Forgot your password?
typodupeerror
AMD The Almighty Buck Hardware Technology

AMD Hits Milestone in Server Market 215

Posted by CowboyNeal
from the moving-on-up dept.
DontClickHere writes "According to data from Mercury Research, AMD has finally cracked the 10% mark in x86 instruction set server CPUs. AMD's Chairman had hoped that their server sales would hit 10% at the end of 2004, but they had only reached 5.7%. Some of this gain can be attributed to AMD's introduction of dual core chips in April this year. With Intel only due to ship dual core chips for low end servers later this year, AMD has been handed a golden opportunity to take a larger share in the server market."
This discussion has been archived. No new comments can be posted.

AMD Hits Milestone in Server Market

Comments Filter:
  • Re:Obligatory. (Score:2, Informative)

    by m3rajk (670560) on Friday July 29, 2005 @08:04AM (#13193611)
    not suprising, AMD has been making products that have better quality than Intel since th eAthlon was first launched. The fact I work in a place where 75% are ex-DEC hardware engineers and found out that three fourths of the alpha processor people left before Intel got their hands on that, combined witht he fact half odf them went to AMD and the fact I know someintel people and know that the Xenon is the ONLY chip to have ANY alpha technology integrated says a lot. AMD has been making AMAZING strides in graphics ability. Alpha basically handled ALL Intel x86 and apple stuff better than both of those, was the frist TRUE 64 bit chip. even now intel Itanium is a massive change fromtheir standard, and wwas released AFTER the AMD 64 bit. it's only a matter of time until AMD starts seeing their marketshare increase as the general populace starts to realize that Intel HAS ALREADY become the dinosaur that DEC was.
  • Re:Obligatory. (Score:5, Informative)

    by orz (88387) on Friday July 29, 2005 @08:11AM (#13193646)
    even now intel Itanium is a massive change fromtheir standard, and wwas released AFTER the AMD 64 bit.

    The Intel Itanium was released before the Athlon 64. You're thinking of EM64T-enabled Pentium 4s and Xeons.

    But yeah, AMD got a lot of very good engineers from DEC.
  • Re:Laptops? (Score:5, Informative)

    by RockModeNick (617483) on Friday July 29, 2005 @08:12AM (#13193650)
    Pentium M's are good because they are primarily based on P3 technology, not P4 - the P4 architechture delivers signifcantly less bang per mhz, and thus far the increased top clocks of the P4's are not keeping ahead in actual performance the way they were expected to.
  • Re:Main Reason (Score:5, Informative)

    by Lonewolf666 (259450) on Friday July 29, 2005 @08:24AM (#13193711)
    Not anymore. Intel sells the Pentium D now, which is a dual core Pentium 4. The cheapest model (the Pentium D 820 with 2.8 GHz) is available at Alternate.de for 279 Euros.
    One might suspect Intel of dumping prices here, but it cannot be denied that this is an attractive offer.
  • Re:Main Reason (Score:3, Informative)

    by Erwos (553607) on Friday July 29, 2005 @08:29AM (#13193729)
    "The main reason is that they sell the only 64-bit consumer chip."

    This is just plain wrong. Intel's 6XX series of Pentium4's has the EMT64 (aka, AMD64) instructions as well. Both AMD and Intel are selling 64-bit CPUs now.

    -Erwos
  • Re:Main Reason (Score:5, Informative)

    by fshalor (133678) <fshalor@[ ]cast.net ['com' in gap]> on Friday July 29, 2005 @08:46AM (#13193809) Homepage Journal
    And HP.... hehe

    I'm actually an all AMD shop, except for a few workstations. The only intel machines in the institute are PIII 700's and 900's from before my time there and a set of 6 Dell Precicions 650's (running Debian.). (Which were also the fastest machines in the place 3 years ago when I started.)

    Servers are all AMD MP's with a few AMD opterons rouding out the bunch. Workstations are dual MP's. Desktops are mostly Duron's through XP's .

    Just bought a few 1u tyan machines. (amd opterons) and planning on building up a cluster in a few weeks with about 30 more.

    AMD has won on the campus scene at least.

    Oh, and the desktop machines in my house are all AMD except for a crappy compaq that my bro bought and an iBook g3.

    Kind of funny. Can't believe they only have 10% right now. But it happens I guess. ;)

    Best,
  • Re:Laptops? (Score:5, Informative)

    by wild_berry (448019) on Friday July 29, 2005 @09:19AM (#13194056) Journal
    I don't think you're up to date. The Athlon 64's have been trimmed down for notebook use, called Turion [amd.com], and they have two performance envelopes [amd.com], one at 35 watts and another at 25 watts typical power consumption. The present range is explained here [amd.com].
  • Re:Main Reason (Score:3, Informative)

    by Anarke_Incarnate (733529) on Friday July 29, 2005 @09:26AM (#13194114)
    EM64T is based on AMD64, EPIC is based on Itanium. Why do people still get this wrong? EPIC is not a backwards compatible instruction set.
    The difference is, intel's memory addressing on EM64T is weak by comparison (which has nothing to do with on die memory controllers)
  • Re:Laptops? (Score:3, Informative)

    by Nuffsaid (855987) on Friday July 29, 2005 @09:29AM (#13194143)
    I see a big marketing debacle by AMD, if even part of a supposedly technical oriented public like the Slashdot crowd never heard about the Turion 64 processor! Its power consumption goes as low as 27W, with performance slightly better than Pentium M on a clock-by-clock basis. You can already buy some laptops, and the reviews are good so far.
  • Re:Apple? (Score:3, Informative)

    by Slack3r78 (596506) on Friday July 29, 2005 @10:13AM (#13194497) Homepage
    Your information is out of date. Northwood was already getting to be hotter than AMD, but when Prescott was released (almost 2 years ago now), it blew the doors off everything else in the x86 market in terms of power consumption and heat disipation. Disipation's well over 100W and operating temperatures in the 70-80C range. AMD64 chips, in comparison, generally put out in the range of 35-50W and operate between 30-50C. Personally, I've never seen my A64 3200+ get any higher than around 37C under full load with the stock fan.

    In addition to this, AMD64 chips feature something called "Cool n Quiet." CnQ is basically a fancy name for intelligent dynamic clock scaling. Again, using my 3200+ as an example, when under full load, it runs at 2GHz @ 1.375V. However, when the computer's idling or under light loads (ie: most web browsing, word processing), the CPU drops down to 1GHz@1.0V. When the load's somewhere inbetween, the CPU scales up in 200MHz increments on the fly. It's actually kind of cool to watch happen in a clock speed/voltage monitor.

    So the short answer - AMD's been ahead of Intel in this regard (on the desktop) for quite some time. Prescott took it from AMD being a bit cooler to there being no comparison whatsoever. Hope that helps.

  • by krgallagher (743575) on Friday July 29, 2005 @11:02AM (#13194965) Homepage
    "AMD made very cheap, inferior processors for years."

    AMD made the first 100 MHZ 486 DX4 chips. This was at a time when the Pentium 75 was just entering the market and the 486 DX4 100 was both faster and cheaper. Throughout their history, AMD has always been able to deliver superior performance at slower clock speeds than Intel. They have also been cheaper to purchase. Whle I have always considered AMD CPU's to be economical, I also consider them to be superior chips to anything Intel produces.

  • by optikshell (786466) on Friday July 29, 2005 @11:29AM (#13195185) Homepage
    and I always push AMD processors first. Once I present the benefits of an AMD processor (be it the 64, Sempron, or Turion), the consumer usually goes with the AMD. Occasionally I get the brainwashed retard - "My cousin has a Dell with a Pentium 4, I want a Pentium 4" and can't convince them otherwise. Explaining the benefits of an AMD processor, and letting the customer know that I've been using AMD processors for quite some time usually sways them. AMD... FOR THE NEW REPUBLIC!
  • by dwkunkel (546825) on Friday July 29, 2005 @11:40AM (#13195314) Homepage
    Much to my surprise, Sun has become the preferred source for servers in our data center. The reason is the availablity of AMD Opteron servers [sun.com] from Sun. These are replacing Intel Zeon based servers from HP and IBM. They're running Linux, of course.
  • AMD's NUMA support. (Score:3, Informative)

    by Martin Marvinski (581860) on Friday July 29, 2005 @03:35PM (#13197601)
    Don't forget that AMD's x86-64 bit processors support NUMA (non-uniform memory architecture), where as the Intel EMT64 does not. So if you use an operating system, like Linux which has NUMA support built into the kernel, and Opteron chips and a quality motherboard that has seperate memory for each processor, each processor can have its own dedicated memory! If anyone has the link, there was a benchmark out there that really rocked, with NUMA enabled dual Proc Opteron Server kicking intel's offerings.
  • by larstr (695179) on Friday July 29, 2005 @07:05PM (#13199102)
    IBM has written a paper on this comparing two of their 1U 2 cpu servers, the e326 (AMD) and x336 (Xeon MP). The paper is named "Performance of Two-Way Opteron and Xeon Processor-Based Servers" from April 2005.

    A search on ibm.com does not give me a link to the document and neither does google. I did however find an IBM provided AMD vs XEON linpack-comparison benchmark ftp://ftp.software.ibm.com/eserver/benchmarks/wp_L inpack_072905.pdf [ibm.com], but first benchmark (that I can't locate) was better.

    It clearly shows the advantages of the AMDs NUMA architecture and also other factors.

    NUMA is also available on some enterprise level IBM XEON servers like the x440, x445 and x460 (or the equivelent systems from Fujitsu Siemens or NEC). One thing that is important on these servers is that you should balance each CEC with the same amount of memory or it will greatly affect performance. AMD's NUMA technology is not affected as much as XEON on this (as the mentioned paper shows).

There is no distinction between any AI program and some existent game.

Working...