Options

Opteron delivers workstation performance leadership

edited November 2003 in Science & Tech
<b>Xeon counterparts left in Opteron's wake</b>

[blockquote]In the November issue of Cadalyst Magazine, nine dual powered workstations were reviewed. Two of those were AMD Opteron-based, one was AMD Athlon MP powered, and the remaining six were Intel Xeon configured. For its roundup requirements, Cadalyst asked each vendor to send in a representative system that was equipped with a 3D OpenGL graphics card. So that wasn't a situation, which has been seen in so many reviews before, where a consumer level graphics card had been used instead. Cadalyst's review tests were run on systems that were properly configured for workstation use.

Cadalyst used various benchmarks for its workstation tests. For the AutoCAD tests, it ran its Cadalyst C2001 benchmark test using AutoCAD 2004 . The proe-01 (Pro Engineer) suite from SPEC ViewPerf v7 was used as well. MaxBench 4, a 3D Studio Max benchmark, was the other test suite used. All machines were configured with Windows XP Professional.

"Xeon is hemorrhaging performance because its northbridge-based architecture is out of date. Opteron's ability to scale has made today's Xeon yesterday's technology." Today's up to two-way Xeon platforms are limited to using DDR266 memory. The up to four-way Xeon MP is hobbled with the slower 200 MHz RAM. When DDR400 Opteron support is officially confirmed, the Xeon platform will be up to three memory speed grades behind. I've said this before, but I'll say it again:

"If Intel's 90 nm Netburst products don't pack anything more substantive than more cache and higher frequency, then from where I'm sitting, Opteron's server performance lead could be maintained for at least the next year and through the 90 nm generation." AMD64 technology has truly been combined into a carrier-grade architecture that fully projects its credentials. Intel's disparate technologies perform well in isolation, but when they're brought together as a whole, the total sum doesn't hold a candle to what AMD64 can and will deliver. [/blockquote]Source: <a href="http://inquirerinside.com/?article=12428&quot; target="_new">The Inquirer</a>


Comments

  • ShortyShorty Manchester, UK Icrontian
    edited October 2003
    Telling us what we have been saying since the first announcement of the hammer line of processors ;)

    :D
  • ThraxThrax 🐌 Austin, TX Icrontian
    edited October 2003
    "If Intel's 90 nm Netburst products don't pack anything more substantive than more cache and higher frequency, then from where I'm sitting, Opteron's server performance lead could be maintained for at least the next year and through the 90 nm generation." AMD64 technology has truly been combined into a carrier-grade architecture that fully projects its credentials. Intel's disparate technologies perform well in isolation, but when they're brought together as a whole, the total sum doesn't hold a candle to what AMD64 can and will deliver.

    And guess what Intel told Shorty at the UK Gamer's expo he attended? Hyperthreading II, higher frequencies, and more cache!

    Guess what the P4EE is doing? Higher frequencies and more cache!

    Guess what the Prescott is doing? High frequencies, Hyperthreading II, and more cache!

    Let us all give three cheers for AMD and the introduction of the product which finally shows Intel chips for what they are:

    Poorly engineered products crafted with the intent of fooling consumers and banking upon customer loyalty as the sole methods for longevity.

    Let us also give three cheers for AMD for poking the proverbial hole in Intel's business ideology (Bigger, not really any better, only marginally faster), the fundamentally-flawed principles of their operation which has defied nature of business: That is they've been allowed to produce an objectively inferior product with resounding success.

    <b>No longer.</b>

    AMD has delivered the revolutions: The revolution of a 1GHz clock-deficit arriving to 95% success. The revolution of 99% memory efficiency. The revolution of low-latency memory access. The revolution of high-speed/low-heat processor architecture which lays utter waste upon the competition.

    And they have <b>delivered</b> the revolution of a low entry price. AMD's newest S940 alone is serving the needs of both enterprise and consumer needs with a highly-scalable, and blistering fast architecture that leaves Intel scrambling to cut prices, rush products, and kick the propaganda higher in lieu of any real and viable product to combat the onslaught of the Almighty HAMMER.
  • BlackHawkBlackHawk Bible music connoisseur There's no place like 127.0.0.1 Icrontian
    edited October 2003
    But everybody else doesn't know it. What's really hurting AMD is the lack of advertising.
  • ketoketo Occupied. Or is it preoccupied? Icrontian
    edited October 2003
    The only thing I question (question, mind you, not objecting in any way, just haven't seen anything on the subject) is "highly scalable" in the last paragraph of Thrax's reply. Can anyone point me to something that bears this out?
  • ThraxThrax 🐌 Austin, TX Icrontian
    edited October 2003
    It is highly scalable in four ways:

    1. It's the most consistently linear scaling SMP platform ever. The performance increase from adding additional opterons to the equation presents a clear line upwards in terms of performance benefits. Whereas the Xeon platform tends to vary up and down, the Opteron platform presents a relatively straight performance curve upwards. Simply put, adding more chips to SMP with the Opteron has tangible performance increases that parallel the amount of chips being added.

    2. The clock speed scaling on the Opteron has a very direct relation with performance returns. A 10% speed increase on the Opteron warrants a 9 to 11% increase in performance in all applications.

    3. The architecture of the Opteron and its platform is such that it has many upgrade paths available to it without changing the core functions of the processor. DDRII, QBM possibilities should that ever pan out, additional cache, DDR400, faster Hypertransport, etcetera. All of these could easily lend to significant speed increases in the platform, without major changes to the chip itself.

    4. The K8 is not "All-new." It is similar in many regards to the Barton Athlon. The Barton scales to about 2.2GHz on air cooling, with 10 stages in the pipeline. We can see that it maxes out at around 220Mhz per stage, roughly. We can assume that AMD has done a better job of making processors scale this time around, so let us call it 250MHz per stage once they iron out the teething problems all processors experience. Let us not ALSO rule out the introduction of the Paris-cored Opteron (.09) by the end of next year. It is conceivable that AMD has a full 3.0-3.3GHz to scale up throughout early 2005.
  • ketoketo Occupied. Or is it preoccupied? Icrontian
    edited October 2003
    It's the last point (specifically, MHz) I was most interested in, tho all make sense. No, I'm not a slave to clock speed vs efficiency but let's face it, it's a very relevant and/or interesting topic. Thx Thrax.
  • LeonardoLeonardo Wake up and smell the glaciers Eagle River, Alaska Icrontian
    edited November 2003
    As Black Hawk perceptively stated - this all wonderful, but rather irrelevant if AMD will be hapless as usual with their marketing. I have no doubt about the veracity of the article in Cadalyst magazine; but frankly, how many people view that magazine? 10 or 12?
  • ThraxThrax 🐌 Austin, TX Icrontian
    edited November 2003
    They've already made inroads with all the correct markets for their Opteron:

    Cray
    IBM
    Oracle
    Most linux distributors
    Major OEMs
    NEC corp
    Fujitsu
    and;

    Sun

    That lends incredible credence to the strength of the platform, and is very much advertisement enough. As an enterprise platform, you'll notice both the Xeon and the Itanium receive little to no marketing in high-volume media. Opteron is the same way.
Sign In or Register to comment.