A Good, Inexpensive Single-Core cpu

On Wed, 30 Jul 2008 10:08:57 -0700 (PDT), Robert Myers wrote:

On Jul 28, 1:09*pm, Puddin' Man wrote:
Does Don Knuth know anything about computing?

Donald Knuth knows quite a lot about computing. The question is
whether he is the source to go to as computing moves away from single
core processors, and the answer is that he is not. Even he says so:

"Other people understand parallel machines much better than I do;
programmers should listen to them, not me, for guidance on how to deal
with simultaneity."

I've already discussed Knuth's position. You can read or not read
previous posts as you wish.

It's not enough that Knuth is great at algorithms. He is like
"a religion/culture of one" in that he looks at computing -only-
for computing's sake regardless of what the industry, etc is saying.
He is impervious to hype.

His denigration of the usefulness of multiple cores has to be viewed
in that light. Knuth claims that those who will most benefit from
multiple cores are not typical users,

They're not. Unless you belong to a "multi-core religion", it's
indisputable (for common apps). Did you run a "dualie" years ago? :-)

but we're probably more typical
than he is.

Speak for yourself. Near as I can tell, Knuth's common usage is -very-
similar to mine.

Most users will get some benefit from at least one more
core.

That's not disputed. Becomes a question of degree. Do you imagine
you could measure it (again, for common usage as we've been discussing)?
I'm not talking about a seat-of-the-pants guess or an obviously biased
estimate.

I'd advise buying a single core platform only where size,
noise, or power consumption are critical.

There may well be no practical alternative to multi-core for desktop,
because it appears that m-c is all they want to produce/market.

P


" ... and the bees made honey in the lion's head."
- from "If I Had My Way", Blind Willie Johnson

On Jul 30, 5:50*pm, Puddin' Man wrote:
On Wed, 30 Jul 2008 10:08:57 -0700 (PDT), Robert Myers wrote:
On Jul 28, 1:09*pm, Puddin' Man wrote:
Does Don Knuth know anything about computing?

Donald Knuth knows quite a lot about computing. *The question is
whether he is the source to go to as computing moves away from single
core processors, and the answer is that he is not. *Even he says so:

"Other people understand parallel machines much better than I do;
programmers should listen to them, not me, for guidance on how to deal
with simultaneity."

I've already discussed Knuth's position. You can read or not read
previous posts as you wish.

I wasn't soliciting your wisdom about Knuth. I've already very tartly
discussed the appropriateness of constantly referring to him when the
world of hardware he imagined is rapidly vanishing. We need to pin a
medal on the man and stop referring to him. Time to get a move-on.
Time's a wasting and Knuth has become a distraction.

It's not enough that Knuth is great at algorithms. He is like
"a religion/culture of one" in that he looks at computing -only-
for computing's sake regardless of what the industry, etc is saying.
He is impervious to hype.

That's just plain silly. Mathematicians don't make microprocessors.
Manufacturers do. "I've got some great ideas, if only they'd abandon
backward compatibility." This man is not in touch with reality.

His denigration of the usefulness of multiple cores has to be viewed
in that light. *Knuth claims that those who will most benefit from
multiple cores are not typical users,

They're not. Unless you belong to a "multi-core religion", it's
indisputable (for common apps). Did you run a "dualie" years ago? :-)

but we're probably more typical
than he is. *

Speak for yourself. Near as I can tell, Knuth's common usage is -very-
similar to mine.

Most users will get some benefit from at least one more
core. *

That's not disputed. Becomes a question of degree. Do you imagine
you could measure it (again, for common usage as we've been discussing)?
I'm not talking about a seat-of-the-pants guess or an obviously biased
estimate.

The metric (did I just say that?) that matters most is perceived
responsiveness, and it's not such an easy thing to measure. I don't
get all the responsiveness I want all the time in spite of using a
modern dual core processor, and I've heard many others say the same.
The fact that your second core is idling most of the time doesn't
matter if it keeps your browser from becoming unresponsive at awkward
moments.

I'd advise buying a single core platform only where size,
noise, or power consumption are critical.

There may well be no practical alternative to multi-core for desktop,
because it appears that m-c is all they want to produce/market.

And since manufacturers have become very good at minimizing power
consumption by hardware that idles, what incentive is there for you to
hunt down that single core machine, anyway? Would you buy the single
core processor if it were more expensive, as it should be?

Robert.

Puddin' Man wrote:
They're not. Unless you belong to a "multi-core religion", it's
indisputable (for common apps). Did you run a "dualie" years ago? :-)

Yup. The majority (narrowly) of my professional coding (and much of my
sysadmin experience) experience is on dual-cpu boxes going back to the PPro
(and I did work on a nonstandard multi-socket P-Classic server, ugh),
although at some point in a several more years it will balance out with if I
can't talk my employer into getting the dual-dual or dual-quad for my
desktop

There may well be no practical alternative to multi-core for desktop,
because it appears that m-c is all they want to produce/market.

Just buy cheapie Celerons, if that's all you need. Meanwhile, the cost
difference between a 3ghz E6850 and a hypothetical 3ghz Core Solo is not
going to be fairly significant.

GUIs are inherently parallel.

--
Nate Edel http://www.cubiclehermit.com/
preferred email |
is "nate" at the | "This humorous tagline or quotation is not in service."
posting domain |

On Wed, 30 Jul 2008 15:36:25 -0700 (PDT), Robert Myers wrote:

On Jul 30, 5:50*pm, Puddin' Man wrote:
On Wed, 30 Jul 2008 10:08:57 -0700 (PDT), Robert Myers wrote:
On Jul 28, 1:09*pm, Puddin' Man wrote:
Does Don Knuth know anything about computing?

Donald Knuth knows quite a lot about computing. *The question is
whether he is the source to go to as computing moves away from single
core processors, and the answer is that he is not. *Even he says so:

"Other people understand parallel machines much better than I do;
programmers should listen to them, not me, for guidance on how to deal
with simultaneity."

I've already discussed Knuth's position. You can read or not read
previous posts as you wish.

I wasn't soliciting your wisdom about Knuth. I've already very tartly
discussed the appropriateness of constantly referring to him when the
world of hardware he imagined is rapidly vanishing. We need to pin a
medal on the man and stop referring to him. Time to get a move-on.
Time's a wasting and Knuth has become a distraction.

To yourself no doubt. Not to me.

It's not enough that Knuth is great at algorithms. He is like
"a religion/culture of one" in that he looks at computing -only-
for computing's sake regardless of what the industry, etc is saying.
He is impervious to hype.

That's just plain silly.

What is silly? That he is impervious to hype??

Mathematicians don't make microprocessors.
Manufacturers do. "I've got some great ideas, if only they'd abandon
backward compatibility."

You wanna point to a source for your "quote" of Knuth's words? :-)

This man is not in touch with reality.

Perhaps not with your particular reality ...

I'm not into hero worship at all. It's just that I look and look and
look and I see noone in the industry who has a valid long-term claim
to independent thinking on such subjects. The garbage-hype seems to
affect Knuth like water off a duck's back.

I think Knuth makes a valid point, and I'm not much concerned with
the fact that you cannot see it.

His denigration of the usefulness of multiple cores has to be viewed
in that light. *Knuth claims that those who will most benefit from
multiple cores are not typical users,

They're not. Unless you belong to a "multi-core religion", it's
indisputable (for common apps). Did you run a "dualie" years ago? :-)

but we're probably more typical
than he is. *

Speak for yourself. Near as I can tell, Knuth's common usage is -very-
similar to mine.

Most users will get some benefit from at least one more
core. *

That's not disputed. Becomes a question of degree. Do you imagine
you could measure it (again, for common usage as we've been discussing)?
I'm not talking about a seat-of-the-pants guess or an obviously biased
estimate.

The metric (did I just say that?) that matters most is perceived
responsiveness,

Ahhhhh. Perception! :-)

and it's not such an easy thing to measure. I don't
get all the responsiveness I want all the time in spite of using a
modern dual core processor, and I've heard many others say the same.
The fact that your second core is idling most of the time doesn't
matter if it keeps your browser from becoming unresponsive at awkward
moments.

For Gawd's sake, Bob. In any ordinary circumstance and quite a few
that are not ordinary, your browser is i/o bound and not in the least
ready to gobble cpu cycles from additional instruction processors.

I'd advise buying a single core platform only where size,
noise, or power consumption are critical.

There may well be no practical alternative to multi-core for desktop,
because it appears that m-c is all they want to produce/market.

And since manufacturers have become very good at minimizing power
consumption by hardware that idles, what incentive is there for you to
hunt down that single core machine, anyway? Would you buy the single
core processor if it were more expensive,

All other things being equal, no.

as it should be?

It shouldn't be.

P

" ... and the bees made honey in the lion's head."
- from "If I Had My Way", Blind Willie Johnson

On Jul 30, 11:04*pm, Puddin' Man wrote:
On Wed, 30 Jul 2008 15:36:25 -0700 (PDT), Robert Myers wrote:

It's not enough that Knuth is great at algorithms. He is like
"a religion/culture of one" in that he looks at computing -only-
for computing's sake regardless of what the industry, etc is saying.
He is impervious to hype.

That's just plain silly.

What is silly? That he is impervious to hype??

It's silly on two counts:

1. That Knuth is impervious to hype, or that Knuth doesn't have his
own silly hangups, as manifestly he does. Being a bit of a crank
isn't a sin, and smart people have their moments just like everyone
else.

2. That there are not other similarly informed and talented users
capable of making discriminations equally or even more insightful and
valid.

Mathematicians don't make microprocessors.
Manufacturers do. *"I've got some great ideas, if only they'd abandon
backward compatibility." *

The exact quote:

"(But my MMIX design contains several ideas that would substantially
improve the current performance of the kinds of programs that concern
me most—at the cost of incompatibility with legacy x86 programs.)"

I'm going to be even rougher on the sainted Knuth. Many of us could
design better microprocessors for our own applications if the rest of
the world could be told to take hike. No one but Knuth would be
quoted for such nonsense.

You wanna point to a source for your "quote" of Knuth's words? :-)

This man is not in touch with reality.

Perhaps not with your particular reality ...

You think so? You think there's a chip manufacturer out there who's
going to build a one-off microprocessor to compete with x86? You're
out of touch with reality, too.

I'm not into hero worship at all. It's just that I look and look and
look and I see noone in the industry who has a valid long-term claim
to independent thinking on such subjects. The garbage-hype seems to
affect Knuth like water off a duck's back.

I think Knuth makes a valid point, and I'm not much concerned with
the fact that you cannot see it.

What new information are you conveying here? Do you have any notion
of how many really, really smart people there are in this business?
Knuth's tomes will no longer be so central and he doesn't like it.
*shrug*


The metric (did I just say that?) that matters most is perceived
responsiveness,

Ahhhhh. Perception! :-)

What's this? Argumentation by derision? Most desktop users care
about how responsive the system is. A system that won't track natural
mouse movements is too slow, and, in some circumstances, nearly
useless. Perception matters more than anything else. Knuth's
theoretical considerations (and yours) are the kind of analysis that
might be appropriate for HPC or server applications.

and it's not such an easy thing to measure. *I don't
get all the responsiveness I want all the time in spite of using a
modern dual core processor, and I've heard many others say the same.
The fact that your second core is idling most of the time doesn't
matter if it keeps your browser from becoming unresponsive at awkward
moments.

For Gawd's sake, Bob. In any ordinary circumstance and quite a few
that are not ordinary, your browser is i/o bound and not in the least
ready to gobble cpu cycles from additional instruction processors.

I didn't just pick the example out of the air. It's easy to see
what's gobbling cycles when the system becomes unresponsive. If I'm
having trouble navigating tabs, it's not an I/O problem, and the
system monitor agrees.

I'd advise buying a single core platform only where size,
noise, or power consumption are critical.

There may well be no practical alternative to multi-core for desktop,
because it appears that m-c is all they want to produce/market.

And since manufacturers have become very good at minimizing power
consumption by hardware that idles, what incentive is there for you to
hunt down that single core machine, anyway? *Would you buy the single
core processor if it were more expensive,

All other things being equal, no.

as it should be?

It shouldn't be.

It has to be worth a manufacturer's while to offer a separate
product. Apparently, AMD thinks it's worth it enough to sell three
core chips, which I assume are sold at a lower price. Unless the cost
saving of salvaging dual core chips with one bad core is significant,
then the product you want to buy is just one more SKU to keep track of
and market. The market for such chips is already well-served by other
products, so it's going to be a low-volume product. The only
incentive I can see is that some customers will just demand it, in
which case, they can be expected to pay more.

Robert.

On Wed, 30 Jul 2008 13:35:46 -0500, General Schvantzkopf wrote:


The only reason I'd spring for dual-core for the nickel/dime usage would
be because Intel/AMD have virtually abandoned the market for good,
inexpensive single-core cpu's (for desktop). I will assume they have
done so until I have evidence to the contrary.

The Atom and the C7 are exactly what you are looking for. I've played
with the $180 Everest C7 box and it's runs Fedora just fine. If all you
want to do is surf the web, do e-mail and run OpenOffice, these
processors are more than good enough. For basic desktop computing CPUs
have been fast enough for years. The current crop of netbook processors
like the Atom, and the C7, can do the job on a few watts of power. You
can't run Vista on these things but you can run any distro of Linux that
you want. Microsoft has kept a barebones version of XP around for use on
netbooks so they wouldn't be left out of the netbook market.

I'm not a Linux guy.

Mea culpa. I should have explicitly queried for "A Good, Inexpensive Single-
Core cpu" for desktop, implying that at least moderately robust chipset,
motherboards, etc for desktop existed.

I should add that the appeal of these processors is their extremely low
power consumption, not their price. If you look at pricewatch you'll find
bottom of the line dual core AMD systems for around $200, about the same
as the Everest C7 box.

And that would also be an issue.

Thx,
P

"I Ain't Blind, I Just Don't Wanna See"
- the title of a tune by Little Joe Blue, maybe 1966

You don't understand.

Knuth is not the issue.

I am not the issue.

You are not the issue.

The issue is cpu cycle utilization.

Now. The net has long been rife with experts and genii who are
looking for a ****ing contest. I am just a po' feller who counts
idle cpu cycles.

So I cannot further discuss the matter with you.

Goo-bye, S'Long.

Puddin'

"I Ain't Blind, I Just Don't Wanna See"
- the title of a tune by Little Joe Blue, maybe 1966


On Wed, 30 Jul 2008 21:16:51 -0700 (PDT), Robert Myers wrote:

On Jul 30, 11:04*pm, Puddin' Man wrote:
On Wed, 30 Jul 2008 15:36:25 -0700 (PDT), Robert Myers wrote:

It's not enough that Knuth is great at algorithms. He is like
"a religion/culture of one" in that he looks at computing -only-
for computing's sake regardless of what the industry, etc is saying.
He is impervious to hype.

That's just plain silly.

What is silly? That he is impervious to hype??

It's silly on two counts:

1. That Knuth is impervious to hype, or that Knuth doesn't have his
own silly hangups, as manifestly he does. Being a bit of a crank
isn't a sin, and smart people have their moments just like everyone
else.

2. That there are not other similarly informed and talented users
capable of making discriminations equally or even more insightful and
valid.

Mathematicians don't make microprocessors.
Manufacturers do. *"I've got some great ideas, if only they'd abandon
backward compatibility." *

The exact quote:

"(But my MMIX design contains several ideas that would substantially
improve the current performance of the kinds of programs that concern
me most—at the cost of incompatibility with legacy x86 programs.)"

I'm going to be even rougher on the sainted Knuth. Many of us could
design better microprocessors for our own applications if the rest of
the world could be told to take hike. No one but Knuth would be
quoted for such nonsense.

You wanna point to a source for your "quote" of Knuth's words? :-)

This man is not in touch with reality.

Perhaps not with your particular reality ...

You think so? You think there's a chip manufacturer out there who's
going to build a one-off microprocessor to compete with x86? You're
out of touch with reality, too.

I'm not into hero worship at all. It's just that I look and look and
look and I see noone in the industry who has a valid long-term claim
to independent thinking on such subjects. The garbage-hype seems to
affect Knuth like water off a duck's back.

I think Knuth makes a valid point, and I'm not much concerned with
the fact that you cannot see it.

What new information are you conveying here? Do you have any notion
of how many really, really smart people there are in this business?
Knuth's tomes will no longer be so central and he doesn't like it.
*shrug*


The metric (did I just say that?) that matters most is perceived
responsiveness,

Ahhhhh. Perception! :-)

What's this? Argumentation by derision? Most desktop users care
about how responsive the system is. A system that won't track natural
mouse movements is too slow, and, in some circumstances, nearly
useless. Perception matters more than anything else. Knuth's
theoretical considerations (and yours) are the kind of analysis that
might be appropriate for HPC or server applications.

and it's not such an easy thing to measure. *I don't
get all the responsiveness I want all the time in spite of using a
modern dual core processor, and I've heard many others say the same.
The fact that your second core is idling most of the time doesn't
matter if it keeps your browser from becoming unresponsive at awkward
moments.

For Gawd's sake, Bob. In any ordinary circumstance and quite a few
that are not ordinary, your browser is i/o bound and not in the least
ready to gobble cpu cycles from additional instruction processors.

I didn't just pick the example out of the air. It's easy to see
what's gobbling cycles when the system becomes unresponsive. If I'm
having trouble navigating tabs, it's not an I/O problem, and the
system monitor agrees.

I'd advise buying a single core platform only where size,
noise, or power consumption are critical.

There may well be no practical alternative to multi-core for desktop,
because it appears that m-c is all they want to produce/market.

And since manufacturers have become very good at minimizing power
consumption by hardware that idles, what incentive is there for you to
hunt down that single core machine, anyway? *Would you buy the single
core processor if it were more expensive,

All other things being equal, no.

as it should be?

It shouldn't be.

It has to be worth a manufacturer's while to offer a separate
product. Apparently, AMD thinks it's worth it enough to sell three
core chips, which I assume are sold at a lower price. Unless the cost
saving of salvaging dual core chips with one bad core is significant,
then the product you want to buy is just one more SKU to keep track of
and market. The market for such chips is already well-served by other
products, so it's going to be a low-volume product. The only
incentive I can see is that some customers will just demand it, in
which case, they can be expected to pay more.

Robert.

Puddin' Man wrote:
Does Intel (or AMD for that matter) make A Good, Inexpensive Single-
Core cpu that doesn't heat the entire county (like Prescott)?? :-)


Apparently not, but some motherboards allow you to disable cores through the
BIOS.

Note that Windows may not work if you install it with [one | two] cores
enabled and then try to boot it with [two | one] core enabled.

But I read that if you're doing multi-threading in your programming and
using those programs on other computers which might have multi-core
processors installed, it is safer to do the testing on a multi-core system
so that any threading issues are more likely to appear.

On the positive side, "Nehalem" processors will be available in a few months
with perhaps as many as eight cores d&c.

Andrew

On Jul 31, 9:58*am, Puddin' Man wrote:


The issue is cpu cycle utilization.

Now. The net has long been rife with experts and genii who are
looking for a ****ing contest. I am just a po' feller who counts
idle cpu cycles.

If you don't want a ****ing contest then don't make extravagant claims
based on your opinion of one man and don't demand that your
correspondent produce "exact quotes" that you should have been able to
find yourself by reading the article you cited.

The issue *isn't* average utilization, which you call CPU cycles, if
the system is being used interactively. If you want to persist in
seeing it that way, that's your business.

As to your snarky remarks about genii and experts, I make no claim to
be either. In the name of meeting your personal needs, you've spread
unhelpful misconceptions in a place where it might matter. The future
is concurrent processing, and the sooner everyone starts operating
that way, the better off we'll be. As it stands, the state of the art
is pathetic and Knuth is a negative contributor.

Robert.

I should add that the appeal of these processors is their extremely low
power consumption, not their price. If you look at pricewatch you'll
find bottom of the line dual core AMD systems for around $200, about the
same as the Everest C7 box.

And that would also be an issue.

Thx,
P

My point was that there are dual cores available at an extremely low
price also, as long as you are willing to use Linux. Pricewatch has
Athlon X2 4200+ boxes with no OS listed for as little as $165. That's a
512M box with a 40G drive, it fully capable of running any Linux distro
although you would be better off spending an extra $10 and up it to 1G of
RAM. That box wouldn't be adequate for Vista (which costs more that the
whole box anyway) but it's perfectly adequate for a basic desktop Linux
system. If will run XP fine also except that MS won't sell you a new copy
of XP anymore unless you also agree to buy an expensive version of Vista
(XP downgrade rights aren't available with Vista Home).

There isn't a lot of downside to using dual core processors, even if one
core would do, except for the power consumption. The Atom addresses the
power issue but once you get down to the bottom end of the performance
spectrum the CPU only makes up about 25% of the system price. The reason
that Intel is building Atoms is to enable a new class of ultraportable
devices not to reduce the cost of regular desktop systems.