Why AMD is right by doing XP#### Perf ratings.

This is the place where all heated debates shall reside. Non-tech topics allowed. Personal attacks will not be tolerated. "Enter at your own Risk".

Moderator: The Mod Squad

Why AMD is right by doing XP#### Perf ratings.

Postby Peanya » Thu Apr 17, 2003 9:19 pm

All the time, I hear people whining about senseless things - "if you take the heatsink off of an older AMD, you'll fry it!!! Not with Intel!!" and "Intel honestly reports their speeds whereas AMD doesn't"
Well, wake up! AMD DOES report their speeds! Everytime you go look up ANY site selling an XP processor, they ALWAYS post the true clock speeds. So why the performance rating? Why do we complain about that too? Be patient while I tell you my story...
About 6 years ago, I was selling audio equipment at an electronics retailer. Something I learned about amplifiers was that you cannot judge how powerful it was by looking at the wattage rating of it. There were some Harmon Kardon amplifiers rated at 40 watts per channel which would play music louder, and clearer than a Pioneer rated at 100 watts per channel. Why? Wattage of Amplifiers is measured on the same scale, which is to determine the power an amplifier has. Let's look into this a little more: 100W per channel RMS at 1KHz into 8 ohms. vs 40W per channel RMS at 1KHz into 8 ohms. That's just a tiny bit of information actually! The full audible spectrum of sound is considered to be 20Hz to 20,000Hz What about the power output of the rest of those frequencies? Also take into consideration that at varying frequencies, the impedance of each speaker changes. It is a proven fact that an average "8 ohm" speaker can draw up to 15 amps of current!! WOW! Now if you ask any electrical engineer if you can run that much current thru an 8 ohm load, they'll tell you that can't happen, and you'll not even get 1 amp out of it. Poor, niave engineer! He's forgetting the fact that the speaker is moving within a magnetic field and producing its own current and current draw.
As it turns out, the pioneer can put out about 5 amps of power, whereas the Harmon Kardon can put over 35 amps! The Harmon Kardon is rated at 40 watts, so maybe they should say XP200W+! And the Pioneer, if we were going to be fair, should be rated at XP15W+!
It boils down to this: The XP simply gets more work done per cycle overall than the P4. So AMD accounts for this, cause there is no standardized rating system for work output of CPU's. Maybe Intel should do a performance rating system too! P42200+ 3.0GHz Wouldn't sound too pretty in all their marketing pazaaz and hype. And don't get me started on their chipset hype fiasco!
Peanya
Black Belt 5th Degree
Black Belt 5th Degree
 
Posts: 6055
Joined: Fri Feb 07, 2003 7:43 pm

Postby Valour » Thu Apr 17, 2003 9:26 pm

Ignoring the errors in your logic, I'll say this: the XP rating system is misleading. It doesn't always accurately portray how much work is done compared with a P4's speed rating. When the XP3000+ came out it became apparent that the performance ratings were a little weird, being that it didn't offer a consistently incremental performance increase over the XP2800+.

Do you even know how the AMD performance rating is calculated?

-Jem
"The proverb says that Providence protects children and idiots. This is really true. I know because I have tested it." -Mark Twain

I wrote this book: www.herotale.com
And here's the next one: www.emeraldblackbird.com
Valour
Black Belt 2nd Degree
Black Belt 2nd Degree
 
Posts: 2893
Joined: Sat Oct 05, 2002 6:24 pm
Location: Rochester, NY USA

Postby Peanya » Thu Apr 17, 2003 9:54 pm

I see no fault in the logic... Can you show me definitive results where a 3.06 P4 clearly outperforms an XP3000 in EVERY test?
Peanya
Black Belt 5th Degree
Black Belt 5th Degree
 
Posts: 6055
Joined: Fri Feb 07, 2003 7:43 pm

Postby evasive » Fri Apr 18, 2003 2:19 am

Not quite. Can you show me one where it is the other way around? Not likely. Give the gamers their AMD, let us bussiness people use the P4 and be done with it. Does not mounting the heatsink properly count as running without one? If so, do this test with both a P4 and any AMD Athlon.
We hate rut, but we fear change.
********************************
System error, strike any user to continue...
evasive
Mobo-fu Master
Mobo-fu Master
 
Posts: 37389
Joined: Sun May 06, 2001 12:01 am
Location: Netherlands

Postby Valour » Fri Apr 18, 2003 10:27 am

Peanya wrote:I see no fault in the logic... Can you show me definitive results where a 3.06 P4 clearly outperforms an XP3000 in EVERY test?


That has nothing to do with it; I'm not comparing CPU performance at all. Rather I'm taking issue with your argument. You have no idea what the XP rating means or how it's calculated, yet you're extolling its benefits. I agree, performance ratings are a good idea, but AMD's ratings are voodoolike and mysterious. AMD talks about "more work per cycle" but what work are we talking about? Under what conditions, using what programs? What does this mean? How is this measured? Can you answer any of these questions, or were you just blindly supporting "your team?" :?

-Jem
"The proverb says that Providence protects children and idiots. This is really true. I know because I have tested it." -Mark Twain

I wrote this book: www.herotale.com
And here's the next one: www.emeraldblackbird.com
Valour
Black Belt 2nd Degree
Black Belt 2nd Degree
 
Posts: 2893
Joined: Sat Oct 05, 2002 6:24 pm
Location: Rochester, NY USA

Postby Peanya » Fri Apr 18, 2003 12:03 pm

Amd gets their performance ratings by their "instrustions per cycle". While you assume I don't know how they come about it, you overlook a few things - I'm not trying to get into a technical argument here, but merely giving a simpler analogy to the whiners about how the rating system is "unfair" and a P4 showing its true operating frequency is all high and mighty. Yes, their multiplier to get their performance rating system DOES change due to FSB and cache, which in turn affects their multiplier. This does go against what they said about thier IPC, however, they are still at least giving a fair rating in comparison to a P4.
I'm not saying that the AMD is clearly superior to the Intel, nor Intel to AMD. I'm trying to point out the fact that they both perform similar, and that too many people try to run down AMD for this, when they are merely trying to combat a company with a virtually unlimited advertising budget in comparison to their budgets in this area.
BOTH systems are good. Personally, I can go out, and get an XP3000 barton for 320, or a comparable performing P4 3.06GHz for 440. I'd keep the money in my pocket. Most new buyers see the "big name" and think that it's better, and then hear people run down AMD, which misleads them into thinking that AMD isn't as good, or they're trying to rip them off. This is unfair in my opinionn :)
Peanya
Black Belt 5th Degree
Black Belt 5th Degree
 
Posts: 6055
Joined: Fri Feb 07, 2003 7:43 pm

Postby Tulatin » Mon Apr 21, 2003 10:08 pm

The purpose of the Performance ratings on the Athalon Xp is to show a relative performance for the chip. Just like an Athalon XP 1700+ Performs Like a 1.7 Ghz P4. This is like before, when clock speeds were equal - I can remeber the Max PC duel of The Athalon 600 and the P3 600 - both efficient architectures - but they triumph in different areas...

The only reason that the P4 is so fast is that Intel chose the brute force aproach (NVida does this too...) more mhz=better performance. By doing this, they sacrificed features (floating point units...) and majorly lost eficency. It does seem weird that a chip clocked at 2.25 Ghz (3000+) can compete with a chip Clocked at 3.06 Ghz (No PR Here). What seems odder is the SiSoft Sandra PR feature - my 2 Ghz chip gets 2200+ when oc'd , 1900+ when not. Maybe AMD has decided to use this to name their chips :twisted:
Tulatin
Enlightened Master
Enlightened Master
 
Posts: 14664
Joined: Wed Dec 11, 2002 2:29 pm
Location: Canada

Postby hypnotik » Tue Apr 29, 2003 3:32 am

Actually AMD claim the PR system rates the performance compared to an Athlon Thunderbird hence a XP1800 running at 1.53Ghz runs at the relative speed of a T-bird @ 1.8Ghz.
"I disapprove of what you say, but I will defend to the death your right to say it"

-Voltaire
hypnotik
Black Belt
Black Belt
 
Posts: 584
Joined: Tue Oct 22, 2002 9:07 pm
Location: Melbourne, Australia

Postby tommie j » Tue Apr 29, 2003 9:14 pm

i found a great benchmark test pitting the amd 3000 on an nforce board against the p43.ghz on the canterwood 800 board


www.ocworkbench.com/2003/others/barton4 ... /page1.htm
I bailed, moved to puerto rico ,lifes too short!
tommie j
Black Belt 1st Degree
Black Belt 1st Degree
 
Posts: 1035
Joined: Wed Feb 19, 2003 1:36 am
Location: cape cod,to puerto rico

Postby trophyhunter_1 » Wed May 14, 2003 12:13 am

I favor AMD CPU's but will agree the latest bartons are not scaling with the ratings, but on the other hand a 3.0 P4 does not give a 50% increase in performance over a 2.0 P4.

And how come a 2.0 northwood core mops the floor with a 2.0 willamette core? Intel pushes mhz which is not a true indication of performance either. Also, it depends on whose benchmarks you look at, BAPCO and other mainstream benchmarks are said to be optimized for SSE2 which puts the AMD chips in a hole. But how many software package out there are using SSE2 optimization?

If you look at PC World's Worldbench, they will tell you AMD is the definate leader. I think what it all boils down to is both companies are producing high performance chips that will meet or exceed the demands of 99% of the users.
Last edited by trophyhunter_1 on Tue May 20, 2003 1:27 pm, edited 1 time in total.
trophyhunter_1
Anti-Static Strap
Anti-Static Strap
 
Posts: 489
Joined: Wed Apr 16, 2003 3:33 pm
Location: Ohio

Next

Return to The Hundred Year War

Who is online

Users browsing this forum: No registered users and 1 guest