Intel chipsets are the most stable?

[Dave C.]

What experts? Tomshardware might as well change name to intelslapdog.com

Funny that anandtech and sharky extreme and many other hardware sites agree
with tomshardware. Are they all intelslapdog.com? -Dave

 

[Ykalon]

Not really a flame war. Just a well-deserved smackdown of an obvious AMD
shill. I'm a huge AMD fan myself, but it's insane the way someone keeps
bashing Intel. Just seeking a little balance is all. -Dave

That was your biggest lie. You are NOT an AMD fan if you claim Intel is
faster than AMD64 in games when all evidence says different. Just read a
WHOLE test next time. AMD wins no mater what API is used. P4 wins in
Comanche 4 thats it. P4 EE doesn't cost as no regualar mortal can afford
one. An Athlon64 FX-53 costs a fraction of what a P4 EE costs and it is
very close behind (or in front of) P4 EE. Besides a lot of the tests are
very GPU bound ( especially in DX9 games like GunMetal) so you really
shouldn't count those tests. If you had run them again a lot of
positions would switch.

If you discount P4 EE AMD wins ALL game tests at anandtech.com

 

[Dave C.]

That was your biggest lie. You are NOT an AMD fan if you claim Intel is
faster than AMD64 in games when all evidence says different.

Huh?!? I'm just repeating what the experts, including anandtech, report.
Are all the experts liars, also? If you want to call me a liar, you'd
better be able to prove that tomshardware, anandtech and sharky extreme are
liars, also. Good luck on that. -Dave

According to www.pricewatch.com, same price range at the moment would be:

P4 3.2 Prescott vs. Athlon64 3200+ or

P4 3.4 Prescott vs. Athlon64 3400+

Beyond that range, you can pay up to several hundred dollars for either an
Intel or AMD chip, but hardly anybody gives a damn about those chips, as
hardly anybody spends as much on a processor as they do on the entire rest
of their system combined.

So the P4 3.2/3.4 and Athlon64 3200/3400 would be the best indicators of who
has the best bang for buck, at the moment.

Gaming: OpenGL: The Intel chips are much faster
Gaming: DX8: The AMD chips are faster, no doubt about it
Gaming: DX9: It's virtually a tie, as the AMD chips are two to three
TENTHS of a percentage point faster than Intel.
So on the gaming benchmarks, that's one win for Intel, one win for AMD and
one tie.
GAMING OVERALL: TIED

Business Applications: Office Applications: Intel blows AMD away
Business Applications: Internet Content Creation: Intel blows AMD away
Business Applications: Overall: Intel blows AMD away

Video Encoding: This one is so lopsided, AMD should have thrown in the
towel before entering the ring. Intel wins by a landslide.

Audio Encoding: Again, Intel wins by a landslide

Synthetic Benchmarks: (PC Mark 2004): Here, Intel blows AMD away on both
*CPU* and memory benchmarks

Even at the same price for CPU, an Intel system can be cheaper to
build, as the P4 boards are more mature at this point, and thus there are
better bargains to be found. Considering that an Intel system will likely
be cheaper to build and WILL perform better on all benchmarks except DX8,
it's kind of a no-brainer as to which chip to build with, at the moment.

http://www.tomshardware.com/cpu/20040322/index.html

The following is an article on the Athlon 64 2800+. But more interesting
is,
the benchmarks included in the article are a GREAT comparison of the 3.2GHz
P4
processors with the Athlon64 3200+. In this article, these two processors
are
pretty evenly matched, with Intel being faster on some benchmarks, and AMD
being faster on others.

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2038&p=1

Now lets look at what Sharky Extreme has to report in their article about
the
3.4GHz Prescott processor. This one has benchmarks that are a great
comparison
of the 3.4GHz Intel chips with the Athlon64 3400+. Here, you have to be
careful,
as Sharky doesn't organize their charts in order of fastest to slowest. And
on
some charts, LOWER scores are better. But if you read all the benchmarks,
you
will again notice that the two chips are pretty evenly matched, with AMD
faster
on some and Intel faster on others.

http://www.sharkyextreme.com/hardware/cpu/article.php/3261_3329681__1

Intel is better than AMD, at the moment. The only way AMD could change that
would be to drop their prices by 30% or better. -Dave, updated 10/2/04

 

[Ykalon]

Funny that anandtech and sharky extreme and many other hardware sites agree
with tomshardware. Are they all intelslapdog.com? -Dave

Disregarding P4 Extremely Expensive Edition they do NOT agree.

 

[Ykalon]

Huh?!? I'm just repeating what the experts, including anandtech, report.
Are all the experts liars, also? If you want to call me a liar, you'd
better be able to prove that tomshardware, anandtech and sharky extreme are
liars, also. Good luck on that. -Dave

According to www.pricewatch.com, same price range at the moment would be:

P4 3.2 Prescott vs. Athlon64 3200+ or

P4 3.4 Prescott vs. Athlon64 3400+

Thing is I have READ those reviews and they do NOT say what you claim
they say. P4 Extremely Expensive Edition at the top or right after FX-53
and when a lot of AMD64's and FAAR down the list a P4 that is NOT an EE.
Do yourself a favor and READ the reviews you link to.

 

[Dave C.]

Thing is I have READ those reviews and they do NOT say what you claim they

say.

Oh man, maybe you should look into night school. Really. What is UP with
all these morons claiming that what is published is not published? -Dave

 

[Never anonymous Bud]

So the P4 3.2/3.4 and Athlon64 3200/3400 would be the best indicators of who
has the best bang for buck, at the moment.

Gaming: OpenGL: The Intel chips are much faster

Hmmm, at tomshardware.com, in a recent review,
using Quake at 1024x768, the P4 3.4gnz hit 234fps, an A64-3400 hit 229.

NOT 'much faster', barely 1% difference. Statistically a tie.
(I consider 2% difference to be a tie,
5% to be a marginal win, 10% or more is a clear win).

Gaming: DX8: The AMD chips are faster, no doubt about it

Wolf, A64 hit 156.1, P4 hit 156.4, NO difference.
Comanche, A64 did 70.38, P4 did 66.33. Slight win, AMD
Unreal, A64 did 147.11, P4 did 127.01, clear win, AMD.

Gaming: DX9: It's virtually a tie, as the AMD chips are two to three

3DMark graphics, A64 at 6607, P4 at 6611, tie.
3DMark CPU, A64 at 747, P4 at 735, tie.
Aquamark 3, A64 at 105.69, P4 at 107.6, tie.
FarCry, A64 at 220.8, P4 at 207, slight edge, AMD.


Well, I have NO idea where you're getting YOUR info,
but many here claim Toms Hardware is an Intel shill,
yet the numbers CLEARLY show AMD at least keeping pace
with Intel, and often beating them.

But I DO remember John Corse,
and the clearly FALSE benchmarks he posted quite often.

Even when he DID post the right numbers,
he'd almost always mis-interpret them.

 

[Roman Werpachowski]

IME, large corporations with big budgets nearly always go with the
most prominent vendor, whether or not he has the best product, the
rationale being that, if the product fails to perform as expected,
then the person who authorised its purchase cannot be seen to have
gambled.

"Noone has ever been fired for going IBM", eh?

 

[Ykalon]

Oh man, maybe you should look into night school. Really. What is UP with
all these morons claiming that what is published is not published? -Dave

And you should visit your optician. You clearly need your vision checked.
PROOF you have not read the tests you link to.
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2038&p=3
Business Winstone 2004

1. AMD Athlon64 FX53
2. AMD Athlon64 FX51
3. AMD Athlon64 3400+
4. AMD Athlon64 3200+
5. Intel Pentium 4 EE 3,4 GHz (EE = Extreme Edition, popularily known as
Extremely Expensive)
This is followed by A64 3000+ and another Extremely Expensive P4. After
that both the slowest A64 and even the old XP3200+ is ahead of all
"normal" P4's. This is clearly NOT a landslide win for Intel as you said.

Content Creation just minor shifts AMD still WAY ahead mostly.

Sysmark with the exception of AMD64 FX53 and Extremely Expensive this is
a clear win for P4. But that is because the tests are optimised for SSE3
and Hyper Threading which AMD64 lacks at the moment.

DX9

Aquamark fps

1. A64 FX53
2. P4 EE 3,4 GHz
3. P4 EE 3,2 GHz
4. A64 3400+
5. A64 FX51

But lets call this a tie since it is really only a difference of 2,8 fps
(disregarding the XP3200+) between the fastest and the slowest.

Aquamark CPU score goes to Intel clearly.

Halo landslide win for AMD if you just look at the positions but again
only a difference of about 1 fps between the slowest and fastest so lets
call it a tie.

Same with GunMetal just too close to call really = tie.
But my guess is that if they had used a 6800 Ultra instead of a 9800 PRO
AMD would have a huge win in nearly all dx9 tests.


DX8

Unreal Tournament 2003

No contest AMD wins both flyby and botmatch. Intel not even close

Warcraft 3 Same as with UT2k3

OpenGL (which you claim Intel wins)

Quake 3 I call for AMD (discounting the Extremely Expensive edition P4)

Jedi Knight AMD wins again

Wolfenstein Once again AMD

Where is the test that shows Intel performs better in OpenGL than AMD?


DivX Encoding goes to Intel

3D Rendering

3DStudio = Intel by quite a big margin
Lightwave Intel again but closer this time.

Development

Q3 Source compile = landslide win for AMD

All in all AMD won almost all game tests and quite a lot of the other tests.

 

[Ykalon]

Oh man, maybe you should look into night school. Really. What is UP with
all these morons claiming that what is published is not published? -Dave

I wonder what is up with all these morons that claims that what is NOT
published is published.....

 

[JK]

A few _cheap_ corporations

This is from the AMD website.

"Twenty-five percent of the Fortune Global 100 companies now use AMD Opteron™
processor-based systems to run critical enterprise applications. These organizations
include industry leaders in banking, insurance, pharmaceuticals, manufacturing, energy and
telecommunications. "

or bureaucracies will use AMD and off brand
chipsets....Check out the banks who need no fault tolerances....

Intel based IBM........

Not all of them. Some use Sun Opteron based systems.

 

[nobody]

On Fri, 15 Oct 2004 10:04:50 -0400, Tony Hill

....snip...

However when nVidia entered the game, the rules changed somewhat.
nVidia right out of the gate had VIA and ALi beat cold in terms of
driver quality and their chipsets were used on higher-end products
that SiS (if you use low-end crap components on a motherboard with a
shitty design, it really doesn't matter how good the chipset is, your
board will still suck). The result of this was two-fold: first off it
gave a real, viable competitor to Intel for the most stable chipsets,
and secondly it really forced VIA to pick up their socks. While I'm
still no big fan of VIA chipsets, my understanding is that their
latest couple versions have been rather significantly better than
where they were two years ago.


Also, Intel is hardly without their own faults as well. While some of
their chipsets have been good, they have had their own sets of
problems, ranging from the extremely problematic memory interface of
the i820 chipset to the very poor quality of the early i810 drivers,
and pretty much all of us who were dealing with PCs back in the late
'96/early '97 time frame remember incredibly problematic ATA drivers
for the PIIX4 southbridge (this caused many people to have to format
and re-install their entire OS just because they installed patches and
drivers in a different order than was required).


Personally, if I were to build a system using an Intel processor, I
would probably stick to an Intel chipset simply because the only
advantage of non-Intel chipsets is about a $5 price savings (ie
nothing). On the other hand, my last 4 motherboard + CPU combos have
been using AMD processors, obviously all of which used non-Intel
chipsets (2 x SiS, 1 VIA and 1 nVidia). If I were to buy a new system
today, it would have an AMD processor in the thing and an nVidia
chipset on the motherboard, because IMO they are now the leaders in
terms of chipset driver quality, not Intel.

I could agree that early VIA chipsets were not flawless in terms of
stability, performance, features, whatever else you look at. However,
these chipsets were found in mostly low-end boards, and the makers
thereof cut every corner to bring the price point down. With $hitty
design and $hitty components, these boards could be nothing better
than $hit. Moreover, these boards were mostly used together with the
lowest of the lowest priced components that had their own stability
issues and $hitty drivers, whereas Intel boards tended to be used with
higher-end components. Yet, even then there were some boards that,
when set up correctly and with at least marginally good PSU, video,
and NIC, were quite decent and stable performers. Case in point - my
1998-ish FIC VA503+ board that is still alive in my second system with
k6-2+ overclocked to 600MHZ - I use it to browse some iffy sites when
I would not want to risk infecting my main system with some crap. By
the way, that one - dual Opteron on MSI board with VIA KT800 chipset -
has yet to show a blue screen. Yes, a few times I had app crashes -
when I debugged .NET apps that make calls to unmanaged functions ;-).
When I do same things at work, the all-Intel IBM NetVista PC usually
shows BSOD.

 

[Dave C.]

And you should visit your optician. You clearly need your vision checked.
PROOF you have not read the tests you link to.
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2038&p=3
Business Winstone 2004

1. AMD Athlon64 FX53
2. AMD Athlon64 FX51
3. AMD Athlon64 3400+
4. AMD Athlon64 3200+
5. Intel Pentium 4 EE 3,4 GHz (EE = Extreme Edition, popularily known as
Extremely Expensive)

OH I get it now. You are evaluating processors on the basis of who the ****
cares about cost? If I wanted to pay more for a processor than I did for
the last car I purchased, then YES, the benchmarks might support your point
of view. But I have claimed all along that Intel processors are a better
deal at the price point most people are building at, in terms of "bang for
buck". That is still true. -Dave

 

[JK]

While those who spend over $500 for a cpu may not be a large percentage
of pc buyers, there are many gamers who will spend around $300 to get an
an Athlon 64 3500+.

 

[Ruel Smith]

Hmmm, at tomshardware.com, in a recent review,
using Quake at 1024x768, the P4 3.4gnz hit 234fps, an A64-3400 hit 229.

NOT 'much faster', barely 1% difference. Statistically a tie.
(I consider 2% difference to be a tie,
5% to be a marginal win, 10% or more is a clear win).


Wolf, A64 hit 156.1, P4 hit 156.4, NO difference.
Comanche, A64 did 70.38, P4 did 66.33. Slight win, AMD
Unreal, A64 did 147.11, P4 did 127.01, clear win, AMD.


3DMark graphics, A64 at 6607, P4 at 6611, tie.
3DMark CPU, A64 at 747, P4 at 735, tie.
Aquamark 3, A64 at 105.69, P4 at 107.6, tie.
FarCry, A64 at 220.8, P4 at 207, slight edge, AMD.


Well, I have NO idea where you're getting YOUR info,
but many here claim Toms Hardware is an Intel shill,
yet the numbers CLEARLY show AMD at least keeping pace
with Intel, and often beating them.

One thing stands out, though. With a 64 bit OS, it's expected that the
Athlon 64/FX/Opterons will get roughly a 20% speed increase. And, Microsoft
is considering making the upgrade to Windows XP-64 free to those that have
a valid XP license and a 64 bit processor. They expect the OS to be
available 1st or 2nd quarter in '05. You can download and run the beta,
now.

 

[Ykalon]

OH I get it now. You are evaluating processors on the basis of who the ****
cares about cost? If I wanted to pay more for a processor than I did for
the last car I purchased, then YES, the benchmarks might support your point
of view. But I have claimed all along that Intel processors are a better
deal at the price point most people are building at, in terms of "bang for
buck". That is still true. -Dave

Ok lets just use A64 3200 and P4 3,2 Prescott.

Business Winstone = Clear win for A64 3200+
Content Creation = AMD again.
All Sysmark 2004 tests goes to P4.

Aquamark = tied
Aquamark CPU = P4
Halo = tied
GunMetal = tied

Both UT2k3 tests goes clearly to A64 3200+
Warcraft 3 slight edge to A64 3200 but I think it should be considered a
tie.

Q3 = A64 3200+
Jedi Knight = A64 3200+
Wolfenstein = A64 3200+

ALL 3 are OpenGL tests which you claim P4 is superior in. Sorry but you
have NO evidence that shows that.

 

[Franklin]

I could agree that early VIA chipsets were not flawless in terms
of stability, performance, features, whatever else you look at.
However, these chipsets were found in mostly low-end boards, and
the makers thereof cut every corner to bring the price point
down. With $hitty design and $hitty components, these boards
could be nothing better than $hit. Moreover, these boards were
mostly used together with the lowest of the lowest priced
components that had their own stability issues and $hitty
drivers, whereas Intel boards tended to be used with higher-end
components. Yet, even then there were some boards that, when
set up correctly and with at least marginally good PSU, video,
and NIC, were quite decent and stable performers. Case in point
- my 1998-ish FIC VA503+ board that is still alive in my second
system with k6-2+ overclocked to 600MHZ - I use it to browse
some iffy sites when I would not want to risk infecting my main
system with some crap. By the way, that one - dual Opteron on
MSI board with VIA KT800 chipset - has yet to show a blue
screen. Yes, a few times I had app crashes - when I debugged
.NET apps that make calls to unmanaged functions ;-). When I do
same things at work, the all-Intel IBM NetVista PC usually shows
BSOD.

Ironically, I once had an MSI board with Via 133 chips. You know,
the ones with the buggy firmware in the Via chips which they replaced
by the 133A.

It put me off MSI after that because their website and tech forums
were so poor.

 

[Johannes H Andersen]

It was much more of a contest then then. Now AMD is beating Intel in desktop
performance by such a large margin.

Your AMD bias is so obvious that it's easily disregarded.

 

[JK]

LOL! Look at the benchmarks.

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2065&p=6

 

[George Macdonald]

A few _cheap_ corporations or bureaucracies will use AMD and off brand
chipsets....Check out the banks who need no fault tolerances....
Intel based IBM........

If the banks have any sense at all, and I believe they do, they will most
certainly not be running mission critical servers on distributed
x86/Windows platforms. That's a role which is still best filled by an IBM
mainframe.

As for branch servers and desktops, Dell is probably the choice and of
course that means Intel x86, for the moment but not for any reasons of
technical excellence or superiority. If I were a bank IT decision maker
though, I'd be wary of placing such faith in an organization with such a
fragile business model.

As for fault tolerance I dunno where you got that from - Intel doesn't have
it in any form, in house - they get that from 3rd parties and specialist
OEMs... just like AMD does.

I don't follow why you'd think "cheap" would be associated with the use of
AMD CPUs - a throwaway comment that.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??