A8N-E and dual core Athlon X2

  • Thread starter Thread starter James
  • Start date Start date
J

James

Well, I have the 1005 BIOS now and I'm thinking about upgrading to a
4400+ dual core Athlon CPU.

My question is, is the performance increase worth the $650 for the chip?

I would also like to upgrade my OS to Windows XP64, but I'm having
trouble finding 64-bit drivers for everything.

Has anyone seen any benchmarks for the dual core X2 under XP vs XP64?

James
 
James said:
Well, I have the 1005 BIOS now and I'm thinking about upgrading to a 4400+
dual core Athlon CPU.

My question is, is the performance increase worth the $650 for the chip?

I would also like to upgrade my OS to Windows XP64, but I'm having trouble
finding 64-bit drivers for everything.

Has anyone seen any benchmarks for the dual core X2 under XP vs XP64?

James

I have seen benchmarks posted on the web, but don't recall where. Try
Google.

It really depends on the application as to whether it is worth it. For most
desktop uses, it is not worth it because most applications cannot exploit
dual processors.
 
Mark A said:
I have seen benchmarks posted on the web, but don't recall where. Try
Google.

It really depends on the application as to whether it is worth it. For
most desktop uses, it is not worth it because most applications cannot
exploit dual processors.

Yes, but it's kind of nice to burn a CD/DVD and surf the web at the same
time. Each core has it's own workload, hoohay!
And, these dual cores, except for games, are ripping up benchmarks.

john
 
name said:
Yes, but it's kind of nice to burn a CD/DVD and surf the web at the same
time. Each core has it's own workload, hoohay!
And, these dual cores, except for games, are ripping up benchmarks.

john
My AMD64 3500+ can burn a CD/DVD and I surf the web at the same time without
any noticeable loss of speed. There are no rational arguments that will
convince a geek that the dual core X2 is not absolutely necessary for their
personal happiness and eternal salvation.

But please don't insult my intelligence and try to convince me that more
than a handful of all PC users will noticeably benefit enough to justify the
price of a dual core CPU with today's software applications . It makes me
very angry when you insult my intelligence.
 
What processor do you have now in your A8N-E board?

Are you happy with the A8N-E board?

Any OC'ing on the board.

There are no reviews of this A8N-E anywhere which is so odd for a mainstream
board, especially a name brand like ASUS. I've read in forums that many
people are happy with the A8N-E.
There are so many reviews of the DFI Ultra D. DFI must send them out to
everyone to review.

I just bought one today but am waiting for an XP-90 cooler I ordered over
the internet. I have a FX53 chip to put in it.
 
My FX-53 burns CD's/DVD's and surf's the web like a rocket. If I was
into CAD or doing intensive video/graphics editing I might consider a
dual core cpu but only after my current processor proved to be inadequate.

Mark said:
My AMD64 3500+ can burn a CD/DVD and I surf the web at the same time without
any noticeable loss of speed. There are no rational arguments that will
convince a geek that the dual core X2 is not absolutely necessary for their
personal happiness and eternal salvation.

But please don't insult my intelligence and try to convince me that more
than a handful of all PC users will noticeably benefit enough to justify the
price of a dual core CPU with today's software applications . It makes me
very angry when you insult my intelligence.

--
**********************************************************************************************************************************
David F. Mishiwiec Sr.

"It is possible to support the members of our armed services while opposing the foreign policy that has them placed in harms way."
**********************************************************************************************************************************
 
Wookie said:
What processor do you have now in your A8N-E board?

Are you happy with the A8N-E board?

Any OC'ing on the board.

There are no reviews of this A8N-E anywhere which is so odd for a
mainstream board, especially a name brand like ASUS. I've read in forums
that many people are happy with the A8N-E.
There are so many reviews of the DFI Ultra D. DFI must send them out to
everyone to review.

I just bought one today but am waiting for an XP-90 cooler I ordered over
the internet. I have a FX53 chip to put in it.
There are now 4 different versions of the Asus A8N series (E, SLI, SLI
Deluxe, and SLI Platinum) so it is a little redundant to include them all in
reviews.

The A8N-E uses the same OC bios and software as the others and is reportedly
fairly good. I don't OC because my 3500+ has lots of power and I don't have
time to mess with it. Maybe in few years I will crank it up before upgrading
to a faster system.
 
James said:
Well, I have the 1005 BIOS now and I'm thinking about upgrading to a
4400+ dual core Athlon CPU.

My question is, is the performance increase worth the $650 for the chip?
When it comes to price/performance ratio the powerchips are always bad
compared to the mainstreamchips.

The question is; do you really need this power?

If Yes, then take it, if No then wait until you need it. In a year dualcore
CPUs are quite familiar, and you can get the same CPU maybe for half the
price.
I would also like to upgrade my OS to Windows XP64, but I'm having
trouble finding 64-bit drivers for everything.
Have you more than 4 Gb memory? If not XP64 gives you no real advance, only
trouble to get all of your hardware working, cause its too new.
Has anyone seen any benchmarks for the dual core X2 under XP vs XP64?
No, but on a standard home-PC I don´t believe that there is much difference.

Conclusion, you do not really need always the newest and hotest on the
hardware market, if you are not a real poweruser or have to much money or
you like it very much to play the beta tester for the industry.

Tschüß
Chris
 
Christoph Spies said:
When it comes to price/performance ratio the powerchips are always bad
compared to the mainstreamchips.
It is not just a question of power. It is also a question of which software
can take advantage of two processors at once. Right now, there are not many
desktop applications that can do that, and frankly, I don't think that will
change too much in the near future. The dual core CPU's are better suited to
servers where multiple CPU intensive applications are running at one time.

But if you are a multi-media or design professional that spends all day
working with a multi-threaded application like Adobe Photoshop or certain
CAD applications, then it may be worth it to have a dual core system. Many
of these type users already have (or least tried) dual processor systems, so
they know whether there is a benefit for their workload to have 2 CPU's
available at one time.
 
Wookie said:
What processor do you have now in your A8N-E board?

Athlon 3500 Venice
Are you happy with the A8N-E board?

I've owned many Asus MBs over the years and I really like the A8N-E. So
far it has been a rock solid performer. I am especially impressed with
how stable it has been compared with other Asus MBs I have owned.
Any OC'ing on the board.

No. My past experience with this has shown me that running on the edge
of the envelope risks instability...and I like stability ;) IOW, the
slight performance gains aren't worth the risk of lockups. Not that I
don't like blazingly fast computers - that is why I bought the A8N-E -
but for me it must operate reliably at the same time.
There are no reviews of this A8N-E anywhere which is so odd for a mainstream
board, especially a name brand like ASUS. I've read in forums that many
people are happy with the A8N-E.

You can add one more to that list.
I just bought one today but am waiting for an XP-90 cooler I ordered over
the internet. I have a FX53 chip to put in it.

I think you will be very happy with your choices... As for me, well, I
think I'm going to spring for the Athlon 4400 Toledo. I know it is $650
minimum at this time, but I'm a big kid who likes to blow his hard
earned cash on cutting edge toys. No good reason other than that.

James :o)
 
Mark said:
It is not just a question of power. It is also a question of which software
can take advantage of two processors at once. Right now, there are not many
desktop applications that can do that, and frankly, I don't think that will
change too much in the near future. The dual core CPU's are better suited to
servers where multiple CPU intensive applications are running at one time.

Are you absolutely positive that the dual core CPUs are only an
advantage to applications that are designed to use them? If so, then the
extra CPU core just sits there and does nothing otherwise?

James
 
James said:
Are you absolutely positive that the dual core CPUs are only an advantage
to applications that are designed to use them? If so, then the extra CPU
core just sits there and does nothing otherwise?

James
Yes. The extra CPU core just sits there unless you are running two
applications at once, or a single multi-threaded application. It is very
hard to design multi-threaded applications since the programmer has to split
the workload into two (or more) pieces and then reassemble the results back
into a single result. There is extra overhead inherent in this type of
application design (reassembling the pieces into a single result), so it is
not used unless the tasks are fairly intensive.

For most desktop users, given a fixed amount of money to spend, they would
see much better performance with a single faster CPU, than a multi-core CPU.
As I said previously, there are some exceptions with a some professional
multi-media and design applications specifically designed for with
multi-threading.

If you Google this subject, you can find reviews that back up what I said.
Some applications (like Office) may actually slower with dual-core CPU's.
 
James said:
Are you absolutely positive that the dual core CPUs are only an advantage
to applications that are designed to use them? If so, then the extra CPU
core just sits there and does nothing otherwise?

James
In the same manner that software has to be "aware" to take advantage of
hyperthreading on an
Intel proc? Even though there is no improvement in most applications with
hyperthreading enabled, Intel would have you believe it is the greatest
thing that ever happened to computing. People foolish believe the hype and
spend their hard earned dollars on an Intel HT proc...so sad.

However, it is much easier to port apps to take advantage of dual core than
it is to make it aware of hyperthreading...add this to the huge superiority
that AMD has with the hypertransport bus (which has nothing to do with
hyperthreading), and AMD is a better choice. In the long run, apps such as
video editing software, CADD programs, and other processor intensive
software will directly benefit from the dual core technology. The server
argument is no longer a valid one; Windows XP Professional x64 Edition is
based on Server 2003 x64 code, it will benefit the user running it on a
consumer dual core platform. The same will be true of Longhorn when it is
released.

In short, don't count dual core out...it's just getting started.

Bobby
 
Mark A said:
Yes. The extra CPU core just sits there unless you are running two
applications at once, or a single multi-threaded application. It is very
hard to design multi-threaded applications since the programmer has to
split the workload into two (or more) pieces and then reassemble the
results back into a single result. There is extra overhead inherent in
this type of application design (reassembling the pieces into a single
result), so it is not used unless the tasks are fairly intensive.
Not quite.
The system will still do it's own 'housekeeping' on the second processor.
So IDE I/O, software RAID calculations etc., will give a slight advantage
over a single processor. The biggest gain in this form, is when you have a
single application that likes to hog 100% processor time, when you will
still see the a fast keyboard response with the dual core system. Software
RAID, and file compression (if using a compressed drive on XP), are two
things that give more gain. However these all come at a slight 'cost' from
the more complex kernel.
For most desktop users, given a fixed amount of money to spend, they
would see much better performance with a single faster CPU, than a
multi-core CPU. As I said previously, there are some exceptions with a
some professional multi-media and design applications specifically
designed for with multi-threading.
The 'rule of thumb', is that when using non multithreaded applications,
you will see between 20%, and 40% performance gain, unless the application
is unusual, and is almost entirely 'memory based', when you will see a
slight loss because of the extra kernel overhead. Unfortunately, some
Windows applications are this way inclined... The biggest gain though
comes with multiple applications (which is a thing very rarely 'done' in
Windows - how often do you leave an application 'processing' a major
mathematical task, while you run something else? - generally Windows users
'task jump', running just one task at a time, but with several 'active'),
or with a properly written MP application.
If you Google this subject, you can find reviews that back up what I
said. Some applications (like Office) may actually slower with dual-core
CPU's.
Though, they still respond better to user input.
Generally, unless running an MP application, expect small gains, rather
than large ones.

Best Wishes
 
NoNoBadDog! said:
In the same manner that software has to be "aware" to take advantage of
hyperthreading on an
Intel proc? Even though there is no improvement in most applications with
hyperthreading enabled, Intel would have you believe it is the greatest
thing that ever happened to computing. People foolish believe the hype and
spend their hard earned dollars on an Intel HT proc...so sad.

However, it is much easier to port apps to take advantage of dual core
than it is to make it aware of hyperthreading...add this to the huge
superiority that AMD has with the hypertransport bus (which has nothing to
do with hyperthreading), and AMD is a better choice. In the long run,
apps such as video editing software, CADD programs, and other processor
intensive software will directly benefit from the dual core technology.
The server argument is no longer a valid one; Windows XP Professional x64
Edition is based on Server 2003 x64 code, it will benefit the user running
it on a consumer dual core platform. The same will be true of Longhorn
when it is released.

In short, don't count dual core out...it's just getting started.

Bobby
The reasons why servers are more appropriate for dual core CPU's is because
servers (especially application servers and database servers) typically
handle multiple client requests at one time, and each one can be handled by
a separate CPU without having to multi-thread any one individual client
request (which most applications cannot do). This is the same reason why
many servers typically have 2, 4, ,8 (or even more) completely separate
CPU's. I am not talking about file or print servers, which can usually
operate fine with one CPU because they are I/O bound and not CPU bound.

I agree that there is nothing special about the OS, since Windows XP can
multi-task quite well. It is just that desktop systems do not usually have a
large number of CPU intensive simultaneous processes executing at once.

For a given expenditure on CPU chips, 99% of desktop users will benefit more
with a faster CPU (and larger cache) than a dual core CPU.
 
Roger Hamlett said:
The 'rule of thumb', is that when using non multithreaded applications,
you will see between 20%, and 40% performance gain, unless the application
is unusual, and is almost entirely 'memory based', when you will see a
slight loss because of the extra kernel overhead. Unfortunately, some
Windows applications are this way inclined... The biggest gain though
comes with multiple applications (which is a thing very rarely 'done' in
Windows - how often do you leave an application 'processing' a major
mathematical task, while you run something else? - generally Windows users
'task jump', running just one task at a time, but with several 'active'),
or with a properly written MP application.

Based on benchmarks I have seen, the 20-40% increase in performance of non
multi-threaded applications with dual core is not accurate. Some
applications actually run slower on a dual core CPU than on a single
processor with the same speed. Remember that if you have two processors,
they have to share the system memory, and sometimes the CPU cache (depending
on the CPU design).

Bottom line is that for a given amount of money spent on a CPU, 99% of
desktop users will see much better performance with a faster CPU (and more
cache) than a slower multi-core CPU.
 
There are no rational arguments
that will convince a geek that the dual core X2 is not absolutely
necessary for their personal happiness and eternal salvation.

Now that's really true and rational :-)

I am fighting it because I know my 3500+ is perfectly adequate for my
needs!

John

Please remove "NO-SPAM" if sending email.
 
John Hollingsworth said:
I am fighting it because I know my 3500+ is perfectly adequate for my
needs!

John
If you want to spend more money on a CPU upgrade, get a faster (and more
cache) single processor CPU.
 
Mark A said:
Based on benchmarks I have seen, the 20-40% increase in performance of
non multi-threaded applications with dual core is not accurate. Some
applications actually run slower on a dual core CPU than on a single
processor with the same speed. Remember that if you have two processors,
they have to share the system memory, and sometimes the CPU cache
(depending on the CPU design).
You will see that I mention exactly this.
Bottom line is that for a given amount of money spent on a CPU, 99% of
desktop users will see much better performance with a faster CPU (and
more cache) than a slower multi-core CPU.
Yes, with a couple of 'exceptions'. With a single processor machine, you
can get the 'delightful' situation, where something like an old DOS
application, polls the keyboard, and uses 100% processor time doing
effectively nothing. You end up having to hit a key, and wait a few
minutes for Windows to eventually respond. With a MP machine this does not
happen, and the response is nearly instantaneous. The same thing happens
unfortunately with some Windows applications (MS is great at writing ones
that do this - try having a network problem while downloading a file). The
MP machine will often benchmark quite poorly, but the responsiveness to
the user, makes it feel faster than the benchmarks imply.

Best Wishes
 
Roger Hamlett said:
Yes, with a couple of 'exceptions'. With a single processor machine, you
can get the 'delightful' situation, where something like an old DOS
application, polls the keyboard, and uses 100% processor time doing
effectively nothing. You end up having to hit a key, and wait a few
minutes for Windows to eventually respond. With a MP machine this does not
happen, and the response is nearly instantaneous. The same thing happens
unfortunately with some Windows applications (MS is great at writing ones
that do this - try having a network problem while downloading a file). The
MP machine will often benchmark quite poorly, but the responsiveness to
the user, makes it feel faster than the benchmarks imply.

Best Wishes
Yes, there are some minor cases of multi-threading where the application
hands off work to one of the subsystems (disk, audio, etc), however these
are not long lasting CPU intensive tasks in most situations (especially if
you have a decent MB and separate audio card).

Bottom line is that for a given amount of money spent on a CPU, 99% of
desktop users will see much better performance with a faster CPU (and more
cache) than a slower multi-core CPU.
 
Back
Top