:
>>Cameron Ward wrote:
>>
>>> There is no hype to dual processors!!!!! sure for running one
>>> application you don`t see that much improvment, but the whole point is
>>> that you can have two processors doing two completely different things
>>> with half the lag to your pc.... dual processors aren`t speed boosts
>>> they are performance boosts.
>>
>>IF you have two CPU-bound tasks to be performed. Most users rarely have
>>_one_ CPU-bound task.
>
> The software engineering community is starting to acknowledge the gap.
> With single CPU performance doubling every 18 months, there was relatively
> little need to think about writing individual applications that are
> multi-threaded. These days, with single CPU performance growth slowing
> and chip vendors going multi-core, people are starting to say things like
> this:
>
> http://www.gotw.ca/publications/concurrency-ddj.htm
>
> 30 years from now, single CPU machines might seem as quaint and niche as
> mainframes and discrete transistors do today.

You do grasp the concept of "CPU-Bound" do you not? If the CPU is spending
most of its time waiting for user input, which is the case for most
end-user applications of interest to consumers, then having two or ten or a
billion CPUs waiting isn't bringing anything to the party. If the
performance is limited by CPU performance (that is what "CPU-bound" means),
which it seldom is (the only tasks other than games that I can think of
that have any mass appeal would be video renders), _and_ if the task is
amenable to parallel processing (all tasks are not), _then_ duals will get
a workout.

This is nothing new--they were teaching concurrent programming in the
standard computer science curriculum 20 years ago, and the second computer
I ever worked with back in the late '70s had IIRC 24 processors.

> -Mike

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)
Anonymous
a b D Laptop
June 27, 2005 5:26:20 PM

Archived from groups: comp.sys.laptops (More info?)

In article <d9p9io12fum@news1.newsguy.com>,
J. Clarke <jclarke.usenet@snet.net.invalid> wrote:
...
>You do grasp the concept of "CPU-Bound" do you not?

I skimmed the original article and mis-spoke. My apologies. I should have
hesitated to put my mouth in 'drive' a little longer.

>the second computer
>I ever worked with back in the late '70s had IIRC 24 processors.

Out of curiosity, What was it?

-Mike
--
http://www.mschaef.com
Anonymous
a b D Laptop
June 27, 2005 7:43:56 PM

Archived from groups: comp.sys.laptops (More info?)

J. Clarke wrote:

> Depends on what you're doing. Some tasks are amenable to multiprocessing,
> others aren't. The main benefit of duals IMO is the ability to handle two
> interrupts at the same time. For general use there's no real difference in
> performance, but sometimes there is a difference in responsiveness.

Responsiveness is precisely the main point I was addressing with my
initial comment on dual-cores.

For a given level of "computer performance", the fact that you can two
CPUs doing things at the same time certainly improves things in the
responsiveness front. True that if there is a severe bottleneck in
the rest of the resources, the second CPU won't even have a chance to
start working soon enough to make the whole thing go extra-smooth.

My question, and I guess an interesting question for the OP, is the
following: is the bottleneck in memory/buses/disks really that severe
on a standard notebook?


As for tasks that are amenable to multiprocessing -- aren't you
overlooking the fact that most users (even the regular, plain vanilla
average Joe that does e-mail, browsing, and Office'ing) run more than
one application at a time (you fire the MP3 player and at then switch
to your e-mail; or start a spellcheck while you were downloading
something AND playing MP3, etc.). True that in some of these cases,
the tasks are entirely IO-bound, and thus the responsiveness or
performance boost from a multi-CPU is not that great compared to the
natural multi-tasking that can be done on a single CPU.

For the OP: perhaps it would be interesting to try to find out if
dual-cores will *really* make a difference in the overall performace
of a Notebook -- I certainly proposed so in my previous post, but I'm
starting to wonder if I'm perhaps mis-estimating the issue. Experts
out there? Any comments? (yes, the general issues were already
commented, but what about more specific estimates/answers to the
above question? anyone?)

Carlos
--
Anonymous
a b D Laptop
June 27, 2005 8:59:58 PM

Archived from groups: comp.sys.laptops (More info?)

MSCHAEF.COM wrote:

> You can't really talk to the history of laptops without mentioning the
> Dynabook, conceived at the Xerox PARC.

Oh my GOD!!!! I can't believe this!!

Those Palo Alto Research Center guys have been my ultimate heroes for
quite a while... Everything that everyone in the technology world
claims credit for, is really due to those Xerox guys back in the 70s
(including graphical interfaces, which sadly, most people think it is
thanks to Billy-boy and his gang -- the presumably serious people
attribute it to Apple).

Now it turns out that Notebooks are also their invention?!!!

Those guys *really* are my heroes!

Give me the X !!! Give me the E !!! Give me the R !!!! :-)


Carlos
--
Anonymous
a b D Laptop
June 27, 2005 11:09:47 PM

Archived from groups: comp.sys.laptops (More info?)

Carlos Moreno wrote:

> J. Clarke wrote:
>
>> Depends on what you're doing. Some tasks are amenable to
>> multiprocessing,
>> others aren't. The main benefit of duals IMO is the ability to handle
>> two
>> interrupts at the same time. For general use there's no real difference
>> in performance, but sometimes there is a difference in responsiveness.
>
> Responsiveness is precisely the main point I was addressing with my
> initial comment on dual-cores.
>
> For a given level of "computer performance", the fact that you can two
> CPUs doing things at the same time certainly improves things in the
> responsiveness front. True that if there is a severe bottleneck in
> the rest of the resources, the second CPU won't even have a chance to
> start working soon enough to make the whole thing go extra-smooth.
>
> My question, and I guess an interesting question for the OP, is the
> following: is the bottleneck in memory/buses/disks really that severe
> on a standard notebook?
>
>
> As for tasks that are amenable to multiprocessing -- aren't you
> overlooking the fact that most users (even the regular, plain vanilla
> average Joe that does e-mail, browsing, and Office'ing) run more than
> one application at a time (you fire the MP3 player and at then switch
> to your e-mail; or start a spellcheck while you were downloading
> something AND playing MP3, etc.). True that in some of these cases,
> the tasks are entirely IO-bound, and thus the responsiveness or
> performance boost from a multi-CPU is not that great compared to the
> natural multi-tasking that can be done on a single CPU.

How often are those applications _doing_ anything? Having Explorer,
Outlook, and Office on the desktop doesn't mean that any of them are doing
anything--check the CPU utilization and you'll find that they sit at zero.
So another billion processors doing nothing doesn't help. As for MP3s, XMMS
is currently showing 0.75% on a 1.4 GHz box. The only media playback that
I've encountered that results in high enough utilization for contention to
be an issue is HD playback. Before I got a board with hardware encoding
analog recording put a bit of load on the machine--HD recording seldom goes
above 10%.

> For the OP: perhaps it would be interesting to try to find out if
> dual-cores will *really* make a difference in the overall performace
> of a Notebook -- I certainly proposed so in my previous post, but I'm
> starting to wonder if I'm perhaps mis-estimating the issue. Experts
> out there? Any comments? (yes, the general issues were already
> commented, but what about more specific estimates/answers to the
> above question? anyone?)
>
> Carlos
> --

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)
Anonymous
a b D Laptop
June 27, 2005 11:14:54 PM

Archived from groups: comp.sys.laptops (More info?)

MSCHAEF.COM wrote:

> In article <d9p9io12fum@news1.newsguy.com>,
> J. Clarke <jclarke.usenet@snet.net.invalid> wrote:
> ...
>>You do grasp the concept of "CPU-Bound" do you not?
>
> I skimmed the original article and mis-spoke. My apologies. I should have
> hesitated to put my mouth in 'drive' a little longer.
>
>>the second computer
>>I ever worked with back in the late '70s had IIRC 24 processors.
>
> Out of curiosity, What was it?

CDC 6600. My memory was not correct--it only had 10.

> -Mike

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)
Anonymous
a b D Laptop
June 28, 2005 1:03:52 AM

Archived from groups: comp.sys.laptops (More info?)

In article <d9q26m011nu@news3.newsguy.com>,
J. Clarke <jclarke.usenet@snet.net.invalid> wrote:
>MSCHAEF.COM wrote:
...
>CDC 6600. My memory was not correct--it only had 10.

Small world. I had a few summer internships at a company that had a few
Cyber 170's and 180's. Those were the closest (last, I think) descendents
of the 6600 architecture. In 1990 they ran a process control system that
sampled 50,000 points (spanning the city of Houston) every couple of
seconds. It was an amazing thing to see, even by 2005 standards.

-Mike
--
http://www.mschaef.com
Anonymous
a b D Laptop
June 28, 2005 5:50:48 AM

Archived from groups: comp.sys.laptops (More info?)

MSCHAEF.COM wrote:

> In article <d9q26m011nu@news3.newsguy.com>,
> J. Clarke <jclarke.usenet@snet.net.invalid> wrote:
>>MSCHAEF.COM wrote:
> ...
>>CDC 6600. My memory was not correct--it only had 10.
>
> Small world. I had a few summer internships at a company that had a few
> Cyber 170's and 180's. Those were the closest (last, I think) descendents
> of the 6600 architecture. In 1990 they ran a process control system that
> sampled 50,000 points (spanning the city of Houston) every couple of
> seconds. It was an amazing thing to see, even by 2005 standards.

Nice machines. I was cleaning up a while back and came upon some old
listings--one of them was a core dump from either the 6600 or the Cyber
205. I looked at it and marveled that I used to be able to read them.

>
> -Mike

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)
!