Sign in with
Sign up | Sign in
Your question

P4 2800Mhz! Intel just loves it!

Last response: in CPUs
Share
August 26, 2002 5:31:00 AM

jeez.... Intel loves waiting for the right times dont they... They knew...they knew 2600+ XP was coming out. They can afford to wait since they have the fastest chip out there. But not AMD, AMD needs to get out his chips to be a bit more competitive. and just as we thought AMD got the throne... BAM! INTEL STRIKES BACK! Although we knew the new intel's were coming out soon, it was obviously not coincidence that it came out just a few days after AMD's flagship cpu's...Intel saw, came, and conquered. But some of the AMD overclocked chips didnt do bad...but I feel bad for AMD for having such a hard time...It happens every time.
But after all, price/performance crown goes to AMD... but still.... Its gonna be very hard for AMD. IMO, the only way that AMD can take over is as soon as Intel delays something or makes a mistake...that will be the AMD's chance to truly take the crown (at least for a few months) OR AMD comes out with something UBER COOL. We will see

Comments? Feelings? Let it all out here brother...

real philosophy of life: "do onto others what you dont want them do onto you"

More about : 2800mhz intel loves

August 26, 2002 6:28:26 AM

Maybe this is a little harsh, but: it was well known that Intel would announce 2.8G around or before Sept 1, but T-Bred 2600+ was a real surprise to everyone. It is therefore pretty clear that AMD tried to steal some thunder from Intel, not the other way around. It is also obvious to me that AMD paper launched 2600+ 5 days before a real announment of P4 2.8G so as not to be embarassed by Intel's 1G lead ahead itself: P4 2800- AXP 2200+ (1800M) =1G.
August 26, 2002 7:13:03 AM

you are right on the customer (which is most of us) side point of view. but you gotta know that sometimes release dates can change quickly. intel's chip was scheduled around or before sept 1. thats true, but they didnt specify EXACTly what day. that gives intel a little room to work for. Intel is not dumb, so they had their ways to know what day the 2600+ XP was going to released. IMO, i dont think AMD would dare steal thunder from intel b/c AMD knows that Intel can just beat them (in this case, 2800mhz). AMD has no other option but to release earlier. Remember, All this is my opinion so im not trying to start an argument. but because I have seen this (AMD vs Intel) (ATI vs NVIDIA) this kind of stuff happens a lot (where the powerful waits for the weak to release first) I concluded that Intel did the same again. Again, i might be wrong but you gotta admit that intel has done it before, and IT could be a possiblity that he did it again. We will never know what intel and amd knows about each company and "REAL" release dates. I dont think matisaro appreciated your retirement from his thread because his is the "ORIGINAL ARTICLE THREAD" but you made the right choice.

real philosophy of life: "do onto others what you dont want them do onto you"
Related resources
a b à CPUs
August 26, 2002 5:43:47 PM

It's a shame. And it's not that AMD CAN'T compete at the top, but that they WON'T. That's what they said in a press release several months ago. They said they were aiming at the much larger midrange market.

<font color=blue>You're posting in a forum with class. It may be third class, but it's still class!</font color=blue>
August 26, 2002 7:16:42 PM

is it me or did the 'tom's blirp' thread disappeared.

<font color=green> there's more to life than increasing its speed -Ghandi</font color=green>
August 26, 2002 7:27:11 PM

Great news for you, huh?! You are very happy with it, right?
Probably you wanna AMD out of the race. I wanna see your face if AMD goes out of the market, and your lovely Intel doubles the prices. This is exatly what will happen if Intel stays alone.

<b>Before getting angry to the (d)evil, just think about who CREATED it!</b> :mad: 
August 26, 2002 7:38:42 PM

Just read the article on the 2.8GHz P4 now while i really like THG for its impartial testing...and i do agree with the results that come out. Howvever I disagree with the way the data is obtained and presented. First of let me say I'm neither for or against Intel or AMD i just want the 'best' chip i can get for my (limited) budget.

Right now to my point...the 2.8GHz 133FSB P4 is the fastest CPU out at present, the results speak for themselves. What i disagree with is the way you compare P4s to Athlons. Granted the point of the article is to show which CPU is the fastest using the best kit each has to offer in a series of tests, however it the comparison part which i think is flawed.

My understanding in comparison tests is to compare similar products under similar conditions i.e. use the lowest common denominator. In yours P4 are tested with RD-RAM which is acknowledged to be 'superior' to DDR. Wouldn't a fairer test be to compare using the same memory type to maintain system parity during testing and then use RD-RAM to show what the optimum results would be for P4s. The way your article reads its skewed towards P4 by the way the tests are done. Let me just say again its not the results that bother me its the way then tests are done and presented.

Let me put it another way, compare the P4 with and without RD-RAM with the Athlon XP otherwise you are, whether you intend or not, presenting a biased view of P4s vs AXP.

Lets compare apples with apples and not apples and fruit salad.

Of a personal note, what i would like to see is an article which addresses the CPU clock speed issue. Would it be possible to re-rank your charts to compare similar speed CPUs from AMD and Intel so as to give comparison of clock speed vs performance. This would provide a more direct comparison of CPUs for even novices to understand.

All in all you website is still my favorite for PC news and Mr Pabst re: Tom's Blurb, don't let the b@$tards grind you down.

BananaSkin
August 26, 2002 7:58:59 PM

tersagun, you obviously didnt understand my article fully. Au contraire monsieur, i want amd so both companies can offer competitive chips at competitive prices. i even said that i felt bad for amd. please read my post carefully.

real philosophy of life: "do onto others what you dont want them do onto you"
August 26, 2002 8:02:15 PM

Sorry then, my English isn't so well. But i still feel you are soo happy that Intel is far in the lead now.

<b>Before getting angry to the (d)evil, just think about who CREATED it!</b> :mad: 
August 26, 2002 8:10:53 PM

"but I feel bad for AMD for having such a hard time"

thats what i said.

real philosophy of life: "do onto others what you dont want them do onto you"
August 26, 2002 8:25:25 PM

BananaSkin, your logic is flawed for at <i>least</i> two reasons:

1) AMD and Intel are <i>already</i> apples and oranges. Each chip's actual arcitecture is extremely different. As I understand it, even the manner in which they process x86 instructions differs. The <i>only</i> similarity is that they both process the same generic instruction set. (Though the detailed instruction sets (SSE2, 3D Now!, do differ.) So there basically is no true good way to compare them as 'better' or 'worse' because each performs very differently per individual application.

2) How can you declair one processor 'the best' when it's competition is <i>intentionally</i> being handicapped by slower RAM?

It isn't Intel's fault that AMD refuses to embrace superior bandwidth. Instead of trying to run Intel's CPU with DDR to match the Athlon, why don't you ask why you can't run the Athlon with RDRAM to match the Intel system?

If anything, it is a flaw in AMD's CPUs. Hence, it should be AMD's problem, not Intel's, and thus AMD's benchmarks that suffer, not Intel's.

P.S. It's technically DRAM, so if you put the dash anywhere, it would be R-DRAM and (DDR) S-DRAM.

<pre><A HREF="http://www.nuklearpower.com/comic/186.htm" target="_new"><font color=red>It's all relative...</font color=red></A></pre><p>
August 26, 2002 8:35:05 PM

Quote:
Probably you wanna AMD out of the race. I wanna see your face if AMD goes out of the market, and your lovely Intel doubles the prices. This is exatly what will happen if Intel stays alone.

You say that as though it were a <i>bad</i> thing. ;)  Heh heh.

Seriously though... <i>If</i> AMD were to get knocked back down to the status of VIA, it would not be all bad.

First of all, Intel would stop ramping up CPU speed so much. So then the $2000 you spend on a P4 2.8GHz would in two years time only be like $1500 for a P4 2.8GHz, or $2000 for a 3.2GHz. Thus, your system would remain state-of-the-art for considerably longer and computers would no longer be such a considerably waste of an investment. (From a financial point of view.)

Secondly, if computer <i>hardware</i> technology growth were to stagnate, then <i>software</i> would finally have to compensate. Programmers would start to actually optimize to make the absolute most out of what is available in hardware. This would both require the general computer programmer to improve in skill dramatically, but also result in plain old better software that <i>doesn't</i> waste your resources like a spend-thrift wife.

Third, how many hazardous-material PC parts get tossed into landfills around the world instead of being processed properly to protect our environment? A slowdown in PC technology would actually improve this pollution as less and less PCs would be thrown away.

There are other benefits as well. But my point is that were Intel to suddenly be the only CPU manufacturer worth a darn (or even were AMD to take that place) there would be a great number of benefits that would offset the immediate downfall of an increase in price.

In the end, neither way is explicitely 'good' or 'bad'.

<pre><A HREF="http://www.nuklearpower.com/comic/186.htm" target="_new"><font color=red>It's all relative...</font color=red></A></pre><p>
August 26, 2002 9:11:42 PM

An interesting look at things.

Also, with that race slowing down though, you would slow down the consumption of new other parts of hardware. Hard Drives, CD/DVD drives, Sound Cards, Video cards, Motherboards, Memory, and basicly, it would stall a large and lucrative industry beyond the CPU market.

That, in turn, could cause a major decline in the market both in the US and worldwide. It wouldn't be the Dot Com crash, especialy because you would get far fewer overpaid people demanding too much money for their next job (One of the dumber problems of the Dot Com crash, Medeocre web designers unwilling to accept sup $100,000 a year jobs because that was what they were used to)

Cometition is good for the consumer, but it is also good for the market, because it creates jobs.

The Boogie Knights: Saving beautiful monsters from ravoning princesses since 1983.
August 26, 2002 9:22:11 PM

Errmm...if AMD and Intel were indeed apples and oranges, then why compare? Yes, both CPU are architecturally different, process differently but thats not the point, they both do the same job so they can be compared i.e. the end result matters.

If the test was about whole systems then by all means compare Intel with DRAM and AMD with DDR. But this article was titled P4 2.8GHz vs Athlon 2600+ so comparisons should be done on a level playing field so to speak.

And so for the same reason i can say the P4 2.8GHz is currently better than the AXP 2600+ with the best associated parts because it provides a faster perfomance.

I'm not arguing the results i'm just querying the testing procedure. Granted Intel's use of DRAM gives its a significant perfomance boost but thats not my point, why compare CPUs when you intentionally introduce a non-static variable like different memory. If you get my drift...its not good scientific testing procedure.

The test should have been a comparison of the P4 vs AXP perfomance with the same memory and then compare those results with P4 with DRAM. The results still would stand with P4 with DRAM being the fastest combination out there but the P4 vs AXP comparison would then make more sense in relative terms.

Sorry if i didn't make myself clear the first time.

BananaSkin
August 26, 2002 10:54:38 PM

You all seem to be forgetting the most basic and important thing. Benchmarks were not created so you fanboys can bicker. They were created so that users would know how they'd get the best performance. If the P4 is capable of reaching a better performance level with RDRAM, then the P4 setup with RDRAM is the choice for the performance-crazed. End of story. There is no fair or leveled playing field. Such concepts are rediculous in something that was meant to educate the consumer on which setup would be better performing. The fact that you people have turned it into a war of the processors doesn't mean the reviews have to abide by your betty need for something to bicker with.
If you want to know which setup would perform better with a certain type of memory. Look up the results for that. There are plenty of reviewers (such as Ace) who used DDR 333 with the P4. However, if you want the absolute best performance, you'll have to use a P4/RDRAM combo. It's not about fair, it's not what you feel should be leveled so you can feel some pride in a particular piece of silicon, it's about facts and end results for the people who BUY this stuff.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
a b à CPUs
August 27, 2002 12:32:17 AM

Hmm, time to start in inquest.

<font color=blue>You're posting in a forum with class. It may be third class, but it's still class!</font color=blue>
August 27, 2002 2:17:45 AM

It all comes down to: How much money you got to spend in your wallet? That will be THE decisive factor on what you are going to buy.

I.E, if someone asks me "what parts should I buy for my computer?" I wouldnt ask them "do you want AMD or INTEL? do you want ATI or NVIDIA?" I would ask them "How much you got to spend on the rig?"

real philosophy of life: "do onto others what you dont want them do onto you"
August 27, 2002 2:26:19 AM

crash, i think we're the only one that realize this! lol!

<font color=green> there's more to life than increasing its speed -Ghandi</font color=green>
August 27, 2002 2:35:46 AM

Agreed with Grassapa. But if you want for the best "bang for your buck" go with AMD... look at those benchmarks all most ALL of them show that the AXP with sometimes LOWER Mhz than the INTEL counterpart is a lot higher, in all most ALL, this AXP2600+ is trying to compete with a CPU that has a mhz running way out of it's range, it's amazing it can even be questioned to compare.
August 27, 2002 2:41:06 AM

Well, that's not true though. When gamers and people who use computers for cpu/gpu intensive activities (ie "enthusiasts") buy computers, it's going to be either out and right the best thing ever, or it's going to be some combination based mainly on a constrained budget.

Sure, it's not so different among the average computer purchaser, but there are differences. When the average person buys a car, they want either something that does 0-60 in under 5 seconds, or they want something that looks good and doesn't break easily (you want a car that runs above most everything else, right?). When buying computers, i'm sure a lot of people buy Intel, because the number of AMD complaints heavily outnumbers Intel complaints. It's the same reason people don't want celerons as often as something from the pentium line. And since Intel advertises the p4 chip as really great internet and gaming chip for everyone (I can't really tell what AMD tries to market the Athlon as...), people are drawn in. Hondas and Toyotas sell better than Cheverolets and Fords. Even though i personally think a camaro looks better and is more fun to drive than my 2002 toyota camry sedan, i know that I won't have to pay for much on my camry pre-100,000 miles, regarding maintance. Likewise, I know my 1.7 P4 can take the beating I give it with ease, while still lasting until i wish to replace it (and then long after as an extra chip for possibly a server or linux box).

Marketing and track records signficantly help companies. Likewise, personal experience is important. If someone buys their first honda civic and the transmission goes dead after a year, the owner is not going to be too fond of civics, even though they have an incredible track record.

So I don't think it can be said that it's just a price thing. I'm still willing to pay extra to know my parts are going to last. If Moore's Law gets bent and chips start doubling the # of transistors in 6 months, maybe then I'll change my purchasing strategy :) 

Athlons and Pentiums are just melted rock. Who’s rock is better? Who cares, let’s play some games
August 27, 2002 2:59:11 AM

Well, I would have to say that spending $1000 (canadian) on a 2.8 ghz P4 when I could have an XP2600 for about 1/3 of that, I know what I would get. Also, when is Intel going to learn that cache is good??? All they do is play to the marketers, while AMD has been for the Geek.
August 27, 2002 3:10:24 AM

AMDchris i dont have an idea of what u are talking about, but if you have money you buy INTEL, if you dont have much money you buy AMD, its simple like that.
intel offers more performance for more money
amd offers very competitive performance for less money

real philosophy of life: "do onto others what you dont want them do onto you"
August 27, 2002 4:24:25 AM

Quote:

There are other benefits as well. But my point is that were Intel to suddenly be the only CPU manufacturer worth a darn (or even were AMD to take that place) there would be a great number of benefits that would offset the immediate downfall of an increase in price.

I hope your "communist" (one-world) views never come to pass. If you were in charge for the last 15 years we would still be working and playing games in DOS and PC's would be limited to the very few. The garbage produced by obsolete electronics may very well be a problem but slowing down the development of technology is not the answer komrade.



<font color=green>Tbred or Northwood?? Anybody have a quarter?</font color=green>
August 27, 2002 4:58:14 AM

Quote:
AMDchris i dont have an idea of what u are talking about, but if you have money you buy INTEL, if you dont have much money you buy AMD, its simple like that.
intel offers more performance for more money
amd offers very competitive performance for less money


The 2.53 P4 is $249 on pricewatch and is offering comparable performance to the 2.13 Athlon which isn't even out yet and will probably come out at about $250. It's not longer a clear win as far as price/performance. Although at the lower budget range, the Athlons are still cheaper compared to P4's in its performance range but the price difference is somewhere around $50 maximum.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
August 27, 2002 6:30:08 AM

thats nice... But over here in Auzzie land the price of high end P4's <b>DO NOT</b> reflect newegg or pricewatch prices.

current prices from a popular online auzzie store: (prices include cooler)

XP1800+ = $198
XP1900+ = $225
XP2000+ = $247
XP2100+ = $308
XP2200+ = $396

P4 1.8A = $385
P4 2.26B = $544
P4 2.4B = $874

P3T 512k 1.4 = $660.

price equity my tush! :frown:

<b>Due to Customer Complaints, this sig has been witdrawn from public use. Thankyou. :lol:  </b>
August 27, 2002 6:50:02 AM

After reviewing the prices and all, i think you are right imgod2u. The 2.53 really went down on its price and it is very competitive to the newest athlon. The only scenario in which your point might not be valid is that rdram is expensive compared to ddr and rdram is much needed by the 2.53 to perform well. In any case, AMD is getting it tough man..... i dont know how he will manage to survive. Maybe the Hammer might save AMD's ass for a few months until prescott comes out...

wait a minute....why does the 2.4ghz cost 50 bucks more than the 2.53ghz?

real philosophy of life: "do onto others what you dont want them do onto you"
August 27, 2002 1:10:51 PM

Cause Intel dropped a wad on the 2.53's price tag so that it may compete better. The rest of the line remained relatively unchanged. At least, that's my guess.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
August 27, 2002 5:13:22 PM

Firstly, I don't know if you even bothered to read my previous posts imgod2u but i am not a 'fanboy' as you so put it. I personally don't care if i use Intel or AMD (i've used both in the past and will probably do so in the future) i want to know whats good for my money.

Secondly i'm more interested in the testing procedure and display of the results. Benchmarks are by their very nature intended to be 'fair' or in a better word impartial so as not to prejudice the results. Even the best experiments conducted with perfect accuracy can be ruined by poorly displayed results which can bias the findings one way or the other. Thats the crux of my point.

I come from a scientific background so i'm familiar with experimental testing procedures. One of the most fundamental things in a write up of an experiment is to include experimental setup details. This opens the forum up to discuss and criticise in a constructive way.

Let me say it again P4 with DRAM is the best combination currently for top of the range perfomance. That was never in question. I was merely pointing out the rest of the results were displayed in a way i thought to be biased. I personally would have ranked the tables according to true CPU clock speed and ran P4 and AMD with DDR and then compare those results to P4 with DRAM. After all most people do use DDR rather than DRAM and i would rather have the whole truth rather than half to base my decisions on...what about you?

My posts were not intended to become an Intel vs AMD debate, i don't care. Its the way the tests were done and how the results were shown that i'm criticising. I'm sure Mr Pabst, as a scientist would agree, any published results are open to debate, sensible debate and not one that degenerates into name calling or labelling.

BananaSkin
August 27, 2002 6:18:58 PM

Quote:
Firstly, I don't know if you even bothered to read my previous posts imgod2u but i am not a 'fanboy' as you so put it. I personally don't care if i use Intel or AMD (i've used both in the past and will probably do so in the future) i want to know whats good for my money.


I made no direct references to you. I was addressing the entire bickering crowd as a whole. The fact that you took it personally says something though.

Quote:
Secondly i'm more interested in the testing procedure and display of the results. Benchmarks are by their very nature intended to be 'fair' or in a better word impartial so as not to prejudice the results. Even the best experiments conducted with perfect accuracy can be ruined by poorly displayed results which can bias the findings one way or the other. Thats the crux of my point.


Accuracy is a far cry from what you would consider "bias". First and foremost, as I said benchmarks were not created so people could bicker around about "mine is better". They were created to show performance in things that people do. If a benchmark consists of programs that people don't use, then it's not a very important nor relevant benchmark. If 95% of the people in the world, including the enthusiaste community, used a certain software that excelled due to it being hand-optimized for the P4 and performs abismal on the Athlon, it really doesn't matter whether you feel it is fair or not. That is an application that is commonly used and the P4 provides a better performance, period. I'll bet the people who buy the P4 and use that software couldn't give a hoot whether you or anyone else felt a certain benchmark was fair or not. Of course, I'm exaggerating. I doubt there is even software out there that 95% of all people use (with the exception of Windows or Office). And I very much doubt that such software is specially hand-optimized to run well on the P4 and not the Athlon. However, the point still exists. Benchmarks are useful if and only if they accurate portray relative performance in <b>commonly used</b> applications. This is why I frown upon synthetic benchmarks because they really don't mean anything. Do you play Sandra? Shouldn't better performance in, oh say, 3dsmax or Lightwave be more important that something that has no other purpose than to generate numbers? Would it matter if 3dsmax was hand-optimized for the P4? People use it, they benefit from having a P4. End of story.

Quote:
I come from a scientific background so i'm familiar with experimental testing procedures. One of the most fundamental things in a write up of an experiment is to include experimental setup details. This opens the forum up to discuss and criticise in a constructive way.


The testing methods are always being questioned, never the testing field. You can protest that someone tested using inaccurate or invalid methods such as estimating at a too imprecise point or not using a valid clock or something to that nature. The test itself is still a valid test if said things are accurate. Whether you feel the test is relevant or not depends on whether the test results affect you. Again, these tests were not meant so you can argue "mine is better". These tests were meant to show how applications that apply to you would perform. There is no "fair". There's only relevance.

Quote:
Let me say it again P4 with <b>R</b>DRAM is the best combination currently for top of the range perfomance. That was never in question. I was merely pointing out the rest of the results were displayed in a way i thought to be biased. I personally would have ranked the tables according to true CPU clock speed and ran P4 and AMD with DDR and then compare those results to P4 with <b>R</b>DRAM. After all most people do use DDR rather than <b>R</b>DRAM and i would rather have the whole truth rather than half to base my decisions on...what about you?


Again, there is no such thing as biased in test results, only in methodology. Such things are an emotional factor, it has nothing to do with the experiment setup. The person performing it could be biased and the only way in which that would ever show up is if the results were invalid. If the reviewer purposely reported the wrong results or used software that was somehow different than what he reported to have used. Are you saying you doubt Tom's testing skills? Or Anand's? How about Aces?
If you don't feel performance of the P4 with RDRAM applies to you then the benchmark is irrelevant. It's still valid, the numbers are there and they are, and I will give Tom the benefit of a doubt here, accurate. Whether you feel that applies to you or not is your own judgement and in no way makes the test any less valid. Again, these tests were not made so you can bicker about which is better and whether it is fair to compare one to another. They were made to state a few simple facts, which setup performs better than the other. The P4/RDRAM setup performs better than the Athlon/DDR SDRAM setup. That's all the numbers on that article tells. The fact that you (and I say you in general to all the people bickering) have turned it into some kind of cry of dispair and unjust proportions has no impact whatsoever on the validity of the test.

Quote:
My posts were not intended to become an Intel vs AMD debate, i don't care. Its the way the tests were done and how the results were shown that i'm criticising. I'm sure Mr Pabst, as a scientist would agree, any published results are open to debate, sensible debate and not one that degenerates into name calling or labelling.


The only difference is the debate should be about the test. Not bickering about something completely irrelevant. Such as "fairness" as you call it. The test was a simple comparison of different setups, nothing more. If Tom had claimed that one CPU was clearly superior to the other in and by itself, then you'd have something to bitch about. But he didn't, and neither did any other review I've read (except maybe the AMD fansites claiming the Athlon as "clearly superior").

As far as comparing CPU's directly, let me put it this way. Do you feel that when the tests for the Thunderbird Athlons vs the P3 CuMines/Tualatins were unfair? The P3 was using PC133MHz memory while the T-birds were given DDR SDRAM memory. Was that unfair? I don't think so. The T-birds had the extra FSB to take advantage of DDR. That was a feature of the CPU. You might as well start claiming that the extra FPU power the Athlon has is unfair in tests that are x87 FPU intensive. It's a <b>feature</b> of the CPU. It's one of the benefits you <b>buy</b> it for. Why don't we start crippling down everything on the CPU and start comparing how well each handles legacy 16-bit x86 code? Maybe then, finally, you'd be satisfied that the test was "fair". Of course, nobody runs much of 16-bit x86 code classic code anymore but that shouldn't matter right? All that matters is so you can finally have a test result in which any feature of a CPU that puts it ahead of the other is disabled and you can finally call it "fair".

"We are Microsoft, resistance is futile." - Bill Gates, 2015.<P ID="edit"><FONT SIZE=-1><EM>Edited by imgod2u on 08/27/02 11:28 AM.</EM></FONT></P>
August 27, 2002 6:57:18 PM

Quote:
I hope your "communist" (one-world) views never come to pass. If you were in charge for the last 15 years we would still be working and playing games in DOS and PC's would be limited to the very few. The garbage produced by obsolete electronics may very well be a problem but slowing down the development of technology is not the answer komrade.


LOL.. i told him weeks ago that if all of mankind were like him we'd still be in the stone-age. He was only surmising though, nothing actual fact based, i.e., if the AMD went out of business and somehow intel were the only mainstream CPU manufacturer your system still wouldn't last years before being old and outdated. Intel would still produce different chips in order to get annual revenues. Without a direct mainstream competitor, the US Government would finally be able to launch a monopolistic practice suit against intel, with Intel passing the costs surely on to consumers. Also memory and GPU technology would still be evolving, as it always will, and Microsoft wouldn’t know how to optimize sh!t if their life depended on it, and god bless communism and the RIAA!



"Opinions are like assholes, everyone has one"
August 27, 2002 7:10:39 PM

BananaSkin, look at it this way.

Say you have two cars. Car A is designed to use standard unleaded petrol. Car B has a special engine designed to use racing fuel.

Now, some schmoe comes along and makes a fuel converter to let Car A run on standard petrol, but at a lower performance.

If you were to compare the speed of the two cars, would you run them both on standard petrol just because they both can now? Or would you run each engine as it was <i>designed</i> to be run? (Thus Car A with high quality standard petrol and Car B with racing fuel.)

A <i>good</i> scientist would see that the <i>only</i> fair comparison is to run each as it was designed to run.

<pre><A HREF="http://www.nuklearpower.com/comic/186.htm" target="_new"><font color=red>It's all relative...</font color=red></A></pre><p>
August 27, 2002 7:24:00 PM

Quote:
I hope your "communist" (one-world) views never come to pass. If you were in charge for the last 15 years we would still be working and playing games in DOS and PC's would be limited to the very few. The garbage produced by obsolete electronics may very well be a problem but slowing down the development of technology is not the answer komrade.

kusek, that was a pretty pathetic post.

1) Your obvious prejudice is just plain sickening.

2) Communists are not any more about a 'one-world' view than Capitalists are.

3) A <i>good</i> government <i>should</i> have elements of Communism inherant to the design. Otherwise who would feed the poor? Who would show mercy to the downtrodden? Are you saying that in your 'cut-throat' world we should just kill the weak because they only drain our resources?

4) Communism aside (since I make no claims to hold such a political distinction, and am, in fact, a disbeliever in humanities ability to <i>ever</i> develop a perfect form of government) I never once said that this was the way I would prefer the world to go. I merely said that there are several benefits which most people completely overlook. Obviously, you casually neglected my statement of "<font color=blue>In the end, neither way is explicitely 'good' or 'bad'.</font color=blue>" But then of course you did, because you appearantly lack the ability to perceive and cogitate on any but the most basic of levels.

<pre><A HREF="http://www.nuklearpower.com/comic/186.htm" target="_new"><font color=red>It's all relative...</font color=red></A></pre><p>
August 27, 2002 7:33:43 PM

Quote:
An interesting look at things.

Also, with that race slowing down though, you would slow down the consumption of new other parts of hardware. Hard Drives, CD/DVD drives, Sound Cards, Video cards, Motherboards, Memory, and basicly, it would stall a large and lucrative industry beyond the CPU market.

I disagree emphetically. Look at the race that went on for new and amazing graphics and sound during Intel's reign in the early x86 era. Look at how all of the printers and scanners kept improving as well, competing like mad. There was a rapidly advancing market for computer components other than the majestic CPU before, and still.

So it is my belief that should the CPU development stagnate, it would only result in an additional push into the computer peripheral markets. In effect, the lost jobs from one group would be consumed by a new push for the parts of the PC that aren't directly tied to the CPU.

Gratned, I still generally prefer the current method of CPU advancement. However, should that suddenly change, I do not believe that it would be the harbringer of the apocalypse that most people make it out to be. It would merely be a different - but still acceptable - way of life.

<pre><A HREF="http://www.nuklearpower.com/comic/186.htm" target="_new"><font color=red>It's all relative...</font color=red></A></pre><p>
August 27, 2002 7:38:15 PM

Quote:
Hmm, time to start in inquest.

>>>Adjusts the brin of his "witch's" hat.<<<

If anyone even so much as picks up a torch and pitchfork, I'm turning them into a newt!

<pre><A HREF="http://www.nuklearpower.com/comic/186.htm" target="_new"><font color=red>It's all relative...</font color=red></A></pre><p>
August 27, 2002 7:41:46 PM

I really don't think he was saying the stronger shall triumpth, the weak shall perish.

BananaSkin, I think the way to tackle it is to show more emphasis on how much a system containing each cpu is going to cost, I would like to see a couple charts showing the peformance/price of intel cpus vs athlons considering the price of motherboards and ram, at the very least the complete cost of the system used to benchmark could be estimated. It would though, as has been said really be a weapon agasint fanboys -for the reason they don't consider the cost, after all they don't buy the things, only argue about them :) .
August 27, 2002 7:44:05 PM

Actually, it would be "fair" if they did test the P4 with DDR. Why? Because you're testing the Athlon with the best memory out there for it, and you should do the same with the p4. You can't just cripple the P4 and call it "fair". A truly fair comparison is when both CPU's are running the BEST platform that you can possible get for it, and all the best parts that are associated with the CPU. I have yet to see a truly "fair" bench. Even Tom's has discrepencies, but Toms benches, along with Anand's are as close as you can get to being fair.

imgod2u, how many consumers <i>actually</i> look at benchmarks. If you exclude overclockers, gamers, and basically the whole tech community, the "average" consumer doesn't even know what benchmarks are. But if you're referring to consumers such as enthusiasts or techies, than we have a right to bicker, because benches are either biased, unfair, or both. they may have other problems, too. I've never seen a "perfect" bench, because nobody is perfect, but Tom's and Anand's are close. If the benchmarks are unfair, it defeats the whole purpose of benchmarking CPU's, becuase to get a fair comparison, you have eliminate as many independant variables as you can, since the p4 and tbred are so different in architecture. You'll never get a truly fair bench, but that doesn't mean that you should act like homer simpson and give up on the bench entirely, or like van smith and produce a very biased bench. I mean if the bench is unfair or biased, then you might actually get the <b>wrong</b> conclusion from the bench if you're looking for the best product (amdmb and vanshardware are good examples). On those 2 sites, the 2600+ is shown to repeatedly stomp all over the 2.5B, and we all know that's rubbish. So, you'll get results and the "facts", only if effort is made to try to make the bench as fair as possible. As you can see, sites like <A HREF="http://www.amdmb.com" target="_new">http://www.amdmb.com&lt;/A>, or <A HREF="http://www.vanshardware.com" target="_new">http://www.vanshardware.com&lt;/A> actually put alot of effort into their benches to make them biased and <b>unfair</b>.

Grassapa, it all comes down to not only how much you want to pay, but whether you want a quality product or not. Most people on these forums these days thinks about only price/performance, but they don't seem to think too much about the quality of the product. In other words, do people really think whether they are getting a quality product for their money or not. No, most don't. And that's why so many people dismiss Intel CPU's nowadays.

AMDChirs, have you been living under a rock or something? Intel HAS realized that cache is good. Jeez man, Banias will have 1MB of L2 cache, Dothan will probably have 2MB of L2 cache, Prescott will have 1MB of L2 cache, Mckinley can have up to 3MB of L3 cache, Gallatin will have over 1MB of L2 cache, Madison will have up to 6MB of L3 cache, Deerfield will have up to 3MB of L3 cache, Nocona will have at least 1MB of L2 cache, and Montecito will have a gigantic <b>12MB of L3 cache</b>. ALL of these caches will (and are) on-die, so they all run at full speed. Please don't tell me Intel hasn't realized cache is good. BTW, if you don't know what any of these Intel CPU's are, just ask, and I'll explain.

- - - - -
Montecito - successor to Madison, and one monster of a CPU.
August 27, 2002 7:48:49 PM

imgod2u i never said the test results, conditions or tests were in question, i merely stated that the test field does not address a certain and significant variation of products available. P4 with DDR ram is a very real sector of the market. Let me put it this way, how would Tom like his test scores mis-quoted in an advertisement by a PC supplier who states the P4 inside his computer will make his computer the best around. The dealer also offers 512MB of RAM which is on offer. The unlucky punter buys into his words and makes a purchase thinking he's got the best kit around unknowingly his kit came with DDR. Thats my point...the test scores should show the P4 DDR combination as a comparison. Its not about AMD or Intel its about a complete picture.

The fact that you jump on the 'fair' word is perhaps a mistake on my part but i did stress that impartial is a better word to use in this matter. All testing should be done impartially and without bias, i have no doubt that the tests were done with extreme care but that doesn't excuse the ommission of a test combination. Thats what my analogy with experimental setup data refers to, an ommission of the P4 with DDR data is IMO a mistake.

And as for no such thing as a bias in test results...tell that to Phillip Morris and all the tobacco companies which did all those tests to prove smoking tobacco was not addictive or a significant danger to your health. Many a scientist has been caught out trying to fiddle their result to prove one thing or another. Pharmaceutical companies try to do it all the time thats why governments have regulatory bodies. I was taught the truth is the truth, no matter how good, bad or indifferent your results stand. An ommission is a bias away from the whole truth.

Now i can understand where you are coming from, and i do agree with you about the endless bickering about 'mine is better', its boring. What i was hoping to do was open some eyes about what you see is not always whats out there. I hope you can understand where i'm coming from. I'm not disagreeing with anything you said its just that i don't think you understand what i'm saying.

And about the 'fanboy' thing, i accept you may not have referred it personally to me but from what i've read in some web forums its not used in a kind way whether to an individual or a group, it smacks of a patronizing attitude. I'd prefer reasoned argument rather than classification.

BananaSkin
August 27, 2002 8:11:01 PM

slvr_phoenix if you compare your analogy to my point i would run Car A and B with standard petrol and then compare those results to Car B on racing fuel (btw it was Car B with the racing fuel).

Then you get three sets of data to compare the relative merits of three different test combinations. A scientist would then get the whole truth on the matter.

BananaSkin
August 27, 2002 8:15:17 PM

Quote:
AMDChirs, have you been living under a rock or something? Intel HAS realized that cache is good. Jeez man, Banias will have 1MB of L2 cache, Dothan will probably have 2MB of L2 cache, Prescott will have 1MB of L2 cache, Mckinley can have up to 3MB of L3 cache, Gallatin will have over 1MB of L2 cache, Madison will have up to 6MB of L3 cache, Deerfield will have up to 3MB of L3 cache, Nocona will have at least 1MB of L2 cache, and Montecito will have a gigantic 12MB of L3 cache. ALL of these caches will (and are) on-die, so they all run at full speed. Please don't tell me Intel hasn't realized cache is good. BTW, if you don't know what any of these Intel CPU's are, just ask, and I'll explain.

You just beat juin's record of talking about cores so far away from ever being heard!
Lol, especially at the 12MB cache, damn that must take lots of transistors, and a lot of core space, and HEAT! Intel finally does something for the city hall, one CPU powers an entire neighborhood, for one price only, yay!


--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!
August 27, 2002 8:35:38 PM

Quote:
imgod2u, how many consumers actually look at benchmarks. If you exclude overclockers, gamers, and basically the whole tech community, the "average" consumer doesn't even know what benchmarks are. But if you're referring to consumers such as enthusiasts or techies, than we have a right to bicker, because benches are either biased, unfair, or both. they may have other problems, too. I've never seen a "perfect" bench, because nobody is perfect, but Tom's and Anand's are close. If the benchmarks are unfair, it defeats the whole purpose of benchmarking CPU's


And what exactly is the purpose of the benchmarks? Is it not to tell what performance you'll get with whatever solution you choose in programs in which you'll use? Be you an enthusiaste, an overclocker or a general consumer? Are you seriously saying that the benchmark's purpose was so that you can bicker about "mine is better" in forums?

Quote:
becuase to get a fair comparison, you have eliminate as many independant variables as you can, since the p4 and tbred are so different in architecture. You'll never get a truly fair bench, but that doesn't mean that you should act like homer simpson and give up on the bench entirely, or like van smith and produce a very biased bench. I mean if the bench is unfair or biased, then you might actually get the wrong conclusion from the bench if you're looking for the best product (amdmb and vanshardware are good examples). On those 2 sites, the 2600+ is shown to repeatedly stomp all over the 2.5B, and we all know that's rubbish. So, you'll get results and the "facts", only if effort is made to try to make the bench as fair as possible. As you can see, sites like http://www.amdmb.com, or http://www.vanshardware.com actually put alot of effort into their benches to make them biased and unfair.


What about the purpose of benchmarks calls for what you say is "fair"? Benchmarks are there to tell you which solution would be the fastest. Period. It is not there so you can play "mine is better" with CPU's with other people in the forums. If a set of benchmarks reported than an overclocked Athlon at 2.9 GHz beat a 1.6 GHz P4 at stock speed, it is still valid. Irrelevant I'll give it that but there are plenty of benchmarks out there. If this one doesn't apply to you don't read it.

Quote:
imgod2u i never said the test results, conditions or tests were in question, i merely stated that the test field does not address a certain and significant variation of products available. P4 with DDR ram is a very real sector of the market. Let me put it this way, how would Tom like his test scores mis-quoted in an advertisement by a PC supplier who states the P4 inside his computer will make his computer the best around. The dealer also offers 512MB of RAM which is on offer. The unlucky punter buys into his words and makes a purchase thinking he's got the best kit around unknowingly his kit came with DDR. Thats my point...the test scores should show the P4 DDR combination as a comparison. Its not about AMD or Intel its about a complete picture.


So you're saying the RDRAM setup tests are irrelevant because there are more people with DDR setups out there. Well ok, there are plenty of RDRAM vs DDR benchmarks out there. People who have DDR can clearly see that that the RDRAM setup would perform better. Anyone, anywhere can be quoted out of context. If Tom had used DDR SDRAM in his benchmarks then some AMD PR guy could just as easily use it as saying "the Athlon is faster" without mentioning what the rest of the system is. Toms benchmarks, as well as all the others, mentioned what setups they were using. So you tell me, what's are you complaining about? There are DDR benchmarks out there. Check out Aces review.

Quote:
The fact that you jump on the 'fair' word is perhaps a mistake on my part but i did stress that impartial is a better word to use in this matter. All testing should be done impartially and without bias, i have no doubt that the tests were done with extreme care but that doesn't excuse the ommission of a test combination. Thats what my analogy with experimental setup data refers to, an ommission of the P4 with DDR data is IMO a mistake.


If you find the benchmark's fields irrelevant to you then it is just that, irrelevant. Doesn't make it any less valid. A person is impartial because he chose to use the best performing platform for each? I have no problem with putting in DDR benchmarks for the P4. But making a solid claim that something was done "biasedly" simply because they didn't bother with a slower-performing setup for one CPU is hardly justified.

Quote:
And as for no such thing as a bias in test results...tell that to Phillip Morris and all the tobacco companies which did all those tests to prove smoking tobacco was not addictive or a significant danger to your health. Many a scientist has been caught out trying to fiddle their result to prove one thing or another. Pharmaceutical companies try to do it all the time thats why governments have regulatory bodies. I was taught the truth is the truth, no matter how good, bad or indifferent your results stand. An ommission is a bias away from the whole truth.


Fiddle with their results. In other words the test was invalid. As I said, if the tests are done with care and accurately, there is no "unfairness" as long as information is not tampered with. Are you saying Toms or Anand's tampered with their results?

Quote:
Now i can understand where you are coming from, and i do agree with you about the endless bickering about 'mine is better', its boring. What i was hoping to do was open some eyes about what you see is not always whats out there. I hope you can understand where i'm coming from. I'm not disagreeing with anything you said its just that i don't think you understand what i'm saying.


My whole point is that the testing setup platforms are clearly indicated on the benchmarks. You can clearly see that it is an RDRAM setup with the P4 and that it performs better. And from the other benchmarks on the web, you can clearly see that a DDR setup wouldn't perform as well. No information is hidden and nothing is "pulled over your eye".

Quote:
And about the 'fanboy' thing, i accept you may not have referred it personally to me but from what i've read in some web forums its not used in a kind way whether to an individual or a group, it smacks of a patronizing attitude. I'd prefer reasoned argument rather than classification.


From the posts I've read, I made a personal judgement and decided that most were fanboys. Hence a generalization.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
August 27, 2002 8:37:12 PM

When your Montecito will be released? 20 years later?
August 27, 2002 8:44:18 PM

The only problem i can see with that DesperateDan is the constant changes in component costs. Thats why i'd like to see true clock speed vs performance be use tables, grouping similar clock speeds together. A futher step would be league tables of clock speed vs performance groupings. These could be constantly re-ranked according to test results and importantly current price.

Could this be the holy grail to PC consumers?

But seriously that would require hard work to assemble and dedication to maintain. What about it Mr Pabst?

BananaSkin
August 27, 2002 8:48:13 PM

How would that be helpful? All you're showing is what everyone already knows. One CPU performs better per clock than the other. Doesn't really provide any relevant information. Why the focus on the clockspeed of each CPU?

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
August 27, 2002 9:05:07 PM

But aren't you falling into your own trap about saying one CPU performs better per clock cycle than the other? Performance as we have already said depends on RAM as well.

Focusing on clock speed would cut the out the confusing PR
system AMD uses to rate its CPUs, and then give a more relative view of performance.

And a league table system would be helpful to show to the vast masses who don't know as much as you or me. How many friends, family and relations do you know that have asked you about computer advice because they are interested in buying a new system/component? I've had loads ask this or that and if i don't know i research. Hence its important to show test results for all common variations. I know lots of people who know little or nothing about computers who would find such a table useful information printed out with an explanation. Remember its not just for readers of THG but for readers to share with others who don't.

BTW when i mean fiddle you can also read into that to mean omit test results not just cheat. Like the tobacco companies not including addictive test and cancer results to their findings or dropping those results which did not fit their findings. This is not a slur on THG its just an example.

As for the setup data, thats my point it shows the P4 with DDR tests weren't done. Therefore no results, ergo IMO an incomplete series of test data.

Let me make this even clearer the benchmarks are ok, theres nothing wrong with them. All the tests done were fine, the tests results stand on their own merit but IMO lack certain relevance without the whole picture (P4 with DDR data). P4 with RDRAM is the undisupted performer on the current market.

Your right no one pulled the wool over my eyes, IMO i just saw something was missing which i think is relevant to the whole picture.

BananaSkin

<P ID="edit"><FONT SIZE=-1><EM>Edited by BananaSkin on 08/27/02 05:39 PM.</EM></FONT></P>
August 27, 2002 9:08:40 PM

WOW, I actually beat a record on this forum? BTW, Montecito is the successor to Madison, due out in <b>H1 2004</b>, supposed to be made on 0.13 process, and is estimated to have 1 billion transistors (800 million of them are the cache). As to heat and power, I think Intel will try to keep it to a mininum, even though it will be high.

Hey, Spitfire, I'd advise you to not make fanboy comments like that. Some members of this forum might get angry at those kinds of comments.

imgod2u, It seems to me that you just ignored what I said in my previous post. First of all, I'm an enthusiast, and former OCer. Second, I did NOT say that we should bicker about "mine is better" in the forums. Stop twisting around my sentences. You yourself said that the point of benchmarks is to show people what sort of performance a CPU offers in real world programs. I don't know about you, but it seems <b>logical</b> that you should show the maximum performance of each CPU, in other words, it would be stupid to bench a CPU at 50% of it's performance potential. And yes overclocked CPU benches are valid, but they *should* be shown in comparison with the stock CPU to see what you're going to get for overclocking. If a CPU is not benched at full potential, then you'll never know the REAL performance of the CPU, will you? And not benching a CPU at full performance poetential is <b>misleading</b>, and defeats the purpose of the bench, because you're looking at the performance of a CPU in a particular program running at X% of it's full performance . People then play guessing games to see what the REAL performance is of the CPU, not half, or a quarter. If you like misleading benches, fine. If you like unfair benches, fine. And benchmarks are a COMPARISON. If you see benches of a 2.9 athlon vs. a 1.6 p4 (stock), then that's a bad bench because you don't know what the performance of the Athlon (at stock speed) is to begin with. So, if the Athlon is not tested at stock, then people get the impression that the Athlon is better, hands down. Overclock benches are valid, as long as they have the same CPU at BOTH stock speed, and the overclocked speed.

- - - - -
Montecito - successor to Madison, and one monster of a CPU. <P ID="edit"><FONT SIZE=-1><EM>Edited by Dark_Archonis on 08/27/02 05:10 PM.</EM></FONT></P>
August 27, 2002 9:35:45 PM

Quote:
Most people on these forums these days thinks about only price/performance, but they don't seem to think too much about the quality of the product. In other words, do people really think whether they are getting a quality product for their money or not. No, most don't. And that's why so many people dismiss Intel CPU's nowadays

What is the defination of "Quality" to you?

Quote:
Montecito is the successor to Madison, due out in H1 2004, supposed to be made on 0.13 process, and is estimated to have 1 billion transistors (800 million of them are the cache). As to heat and power, I think Intel will try to keep it to a mininum, even though it will be high.

0.13 micron process and 1 billion transistors. We may see a kilowatt processor in H1 2004! :lol: 

Dont mind, it's a joke. I actually want to tell that you are over excited about Montecito. Calm down please!
August 27, 2002 10:29:50 PM

why would intel come back to micron .13 when they are slowly moving towards .09 micron? i believe montecito will be .09 micron dark, check that out again

real philosophy of life: "do onto others what you dont want them do onto you"<P ID="edit"><FONT SIZE=-1><EM>Edited by grassapa on 08/27/02 06:32 PM.</EM></FONT></P>
August 27, 2002 10:46:19 PM

Quote:
But aren't you falling into your own trap about saying one CPU performs better per clock cycle than the other? Performance as we have already said depends on RAM as well.


Excuse me? Of course one CPU performs better per clock than another. The difference can vary depending on memory bandwidth and the type of subsystem the CPU is provided with but up until very recently, the P4's per clock performance even with its best stock memory subsystem does not have as high an IPC as the Athlon with its best stock memory subsystem (this includes a 133MHz DDR FSB with 200MHz DDR memory, which is stock).

Quote:
Focusing on clock speed would cut the out the confusing PR
system AMD uses to rate its CPUs, and then give a more relative view of performance.


Relative <b>per clock</b> performance, which is a far cry from actual performance in applications. If one CPU has a better IPC but comes at best at 2.13 GHz and another can do less IPC but comes in at best 2.8 GHz, then of what value is per clock performance? What makes clockrate so important to be the standard to be judged by? Why must a CPU be judged by its MHz or how well it performs per MHz? Why is it relevant?

Quote:
And a league table system would be helpful to show to the vast masses who don't know as much as you or me. How many friends, family and relations do you know that have asked you about computer advice because they are interested in buying a new system/component? I've had loads ask this or that and if i don't know i research. Hence its important to show test results for all common variations. I know lots of people who know little or nothing about computers who would find such a table useful information printed out with an explanation. Remember its not just for readers of THG but for readers to share with others who don't.


They understand bar graphs quite well don't they? You put a 2.13 Athlon against a 2.2 P4 and show the bar graph of the Athlon as being faster, are they not going to understand that? You just gotta ask yourself, why isn't there a reason to compare a 2.13 Athlon with a 2.8 P4? They're both the top models of their respective family of CPU's and are currently the flagship CPU's. Why? The price. That's what it all comes down to. That's the resource that has to be conserved. Why is using less MHz to achieve performance so important? What is a MHz? Have you seen one? Do you get a box of it when you buy the CPU and have to conserve using it? Does it matter how much you use as long as you get the performance you want? On the other hand, the price is something to be conserved. And as price adjusts, so will comparisons. Nothing ever stays the same.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
August 27, 2002 11:16:25 PM

I don't really know what the point in clock speed v performance is, I think the cpus performance in a certain benchmark agaisnt the price of it would be much more usefull, this of course could be done for a variety of benchmarks as results for benchmarks vary.
August 28, 2002 12:30:39 AM

Quote:
imgod2u, It seems to me that you just ignored what I said in my previous post. First of all, I'm an enthusiast, and former OCer. Second, I did NOT say that we should bicker about "mine is better" in the forums. Stop twisting around my sentences. You yourself said that the point of benchmarks is to show people what sort of performance a CPU offers in real world programs.


No, I said the point is to show performance with <b>setups</b> with applications that apply to the people who buy it. Different setups will offer different performance at different values in different applications. Those are the variables. Benchmarks show what the result would be in applications that apply to the people reading and buying this stuff which combination of variables would work out to how well a performance. And the person can then decide whether said thing is worth the price.

Quote:
I don't know about you, but it seems logical that you should show the maximum performance of each CPU, in other words, it would be stupid to bench a CPU at 50% of it's performance potential.


If 75% of all users who buy the CPU are using such a setup that would, as you said, provide "50% of the potential performance" then the benchmark is indeed relevant. Because those people need to know either 1. that setup will not offer significant performance or 2. that setup will offer nearly as much performance (which is the real case) but at a lower price.
Again, like I said before, there is no "fair", only relevant. In this case, benchmarking a P4 setup with DDR memory would be just as relevant, if not more, than benchmarking a P4 with an RDRAM setup because more people who actually read these articles consider DDR setups over Rambus.

Quote:
And yes overclocked CPU benches are valid, but they *should* be shown in comparison with the stock CPU to see what you're going to get for overclocking.


Doesn't make it any less "fair". It just makes it irrelevant because you don't know what the P4 is capable of overclocking to with the same price spent on cooling. Because if you were going to buy that cooling setup anyway, it'd be better to know whether it would bring more performance to you on a P4 or Athlon. However, that was not the scope of the article. If and when an article would actually compare an Athlon overclocked to 1.9 GHz vs a P4 at stock 1.6 GHz, it would more than likely not be within the scope of comparing P4's with Athlons or even P4 setups with Athlon setups. It would more than likely be just using the P4 as a reference point.

Quote:
If a CPU is not benched at full potential, then you'll never know the REAL performance of the CPU, will you?


If 75% of the people who will actually buy the CPU (and they're the ones these things are suppose to inform) consider setup number 2 which doesn't bring the "full performance", then the scope of the article, in order to be most informative, should include setup number 2. It's not about fair or just, it's about being informative and having useful information.

Quote:
And not benching a CPU at full performance poetential is misleading, and defeats the purpose of the bench


The purpose of the bench is to tell what a certain combination would perform like using reference points. If an article compared a P4 on a DDR setup it will be trying to show what that particular setup is capable of. That's it, information. Whether that setup is worth it or not is up to the people who buy it to decide. Benchmarks are facts, nothing more. They're not there in and of themselves to convince anyone which is the smarter choice nor are they there to tell you what a particular CPU's full performance is. If that were the point of the article then the article should've been labeled "The full ability of the <insert CPU name here>".

Quote:
because you're looking at the performance of a CPU in a particular program running at X% of it's full performance . People then play guessing games to see what the REAL performance is of the CPU, not half, or a quarter.


You're still thinking of CPU's as independent things to play "this one's better" with. They're not. It's the system that matters because that's what causes overall performance. Tom in his reviews of CPU's decided to use RDRAM setups because he wanted to show P4/RDRAM setups compared to Athlon/DDR setups. Someone other reviewers may choose to compare other setups. They're all just information.

Quote:
If you like misleading benches, fine. If you like unfair benches, fine. And benchmarks are a COMPARISON. If you see benches of a 2.9 athlon vs. a 1.6 p4 (stock), then that's a bad bench because you don't know what the performance of the Athlon (at stock speed) is to begin with. So, if the Athlon is not tested at stock, then people get the impression that the Athlon is better, hands down. Overclock benches are valid, as long as they have the same CPU at BOTH stock speed, and the overclocked speed.


Fair? Why don't we start disabling everything on the CPU so that they're both CPU so that they all just run classic x86 code? Would it be relevant? No, because nobody does that when they buy their setups. Do people buy DDR setups with P4's? Yes. Therefore, it is relevant because it is informative for those people. Do people buy RDRAM setups with P4's? Yes. Therefore, it is relevant because it is informative to those people. These reviews are just what they are, reviews of that particular processor, they're not there so people can go "we need to compare the Athlon design to the P4 design". If that were the case, then they should mention other factors such as memory subsystems and show the benefits and weaknesses of each design. But I think you all know it by now.
If the scope of the review was "how high the Athlon can overclock" why do you always have to bitch about which Intel CPU they had in there? Asside from being a reference point, what possible other use could an Intel CPU be? The article was not intended to fuel the P4 vs Athlon bickering, it was intended to test the Athlon at its maximum. Why does everything have to be P4 vs Athlon?

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
!