P4 is the fastest machine at madonion -

[gjwild]

<< But in the real world, the Ultra will keep up with the GF3 in FPS under most conditions >>


What the hell, have you owned a GeForce3 and compared them at say 1600x1200 32bit?
Man they are not close believe me. And why do you Zealots get so upset because Intel is leading the MHz war? And since when is water cooling an extreme method of cooling the chip? I was very stable at 2GHz with the retail/HSF before water cooling. And what resolution are you talking about, 9+++ scores?

 

[blindspot]

forcesho, I wonder how much you've lost recently after the recent NASDAQ crash. Got rid of AMD at 30s. The outlook for the chips is not that great. Even at these bargain prices and low P/Es, the chances of getting an upsurge is dismal. Without any good news coming, both AMD and Intel stock prices are gonna go down, probably in the sub-20's. However, AMD might have a slight bit of advantage, considering that it's been gaining marketshares, not losing. So good luck to all of us traders!

 

[Pabster]

zzzt wrote:

&quot;A score of 9415 at 2272 MHz??? 8161 is the highest athlon score at 1599MHz. (for some reason MP's show no results although Pabster has submitted his MP results)&quot;

Well, d-oh, those benchies were ran at 640x480. How practical is that? 3DMark2K1 does not recognize my CPU, for some reason, so it lists &quot;Unknown&quot;. Probably has something to do with the 8K7A not fully supporting Palomino.

&quot;That's a 15% performance difference... Now I love the athlon, and run it myself, and I understand that you're looking at a good $200 more to build a P4(not including a new case because for 4 bucks you can get an adapter to let your normal ATX PS run a P4 board), but 15% is a heck of a lot!&quot;

Again, if you like to play your games at 640x480, go right ahead. Let's look at the numbers @ 1024x768x32, i.e. the DEFAULT BENCHMARK. I don't see the 15% there...

&quot;Are P4's running at these speeds unstable? Why have I not seen these 2GHz+ chips in ANY articles? I've seen plenty of benchmarks showing OC'd athlons against 1.2, 1.5, and 1.7GHz P4's, and am feeling a bit cheated when we don't get to see the OC'd P4's also.&quot;

First off, not all 1.7 P4 parts will hit 2+ GHz. Like anything else, it is hit-and-miss. Second, P4 sales are rather slow. And the reality is, most of the P4s sold are OEM and Retail units, not DIYers (like most of us.) Therefore, you aren't going to see a lot of these overclocked P4 rigs on &quot;display&quot;.

&quot;Unless my thinking is flawed in my above statements, a parallel between the AMD-Intel and the Mac-PC arguements can be made. How many times have you heard a mac guy say,&quot;The G4 can do sooo much more per clock cycle than any PC?&quot; We always responded with,&quot;Who cares - you're still slower.&quot;

I'm not following. No one here has yet disputed that 2.2GHz P4s will outpace even the fastest O/C AMD machines. But one must take in to account the clock speed advantage the P4 has. 2200MHz against what, 1700MHz tops? That's rather significant, I'd say.

&quot;Does anyone know of any reliable benchmarks showing a 2.2+ GHz P4 against a 1.5+ GHz Athlon? Any information on stability? When 6 P4's are faster than the fastest Athlon, I really have to wonder...&quot;

You're looking at one benchmark. Try a few more. It's hit-and-miss. Athlon wins some, P4 wins some. There's really no &quot;clear&quot; leader.

&quot;P.S. I understand that 3dmark 2001 is not the only benchmark, but 15%...&quot;

Let me reiterate: you are looking at one benchmark, and it was ran at 640x480. Think about it.

 

[Pabster]

Fkloster wrote:

&quot;Tough one to prove, especially with newer games like Sacrifice, Giants, NASCAR 4, MBTR, Tribes 2, ect that I would argue run faster on the P4 but have limited Benchmarking results published. The future 'Max Payne' will smoke on the P4 as it will be written to take full advantage of DX 8 &amp; SSE2.&quot;

That's exactly the kind of statement that provokes a flame shoot. You're inferring performance on a product still in development. You can't say that it will &quot;smoke&quot; because you haven't even seen it yet. DX 8 won't help one processor more than another; that is GPU-dependent. SSE2 remains to be seen. I wouldn't count my chickies before they're hatched.

 

[Pabster]

Ben wrote:

&quot;But in the real world, the Ultra will keep up with the GF3 in FPS under most conditions.&quot;

That's just not true. In 16-bit resolutions (especially lower ones) they are about equal, to the Ultra beating the GF3 by a frame or two. Once you go to 32-bit resolutions, especially high resolutions, GF3 has a sizeable advantage over the heaviest of overclocked GF2 Ultras.

 

[BreakApart]

Pabster i totally agree with you.(nice post)

The Mhz -vs- % increase: Heck, that's exactly what i was trying to point out in my first post. Somehow it got turned around, and i was compared to a MAC zealot. Ohh well...

Personally i think it's great these die-hards are pushing the CPUs to the limit like that.

Intel or AMD in my home systems I could care less, I'm all for the best bang for the buck!

 

[nyati]

Who else is having problems logging on the MadOnion? I saw one guy with the &quot;you are not logged on&quot; in a previous message. This is really strange. My work PC on NT can still access it through its cookie, but my home PC can't on WinME or Win2K. Any ideas? thoughts? rants? flame?

This has me stumped.

 

[nyati]

Never mind. My date was set to 2007, did I fatfinger it at some point? I don't remember. Ah well, problem solved.

 

[Quickfingerz]

You can't beat the Memory bandwith of a P4

 

[Pabster]

QuickFingerz:

&quot;You can't beat the Memory bandwith of a P4&quot;

Nor the high latencies inherent in RDRAM

(I've just put on the flame suit. Gentlemen, do NOT start your engines!)

 

[Pers]

Pabster no need for a flame suit...you simply stated a fact.

 

[Rectalfier]

I just built a tricycle in my backyard, some day I will hit 100mph with it. Right now I'm getting really tired doing all that peddling.

Next I'm going to release the Idunnowjaqsquatium running at 5000Terahertz, With the new FNBS instruction set.

All you naysayers will be eating your shoes for breakfast.

No Really Seriously.

 

[Zzzt]

Pabster wrote:



<< Again, if you like to play your games at 640x480, go right ahead. Let's look at the numbers @ 1024x768x32, i.e. the DEFAULT BENCHMARK. I don't see the 15% there... >>



Of course not. at 640x480, the limiting factor is not the video card, but the CPU. The vid card is becoming more of a limitation as the resolution goes up. If you check the scores at 1600X1200 the scores become almost identical. Again, the video card is the limitation. When the Kyro3 or GeForce4 or whatever it is that's new in the next few months arrives, you'll see the spread look more and more like the spread at 640x480... At least in this benchmark.

I've been following your posts, Pabster, and I truly respect and admire you. Please don't consider any of this a flame in any way; I'm just feel that there might be a feeling among AMD lovers similar to that of mac lovers.

You said:



<< I'm not following. No one here has yet disputed that 2.2GHz P4s will outpace even the fastest O/C AMD machines. But one must take in to account the clock speed advantage the P4 has. 2200MHz against what, 1700MHz tops? That's rather significant, I'd say. >>



Isn't that a bit of a bad arguement? Isn't that a mac arguement?
&quot;(fill in AMD or Mac here) can do more per clock cycle.&quot; - So what? (Fill in Intel or PC's here) is/are still faster.

We've seen all kinds of benchmarks of OC'd Athlons against P4's. Why hasn't the P4 been given the same advantage? That is the reason behind my stability question.

 

[Ben]

I love people that dissect posts and then argue about facts to death, making it look like everyone is wrong in some way or another except for themselves.

The point was (Pabster) that the GeForce3 gets additional points for features, rather that speed. Maybe the P4 has the same advantage over the Athlon in relation to 3DMark 2001.

 

[christoph83]

Blindspot .Remember we are 3 months away from the computer buying season and plus windows XP is coming out. But your correct, the upside potential in Intel and AMD stock is much lower than maybe some other compaines. Both good investments.

 

[Rankor]

&quot;...designed for 10 Ghz...&quot;

There's a difference between designed for and obtaining sufficient yield at those speeds.

I'll believe it when I see it, no earlier.

 

[dowxp]

*yawn*. i dont really care anymore... . . ..

 

[Sugadaddy]

3Dmark is a horrible benchmark. It is useful for checking your system, and it shows when something's not set up properly, but it shouldn't be used to compare different products. You do a bench with a GF2 Ultra overclocked beyond belief, then you do another one with a stock GF3, and it like doubles your score. A good benchmark scales linearly, it doesn't jump just because it recognizes the name of a product. It has always ran faster on Intel CPUs anyways.

And it is a synthetic benchmark, yes it has a game you can try for 2 minutes, but if you're playing that thing all day, you have a problem...

BTW, you can play a free game (without registering) in 3dmark2001. Just enter &quot;Holy Cow!&quot; in the project name box.

 

[Dravic]

just an FYI, 3dmark IS(or has) A GAME.. and I'm NOT talking about the registed version

there is an &quot;easter egg&quot; game in 3dmark2001

OPen up 3dmark2001

click the edit button under the project section

name the project

Holy Cow!

exactly like that, click ok

click on gamed demo.. and enjoy killing the cows..

&quot;t&quot; for tracking and firing.. mouse moves camera angles..

 

[Degenerate]

<< 1.6G -vs- 2.2G = 27% clock difference, yet ONLY a 15% performance increase.
Heck, that doesn't even include the price difference.
Sad, sad day when people think this type of performance ratio is a good thing... >>




<< But one must take in to account the clock speed advantage the P4 has. 2200MHz against what, 1700MHz tops? That's rather significant, I'd say.
>>


hell, so? when AMD can slap out an 2.2 then talk about it. the fact is that Intel can produce a 2.2 and AMD can do a 1.6. Period. (yes, un OC'ed it is 1.7 and 1.4)

 

[insane]

Right on Degenerate
I have a AMD so what. I am not proud of it it is suck
I will switch to p4 as soon as 478 appear which is rulez
AMD/VIA combo really is the worst cpu/chipset is definitly the worst combo EVER

 

[julianf]

my two PIII's is currently running stable at 67Degrees C. sorry, but i havn't got my aquacool yet. give me an AMD and let's see for how long it'll do that...

 

[formulav8]

<<


<< I get up to 5% to 10% different results everytime I run it... >>



Assuming you are getting atleast a score of 5000 (prolly more like 6500) with that rig of yours, you are proclaiming that your score jumps from 250 to 500 point swings every time you run the bench? I call Bullsh|t on the unbelievable statement formula. That is a lie. :| >>




10% may be stretching it alittle bit, but 5% isn't.


Jason

 

[grant2]

The Law of Diminishing returns only applys to the same object/concept when subjected to growth.
These (2) CPUs are NOT the same!

You are the one making it sound like that 2 different CPUs should for some reason not only scale exactly the same, but should scale linearly. What makes a clock: performance ratio such a holy grail?

But fine, if you insist, let's compare an AMD athlon scaled from 1200 -> 1600. That's a massive 33% clock increase. Yet its webmark 2001 score only rises 18%. Its 640x480 q3 score rises a measly 14%. (Benches)

I was comparing an Athlon -vs- P4, That it took 27% more Mhz in the P4 to beat the 1.6G Athlon by only 15%.

I am comparing an Athlon -vs- Athlon, That it takes 33% more Mhz in the Athlon to beat the 1.2G Athlon by only 14-18%.

Gee, it looks like EVERY consumer processor is &quot;sad&quot; in your estimation, what DOES it take to satisfy you?

 

[SpideyCU]

&quot;And since when is water cooling an extreme method of cooling the chip?&quot;

Man, I gotta tell ya, I don't want to know your version of 'extreme' method of cooling a chip. Gonna bring out some cryo stuff?

&quot;Pabster no need for a flame suit...you simply stated a fact.&quot;

You'd be surprised how a proponent of either side can turn a fact into a war. &quot;That may be so, but...!&quot; I'm just happy I haven't taken a &quot;side&quot; in the chip world, even though I am (gasp) running two T-bird systems. All I openly hate is Microsoft. But even that's become uneventful since Microsoft-loathing has become taboo. I mean, heck, this whole THREAD was started simply with the stating of a fact, and look what happened. It doesn't take much man!

I think being for or against a certain chip &quot;just because&quot; it came from a certain company is ridiculously childish. To a point I have to fault the Intel lovers here because anytime someone says something positive about AMD they're labelled a 'zealot'. Seems objective opinions aren't often found here. Now ya know why I won't take sides. Especially when people start talking like they're in a gang war in the bad part of town...then you know it's quickly going downhill!