• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Dual Shootout at Tomshardware



<< Is this a typo? >>


Yep.


<< The Xeon 2.2 GHz pretty much whacked the Athlon MP 2000+. >>


I wouldn't say whacked. The 2.2 GHz shows the same slight lead over the 2000+ that is seen with single processors in all other single benchmark tests. Who cares about a few %...
 
Anyone know when we will see Divx5 benchmarks at Tom's?


and.. wouldnt it be nice if Anand started benchmarking divx encoding to?
 
Once you throw out pointless benchmarks like SiSoft 2002 and Sysmark 2002, you'll see that the 2.2GHz Xeon (HT enabled) is neck and neck with the AMP 2000+, except in lightwave where the P4 has always fared extremely well.
 
HT is nice... for me to poop on!


actually... it whips ass for seti@home... dual p4 boxes running 4 units at a time is surprisingly fast
 
Considering that the cheapest Xeon 2.2 on Pricewatch is $646, I'll stick with my Tiger MP.

Thanks anyway, though.

🙂

Viper GTS
 
Those Xeon's have over 1ghz clock speed advantage and can't do no better then that against the mp cpu's. Not very impressivce if you ask me. I know mhz isn't everything but with that much more I would expect it fair much better. Oh well, maybe its just me.
 
"Who cares about a few %..."

More than half the non synthetic benchmarks were won by more than 10%, a couple by over 30%, those aren't trivial amounts, especially when you consider how fast the SMP AMD system is that the Xeon system bested. No home user is going to put together an SMP Xeon system, the market those systems are targetted for are not filled with price conscious shoppers.
 


<< Those Xeon's have over 1ghz clock speed advantage >>

wrong

Xeon is at 2.2 GHz. MP is at 1.67 GHz. That's a 533 MHz difference, not 1 GHz.



<< Once you throw out pointless benchmarks like SiSoft 2002 and Sysmark 2002, you'll see that the 2.2GHz Xeon (HT enabled) is neck and neck with the AMP 2000+ >>

wrong

MP3 Encoding - Xeon 2.2 x2, 95 seconds, MP2000+ x2, 128 seconds. Xeon is 26% faster

DIVX Test - Xeon 2.2 x2, 60.47 FPS, MP2000+ x2, 54.71 FPS Xeon is 10.5% faster

Lightwave - Xeon 2.2 x2 152.9 sconds, MP2000+ 223.3 sconds. Xeon is 31.5% faster

Xeon 2.2 x2 looses Cinema by about 1%, wins 3DSMax by less than 1%

Sysmark Content Creation: Xeon is almost 50% faster
Sysmark Office Performance Xeon is 12.5% faster
Overall Dual Xeon 2.2 is almost 31% faster than Dual MP 2000+

Generally speaking, 10% or more is noticable on Office apps, and even 5% would be noticable on multimedia apps that take several minutes to process data. This is hardly neck and neck. Undeniable Conclusion: Athlon MP got an @ss kicking.

btw I don't look at sisoft numbers. sisoft is a lousy benchmark.
 


<<

<< Those Xeon's have over 1ghz clock speed advantage >>

wrong

Xeon is at 2.2 GHz. MP is at 1.67 GHz. That's a 533 MHz difference, not 1 GHz.



<< Once you throw out pointless benchmarks like SiSoft 2002 and Sysmark 2002, you'll see that the 2.2GHz Xeon (HT enabled) is neck and neck with the AMP 2000+ >>

wrong

MP3 Encoding - Xeon 2.2 x2, 95 seconds, MP2000+ x2, 128 seconds. Xeon is 26% faster

DIVX Test - Xeon 2.2 x2, 60.47 FPS, MP2000+ x2, 54.71 FPS Xeon is 10.5% faster

Lightwave - Xeon 2.2 x2 152.9 sconds, MP2000+ 223.3 sconds. Xeon is 31.5% faster

Xeon 2.2 x2 looses Cinema by about 1%, wins 3DSMax by less than 1%

Sysmark Content Creation: Xeon is almost 50% faster
Sysmark Office Performance Xeon is 12.5% faster
Overall Dual Xeon 2.2 is almost 31% faster than Dual MP 2000+

Generally speaking, 10% or more is noticable on Office apps, and even 5% would be noticable on multimedia apps that take several minutes to process data. This is hardly neck and neck. Undeniable Conclusion: Athlon MP got an @ss kicking.

btw I don't look at sisoft numbers. sisoft is a lousy benchmark. >>





Adding both cpu's which = total mhz which Xeon = over 1ghz advantage. So I = right - U = wrong. And as my first statement says I think the Xeon looks aweful for having such a high clockspeed advantage.
 


<< Adding both cpu's which = total mhz which Xeon = over 1ghz advantage >>



formulav8 - You are still wrong. For an SMP system adding the MHz of the processors is simply invalid. A dual 600 MHz Katmai system for example won't perform as well as a 1200 MHz Tualatin in the vast majority of cases.

To say 600 MHz processor x2 = 1200 MHz is false. The CPUs are still running at 600 MHz.

And the MHz : MHz ratio of Xeon : MP is still the same even under your strange conclusion.
 


<< Anyone know when we will see Divx5 benchmarks at Tom's?

 >>



He already has them, you just have to do a little hunting around. Most of his articles are available for days before they actually hit the front page... linky
 


<< Sysmark Content Creation: Xeon is almost 50% faster
Sysmark Office Performance Xeon is 12.5% faster >>


Hmmm, does anyone else find it funny that, although Sysmark 2002 is supposed to enable SSE for the Athlons, when compared to Sysmark 2001 (patched by Anandtech to use SSE) the Athlon loses considerable performance. For instance, the athlon XP1800, which beat the P4 2.0 by ~4% in Overall System Performance in 2001, now loses to the same p4 2.0 by ~8% in 2002. That is a 12% performance delta. A 14% variance in this particular benchmark is nominal to the difference between a p4 1.6 and p4 2.0. So between versions of the Sysmark benchmark suite, Intel is able to gain what equates to 4 speed grades of benchmarked performance over AMD, this IMO is bunk.

Links:
2001
2002
 
wrong
MP3 Encoding - Xeon 2.2 x2, 95 seconds, MP2000+ x2, 128 seconds. Xeon is 26% faster

Let me first say that I'm no "CPU fanboy" (in fact, the fastest AMD in my house is a 133 MHz 486 right now).

Regardless, I'm furious about that T&ouml;ppelt guy who's doing the CPU reviews at THG nowadays. His selection of apps to benchmark seems to be influenced by an Intel paycheck, not by "Real Life" (tm) and this makes the entire results pretty questionable, IMO.

Lemme explain: Before this latest dual shootout, the THG MP3 encoder of choice was LAME. Best sound quality available in MP3 encoding anywhere, reasonably optmised for both types of CPU and best of all free and therefore hugely popular.

Now they are using Magix MP3 Maker. Notice something about that URL? BTW the thing is also slower and the SMP capabilities pale in comparison to GoGo, which they should be using if they wanted maximum encoding speed on both platforms.

DIVX Test - Xeon 2.2 x2, 60.47 FPS, MP2000+ x2, 54.71 FPS Xeon is 10.5% faster

Where is the Intel and AMD optimised DivX 5 CoDec? I thought T&ouml;ppelt was a video nut (they claimed he was when presenting THG video No 1). Strange they switch to a rather obscure MP3 encoder for the audio tests, but somehow miss the rather popular new release by DivX.

Lightwave - Xeon 2.2 x2 152.9 sconds, MP2000+ 223.3 sconds. Xeon is 31.5% faster

Newtek is an Intel shop, they have been working and co operating in marketing since at least the days of the PPro... No surprise that Lightwave doesn't use SSE on the Athlon XP at all (Palomino and TBird score the same at equal frequency, both are beaten by a Tualaron).

Sysmark Content Creation: Xeon is almost 50% faster

The only really SMP aware app in there is Windows Media, which, surprisingly, still does not use SSE on the XP even in it's 2002 incarnation.

btw I don't look at sisoft numbers. sisoft is a lousy benchmark.

Agree with that. BTW where is SPEC Viewperf? Wouldn't run on the Xeon or what's the problem?

Edit: yeah, I noticed somthing about the URL too 😉.
 


<<

<< Adding both cpu's which = total mhz which Xeon = over 1ghz advantage >>



formulav8 - You are still wrong. For an SMP system adding the MHz of the processors is simply invalid. A dual 600 MHz Katmai system for example won't perform as well as a 1200 MHz Tualatin in the vast majority of cases.

To say 600 MHz processor x2 = 1200 MHz is false. The CPUs are still running at 600 MHz.

And the MHz : MHz ratio of Xeon : MP is still the same even under your strange conclusion. >>




I know that. I was generalizing by adding the mhz of both cpu's . At how the real world would see it. Only making a statement. I was implying the dissappointment of the performance of the Xeon cpu's. Nothing more. I was expecting much more and instead got much less. I still can't see me spending so much money on Intel. The bank for buck is simply not there.
 
WilsonTung, first realize that this review tested very few real world situations, in fact there were only 4 examples of this (5 actually, but Tom?s dual Xeon configuration couldn?t load Pinnacle Studio 7 for some reason).

Now, lets examine:

1. In Cinema 4D XL 7.303 the AMP 2000+ is ahead of the Xeon 2200 by ~ 1.1%.
2. In 3D Studio Max 4.2 the Xeon 2200 is ahead of the AMP 2000+ by ~ .07%
3. In MPEG-4 Video Encoding the Xeon 2200 is ahead of the AMP 2000+ by 9.5%.
4. In MP3-Audio-Encoding the Xeon 2200 is ahead of the AMP 2000+ by ~ 25.8%.

With Lightwave as the only exception, we see that, on average, the Xeon 2200 was ~ 8.725% faster than the AMP 2000+ in 4 real world examples (SiSoft and Sysmark are pointless obviously). And hell, you could easily stick an AXP 2100+ in there and that would add a little more speed.

I will admit that an 8.7% advantage isn?t the same as ?neck and neck? as I previously stated, but I think you?re still splitting hairs here, especially since Tom only had 4 real world examples (somewhere between 15-25 good real world benchies would be a much better indicator of each platform?s average performance advantage).

And therefore your concluding statement is pretty misleading: Undeniable Conclusion: Athlon MP got an @ss kicking.

EDIT: As a side note (and as Viper GTS mentioned), the Xeon's price/performance needs to improve. An MPX mb and AMP 2000+ CPU is well over 50% cheaper than a i860 mb and 2.2GHz Xeon CPU. And I'm not taking into account that you can run an AXP 2000+ (and even 2100+) in an MPX board for cheaper than an AMP 2000+ (AMP 2100+ is not available yet obviously).
 
Wow. Those benchmarks caught me by surprise. I would have to agree with formulav8 in saying the benchmarks are really dissapointing. With a 533 mhz advantage in each chip one would think these should really open up a nice 12-18% lead over the XPs. When you combine the price, like Viper so nicely pointed out, the benchmarks even look worse. For the cost of a dual xenon rig I could build 2 dual XP rigs. That's madness. Intel may have regained the dual processor kingship, but all they did was shovel more sh*t into the fire, if you catch my drift 🙂
 
Shouldn't we also be taking into account here the issue of COST??? The Xeons CPUs cost AT LEAST $200 more EACH and you better believe that mobo for those Xeons was WAY more expensive than the Athlons too.

I mean hell, if we just wanna talk performance, why not start benchmarking cluster systems with 300 some odd CPUs and 3 terabytes of RAM?

Performance per Dollar, the Xeons get spanked all around that charts. Not to mention I wouldn't even consider MP3 encoding or DivX encoding reasonable benchmarks for these systems, anyone that purchased these systems for DivX or MP3 encoding has either WAY too much money or seriously needs a life.

The only really reasable benchmarks for these would be server applications and the 3D rendering performance.
 
DSTA, it's an Intel world, get used to it. There's no such thing as an AMD SMP optimized application. Intel dominates the market and for obvious reasons gets preferential treatment because of it from software developers.

"With Lightwave as the only exception, we see that, on average, the Xeon 2200 was ~ 8.725% faster than the AMP 2000+ in 4 real world examples"

Huh? How can take out the applications the Xeon had the biggest advantage in? What kind of bogus comparison is that? When you include Lightwave, it's a 13% advantage. That's not an ass kicking, but it is a clear performance advantage.

Cost in this market is irrelevent. Intel couldn't care less that enthusiasts think they are overcharging for Xeon's, because you aren't even remotely in their radar as a potential customer.
 
I don't see why there is any reason to bring up clock speed. Who cares what clock speed the chips run at? It's an apples to oranges comparison between a P4 and Athlon. You think anyone in the business world is basing purchases on IPC ratings? Why doesn't AMD relelase a 2.2GHz XP so these meaningless arguements can stop? Because they can't. And why can't they? Because it is a different architecture than the P4 that was not designed to scale to insane clock rates. It's really sad all the excuses people come up with. AMD had the performance lead throughout most of the Athlons existence, but the pendulum has now swung to Intel's side. With Clawhammer it may swing back to AMD, who knows. Give credit where credit is due, Intel won this round.
 


<<

<< Sysmark Content Creation: Xeon is almost 50% faster
Sysmark Office Performance Xeon is 12.5% faster >>


Hmmm, does anyone else find it funny that, although Sysmark 2002 is supposed to enable SSE for the Athlons, when compared to Sysmark 2001 (patched by Anandtech to use SSE) the Athlon loses considerable performance. For instance, the athlon XP1800, which beat the P4 2.0 by ~4% in Overall System Performance in 2001, now loses to the same p4 2.0 by ~8% in 2002. That is a 12% performance delta. A 14% variance in this particular benchmark is nominal to the difference between a p4 1.6 and p4 2.0. So between versions of the Sysmark benchmark suite, Intel is able to gain what equates to 4 speed grades of benchmarked performance over AMD, this IMO is bunk.

Links:
2001
2002 >>



Conspiracy theories abound... the Athlon does worse in SysMark 2002 then it did in 2001, and worse in 2001 then it did in 2000.... the same pattern held with the K6-2/3, they performed better in 98' then they did in 2000.
It was revealed quite awhile ago that Bapco headquarters is the same as Intel's, Intel owns Bapco the makers of SysMark.

There have been conspiracy theories regarding Bapco's benchmarks for years for that reason
Whether it really is biased or not is debateable... one can find dozens of articles debating the matter back and forth.
Frankly, I just ignore Sysmark, I don't even really give a damn who owns Bapco, or whom is apart of Bapco anymore.... there has long been too much potential for bias within Bapco's benchmarks suites due to their close relationship with Intel.

But then, there are few synthetic benchmarks that I put much faith in anyways.
 


<< I don't see why there is any reason to bring up clock speed. Who cares what clock speed the chips run at? >>



I agree, it makes little difference what clockspeed they run at. Their two vastly different architectures, one would hardly expect them to run similarly in clockspeed.
In the end, performance is what matters.
The comparison is Intel's best vs. AMD's best.
Intel's best may have a 500MHz clockspeed advantage, but it doesnt make any difference because they are comparing the best both manufacturers have to offer.

One can make an argument for the AthlonMP, due to 2X Xeon's + S603 Mobo costs a fair amount more then 2X AthlonMP + Dually SA mobo, but then the market such processors are marketed to is not typically one win which the processor price is the determining factor so even that argument is worth little.
 
I have a Dual Prestonia system (@2G) on a SM mobo. I use it with Premier 6.02 under XP Pro (with a Pinnacle DV500+). Adobe uses all four processes (most of the time). It rocks, it's stable, with or without SMT enabled. Just to round it out, I put in an Adaptec 39160 controller (64bit@66Mhz) with a 73G Cheatah 10K and 36G 15K Cheatah.

It's one of my home systems, I don't use it for any business apps.

I'd recommend it to anyone who meeds a high-end system.

Most of the benchmarks are better for measuring tweaks to the same system: Between systems, there doesn't seem to many/any that are truely universal.

Buy what you like & trust (then STFU). It's not like either side is going to convince the other that their system-of-choice "sucks compared to the XXX."

IMHO, you might as well be talking about religion or abortion, most folks have their opinion set in concrete...no facts, stats, or anecdotal information will be enough to convince the opposing opinion to convert.

You can pretty much count on any post mentioning either architecture to degrade into "Neener neener! poop on you & your opinion."

Take it back to the school yard where it belongs.

FWIW

Scott
 
DSTA, it's an Intel world, get used to it. There's no such thing as an AMD SMP optimized application. Intel dominates the market and for obvious reasons gets preferential treatment because of it from software developers.

Agreed. Intel outsells AMD in the worksatation market by a huuuuge amount and of course they get ISVs to optimise their apps for P4.

But to summarise my previous rant: IMO the apps selected by THG do not represent what is used out there the field. Not only that a dual Xeon is rarely bought to run Magix' MP3 enocoder, it's also that the apparent bias towards P4/SSE2 optimised apps is not "real world". Soo much stuff doesn't even know what MMX is...

I'd want to see numbers for MySQL, Photoshop, Pro Engineer, Sonic DVD authoring stuff and some compiler benchies.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top