Laptops, History and comparrison to desktop counterparts.

[Guest]

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

 

[Guest]

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)

 

[norm]

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 =----

 

[Guest]

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_revisited.htm
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

 

[Guest]

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

 

[Guest]

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)

 

[Guest]

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

 

[Guest]

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
--

 

[Guest]

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
--

 

[Guest]

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)

 

[Guest]

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)

 

[Guest]

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

 

[Guest]

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)