Sign in with
Sign up | Sign in
Your question

AXP VS P3-S, AXP LOSES!

Last response: in CPUs
Share
December 3, 2001 3:17:37 PM

take a look at these german benchmarks that were linked from AMD's other <A HREF="http://www.amdzone.com" target="_new">website</A>.

I notice that the dual AMD AXP and MP are having a heck of a time keeping up with the dual P3-S, a 1.44GHz P3 is kicking the chipsatz out of the AXP at 1800+ (virtual AMD PR speed -/+300 =1.53GHz) in most cases, now how is this possible? isn't the AXP gods gift to the AMDmongrels?

<A HREF="http://www.wiwi-hannover.de/cgi-bin/ubb/ultimatebb.cgi?..." target="_new">P3-S is really P3-S 1900+</A>

"<b>AMD/VIA!</b>...you are <i>still</i> the weakest link, good bye!"

More about : axp axp loses

December 3, 2001 3:28:25 PM

Nice of them to use ECC on the XP system, and overclock the bus on the P3 system to 152. Nice try, Meltdown.

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>
December 3, 2001 3:47:54 PM

The single P4 and the single AMD scores higher than the dual XP... Nice that the P4 scores higher than the dual XP in alot of tests.

dual XP loses ground in alot of tests...

Sad the dual XP loses so much ground in 3Dmark2001 whats up there? Sure you will be real fast at MS word, but when it comes to graphics and video the dual XP is definitly not the way to go.

Props to the slower chip winning, Im sure the AMD fans know what im talking about =)
Related resources
December 3, 2001 4:06:11 PM

Good for the P3. In fact, I bet you could make a P2 233MHz beat a AMD Athlon XP. It's all a matter of optimization. You can optimize an app for a specific platform but not for another. I could make a Athlon 500MHz beat a P4 2GHz or a P41.3GHz beat an Athlon XP 1900+. MHz = part of the equation IPC = not accurate because it changes depending on the app

What I'm saying is that benchmarks prove nothing. There is no right way to do things and it's impossible to say the P4 is better than the Athlon XP at Video editing for example, because that's too broad. It all depends on the particular app.

AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor
December 3, 2001 4:13:22 PM

I've always agreed the PIII-S is a very good design. Shame we can't have this kind of performance on the desktop. The PIII was intels best CPU. P4 was released prematurely, hopefully northwood will be better.

<font color=purple>~* K6-2 @ 333MHz *~
I don't need a 'Gigahertz' chip to surf the web just yet ;-)</font color=purple>
December 3, 2001 4:15:31 PM

>What I'm saying is that benchmarks prove nothing.

what's that? you're sounding like AMD, MHz don't matter blah, blah, now it's...Benchmarks don't matter, blah, blah, blah!

so if benchmarks don't prove nothing, then why do you mongrels keep spewing P4 fud?




"<b>AMD/VIA!</b>...you are <i>still</i> the weakest link, good bye!"<P ID="edit"><FONT SIZE=-1><EM>Edited by AmDmElTdOwN on 12/03/01 01:16 PM.</EM></FONT></P>
December 3, 2001 5:09:25 PM

Quote:


what's that? you're sounding like AMD, MHz don't matter blah, blah, now it's...Benchmarks don't matter, blah, blah, blah!

so if benchmarks don't prove nothing, then why do you mongrels keep spewing P4 fud?

And you sound like a broken telephone. First of all, I never said anything against Intel or the P4, only against trolls. What I meant was that no one benchmark can prove an Athlon XP is superior or a P4 is superior. You need to test hundreds of apps to make an overall decision. That's just not feasible right now. I mean, how would you weigh them? What would each app weigh in the overall score of the benchmark? That's the problem. No single benchmark right now is a good indicator of a processor's overall performance.

AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor
Anonymous
a b à CPUs
December 3, 2001 5:31:55 PM

Thats right you tell him Melty ! Bloody amd pups don't know what they are on about. I've read AMD MANs posts and they are total nonsense.

P.S What are you doing Fri night love bunny?


<font color=red><b>->Melty in my mouth MMmmm...<-</font color=red>
</b>
December 3, 2001 5:45:02 PM

Creeper, you are scaring me!!!

AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor
December 3, 2001 6:26:21 PM

Heh, I decided to look more thoroughly.

Quote:
<i>Translated with Babel Fish</i>: Ooops. what is that? The Sysmark 2001 could not unfortunately pass through on the dual Pentium III-S system. The system remained hanging again and again, fell. We had already often Sysmark conditioned crashes - this time the DVD266u-RN the guilty one is however unique.


Also interesting that both have PC2100 (the AMD has ECC, too), the dual P3 system is at a 152 FSB, but the AMD system beats it in memory bandwidth? That's very hard to believe.

Regardless, the P3s put in a very good performance.

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>
December 3, 2001 6:29:40 PM

I especially like the (translated) quote from <A HREF="http://www.wiwi-hannover.de/cgi-bin/ubb/ultimatebb.cgi?..." target="_new">here</A>:

Quote:

Ooops - which is that? The Sysmark 2001 could not unfortunately pass through on the dual Pentium III-S system. The system remained hanging again and again, fell.

So much for Intel stability/compatibility. It's nice that that an overclocked Tualatin can come close to the performance of a perfectly in-spec, below-max-clock AthlonXP. It's sad that it can't even complete all the benchmarks at its default clock. No professional on earth would accept such crummy stability; I know I don't. :lol: 

It's also funny as hell to see the Pentium 4 2GHz get the crap beat out of it by both CPUs. Haven't you said bandwidth matters more than anything? :wink:

Oh, and no wonder the dual MP loses, in bemchmarks that aren't even SMP enabled and 600MHz below its max specced clock. We both see that a single XP 1800+ on a KT266A creams the P4 even with 200 less "PR points." We can already be pretty certain that a dual MP 1800+ would embarrass dual Xeons, since dual MPs have been beating the snot out of dual Xeons for ages now. :lol: 

Kelledin
<A HREF="http://www.linuxfromscratch.org/" target="_new">LFS</A>: "You don't eat or sleep or mow the lawn; you just hack your distro all day long."
December 3, 2001 8:59:48 PM

lol this has to be the lamest attempt i've seen from you Meltdown. Hmm lets test a DUAL XP's running CAs 2 DDR and a 152fsb. that would make it abit more fair wouldn't you think?

your a god damn intelMongrel, it's also quite pathetic the P4 gets eaten alive! lol what are you gonna do when the p3 is totally phased out? We all know it's intels best CPU!

You complain about trolls and look @ the [-peep-] you post!
Like the saying goes "takes one to know one"
Anonymous
a b à CPUs
December 3, 2001 9:15:43 PM

Um meltdown why do you care about theses benchs man we already know that our products are overall better. Benchmarks equal script running system thats all. Ill perfer my stable, thermal protected, well supported, cool running, slightly slower in hollow benchmark system anyday.

Why feed them benchs when they dont mean nothing to us we are no fault, non problematic systems its more that we know this than what they think. We have no need to fight a crusade with them if the FCC let intel have its way our woes would be over. Till that faitful day comes just let them swim in their own propaganda and half truths. Not to say we dont either but its sounds troll like to say so i figured what the hell.

-Spuddy

<font color=red>Being Evil Is Good. Cause I Can Be A Prick And Get Away With It.</font color=red> :lol: 
December 3, 2001 9:32:39 PM

lol
Spuddy, you crack me up. I pity the people who don't talk to you on Messenger, they miss all the humor in your posts :) 

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>
December 3, 2001 9:39:32 PM

CAN! SOMEONE! PLEASE! TELL! ME! WHY! MELTYS! HEADINGS! ALL! HAVE! TO! LOOK! THE! SAME!?

ANY! CHANCE! OF! NOT! USING! CAPITALS!?

Excuse me for a moment. I need to drive my ergonomic wheely chair over a sheet of bubble wrap!
December 3, 2001 9:48:56 PM

well intel has been doing smp for a long time, and have had more experience and oppertunity to optimize their platform. AMD is new to smp and still has alot of work to be done.

<i>Mommy that dog is trying to jump over that other dog but he keeps not making it</i>
December 3, 2001 11:42:55 PM

Good point Intel_Inside.

ROFL, Spud.. very well put.

Kelledin, "No professional on earth would accept such crummy stability; I know I don't." then why are you in the AMD camp then?? - funnier than Spuds post!

Check this out, I been doing some benchmarking today. This is the reference DDR333 vs the P4T w/ RDR

<A HREF="http://gamershq.madonion.com/compare2k1.shtml?1990471" target="_new">RDR</A>
<A HREF="http://gamershq.madonion.com/compare2k1.shtml?2135781" target="_new">DDR333</A>

Both are straight runs, no saved test, no LODbias adjusts.

Both are at fastest setting.

The DDR333 so far has been a pain to overclock. this 2.0Ghz Wilmette can hit 2.5Ghz on the TH7-II. Only hitting 2175Mhz on the SIS645 chipset.

So far I am disappointed in the DDR333
December 3, 2001 11:55:55 PM

Umm, it's not the AMD processors that are unstable, it's some of the motherboards. If you get stuck with a bad motherboard then stability is an issue. But processor-wise, AMD is equal in stability to Intel. By the way, lhgpoobaa, I've been saying the same thing about AMDMeltdown's writing and he calls me an idiot. Seriously, it's impossible to get these childish trolls to think. Very very sad!! I think I should just give up.

AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor
December 3, 2001 11:59:01 PM

"Xeon will roll it up and smoke it"--isn't that what the Willamette was supposed to do to the Athlon?

Plus, how do you argue with 90% of the benchmarks out there, which put the dual AthlonMP ahead of the dual Xeon, even with a 500MHz disadvantage? Maybe Xeon <i>will</i> roll it up and smoke it sometime, but it sure as hell doesn't right now. :wink:

Quote:
Kelledin, "No professional on earth would accept such crummy stability; I know I don't." then why are you in the AMD camp then?? - funnier than Spuds post!

My CPU can complete all Sysmark tests; The PIII-S apparently can't. Certain other PIII's weren't even able to complete a Linux kernel compile; my CPU handles it faster than any P4 in production. Enough said. :lol: 

Plus, how do you argue with all the professionals getting perfect stability out of 760MP setups? You still haven't managed to refute that. :tongue:

Kelledin
<A HREF="http://www.linuxfromscratch.org/" target="_new">LFS</A>: "You don't eat or sleep or mow the lawn; you just hack your distro all day long."
December 5, 2001 3:41:19 AM

my oced p3 smoked the axp. oh btw the p3 was playing a wav file and the axp was converting vob to mpeg4. hmm is that really fair? what is your problem with being beat? its all just a game. sometimes you loose sometimes you win. its just a bad time for IntEl. just wait for about a million years and maybe intel will beat amd. ok.

<font color=red><b><i>you</i> keep talking and i'll pretend i'm listening.</b></font color=red>
December 5, 2001 6:10:12 AM

Do they have a name for what's wrong with you? Did you even bother to read the last page?

Some translated (by google) comments from page 19: Results and recommendation.

"The comparison is not simple thereby - bench mark showed, which is surely the Athlon XP 1800+ alias the Athlon MP 1800+ the faster of the two CPUs"

- I Think this is pretty clear..



"Both CPUs have pro and cons, however the Pentium iii s stands already nearly at the end of its career, the AMD Athlon MP only at the beginning."

- PIII-S is going to be phased out very soon...



About the Tyan Tiger (AMD) Motherboard:

"In the test the Tyan tiger MP supplied a outstanding achievement, it was very stable, ran reliably and without problems. There were neither driver nor compatibility problems, with many bench mark looked the Tyan tigers thereby completely particularly well."

- Hey, a fast and stable AMD platform!



About the Iwill DVD266u-RN (Intel) motherboard:

"Our point of criticism is here the bad stability, which finally prevents, which clears the Main board our Award. Without these lack, which we cannot explain ourselves, it would fall us easily to kueren the Main board then it is however nearly impossible."

- Not so stable Intel platform though...



Don't throw stones in the glass house, Mr Meltdown...

/* The more you know, the more you realize how little you know */<P ID="edit"><FONT SIZE=-1><EM>Edited by Parma_Endorion on 12/05/01 03:30 AM.</EM></FONT></P>
December 5, 2001 7:00:37 AM

Suggested designations for his condition are "permanent denial", "pre-teen angst", and "bonkers." I believe the most suitable is simply "wet behind the ears."

Kelledin
<A HREF="http://www.linuxfromscratch.org/" target="_new">LFS</A>: "You don't eat or sleep or mow the lawn; you just hack your distro all day long."
December 5, 2001 7:25:34 AM

;o)

/* The more you know, the more you realize how little you know */
December 5, 2001 11:07:43 PM

Strange you do not seem to have responded to any of the points given - that seems to be a trend for you, you post a link, give a bad interpretation of what it was saying, then run away.

Question - do you still like running up to people's front door, ringing the bell, then running away and hiding? I know you are still young, but try and act like there is some hope of you growing up before the end of this decade.

-* <font color=red> Under Offer </font color=red> *-
email for application details
December 6, 2001 12:41:57 AM

"Strange you do not seem to have responded to any of the points given - that seems to be a trend for you, you post a link, give a bad interpretation of what it was saying, then run away."

that's because I'm sitting here laughing at you mongrels!

take a look at the link I provided and you'll notice that the dual tualatin beating the dual AXP(in most test)

the P3-S rocked the AXP which is: clocked a little higher with a fsb266 and a PR of 1800+, now that is sad, if you can't see this then you're an idiot.

now I know some dope is gonna go on about hardware differences and such, please...I didn't run the tests I just noted the scores. get over it!

IMHO, a 512K tualatin @1.6GHz will demolish a AXP PR 1900+, too bad so sad!

"<b>AMD/VIA!</b>...you are <i>still</i> the weakest link, good bye!"
December 6, 2001 1:06:52 AM

Get over it!! You're supporting Intel like you own it!! What is wrong with you? Anyway, replace those dual-processors with a single Athlon XP and a VIA KT266A motherboard then the performance will increase significantly. Also if you take out that ECC RAM and replace with regular DDR RAM then you'll also gain some more performance.

Also, I don't see the P3-S beating the Athlon XP in most benchmarks. In fact, I see is a relative tie. However, that FSB overclock also increases AGP bus performance and PCI bus performance. Anyway, the P3-S is intended to be a server chip, the Athlon XP is not.

AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor
December 6, 2001 2:06:54 AM

Okay, I looked at those stats and saw the following:

Sys 2001 Int. CC, Dual AXP Win (just)
Sys 2001 Office Prod - Dual AXP Win, PIIIS <b>did not finish!!</b>
Business Winstone 2001 o/c PIIIS win
CC Winstone 2001 Dual AXP
Sandra, dual AXP beats PIIIS in 4, ties 1 and PIIIS o/c wins 1.
o/c PIIIS takes 2 Clibench, dual AXP takes 2 as well.
PIIIS pretty much take the dual AXP for Quake and Dronez. 3DMark2001 I have no idea what the testers did since even the single P4 only got 6000. That should be up in the 7000 at least right? Anyway - PIIIS beat the dual AXPs.

So, what do we have

Overall AXPs win 9 sets, PIIIS o/c win 11 and they tie 1. PIIIS platform does not complete 1 set of tests.

I'd like to consider that 6 of the PIIIS wins are for 2 applications, Quake and Dronez at different resolutions. Were we to look at applications wins, we would be 9 for AXPs, 1 tie, 7 for PIIIS o/c.

Now bear in mind that the PIIIS was winning on the basis of overclocking (14%) and competing against a system with ECC memory (workstation/server generally don't overclock and DO use ECC) we see the stock performances are still in the dual AXPs favour. Add to this that a PIIIS 1.26Ghz CPU sets you back a hefty $277, v.s. $181 for AthlonXP 1800+ (AMD saving you $192 on the pair) and I think the dual XP does very nicely, especially since the PIIIS is going bye-bye soon.

-* <font color=red> Under Offer </font color=red> *-
email for application details
!