Intel disagrees with AMD's PR Rating

bront · Mar 4, 2002

I would have called AMD's quest with the PR honest IF they did not also INSIST on the coverup of the standard clockspeeds on MBs and in the OS. Also the confusing paragraph on their own website claims it is not a P4 compairison and then compairs it to the P4.

There is nothing wrong with a rating system to show people which CPU performs better in your own line (Hoover does it with their vacuums, some clean rating). AMD does have an independant company do the testing I believe, so they are not stilting their own numbers simply to look good, they are backing them up by having someone else stick the rating on.

I think this agrees with what you are saying Crashman. You aren't saying that the simple idea of the rating system is in itself not bad, but AMDs implimentation of it is not completely honest (therefore, by definition, dishonest).

This is a non-smoking forum.
If your computer is smoking, please extinguish it immediately.

rickd59 · Mar 4, 2002

It's a nice suggestion, but I don't think the EU can exercise any more influence in these maters than the U.S government. The EU basically creates a single economic unit (helping to stabilize trade, currency, and economic power) with as much power as North America (basically the U.S.). There has to be more important things for governments to worry about....in all honesty, we all know that we can't trust companies...take everything with a grain of salt and try to independently confirm information in more than 1 place. Furthermore, I would think EU reps are just a susceptible to powerful lobbies as are U.S. reps.....Plus ça change, plus c'est la même chose.

-Rick

Crashman · Mar 4, 2002

Yes, I didn't complain when "PR200 at 166MHz" messages were shown in BIOS a few years back.

What's the frequency, Kenneth?

Oni · Mar 4, 2002

(and possibly later models sporting as much as 1 gig of L2)

I know you meant 1 meg of L2, but just imagine a gig of L2 cache!
Holy cow play games off of it, or run the OS off it and that would lead to some fast shiot. It'd probably eliminate ram altogether.

"Why can't I be the man? I mean, I DO have harmony balls..." -epoth

bront · Mar 4, 2002

There is no straight performance formula, as different factors effect different processors differently.

Generaly, as multipliers increase, there is less of an increase in performance on the same FSB speed. The northwood 133 is a good example of that. According to your simple formula, the 100 FSB 2.4 should be the same as the 133 FSB 2.4 P4.

Product ratings will never truely be simple, because processors all tend to do better at some things and worse in others, and also scale very differently.

Also, with the variance in other hardware, it's possable for a variance of a CPU level or two with different hardware (HDs, MBs, Mem, ect), so performance is no longer simply a CPU matter.

This is a non-smoking forum.
If your computer is smoking, please extinguish it immediately.

eden · Mar 4, 2002

The good thing about Athlons is they are levelled in any app. No matter what, it always gives out normal or better performance. The P4 will have "holes" where it will tumble at, which makes the decision of an all-around CPU quite annoying at Intel P4s.
It'll remain that way at 5GHZ, until they decide to give back the P4's units, it will never be the contender it was meant, and it will always have holes.

--
For the first time, Hookers are hooked on Phonics!!

mr_gobbledegook · Mar 4, 2002

The fact is this is the first time I have ever seen Intel actively target and publish documents about a single manufacturer/model of CPU and disputing the claims that the opposing company has made. It is obvious that the PDF was aimed at salesmen in PC retail outlets. The question is why has Intel published such a document ? I never saw such a document when Cyrix adopted their rating system.

Reasons for this document could be:

1 - Inform stupid salesman who don't know the difference between PR rating and Mhz. (Which would be AMD's responsibility).

2 - More confused/curious customers were asking Intel and salesmen what the difference between PR and Mhz was. Therefore Intel decided to step in and explain what the PR modelling system means to them. (Which indicates more customers are actively comparing Pentium PC's to Athlon PC's).

3 -Intel wishes to re-enforce their branding and processor performance(due to insecurites of losing market share to AMD) so the salesman in turn pass this message directly/indirectly onto the consumer.

I agree that both companies main objective is to sell as many processors as possible and may stretch the truth in order to do this. Each company choose to point out their strengths but however fail to disclose their weak points....I suppose that's marketing for you.

It is obvious that each company is strategically trying to get their ideas of what 'performance' means across to the average Joe on the street.....it's a bit like recruiting messengers so that 'the word' is passed around. This means that it has become increasingly more important for the consumer to comprehensively research the product they wish to purchase...i.e how much money they are willing to spend and how fast the processor executes the applications they wish to run from independent sources.

One thing I would add is that brand recognition is a very powerful tool...Intel are famous and respected therefore consumers are more likely to believe what Intel have to say. AMD does not have such a huge recognition which is why they had to get Arthur Anderson to independently verify their PR rating system.

So there you have it AMD's PR rating which is independently backed up by Arthur Anderson or Intel's own statement claiming Mhz is the only viable metric.....Ultimately the customer/salesman decides who to buy from. (Of course the saleman will support the company which provides the most incentive cause they are just in it for the money).

I leave this debate open....

~* K6-2 @ 333MHz *~
I don't need a 'Gigahertz' chip to surf the web just yet ;-)

Intel_inside · Mar 4, 2002

but is it still needed?

I think the pr rating personally was just an attempt to look good next to intel in the numbers game, performance aside. Not an extemely evil marketing ploy compared to some of the things intel or microsoft have done.

My life wasn't complete untill I tried sse-2 optimized pongEdited by Intel_Inside on 03/04/02 04:49 PM.

bront · Mar 4, 2002

I think the pr rating personally was just an attempt to look good next to intel in the numbers game, performance aside. Not an extemely evil marketing ploy compared to some of the things intel or mircorsoft have done.

So, you're actualy saying that Intel has done somethings wrong, and supporting AMD?

Who are you and please don't tell us what you did with the real Intel_Inside

This is a non-smoking forum.
If your computer is smoking, please extinguish it immediately.

mr_gobbledegook · Mar 4, 2002

Once you change something you cant change it back...willy nilly just like that. It would totally confuse everybody even more. AMD invented this PR system so they are basically gonna have to stick to it whether they like it or not.

~* K6-2 @ 333MHz *~
I don't need a 'Gigahertz' chip to surf the web just yet ;-)

Matisaro · Mar 4, 2002

Read the top of page 8 girish.

The normalization you quote was not a basis for the results, as seen in the graph on page 9. The Pr is rated based on the tbird as stated in the first paragraph on page 8, and every other source cited here.

I wish I could figure out how to cut and paste from adobe acrobat!!!

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!

slvr_phoenix · Mar 4, 2002

Geeze. Go away for months and come back and people are still debating AMD vs. Intel. And as usual, whenever AMD does something quasi-honest, it's fine. Yet whenever Intel does something quasi-honest, it's a crime against humanity. Because it isn't like in the end AMD's best chip and Intel's best chip aren't about always neck-and-neck performance wise. And it isn't like either company has ever been looking out for the consumer at the cost of their own profit.

Face it. Both companies are money-grubbing egocentric self-inflated blowhards. It all comes down to that they both produce nearly-equal chips (whether you look at more power per clock-cycle or more clock-cycles per power). And it's up to the consumer to listen to more than just the advertising if they want to make an educated decision.

Neither company has better ethics than the other, nor does either company produce a clearly superior CPU.

<pre>//error-proof coding
void main(){return;}</pre>

Intel_inside · Mar 4, 2002

I am not an amd fan. However some of the things that intel has done, such as the rambus fiasco, were slightly annoying (although rdram is worth it now on the p4/northwood platform)

My life wasn't complete untill I tried sse-2 optimized pong

Matisaro · Mar 4, 2002

I would have called AMD's quest with the PR honest IF they did not also INSIST on the coverup of the standard clockspeeds on MBs and in the OS.

I have an axp, and my motherboard shows me mhz, and so does my os. What insisiting are you talking about?

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!

bront · Mar 4, 2002

Read the origional XP article from THG. Any new motherboard to be approved by AMD had to not show the MHZ anywhere in the POST screen or in the OS. They may have changed that stance since that review though.

This is a non-smoking forum.
If your computer is smoking, please extinguish it immediately.

Raystonn · Mar 4, 2002

Pst. According to the ANSI and ISO specifications, the main function returns an integer. So there is still an error in your sig!

-Raystonn

= The views stated herein are my personal views, and not necessarily the views of my employer. =

Matisaro · Mar 4, 2002

They must have changed it very quickly, because I have never seen a motherboard which does that.

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!

bront · Mar 4, 2002

Girish, on Page 8, they are talking about the P4 1.5 normalization for particular testing that they did in the benchmarks. They also ran a 1.6, 1.7, 1.8 P4 against the 1.5 to show scalability, and that was only for results they had on page 9, not the overall PR Rating. The PR Rating is still a T-bird rating.

This is a non-smoking forum.
If your computer is smoking, please extinguish it immediately.

bront · Mar 4, 2002

Girish, on Page 8, they are talking about the P4 1.5 normalization for particular testing that they did in the benchmarks. They also ran a 1.6, 1.7, 1.8 P4 against the 1.5 to show scalability, and that was only for results they had on page 9, not the overall PR Rating. The PR Rating is still a T-bird rating.

This is a non-smoking forum.
If your computer is smoking, please extinguish it immediately.

AMD_Man · Mar 4, 2002

Pst. According to the ANSI and ISO specifications, the main function returns an integer. So there is still an error in your sig!

lol, maybe that's the point?

AMD technology + Intel technology = Intel/AMD Pentathlon IV; the ULTIMATE PC processor

Raystonn · Mar 4, 2002

I believe he was saying that the only error-free program is one that does nothing. But his program that did nothing also had an error according to the standard. Ironic, is it not?

-Raystonn

= The views stated herein are my personal views, and not necessarily the views of my employer. =

bront · Mar 4, 2002

Just make sure they keep the Athlon's out of the paper shreder.

This is a non-smoking forum.
If your computer is smoking, please extinguish it immediately.

lhgpoobaa · Mar 5, 2002

wow
im reading intelligent and quazi-intelligent posts in the CPU forum....
something must be wrong.

I love helping people in Toms Forums... It reinforces my intellectual superiority!

ZER0 · Mar 5, 2002

<<<If Intel hadn't completely gutted the P4 from the original spec, just to get high GHz speeds, this wouldn't even be an issue.>>>

link

<<<Intel is the company that is trying to confuse the consumer - they're like a car manu selling a 3-cylinder engine that can do 22000 RPM and saying that it will out-perform a V6 that does 16000 RPM. Oh yeah, and they have a special "turbo" called SSE2 that requires a special fuel adative that you can buy for an additional $1000 down at the dealership - if you can get the dealership to even stock the stuff.>>>

they sell their chips at speed rated in mhz just like they've done for over a decade.

<<<AMD is just trying to come up with a decent comparison; Intel is the one refusing to discuss true comparative metrics like MPH/KPH, Horsepower, Torque, etc.>>>

then why with each athlon release does the pr rating increase at a more rapid rate?

<<<Imagine if monitor manus measured their monitors from front to back instead of diagonally across the face - then my 21" CRT would be a 22" Monitor.... Then when companies tried to introduce 18" diagonal LCDs which measure 3" front to back they would be a laughing stock - they would have to come up with a comparative measurement. It is nearly the same thing with the P4 and AthlonXP procs.>>>

how do u figure??? intel is trying to compare mhz to mhz which is a direct comparison. i think the above point goes against you.

<<>>

my advice... keep thinking.

eden · Mar 5, 2002

My advice:

SHUT UP.

--
For the first time, Hookers are hooked on Phonics!!

Intel disagrees with AMD's PR Rating

Distinguished

Distinguished

Polypheme

Distinguished

Distinguished

Champion

Distinguished

Distinguished

Distinguished

Distinguished

Splendid

Splendid

Distinguished

Splendid

Distinguished

Distinguished

Splendid

Distinguished

Distinguished

Splendid

Distinguished

Distinguished

Illustrious

Distinguished

Champion

Share this page