OT Dual core CPUs versus faster single core CPUs?

[[email protected]]

Three problems:

1. Hardware works fine. It's the software that has bugs. Redundancy
won't help that.

WRONG. Errant software is software that has yet to be released into a
product. (at least where I am from)

Proper, fully functioning software is all that goes into such systems.
Then, they can be called a product.

Hell Firmware even. Hell, you could even make the watchdog/overseer
redundant.... or use three, and develop a "minority report" style.

 

[JosephKK]

On Mon, 05 May 2008 09:53:42 -0700, Joerg

Jim Thompson wrote: [snip]

Probably something modeled with a macro rather than devices?

It's all transistors, discretes, transformers and such. No chips.

Round up the usual suspects... the transformer, driven by an ideal
source ??

I wish they would always do that out here. UPS can be a real issue so
when I have a choice I prefer not to use them. Several times they
selected the backyard entrance (closer to the road ...) to drop off
packages and then it started to rain. Other times they claimed that our
address did not exist. Yet it does since almost 40 years.

Ha! The USPS regularly declared me as non-existent, until I gave them
a ration of shit in a letter-to-the-editor... then the postman came to
the door, "It wasn't me, it was a substitute"... sure ;-)

But I agree, only use UPS _and_ USPS when you can afford to have it
get lost ;-)

...Jim Thompson

There is at least one vendor that i cannot deal with, their defective
shippers software refuses a known correct addresses (UPS).

 

[JosephKK]

Jim Thompson wrote:
@removethispacbell.net> wrote:

Jim Thompson wrote: [snip]

Probably something modeled with a macro rather than devices?

It's all transistors, discretes, transformers and such. No chips.

Round up the usual suspects... the transformer, driven by an ideal
source ??

The only time that's a pain is when it entails waiting for a Digikey
order to arrive. Then I tell the shepherd to "keep a good watch". She
immediately understands and moves over to a place where she can see the
road, then almost explodes when the Fedex truck comes around the corner.
This avoids a packages sitting there at the entrance without me knowing
it has arrived.

My dog barks also, but FedEx and UPS always ring the doorbell here,
whether a signature is needed or not.

I wish they would always do that out here. UPS can be a real issue so
when I have a choice I prefer not to use them. Several times they
selected the backyard entrance (closer to the road ...) to drop off
packages and then it started to rain. Other times they claimed that our
address did not exist. Yet it does since almost 40 years.

Ha! The USPS regularly declared me as non-existent, until I gave them
a ration of shit in a letter-to-the-editor... then the postman came to
the door, "It wasn't me, it was a substitute"... sure ;-)

But I agree, only use UPS _and_ USPS when you can afford to have it
get lost ;-)

Just to be fair and give due when it is due... I ordered some parts
from Digikey last Friday. UPS and FedEx are both 4 days by ground at
about $5 for the first pound. I know because I misread the map and
thought it was 3 days only to have a package come a day later than I
wanted. Seems Digikey is up in that Northwest corner of Minn that is
in the 4 day delivery zone for me. Anyway, I wanted it sooner than 4
days and the faster shipping by UPS and FedEx gets expensive real
fast. So I asked for my last shipment to be sent Priority. Today I
asked about a tracking number and got the response, USPS don't give no
stinkin' tracking numbers! I was thinking I had made a mistake since
I realized that now it might take more than 4 days and I wouldn't have
any way to tell where it was. Just after that I went out to get the
mail and found the package in the mailbox! It came in one business
day!!!

So you can't say the PO is all bad.

It has its limitations. Do not expect them to handle anything heavy
or bulky with grace. On the other hand they do all together too well
with junk mail.

 

[JosephKK]

Logic in FPGA's can be incredibly complex, things like gigantic state
machines, filters, FFTs - and they run for millions of unit-years
without dropping a bit. Even Intel CPUs, which are absolute horrors,
are reliable. Big software systems are unreliable, as a power function
of the size of the program. The obvious path to reliability is to run
smaller programs on severely hardware-protected processors,

What else are you going to do with 1024 CPU's on a chip?

John

Multicore has already run up against memory bandwidth limitations. The
quad cores actually deliver less total performance, generally. Even i
can devise a benchmark to make them look good.

 

[Spehro Pefhany]

Again, the real problem in personal computing, or indeed in realtime
systems, isn't performance: it's reliability. Thinking of multicore as
a performance increaser is just playing the same old tired game. Of
course 1024 CPUs are going to be memory bandwidth limited, not to
mention thermally limited, if you run them all full blast. So don't do
that.

Besides, a benchmark usually evaluates the solution of one algorithm.
One of my gripes about Windows is that some processes occasionally
block all other processes for tens of seconds at a time, unless they
block them forever.

John

Yeah, I was looking at a commercial PCI data acquistion card. Looks
like about 5% analog front and AND high speed ADC, and most of the
rest is massive digital circuitry (1E6+ gates) to keep from losing
data due to Winblows.

Best regards,
Spehro Pefhany

 

[Joerg]

On Mon, 05 May 2008 10:16:33 -0700, Joerg



Jim Thompson wrote:
[snip]

Probably something modeled with a macro rather than devices?
It's all transistors, discretes, transformers and such. No chips.
Round up the usual suspects... the transformer, driven by an ideal
source ??





The only time that's a pain is when it entails waiting for a Digikey
order to arrive. Then I tell the shepherd to "keep a good watch". She
immediately understands and moves over to a place where she can see the
road, then almost explodes when the Fedex truck comes around the corner.
This avoids a packages sitting there at the entrance without me knowing
it has arrived.
My dog barks also, but FedEx and UPS always ring the doorbell here,
whether a signature is needed or not.
I wish they would always do that out here. UPS can be a real issue so
when I have a choice I prefer not to use them. Several times they
selected the backyard entrance (closer to the road ...) to drop off
packages and then it started to rain. Other times they claimed that our
address did not exist. Yet it does since almost 40 years.
Ha! The USPS regularly declared me as non-existent, until I gave them
a ration of shit in a letter-to-the-editor... then the postman came to
the door, "It wasn't me, it was a substitute"... sure ;-)

But I agree, only use UPS _and_ USPS when you can afford to have it
get lost ;-)

Just to be fair and give due when it is due... I ordered some parts
from Digikey last Friday. UPS and FedEx are both 4 days by ground at
about $5 for the first pound. I know because I misread the map and
thought it was 3 days only to have a package come a day later than I
wanted. Seems Digikey is up in that Northwest corner of Minn that is
in the 4 day delivery zone for me. Anyway, I wanted it sooner than 4
days and the faster shipping by UPS and FedEx gets expensive real
fast. So I asked for my last shipment to be sent Priority. Today I
asked about a tracking number and got the response, USPS don't give no
stinkin' tracking numbers! I was thinking I had made a mistake since
I realized that now it might take more than 4 days and I wouldn't have
any way to tell where it was. Just after that I went out to get the
mail and found the package in the mailbox! It came in one business
day!!!

So you can't say the PO is all bad.

It has its limitations. Do not expect them to handle anything heavy
or bulky with grace. On the other hand they do all together too well
with junk mail.

When I am not in a rush I use USPS with Digikey as well. Quite fast and
when the package doesn't fit into the mailbox the carrier brings it up
to the house. You do need a secure mailbox though.

 

[Joerg]

Jim Thompson wrote:
On Mon, 05 May 2008 09:24:30 -0700, Joerg

Jim Thompson wrote:
On Mon, 05 May 2008 07:26:19 -0700, Joerg

JosephKK wrote:
]

Does anyone know what the difference is between an Intel Dual-Core and
the Core 2 Duo? Is one 32bit and the other 64?

This here machine has a dual core and it really shows up as two separate
CPUs in the control panel.
Actually, it seems to be power dissipation.
In that respect I was positively surprised. Did a few >1/2h heavy duty
SPICE sims lately and the CPU fan barely picked up speed. It's very quiet.
1/2 hour is NOT "heavy-duty" ;-)

Ok, it's not like your chip designs. Just came across a puzzler. A
circuit works on the screen, puts out lots of energy but doesn't consume
any. I can set the voltage source series resistance to 1M or whatever
and nothing changes. Hey, I might really be on to something here,
solving global warming and all that.

Guess I'll waltz over to the lab and hit that red button on the Weller.
I do not trust sims all that much.
Probably have an extra source in there somewhere, or you AC source is
providing the power. There are no such things as "puzzlers", just
cockpit errors ;-)

No other sources except one low voltage to the small stuff. It cannot
generate the levels coming out. Yeah, it might be a cockpit error but
there is nothing obvious and mostly it's faster to lash up something in
the lab.

The only time that's a pain is when it entails waiting for a Digikey
order to arrive. Then I tell the shepherd to "keep a good watch". She
immediately understands and moves over to a place where she can see the
road, then almost explodes when the Fedex truck comes around the corner.
This avoids a packages sitting there at the entrance without me knowing
it has arrived.

Spice's "initial conditions" can store a lot of energy!

John

Spice is not the do-all solve-all of the world. ...

Certainly not in the analog high-power world. For RF small signal stuff
it's pretty good though.

... It takes some realism
from the cockpit to get useful results.

I thought 100K in series with the lone source would be enough realism.
Shoulda killed it but didn't

With some apps the time spent on models can be much longer than actually
building stuff in the lab. As long as the parts from Thief River Falls
arrive in due course.

 

[Rich Grise]

.

probably be quite unresponsive and might require hitting The Big Red
Switch (or the equivalent nowadays) to regain control.

Oh, you mean the "ANY" key? ;-P

Cheers!
Rich

 

[[email protected]]

Then how did Windows, Word, IE, and Acrobat ever make it in to the
wild?

John

Oh. I thought we were talking about software, as in what those of us
that design products make.

I didn't know that you wanted to throw the horseshit money ploy
scamware that has been rammed down our throats for the last two decades
into the software "pool".

Like I stated "where I come from". Not Billy Ware. What I generate is
software, and it does work in the item BEFORE it gets called a product.
But that's just me.

 

[JosephKK]

The CPU's don't have bugs; the software does.

John

So you have forgotten about the Pentium FDIV bug already?

 

[JosephKK]

You are missing my point. The fact that tasks run on separate
hardware does not mean they don't share memory and they don't
communicate. You still have all the same issues that a single
processor system has. It is **very** infrequent on my system that it
hangs in a mode where I can't get control of the CPU. I am running
Win2k and I let it run for a month or more between reboots. Usually
the issue that makes me reboot is that the system just doesn't behave
correctly, not that it is stuck in a tight loop with no interrupts
enabled. So multiple processors will do nothing to improve my
reliability.



So this indicates that multiple processors don't fix the problem. The
proper use of hardware memory management fixes the problem. No?



That is the big question. I like the idea of having at least two
processors. I remember some 6 years ago when I was building my
current computers that dual CPUs on a mother board were available.
People who do the kind of work that I do said they could start an HDL
compile and still use the PC since they each had a processor. I am
tired of my CPU being sucked dry by my tools or even by Adobe Acrobat
during a download and the CPU nearly halting all other tasks. Of
course, another solution is to ditch the Adobe tools. Next to
Microsoft, they are one of the very worst software makers.

Personally, I don't think we need to continue to increase processing
at this geometric rate. Since we can't, I guess I won't be
disappointed. I see the processor market as maturing in a way that
will result in price becoming dominant and "speed" being relegated to
the same category as horsepower. The numbers don't need to keep
increasing all the time, they just need to match the size of the car
(or use for PCs). The makers are not ready to give up the march of
progress just yet, but they have to listen to the market. It will be
within 5 years that nobody cares about the processor speed or how many
CPUs your computer has. It will be about the utility. At that point
the software will become the focus as the "bottleneck" in speed,
reliability and functionality. Why else does my 1.4 GHz CPU seem to
respond to my keystrokes at the same (or slower) speed than my 12 MHz
286 from over 10 years ago? "It's the software, stupid!" :^)



I am ready to buy a laptop and I am going to get a Dell because they
will sell an XP system rather than Vista. Everything I have heard is
that XP is at least as good as Win2K. No?

Pretty much. I have had XP running for two months continuous. Did
that with 98SE as well.

 

[Dennis]

I am truly amazed in the current generation of P4 chips that the
register colouring and speculative execution doesn't cause more problems.

Why? This is known technology since the 1970's mainframes.

 

[krw]

Why? This is known technology since the 1970's mainframes.

As are the verification techniques.

 

[The Real Andy]

Not trying to decide which is better for everybody. Just interested
in the difference between multiple core CPUs and faster single core
CPUs.

Are there any mainstream applications that would benefit from one
and the other?

My wild guess. Continuous multitasking versus intermittent bursts
(if the bursts usually do not coincide). But I don't know of any
applications to example that.

Thanks.

Funny enough, I have a P4 3.2ghz hyperthreaded box (not real dual cpu)
at home and have a 2.3ghz core 2 duo at work. Both have 2ghz or ram.
Speed differenc is negligable. A few things are faster, notably doing
large builds of software, and SQL server. OTherwise they are the same.

Unless your are doing stuff that can benifit from concurrent
threading, then 2 cores is no better than one. 99% of programmers dont
understand concurrency, and of those who use most cant implement it.

The reason my software builds run faster, is that each core can be
compiling a seperate object concurrently. The reason SQL server speeds
up, is that it can be inserting some data, and updating an index at
the same time. Obviously there is more to it than that, but thats the
general idea.

What most people dont understand these days is that the biggest
performance gains are not had by upgrading processors and adding
memory, its in fact by getting faster hard drives. TRy it sometime