Slashdot Mirror


AMD Launches Athlon 64 FX-57

Kez writes "Today AMD release what could be the fastest x86 processor to date. The FX-57 is the first 90nm Athlon 64 FX from AMD, clocked to 2.8GHz, with 1MiB of L2 cache and support for SSE3. The memory controller has also been tweaked to support mismatched memory module sizes - something some enthusiasts have been crying out for. Hexus.net reviews the new processor, which, in gaming benchmarks, walks all over any of Intel's offerings." There's going to be plenty of reviews I'm sure - if you've found other links, please post them below.

259 comments

  1. AMDZone by luna69 · · Score: 5, Informative
    --
    No gods, no demons, and no masters. Secular Humanism!
    1. Re:AMDZone by James_Duncan8181 · · Score: 1
      Those numbers really just reenforce why there is now a long term move towards AMD - the most important one there being the performance per watt.

      How is it that AMDs highest performing chip is still not as bad heat wise as most of the P4 family? Intel has fallen far...

      --
      "To any truly impartial person, it would be obvious that I am right."
    2. Re:AMDZone by sigaar · · Score: 1

      What bugs me is that so many people I spoke to about the possibility of going with AMD instead of intel, still say "Oh but AMDs run too way hot." Yeah, as opposed too?

      Even my Thoroughbred B core AthlonXP 2400+ in a cheapo case with only one 8cm fan aside from the stock cooler that came with the CPU, harly ever passes 40 degrees celcius under load...

      Speaking of cool CPUs, does anyone know of a benchmark of those dual VIA bords? I think they came with something like 1ghz C3 CPUs or some such - would be very interesting....

      --
      sigaar
    3. Re:AMDZone by arazor · · Score: 1

      In my case stock AMDs fans have more noisier granted this was several years ago. Every stock AMD fan at the time was way noisier then stock Intel fans. My guess would be there that is some sort of connection that users such as myself in thinking or perceiving that more noise equals more heat.

    4. Re:AMDZone by Ed+Avis · · Score: 1

      My Athlon XP2100+ (1733MHz, 133MHz FSB times 13) started crapping out in recent hot weather - not that hot, only 30 degrees Celsius or so - and I had to underclock it to 700MHz (100MHz FSB times 7) to stop things randomly dying. Oh, and take the lid off my PC. I am wondering whether there is any Socket A Athlon that can run safely with only a passive heatsink and no case fan, and whether my BIOS would allow me to clock it slowly enough.

      --
      -- Ed Avis ed@membled.com
    5. Re:AMDZone by sigaar · · Score: 1

      "granted this was several years ago"

      I rest my case ;-)

      --
      sigaar
  2. Toms Hardware Review... by drumgeek · · Score: 5, Informative
    1. Re:Toms Hardware Review... by Anonymous Coward · · Score: 5, Funny

      Although Toms has
      [Next Page]
      decent reviews, there
      [Next Page]
      are times when the reviews
      [Next Page]
      are overshadowed by the
      [Next Page]
      advertising. Not really swayed
      [Next Page]
      by them but I still question it

    2. Re:Toms Hardware Review... by sweetnjguy29 · · Score: 2, Informative

      On a serious note, if those g-d damn advertising pop-ups that masquerade as links piss you off, download the firefox web browser (if you haven't already), and install Greasemoney. Then navigate to http://dunck.us/collab/GreaseMonkeyUserScripts and do a quick search for tomshardware which will lead you to a script called Dumbtext. Then right click and install. Make sure greasemonkey is active and reload tomshardware or any other page that uses intellilinks. Tada...the ads are gone!

      Now only if there was a script to get rid of those damn Next Page links and view the entire article on one sight...fusker, perhaps?

  3. Why would one get this by Quick+Sick+Nick · · Score: 1, Insightful

    Why would I buy this instead of a dual core 4800+? Two is more than one! Even if the two cores are lower clock speeds.

    1. Re:Why would one get this by Sique · · Score: 4, Insightful

      Because this processor is mainly aimed at the gaming community, and games often aren't multithreaded, so the second core would just be idle and convert electricity into heat.

      Interestingly though once there was only one Athlon 64FX, and if a new one came out, its predecessor would then, if remaining on sale, be renamed. This time the old Athlon 64FX (55) remains on sale with the 64FX name.

      --
      .sig: Sique *sigh*
    2. Re:Why would one get this by orz · · Score: 1

      I suspect that many users don't get a lot of benefit from the second core. Which benefits you more depends upon your workload.

    3. Re:Why would one get this by LarsG · · Score: 1

      Two is more than one! Even if the two cores are lower clock speeds.

      Singlethreaded performance.

      --
      If J.K.R wrote Windows: Puteulanus fenestra mortalis!
    4. Re:Why would one get this by AlgebraicRing · · Score: 2, Insightful

      In a windows environment, even though your game might be a single thread, are there not also other programs running in the background? Wouldn't those other threads suck up some of the processor cycles?

      I would think that having a dual core set up, the game can hog one processor while the rest of the OS and other threads can hog the other processor. When I play games, I leave my chat clients open and there's all the crap running in the system tray. All of that can be running on one processor, while the game could run on its own processor.

      Are are you suggesting that its the added functionality of the processor and not how much processing time is available that makes the real difference?

    5. Re:Why would one get this by should_be_linear · · Score: 1

      Dual Core is advertised a lot but keep in mind that it is always better to have 1 core of performance X then 2 cores of performance X/2. For multi threaded apps both machines are same, performance-wise, but for single threaded tasks (still majority) it is big difference. Interesting approach is that of Cell: low level libraries are made to use parallel SPE units. High level apps (like games) are written in normal, single threaded way, but on the CPU-level runs in highly parallel way.

      --
      839*929
    6. Re:Why would one get this by Sique · · Score: 2, Informative

      Programs in the system tray are mainly memory hogs, not processor time consuming. After all those just sit there and wait for something to happen. And interactive programs like chats just don't eat to much processor time either, the average amount of data to process is dwarfed by the amount the game engine is crunching.

      --
      .sig: Sique *sigh*
    7. Re:Why would one get this by LurkerXXX · · Score: 4, Informative

      Pull up task manager and look at how much CPU usage you have when your machine is just sitting 'idle', not running a game or other CPU intensive app. If you have a higher end CPU, it's probably maybe averaging 1-2%. That's not a lot of stuff to toss on the other CPU. The FX chip will run at much more than 2% faster than the dual-core chips available now, so it's the better option for games (99.9% of which won't take advantage of 2 CPUs). In a couple years when more games are written for SMP systems, then the dual-core will be the way to go.

    8. Re:Why would one get this by Bloater · · Score: 1

      Communication.

      If your workload requires synchronisation between CPUs, or only even has one thing that can be done at a time, you don't see *any* benefit. You may even see a reduction in performance. Typically, dual-core is not quite twice as fast even with all else being equal and even with each CPU having its own local memory.

    9. Re:Why would one get this by The_Dougster · · Score: 1
      I don't know, I find myself almost drooling over the upcoming dual core X2 chips. I used to have a system with dual Pentium II 450's and it was pretty sweet, especially in Linux. The OS and system services can run on one of the cores, leaving you with essentially 100% cycles for your game on the second core. From my experience, the dual processor system was very consistently fast, it took a lot of effort to bog it down. You can do things in Linux like run a game on a second X-Server in a different VC and toggle back and forth to your desktop to read the strategy guide in your web browser with no performance hit.

      Pure windows gamers would probably opt for the faster single-core, but I would think that anybody who was running Linux would really benefit from the dual cores. I know what I want for X-mas now this year :-)

      --
      Clickety Click ...
    10. Re:Why would one get this by Koiu+Lpoi · · Score: 1

      So, what's the use of Intel's Pseudo-Dualcore Hyperthreading or whatever, especially if I'm a gamer?

    11. Re:Why would one get this by Nitar · · Score: 1

      What was the FX53 renamed to? I was wondering why they all but disappeared when the FX55 came out.

    12. Re:Why would one get this by Anonymous Coward · · Score: 0

      Do you even know what HT is? Nobody ever claimed it was dual core. That's why they called it HT b/c it allows threaded applications to run more efficiently. I wonder what you think the words "dual core" are for....

    13. Re:Why would one get this by megalomang · · Score: 1

      the second core would just be idle and convert electricity into heat

      You mean I they don't just turn off its power when it's idle???

    14. Re:Why would one get this by shokk · · Score: 2, Interesting

      Regardless of who it's aimed at, this would benefit those of us trying to run multiple apps such as Photoshop, email, web, various centralized management apps, vmware, etc, many of which are vying for CPU time. You know, those of us that are trying to do something productive with our time?
      Ernie

      --
      "Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
    15. Re:Why would one get this by VolciMaster · · Score: 3, Insightful

      The games might not be multithreaded, but the other core can run primary OS tasks. Plus, games will be, more and more, written to take advantage of those extra cores.

    16. Re:Why would one get this by cakesy · · Score: 1

      It means you can play two games at once, of course! What will they think of next...

    17. Re:Why would one get this by Sique · · Score: 1

      It became the AMD64 4200+, as far as I know.

      --
      .sig: Sique *sigh*
    18. Re:Why would one get this by Brian+Stretch · · Score: 1

      I find myself almost drooling over the upcoming dual core X2 chips.

      Not upcoming. They're here and here and here and elsewhere. Yeah, I want one too.

    19. Re:Why would one get this by homer_ca · · Score: 1

      HardOCP tested this against a dual core 4800+ and some other A64's. The FX-57 is still fastest in single-threaded tasks like games. The X2 is looking better to me though. As fast as the A64 4000+ in single CPU tasks plus the huge edge in multitasking.

    20. Re:Why would one get this by plover · · Score: 1
      Sick as it may be, I recently read a possible future consumer application for "hyperthreading" or dual-core CPUs will be: 1 CPU to run your primary app, and the other CPU will run all your anti-virus, anti-spyware, spyware, file-sharing, etc, etc, etc.

      Makes you crazy to see serious people making such stupid suggestions.

      --
      John
    21. Re:Why would one get this by jericho4.0 · · Score: 1
      This would really depend on what those tasks were. If it was important to you to keep crunching SETI@Home while playing games, yes, or any other long-running, processor intensive task.

      --
      "A language that doesn't affect the way you think about programming, is not worth knowing" - Alan Perlis
    22. Re:Why would one get this by jericho4.0 · · Score: 1

      I think you would be much better of with a dual core setup than this for those kind of things.

      --
      "A language that doesn't affect the way you think about programming, is not worth knowing" - Alan Perlis
    23. Re:Why would one get this by jericho4.0 · · Score: 1

      True, for most users. But I suspect that 'power' users could easily benefit. Anyone who hits Alt-Tab a lot (implying running several apps) and runs apps capable of using a significant number of cycles would appreciate the responsiveness.

      --
      "A language that doesn't affect the way you think about programming, is not worth knowing" - Alan Perlis
    24. Re:Why would one get this by MechaStreisand · · Score: 1

      It's worth pointing out that in theory, if the OS was designed right, it would be possible to get good responsiveness out of a single-core chip with twice the clock speed of a dual-core, and you'd still have far better responsiveness with single-threaded apps or even problems that can't be parallelized (these problems do exist, and it's worrying to me that all the CPU manufacturers are having to resort to parallelism which won't scale for some things). But anyway, few OSes are designed this well, and as a result you'll notice a huge improvement with two CPUs. Although I've never used it, I've heard that BeOS was really good in that respect, and Linux as well as Windows and whatever else just suck for letting one application - or a few - slow down the whole machine.

      --
      Disclaimer: IANAL. This post is, however, legal advice, and creates an attorney-client relationship.
    25. Re:Why would one get this by Anonymous Coward · · Score: 0

      A better test would be to run on a dual cpu box the games you normally enjoy and then go back and check to see how busy the second cpu was. There are system level services like networking, file access, sound driverrs, etc that can still benefit from a dual CPU/core system, leaving more cycles for your game.

    26. Re:Why would one get this by composer777 · · Score: 1

      Still, I just got the 4800+, and I must say that I'll never go back to single core. I really can't see how you could tell the difference between a 2.4 Ghz and 2.8 ghz athlon in games, they are both very fast, and having the extra core is worth it. I was able to rip two CD's at once, and play ut2004 without a single hitch. I couldn't even tell that the CD's were being ripped. The only annoying part was when cdex interrupted my game to tell me it was finished, but other than that, it was very fast. Kernel compiles are also faster, and I can now run compute intensive tasks like compiling, ripping, encoding, etc. and play games and surf the net while I'm waiting.

    27. Re:Why would one get this by LurkerXXX · · Score: 1

      Which will return the results stated. 0-2% utilization. I have SMP systems. SMP is great for multimedia, but doesn't do much for games right now. Hopefully that will change by the time DOOM4 and Half-Life 3 come out.

    28. Re:Why would one get this by da · · Score: 1

      Amen to that - I had a dual PII 450 which went to PIII 550 later and I have only just replaced with an AMD64 3000+. It was only when Eclipse 3 came out that I felt I needed (wanted?) to upgrade the hardware. Now, Firebird opens instantly, if only I could get the Eclipse debugger to work in 64-bitness... Prices for a 4200+ X2 should be pretty sweet by Xmas :)

      --
      I reserve the right to be wrong.
  4. Printer-friendly (aka ad-unfriendly) link by theGreater · · Score: 4, Funny

    http://www.hexus.net/content/reviews/review_print. php?dXJsX3Jldmlld19JRD0xMzE3

    This is filler. This is more filler. This is even more filler. This is a rant about the captcha.

    -theGreater.

    1. Re:Printer-friendly (aka ad-unfriendly) link by Anonymous Coward · · Score: 0

      This is a rant about the captcha.

      But you have a karma bonus. I thought the captcha went away when you got a karma bonus?

  5. Re:Mr AMD by Anonymous Coward · · Score: 0

    I can just hear Intel yawning and proceeding with what they do best; discussing large price-slashed processor deals for the industry and Apple.

  6. Whats Up With Hexus.Net? by Cryptacool · · Score: 0, Troll

    This is the third link in recent memory, and their server always goes down. Also for the most part their reviews are pretty bad, can't we wait to post these stories for a reputable site to have a review up?

    1. Re:Whats Up With Hexus.Net? by ryszards · · Score: 2, Insightful

      It doesn't always go down. It only slows down until we tweak the max number of connections for the traffic that's coming in.

      I'm not sure whether hexus.net aliases to somewhere else in your DNS or something, but our reviews are never 'pretty bad', even though I do say so myself. Maybe you're reading another site instead of ours?

      Feel free to point out what you think we suck at though, incase we're missing something and you can help us fix it. Feedback is always welcome, even from Slashdotters.

      --
      - 'sup, G?
    2. Re:Whats Up With Hexus.Net? by Anonymous Coward · · Score: 0

      "Feedback is always welcome, even from Slashdotters.

      What do you mean by "even from Slashdotters?"

      Are you insinuating we're unable to provide meaningful feedback?

      Your reviews are awesome! well researched and aren't full of ad laden crap..*cough* for stuff written by a bunch of cocksucking bastards that is. :P

    3. Re:Whats Up With Hexus.Net? by Hack+Jandy · · Score: 1

      I wish I didn't just blow all my mod points into that Zombie dog article.

      I thought Hexus got bought by some graphics card company? If that's not an unbiased source, then I don't know what is!

      HJ

  7. w00t by qw0ntum · · Score: 5, Funny

    I know I can't wait to do spreadsheets in 2.8Ghz, 64-bit glory. >_> That aside, this is still a pretty awesome development.

    --
    'Every story, if continued long enough, ends in death.' --Ernest Hemingway
    1. Re:w00t by ackthpt · · Score: 1
      I know I can't wait to do spreadsheets in 2.8Ghz, 64-bit glory. >_> That aside, this is still a pretty awesome development.

      Spreadsheets, heck, I can't wait to get this baby on my 56K dial up line and see how fast it can render /. pages!

      that's what it's all really about, isn't it?

      --

      A feeling of having made the same mistake before: Deja Foobar
    2. Re:w00t by vertinox · · Score: 1

      Have you ever done a 50,000 row =vlookup() formula in Excel? I kind of could have used that extra horse power waiting 10 min for it to calculate. ;)

      --
      "I am the king of the Romans, and am superior to rules of grammar!"
      -Sigismund, Holy Roman Emperor (1368-1437)
    3. Re:w00t by Spy+der+Mann · · Score: 1

      I know I can't wait to do spreadsheets in 2.8Ghz, 64-bit glory

      Yay! At last my genetic spreadsheet won't hang my PC! :)

    4. Re:w00t by Zemplar · · Score: 1

      I've had 3+ hour spreadsheet calcs on a loaded P4 2.26GHz, so there certainly is a need for some for this type of horsepower! That's a damn long coffee break when you have work to do.

      Those calcs were pulling data from locally cached databases [since you can't access millions of rows natively in spreadsheets], vlookups, and some 4-line forumlas. I've since moved to a loaded Sun Java Workstation W2100 (dual AMD Opterons) and the same types of calcs are only 20 minutes.

      Point being, Power to the Nerds!

  8. Why is this news? by Anonymous Coward · · Score: 3, Funny

    I stopped caring about CPU advances when the Pentium II reached 233 MHz. I run Windows 98 on my machine wihtout a firewall and I have never had a problem with spyware or viruses. How about we focus research and development on somehting more important, like fuel cells to power my wireless mouse.

    1. Re:Why is this news? by luna69 · · Score: 2, Informative

      Because the market for fuel cells for your mouse isn't driven by the gaming industry, while the processor industry (at the high end, as this CPU is) is. :)

      --
      No gods, no demons, and no masters. Secular Humanism!
    2. Re:Why is this news? by Anonymous Coward · · Score: 0

      Shit, who cares about fuel cells, how the fuck does this "wireless" mouse work?

      Some kind of clockwork perhaps?

      Maybe induction?

      Please tell me, I'm on tenderhooks.

    3. Re:Why is this news? by ettlz · · Score: 1
      How about we focus research and development on somehting more important, like fuel cells to power my wireless mouse.

      Just go down to the local hospital and buy a used pacemaker. Remove the nuclear battery, and hack it into your mouse. Problem solved.

    4. Re:Why is this news? by Krankheit · · Score: 1

      You are the kind of people that are why ISP's like my cable company have to block outgoing ports. Get a router, don't run a zombied machine. 233 MHz is more than enough for an X terminal, but if you have a lot of users on a 233 MHz server, PHP, Apache and Postgre will not perform well. This CPU is for servers.

      --
      Powered by caffeine and sugar; BSD
    5. Re:Why is this news? by ccccc · · Score: 1

      I suspect the people developing the fuel cells probably do care about CPU speed advances. I don't imagine much research of anything would get done if everyone was stuck using P2/233 MHz machines.

    6. Re:Why is this news? by Skye16 · · Score: 1

      It is? I thought the Opterons were for servers. This is for gaming. This is for making punk bitches suck it down. Not XXX bitches sucking it down.

    7. Re:Why is this news? by QMO · · Score: 3, Funny

      Deoxyribonucleic Acid
      All the best mice use it.

      --
      Exam 4/C again. Maybe I'll do better this time.
    8. Re:Why is this news? by Anonymous Coward · · Score: 0

      Or you could just buy a RFID powered mouse instead...

      http://hardware.slashdot.org/article.pl?sid=05/05/ 28/2050227&tid=227&tid=126&tid=1

    9. Re:Why is this news? by SolusSD · · Score: 1

      while you were gone:
      windows xp, open source revolution, highend 3d graphics, agp > 1x, flat panel monitors, broadband, ... ;)

    10. Re:Why is this news? by afd8856 · · Score: 1

      233 is just barely enough for an xserver, not "more than enough". My experience.

      Although it's true that it can be run on a 486 33 Mhz, but mozilla runs really slow.

      --
      I'll do the stupid thing first and then you shy people follow...
    11. Re:Why is this news? by Diakoneo · · Score: 0

      Darn, I never have mod points when I come across the really funny ones like this...

      --
      "Well..here I am..." - Jubal Early
    12. Re:Why is this news? by GigsVT · · Score: 1

      but if you have a lot of users on a 233 MHz server, PHP, Apache and Postgre will not perform well.

      It was enough in 1999.

      Someone should sneak into developer's houses and clock all their systems down so we can get efficient software.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    13. Re:Why is this news? by Eric604 · · Score: 0

      Me neither. I think we should have more modpoints, it also keeps people from posting useless comments like this one.

    14. Re:Why is this news? by Anonymous Coward · · Score: 1, Funny

      Rads aside, I don't think I'd want to touch a "used" pacemaker...

    15. Re:Why is this news? by MHobbit · · Score: 1

      If you could convert the heat a P4 gives off into energy, you'd no longer have to have fuel cells power your wireless mouse.

      --
      Debugging? Klingons do not debug. Bugs are good for building character in the user.
    16. Re:Why is this news? by Anonymous Coward · · Score: 0

      "Please tell me, I'm on tenderhooks."

      Hello. This is your friendly spelling Nazi.

      The word is: 'tenterhooks'

      Yours,
      SN

    17. Re:Why is this news? by drsmithy · · Score: 1
      It was enough in 1999.

      So use software written in 1999.

    18. Re:Why is this news? by GigsVT · · Score: 1

      Every upgrade things get a little shittier, a little more bloated, a lot more buggy. Especially for just running servers.

      The old stuff was buggy when it came out too, but at least it has patches out, and I don't think it was quite as buggy even when new if you compare to something like Fedora, which is very bad lately.

      I use Debian stable in some places too, that usually runs pretty far behind.

      Anyway, whenever I can get away with it, I generally do run older software on servers.

      Our servers right now are a mix of Red Hat (slated for upgrade to Centos), Centos3/4 and a couple Debian stable.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    19. Re:Why is this news? by Anonymous Coward · · Score: 0

      Nuh-uh!

      ATP or Adenosine Triphosphate is what powers cell metabolism!

  9. Confused by setzman · · Score: 3, Interesting

    I've been looking to upgrade (just buying a new mb/cpu/ram combo), but which processor is which out of MANCHESTER, CLAWHAMMER, SAND DIEGO, VENICE, and WINCHESTER??

    --
    C:\>
    1. Re:Confused by Quick+Sick+Nick · · Score: 3, Insightful

      Those are all the core code names. San Diego and Venice are the two newest and best.

      They are the best overclockers and produce the least heat. They use 90nm technology.

      Venice is generally used in the lower lineup (3000+, 3200+, 3500+, etc) And San Diego is used with more expensive processors (4000+, FX-55)

      Those other cores are older and should be avoided unless you want to save some money.

    2. Re:Confused by Skye16 · · Score: 1

      Also, Venice and San Diego are the first cores to incorporate SSE3 instructions. I don't know what that means, but it's one more than SSE2! (seriously, though: SSE3)

    3. Re:Confused by Spacelem · · Score: 3, Informative

      If you are confused, This guide is pretty good at explaining the differences between the current AMD offerings.

    4. Re:Confused by ottffssent · · Score: 3, Informative

      Venice.

      The venice core is the new one. It's a few unimportant percent faster than the others. It's got a few not-terribly-important new features. It's rather more than a few percent less power-hungry. AMD doggedly sticks to its 89W max figure, but Venice maxes out at more like 50W, according to measurements / approximations.

      So, to recap:

      Faster
      Better
      Lower power
      Nicely overclockable

      Buy from a respectable retailer and they'll tell you not only which core you're buying, but the clockspeed, cache size, etc. rather than just the model number.

    5. Re:Confused by EpsCylonB · · Score: 1

      Also, Venice and San Diego are the first cores to incorporate SSE3 instructions. I don't know what that means, but it's one more than SSE2! (seriously, though: SSE3)

      The SSE on my althlon 64 goes up to 11.

    6. Re:Confused by pinkocommie · · Score: 1

      lol @ ur sig. fyi i'm muslim :)

    7. Re:Confused by DarkBlackFox · · Score: 1

      Actually, San Diego is the new one, which is what this article is about. San Diego is essentially Venice with more cache slapped on, and with this new FX 57, faster clock speed (and strained silocon).

      The article is a bit off though, claiming the only San Diego cored chips are the 4000+ and FX 57. I have an Athlon 64 3700+ in my system now, also based on the San Diego core, and it's at a ver good price point (around $320, great mid-high end processor).

    8. Re:Confused by setzman · · Score: 1

      as long as you're not a terrorist you're okay.

      --
      C:\>
    9. Re:Confused by ackthpt · · Score: 1
      Venice.

      The venice core is the new one. It's a few unimportant percent faster than the others. It's got a few not-terribly-important new features. It's rather more than a few percent less power-hungry. AMD doggedly sticks to its 89W max figure, but Venice maxes out at more like 50W, according to measurements / approximations.

      I picked out the very bottom of the line of the new Venice core and look forward to ripping apart my PC over the 4th weekend and planting that baby in there. It's still going to run rings around what I have and use less power, which is all I ask. I couldn't care less about video games (as long as I can still play NetHack)

      Monarch appear a decent source, as they put the CPU on the board and give it a test run before shipping.

      --

      A feeling of having made the same mistake before: Deja Foobar
    10. Re:Confused by Anonymous Coward · · Score: 0

      How come your sig doesn't mention christian terrorists? Seems like you are picking on poeple needlessly.

    11. Re:Confused by kayak334 · · Score: 1

      San Diego is the newest core, no Venice. It's the same as Venice but with more cache.

      CPU chart

    12. Re:Confused by Anonymous Coward · · Score: 0

      "Fuck commie Democrats, fascist Republicans and Muslim terrorists. "

      Shouldn't that be terrorist Muslims, just to be consistent?

    13. Re:Confused by jericho4.0 · · Score: 1

      I think a lot of us are confused. I've been confused since soon after the 486DX. I simply go on a crash course when buying a new system, but then that info is useless 3 months later.

      --
      "A language that doesn't affect the way you think about programming, is not worth knowing" - Alan Perlis
    14. Re:Confused by jericho4.0 · · Score: 1
      "Fuck commie Democrats, fascist Republicans and Muslim terrorists."

      You see, in a post 9/11 world, there are only 3 types of people.

      --
      "A language that doesn't affect the way you think about programming, is not worth knowing" - Alan Perlis
    15. Re:Confused by Deliveranc3 · · Score: 1

      You missed Turion... which is pretty badass as well.

      But Venice and San Diego are both pretty badass.

    16. Re:Confused by coopex · · Score: 1

      Well, going by your handle, then you'd be a commie muslim, and thus there is no problem.

      --
      The road to hell is paved with good intentions.
    17. Re:Confused by whorush · · Score: 1

      dude, your sig is really stupid. i know you're just trying to start s**t. why?

  10. Yeah, but... by allanc · · Score: 4, Funny

    Will a beowulf cluster of these run OSX?

    (I honestly can't tell if I'm being +5 funny or -1 troll some days...)

    1. Re:Yeah, but... by cp.tar · · Score: 1
      Will a beowulf cluster of these run OSX?
      Only in an emulation mode, I'd guess... or a virtual machine...
      Which I guess would not be such a problem for a Beowulf cluster of anything.

      Of course, now we're both -1 Offtopic ;)

      --
      Ignore this signature. By order.
    2. Re:Yeah, but... by JustOK · · Score: 1

      Ask the old people in Korea. They are the only ones who run this.

      --
      rewriting history since 2109
    3. Re:Yeah, but... by HG+Slashdot · · Score: 0

      This is one of those bad days you have been modded +5 funny today :-(

      --
      j0b.org - A famous domain name for sale
    4. Re:Yeah, but... by cp.tar · · Score: 0

      Does that mean that in Soviet Russia, OS X runs Beowulf clusters?

      --
      Ignore this signature. By order.
  11. Apple's "Intel-Macs" will shortly go AMD by Anonymous Coward · · Score: 0, Insightful
    Clearly Apple is using Intel up front for as much financial assistance as possible in making the switch to Macs based on x86 processors. It seems clear that once the first few iterations of machines are launched, over the next two years, Apple will go with the more powerful AMD. The work will have been done. Going with AMD up front would have meant missing out on much assistance from Intel and its deep pockets.

    Mac on Intel.com -> http://www.maconintel.com/


    DLightman

    1. Re:Apple's "Intel-Macs" will shortly go AMD by TheRaven64 · · Score: 4, Interesting

      Did you see the power requirements? 104W for a single core? I really can't imagine Apple preferring this to a Pentium-M derivative.

      --
      I am TheRaven on Soylent News
    2. Re:Apple's "Intel-Macs" will shortly go AMD by Lonewolf666 · · Score: 1

      This might depend on how fast Intel can deliver its dual-core Pentium M in 65 nm (Codename Yonah). If it is available in early 2006 as promised, Apple will have no reason to switch to AMD.
      If it is, however, vaporware, you might be right.

      --
      C - the footgun of programming languages
    3. Re:Apple's "Intel-Macs" will shortly go AMD by gabebear · · Score: 1

      Apple neeeds a new CPU for their laptop and Intel is curently spanking AMD soundly in low wattage CPUs. If Apple didn't need Intels laptop CPUs I wouldn't doubt they would have gone with AMD.

    4. Re:Apple's "Intel-Macs" will shortly go AMD by cybpunks3 · · Score: 1

      So in the meantime, you have PPC Macs that are slower than x86s. Then they will transition to Intel PCs that are slower than AMD PCs. Wonderful.

    5. Re:Apple's "Intel-Macs" will shortly go AMD by ceeam · · Score: 1

      It's the technology, stupid. Early K6's (I recall having a 3.3v one) were heat monsters. Later I had K6-2+ @500 and it could almost run without a cooler. Early Athlons (1.4GHz pre(?)-Thunderbirds) were ouch-hot. The versions before SocketA died were much-much more efficient. OTOH - PIV design is notable for doing way too much extra work (many loops running in advance etc... I don't fully understand modern CPUs but you may try to research it) and that extra work will still heat 'em up no matter what the process. Athlon64 is actually very power-efficient design.

      PS: Before you point it out - I understand that P-M is not P-IV.

    6. Re:Apple's "Intel-Macs" will shortly go AMD by be-fan · · Score: 1

      Couple of points.

      1) This is a high-end CPU. Other CPUs in the same line have TDPs of ~50 watts. This is more than a P-M, and might be an issue on laptops, but is well within the power dissipation that can be quietly handled on a desktop.

      2) AMD reports TDP_max, while Intel reports TDP_avg. A dual-core Prescott is rated at something like 130 watts TDP, but its TDP_max is over 180 watts. While the P-M uses much less power, its TDP_max is going to still be higher than the TDP_avg that Intel reports.

      --
      A deep unwavering belief is a sure sign you're missing something...
    7. Re:Apple's "Intel-Macs" will shortly go AMD by LionMage · · Score: 1

      The only thing certain about Apple's use of a Pentium M derivative is that it will show up in lower-end desktops (Mac Mini) and laptops (iBook, PowerBook). For premium desktop systems, they'll use whatever the market demands for performance. Considering that a lot of bread-and-butter users of PowerMac hardware are Photoshop and digital video users, I don't think a Pentium M is in the cards for Apple's top-of-the-line desktop systems.

    8. Re:Apple's "Intel-Macs" will shortly go AMD by theantipop · · Score: 1

      How is this modded "informative"? It's pure uninformed drivel. If you read any review with a power consumption test, you'd see their entire test benches use just over 190W under full load. Most other Athlon 64's are around 150.

    9. Re:Apple's "Intel-Macs" will shortly go AMD by DrMrLordX · · Score: 1

      That might mean something if Intel actually had a Banias or Dothan(Pentium M)-derived processor available for the desktop segment. Currently, they do not, and they may not until 2007(Conroe). For now, their entire desktop and workstation CPU line chews up power much, much faster than anything AMD offers.

    10. Re:Apple's "Intel-Macs" will shortly go AMD by DrMrLordX · · Score: 1

      Last I heard, Yonah was aimed at the laptop/mobile market, not the desktop market. It won't launch at speeds competative with any desktop AMD offering. Conroe will be the desktop Pentium-M-alike processor, and it won't be out until later. 2007 I think?

    11. Re:Apple's "Intel-Macs" will shortly go AMD by Anonymous Coward · · Score: 1, Interesting

      It looks like Apple will be switching the low-end line of computers to Intel first - which suggests that perhaps there is something suitable for the high-end on Intel's roadmap in a year or two that might not be accurately represented by their current models.

      Even ignoring this, when Apple committed to this deal, I'm certain that Intel's future offerings have been a more significant factor than their current ones. If this isn't the case, Apple has been surprisingly short-sighted. Realistically, they've probably had Intel commit to an agreed (non-public) roadmap with sanctions if they don't deliver.

      It's likely that Apple has agreed to prefer Intel over AMD as long as Intel delivers on their promises. The primary reasons for choosing Intel over AMD have probably not been technological, but their ability to deliver - even when AMD has been ahead in technology, they have occasionally suffered from an inability to deliver in volume. This is probably the main reason why high-volume manufactures (like Dell and now Apple) haven't gone with AMD.

      Some manufacturers with smaller volumes for whom the technology has been important have gone the AMD way (e.g. Sun has been delivering Opteron-based machines for quite some time). The real difference between Intel and AMD today is that Intel still is better "guaranteed to deliver". The end result is that Intel is a better deal for high-volume system integrators. AMD is a better deal for (some) individuals and low-volume high-end vendors.

      Most of my PCs (three out of four) are AMD because they were the most powerful for the price at the time, and I have a couple of Macs, which I desired for the user interface and specific software only available for the Mac.

      I'm the kind of user who'll always choose the best tool for the job, depending on my needs, and who can build his own systems with no trouble.

      Oh, and I don't run MSWin on any of my machines. It's FreeBSD, Linux and MacOS X all the way - there's no MSWin software that I need or want, and MSWin isn't the easiest way to run any of the multi-platform software that I use.

    12. Re:Apple's "Intel-Macs" will shortly go AMD by jericho4.0 · · Score: 1
      What if Apple has simply revesed the problem, and now can't compete on the desktop? Probably still a better place to be for Apple...

      --
      "A language that doesn't affect the way you think about programming, is not worth knowing" - Alan Perlis
    13. Re:Apple's "Intel-Macs" will shortly go AMD by Lonewolf666 · · Score: 1

      A Pentium M does perform quite nicely, even when compared to desktop processors. Consider http://www.tomshardware.com/cpu/20050525/.
      While the results at Tom's Hardware were achieved with overclocking to 2.5 GHz, that clock speed seems not unreasonable for a 65nm version. Now imagine a dual-core version, and Yonah might indeed be impressingly fast.
      That is, if Intel will deliver it on time. If there is a year of delay or so, AMD might cook up something even better in the meantime.

      --
      C - the footgun of programming languages
    14. Re:Apple's "Intel-Macs" will shortly go AMD by Deliveranc3 · · Score: 1

      This is great flamebait congrats.

      First off it's for both cores, Dual cores run significantly cooler than two single cores.

      Second that is equivalent to a low end pentium of the current generation, ouch. Third that's maximum output.

      These things are actually remarkably cool, maybe 5%-10% hotter than the XPs were.

    15. Re:Apple's "Intel-Macs" will shortly go AMD by toddestan · · Score: 1

      Did you see the power requirements? 104W for a single core? I really can't imagine Apple preferring this to a Pentium-M derivative.

      Have you seen the power requirements for the G5?

      I don't see why Apple can't go both ways. Pentium M/Celeron M for the laptops/Mini, AMD64 for the high end towers, and who knows for the iMac/eMac.

    16. Re:Apple's "Intel-Macs" will shortly go AMD by gabebear · · Score: 1

      Intel is still competitive on the desktop even if they currently don't have the "fastest" CPU today. Apple won't be in any worse situation than Dell on hardware and Apple has an OS to differentiate itself from all the other computer manufacturers.

      AMD seems to think Apple was bullied into accepting Intel's offer. I know Intel has some very sketchy discount mechanisms... If this lawsuit is succesful maybe we will see AMD in Apple and Dell desktops.

  12. Remember when Apple/IBM did this...last year? by CptTripps · · Score: 0, Redundant

    REmember when Apple showed everyone a G5....the reaction was "That sucks...it's only 2ghz!"

    Now we have people saying "This is going to be great!" ...but NOW we have Apple going to Intel, and presumably a dual-core CPU. I'm imagining that it'll be the better play.

    But what do I know...I still can't type faster than a 486.

    --


    My .sig can beat up your honor student.
    1. Re:Remember when Apple/IBM did this...last year? by QMO · · Score: 1

      "But what do I know...I still can't type faster than a 486."

      I don't know. My 486 has never even seen Mavis Beacon. Probably can't type worth old stinky beans.

      --
      Exam 4/C again. Maybe I'll do better this time.
  13. Men in Black? by eander315 · · Score: 1

    What is 1 MiB of L2 cache? 1 Million Bytes?

    1. Re:Men in Black? by Mister+Impressive · · Score: 3, Informative

      What is 1 MiB of L2 cache? 1 Million Bytes?

      It's the binary count of bytes, 2^n, instead of the decimal *10^n

      --
      Let the commencement BEGINULATE!
    2. Re:Men in Black? by BetterThanCaesar · · Score: 1

      Mi = Megabinary = 1024*1024
      M = Mega = 1000*1000

      Please, no more flamewars. Hard drive manufacturers destroyed M = 1024*1024 years ago.

      --
      "Stop failing the Turing test!" -- Dilbert
    3. Re:Men in Black? by Anonymous Coward · · Score: 2, Informative

      SI unit for 1 megabyte (1024) is MiB, 1 megabyte (1000) is MB.

      SI units were changed back in the 90s to reflect base-10, the base-2 units were renamed to use the letter 'i'.

    4. Re:Men in Black? by Zo0ok · · Score: 1
      I believe the correct technical notation is:

      MiB = 1024*1024 bytes MB = 1.000.000 bytes

      However, everybody use MB when they mean MiB. The Mi-notation is not very widespread or successful.

    5. Re:Men in Black? by nebulus4 · · Score: 0

      Actually it's called MebiByte (1 MiB = 1048576 Bytes)

      --
      "It would be wrong to refuse to face the fact that everything is fundamentally sick and sad."
    6. Re:Men in Black? by Anonymous Coward · · Score: 0

      And you know what? It was the marketers' fault to begin with. Remember the PC AT with "20 Megabytes of memory"? Well, that "memory" was actualy a 20MB HDD. Because CPU memory is specified in powers-of-two, the HDD, associated as "memory" was in powers-of-two.

      In any case, if you want to be *accurate*, it woudn't be X million bytes, anyway - the number of banks and size of bank means that it will ALWAYS be simpler to note the size in power-of-two mega-bytes. So why make a simpler notation longer by putting a useless "i" in there.

    7. Re:Men in Black? by Anonymous Coward · · Score: 0
      Hard drive manufacturers destroyed M = 1024*1024 years ago

      The bastards! Making things ungeeky, how lame.. Even lusers without knowledge about the binary system can operate a computer these days. Binary was the first thing I learned after PEEK and POKE.

    8. Re:Men in Black? by Ignominious+Cow+Herd · · Score: 1

      However, everybody use MB when they mean MiB. The Mi-notation is not very widespread or successful.

      And let's _please_ keep it that way! This is a solution looking for a problem. I mean, with hard disk sizes in the hundreds of GB and growing, who really cares?

      --
      Lump lingered last in line for brains, and the ones she got were sorta rotten and insane.
    9. Re:Men in Black? by fitten · · Score: 1

      I agree... the jargon was termed a long time ago and there's no real reason to change it other than for some folks to feel important and think they've made their contribution to the world in some small way. If we're going to start arbitrarily redefining things like this, I'll start... When I say "You're a rat bastard idiot", as of today, it "really" means "Hi, have a nice day."

    10. Re:Men in Black? by GigsVT · · Score: 2, Insightful

      As hard disks get bigger, it gets more important, not less important.

      1,048,576 vs 1,000,000 is 4.8% off

      GiB vs GB = 7.3% off

      TiB vs TB = 9.9% off

      PiB vs PB = 12.5% off

      EiB vs EB = 15.2% off

      Anyway, it is important. How long until someone dies because some programmer mixed the two up?

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    11. Re:Men in Black? by MooseGuy529 · · Score: 1
      I mean, with hard disk sizes in the hundreds of GB and growing, who really cares?

      Once we get to terabytes, you'll care. As the units go up, the disparity between binary and metric starts to widen. With megabytes, the difference is 4.6% of your space. With gigabytes, it's 6.9%. With terabytes, it's 9.1%. With petabytes, it's 11.2%. And so on... It gets worse and worse.

      And just because disk space keeps increasing, it's not going to be infinite. There's a saying that "operating systems and files will expand to fit the available media". It's true. If hard drive sizes were suddenly ten times bigger, people would start storing their music files at higher bitrates, storing their pictures as TIFF's or RAW's instead of JPEG's, keeping journaled copies of old files (isn't it odd that with all these increases in space, automatic versioning of files hasn't come back into popularity?), and so on. And when you go to archive some files and they're 153 GB and you happily buy a 160 GB drive and it doesn't fit, that's the problem.

      --

      Tired of free iPod sigs? Subscribe to my blacklist

    12. Re:Men in Black? by hostyle · · Score: 1

      I'll get worried when it hits 100%.

      Whats this? "Warning: You have only 3 Tb of space free on drive C:". I run out and buy a brand new 67 YottaByte drive, install it and then get the exact same message.

      Cause judging from your math it could happen!

      --
      Caesar si viveret, ad remum dareris.
    13. Re:Men in Black? by Ignominious+Cow+Herd · · Score: 1

      Yes, but assuming that humans are at least a little bit intelligent, and that currently the only visible problem trying to be solved by this terminology is Hard Disk capacity 'mis-reporting', can't we assume that eventually most people will catch on? And the ones that don't deserve what they get (either a: a smaller HDD than they thought, or b: government interference to properly label the damn things)? Creating the new terms is giving up!

      (or c: HDD sizes like soft drinks at fast-food places small, medium, large, super-sized. Of course at some point small becomes medium and then we're all screwed.)

      RE: Auto versioning - YES! I installed a Win2003 server at work just so people would be able to recover their own damn files. I'd rather have installed a linux server, but there seems to be no comparable system (yet?).

      --
      Lump lingered last in line for brains, and the ones she got were sorta rotten and insane.
    14. Re:Men in Black? by Anonymous Coward · · Score: 0

      Which is precisely why SI units suck.

      I live in the USA, therefore, I'm allowed to call 2^20 bytes a megabyte, because it hasn't changed in Imperial units. Bytes are on a 2^(n*10) scale, while everything else uses a 10^(n*3) scale. And it's gonna stay that way here. Honestly, I think I've just stumbled on a single good reason to keep Dubya. Hell can now freeze over.

    15. Re:Men in Black? by Anonymous Coward · · Score: 0

      Please, no more flamewars. Hard drive manufacturers destroyed M = 1024*1024 years ago.

      True, but memory manufacturers did not. So if I read "1Mb" in terms of cache memory, I don't think "Oh, shit, that's only 1 million bytes, where'd the rest go??"

      However, when I read "1MiB", I think "Oh, shit, what a tosser ..."

    16. Re:Men in Black? by GigsVT · · Score: 1

      100% according to my math would mean it was half the expected size. You buy a 200 whateverbyte drive and you only get 100 WiB.

      It will happen eventually, assuming unbounded growth. After than it would just go above 100%.

      --
      I've had enough abrasive sigs. Kittens are cute and fuzzy.
    17. Re:Men in Black? by Glonoinha · · Score: 1

      How long until someone dies because some programmer mixed the two up?

      Actually if one of my programmers used that faggoty MiB, Mebibytes, GiB Gibibytes shit and I caught them it would take about 38 seconds for someone to die, because that's how long it would take for me to stab him in the fucking throat 17 times.

      And since it is 1,048,576 vs 1,048,576 (etc.) it is actually 0% off, even at the EB level, because I will have killed the one loser bitch on my team that suggested otherwise. /grin

      --
      Glonoinha the MebiByte Slayer
    18. Re:Men in Black? by hostyle · · Score: 1

      I christen thee Glonoinha the MebiByte Slayer.

      --
      Caesar si viveret, ad remum dareris.
  14. Compare 64bit/32bit apps running on 64bit/32bit OS by Browzer · · Score: 0

    Isn't that what everybody wants to see?

  15. 1MiB != 1024KB by Stalin · · Score: 5, Informative

    For a definition of MiB see this wikipedia article -- http://en.wikipedia.org/wiki/Mebibyte

    And here is a random review that includes the actual cache numbers -- http://www.gdhardware.com/hardware/cpus/amd/athlon 64/fx57/001.htm

    1. Re:1MiB != 1024KB by Stalin · · Score: 1

      At least I don't think they are equal... Gah! confusing notations be damned!

    2. Re:1MiB != 1024KB by Beolach · · Score: 1

      1MiB != 1024KB, but 1MiB == 1024KiB. The wikipedia article is pretty good at explaining this.

      --
      Join moola.com, play games to earn money.
    3. Re:1MiB != 1024KB by OverlordQ · · Score: 1

      Well the Story Submitter is correct, on the HEXUS review it *does* say: 1024KiB data, non-mirrored

      So yes, that's 1 MiB

      --
      Your hair look like poop, Bob! - Wanker.
  16. Unlimeted Powerrrrr! ;) by cp.tar · · Score: 2, Insightful
    I like new CPUs.

    I'm poor, yet I want to buy a new PC.
    Therefore, if new CPUs come out, I can get an old one at a greatly reduced price.

    I do hope I'll be able to afford a 64-bit CPU... otherwise my new computer will be even more outdated as soon as I buy it.

    --
    Ignore this signature. By order.
    1. Re:Unlimeted Powerrrrr! ;) by Anonymous Coward · · Score: 0

      not always the case... very often, they just stop making the old ones, so they're not available for any price. generally the manufacturers want to retain the price points, instead of lowering 'em, they just take product out.

    2. Re:Unlimeted Powerrrrr! ;) by cp.tar · · Score: 1
      Yes... but they only do that with the low-end (or "low-end") CPUs; I'll just buy whatever low-end CPU is available.

      It'll still be at least 6-8 times faster than what I'm using now.

      --
      Ignore this signature. By order.
  17. To think... by ID000001 · · Score: 2, Interesting

    AMD only have about 15% of the market after beating Intel numerious of time tell you how smart the general public are.

    1. Re:To think... by Anonymous Coward · · Score: 0

      AMD doesnt have the production capacity to meet the demand of the general public if they did want to buy AMD.

    2. Re:To think... by Thomas+DM · · Score: 5, Informative

      Well AMD's production capacity isn't as big as Intel's - that's one of the big issues. AMD has an agreement with Intel to use x86, but this means that AMD may only outsource a limited amount of its production to foundries. But the new fab in Dresden, Germany will soon be opened to boost AMD's production capacity.

      Intel also has big deals with huge system builders such as Dell and HP.

    3. Re:To think... by dreamchaser · · Score: 3, Informative

      It has nothing to do with the smarts (or lack of) of the market. It has to do with their capacity and with marketing.

      99% of the users out there don't even need a 2ghz P4, let alone a 2.8ghz FX-57. AMD has long been catering to the enthusiast market which is a very, very small fraction of the overall market.

    4. Re:To think... by bill_mcgonigle · · Score: 1

      AMD only have about 15% of the market after beating Intel numerious of time tell you how smart the general public are.

      All the machines I saw at Sam's Club were AMD-based. HP, mostly.

      Apple has 5%.

      Dell has the other 80% and they're Intel. AMD is most likely not able to supply Dell with the number of chips they need, and they're not well-served to go with an mixture of chips, for supply line reasons.

      (Yeah, I know there are IBM, whitebox, and a few others left, I'm employing hyperbole to make a point.)

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
    5. Re:To think... by bill_mcgonigle · · Score: 4, Interesting

      AMD has long been catering to the enthusiast market which is a very, very small fraction of the overall market

      That's OK as an R&D goal - an enthusiast machine 5 years ago was a 800MHz P3, barely enough to get XP & Office running today.

      If you get your performance out for the enthusiast, next year you can get it into your consumer line, then the year after figure out how to run it in a laptop. At least, traditionally that's how it's done - Intel seems to be kicking ass in the laptop line and bring it up to the desktop - wise from a watts/$/flop perspective. Of course you can argue that was just about the Intel Marketing Department getting hold of R&D for a few years and screwing things up so badly that they're back to the P3 track and Marketing didn't notice what the blokes in Israel were up to.

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
    6. Re:To think... by Flooded77 · · Score: 4, Insightful

      It is all about quantity. People seem to think that the more of something you get in a product, the better. This marketing rationale is used for other kinds of products. Just look at fastfood ('Gorge yourself on our 10lb. McGutBomb Burger!'), automobile ('Scale Mt Everest in your new 3000 horsepower Chevy Truck!') or computer ('You'll need a 3.4Ghz machine to run Word and send email!') advertising. We've become a society of 'More is always better'. I don't think efficiency and quality are not important to the general public anymore.

    7. Re:To think... by cybpunks3 · · Score: 1

      How much capacity do you _need_ in order to satisfy the Mac market, even an Intel Mac market???

    8. Re:To think... by Anonymous Coward · · Score: 0

      They are always available when I need one. It's not like they are not resonably meeting current demand. A chip fab is A LOT of money. They would build one more if the demand or market required it or could support it.

    9. Re:To think... by haggar · · Score: 1

      I thought that large parts of the Dresden fab were already operational? I must have heard wrong. When is the power-up due?

      --
      Sigged!
    10. Re:To think... by Anonymous Coward · · Score: 0

      Well, the problem is that it is software that pushes hardware and not vice-versa like we had years ago with the 8086 where programmers had to tweak every line of code in order to meet memory constraints. Nowadays the code put out is just a mish-mash of whatever is good enough to get the job done while not crashing at the same time. Today's mentallity seems to be "Bah, why optimize it, the customer can just upgrade to a faster machine!". Rather than offering just what the customer needs to get the job done, they add every bell and whistle they can think of that does nothing but bog things down. Sure, you can still do all your desktop/office style work on a PIII 800Mhz system running Windows98, but forget security patches and whatnot once any M$ OS reaches EOL support. Pretty much every vendor seems to follow suit.

    11. Re:To think... by limabone · · Score: 1

      I disagree with your first comment.

      I fix alot of friends and relatives computers, and they all want to do (at least) the following:
      1. Email
      2. Surf the web
      3. Store their digital camera photos
      4. Listen/store/burn music or watch movies
      5. Games

      You don't need a great computer to do 1, 2 or 3, but 4 and 5 can require alot of horsepower, particularly if you are trying to do 4, while doing 1 2 and 3 at the same time.

      This new processor sounds amazing, but I have long given up trying to keep on the bleeding edge. I am still using an overclocked Athlon XP-M 2500 and it's starting to show it's age :(

    12. Re:To think... by Anonymous Coward · · Score: 0

      Can I get a double McGutBomb? Cause I'm really hungry!!

    13. Re:To think... by cheesybagel · · Score: 1

      It is if you are looking for a laptop.

    14. Re:To think... by Anonymous Coward · · Score: 0

      Dude...you do need a 3.4Ghz machine to run word anymore...

  18. Ugh by SilentChris · · Score: 0

    Perhaps someone can explain to me (either from an engineering or economic background) why this new stuff costs so much?

    Just this weekend I saw the new nVidia 7800 and said to myself "It's expensive, but I'd like to get one". Bought the card ($600). Since my current AMD ASUS board didn't support PCI Express, I picked up a new one ($150). The board said it supported Athlon 64 chips, but when I got it it was 939-pins. The chip I was using was 754-pins. Went back to Newegg and got a new San Diego AMD 939-pin chip ($350). Also decided to pick up 2 GB of memory ($250). Now I'm reading they moved to a 940-pin chip (does that extra pin really make a difference), but everything's already ordered.

    This one upgrade that started at $600 ended up being $1350. I'd like to blame it on the video card, but this is probably the most expensive upgrade percentage-wise (almost 70% of the cost to originally make the rig) that I've done to date.

    1. Re:Ugh by Wordsmith · · Score: 3, Informative

      The reason it costs so much is you just outfitted your rig with some seriously serious hardware. You went close to top of the line of your own volition. The top-shelf stuff will always cost you a premium - one that doesn't usually scale to the increase in performance - because you're going for best of the best.

    2. Re:Ugh by nitehorse · · Score: 4, Interesting

      Well, that's what you get for not researching properly before buying things. :)

      So, there are three socket types for AMD64 CPUs. The low-end stuff is Socket 754 - Sempron and Athlon64 and such. S754 is basically a dead end.

      Socket 939 is a much more promising long-term socket for Athlon64 upgrades. Most new S939 boards these days are being made with PCIe, which dovetails nicely with the fact that the high-end graphics card vendors are all moving PCIe as well.

      Socket 940 is for Opterons and Athlon64-FX chips. The FX chips are really expensive and really fast, but more expensive than fast when compared to their S939 brethren. Socket 940 boards come in both AGP and PCIe variants (since S940 is relatively old and predates PCIe, many AGP S940 boards exist).

      As far as I know, there aren't any plans for dual-core chips on the S754 chipsets. However, the Athlon64-x2 chips are S939, and the dual-core Opterons are all S940. If you have an S939 board, it's probably smarter to forget the FX chips and go straight for an A64-x2 once they become a bit more affordable.

    3. Re:Ugh by Skye16 · · Score: 1

      The extra pin doesn't make much of a difference. And I believe you read wrong; it used to be that the Athlon 64's were 754, the Athlon FX's were 940. Eventually they swapped everything over to the 939 (excepting the Opterons, which use the extra pin for the registered memory (I think)). Gamers won't benefit (and may lose out) by using an Opteron (just because of the registered memory), and I assume you're a gamer, with all the high-end hardware you just picked up. You're fine with 939.

      Summary:

      754: Old Athlon 64's
      940: Opterons (and maybe the old FX's - the more I think about it, the fuzzier I get on this)
      939: New Athlon 64's and FX's

    4. Re:Ugh by SilentChris · · Score: 1

      Thanks. Still not sure why they had to make two chips one pin off though...

    5. Re:Ugh by fitten · · Score: 1

      Marketing. S939 chips won't fit into S940 sockets and vice versa. S940 is for server products and S939 is for desktop products. That way they can keep their market segregated. You can do things that make the server stuff more expensive (more testing and verification, etc.) while not impacting your desktop prices as much.

      It's basically the same thing as the Pentium4 vs. Xeon. Little difference between the two physically but one is for servers the other for desktops.

    6. Re:Ugh by shokk · · Score: 1

      Because you're willing to pay for it, even while complaining about it. Welcome to a market economy.

      --
      "Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
    7. Re:Ugh by TobyWong · · Score: 1

      " Well, that's what you get for not researching properly before buying things."

      Speaking of proper research...

      I hope to dear god nobody reads the +5 parent comment and buys a socket 940 mobo for their FX. They are a 939 part that is multiplier unlocked (no ceiling lock like other A64s) and despite what the parent says, they are currently the best desktop CPU money can buy for gaming.

      --
      - Toby
    8. Re:Ugh by Anonymous Coward · · Score: 0

      I hate to tell you this nitehorse (58425) but all of the recent FX chips (53, 55, 57) have been 939 chips!

    9. Re:Ugh by nitehorse · · Score: 1

      Oy.

      Shows me what I get for remembering things wrong after having done *my* research a couple months ago. ::grin::

      Thanks for the correction. Although I still think it's kinda silly to drop a grand on a CPU, but hey, whatever works. :)

    10. Re:Ugh by Anonymous Coward · · Score: 0

      Wrong. A 939 CPU will fit into a 940 socket, but it still won't work because they use incompatible memory. Socket 939 systems use un-buffered DDR SDRAM, and Socket 940 systems use registered (buffered) DDR SDRAM. These are not interchangeable.

    11. Re:Ugh by TobyWong · · Score: 1

      Oh I agree with you... it's certainly out of my league.

      Maybe if I win the lottery some day. ;)

      --
      - Toby
    12. Re:Ugh by bhtooefr · · Score: 4, Informative

      Also, the 940 boards have to be 6-layer (costs more), and they have to have ECC (IIRC), whereas 939 can be 4-layer, and use cheaper, faster non-ECC memory.

    13. Re:Ugh by MechaStreisand · · Score: 1

      Actually, it's registered memory that the 940 boards have to use. Both the 939 and 940 can use non-ECC or ECC memory, but the memory from one will not work with the other.

      It's a shame that some manufacturers will make high-performance registered ECC memory, but they all don't care about performance of unbuffered ECC, which is what I want for my next machine...

      --
      Disclaimer: IANAL. This post is, however, legal advice, and creates an attorney-client relationship.
    14. Re:Ugh by bhtooefr · · Score: 2, Informative

      Here's all of the current sockets that AMD uses:

      462: Newer Classic Athlons, Athlon MPs (server chip), Athlon XPs, Duron, Old Semprons
      754: Old Athlon 64s, Semprons, Turion 64s (mobile chip)
      939: Newer Athlon 64s, most Athlon 64 FX's (53, 55, 57), and in the near future, newer Semprons and 1xx (single CPU only) Opterons
      940: Opterons, Athlon 64 FX-51 and some FX-53s

    15. Re:Ugh by theantipop · · Score: 2, Informative
      Socket 940 is for Opterons and Athlon64-FX chips
      They released the first FX (FX51) on 940, but everyone since then has been 939.
    16. Re:Ugh by pantherace · · Score: 2, Informative

      The original FXs were 940s, However, the newer ones have been 939 (they might also have a 940, but I don't believe so.)

      Also, it's not just one pin removed, the actual layout is different between 939 and 940. (940 also supports smp, that being the main difference.)

      The big difference between 754 and 939/940 is that 939/940 support dual-channel memory. This is important, because if they didn't, while the athlon 64 is relatively non-memory bandwidth hungry (as compared to the p4), performance would suffer with two cores on single channel memory.

    17. Re:Ugh by Jozer99 · · Score: 1

      Doesn't usually?!? So in what cases have you seen a 3.2 GHz processor going for $320 and a 3.8 GHz processor going for $380? It has always been the case that companies know that people who want the best will pay pretty close to anything for it, and that is where the big profit margins are. Intel makes much more selling 300 LB 20 year olds P4EEs than it does selling Dell 2.4 GHz Celerons.

    18. Re:Ugh by orderb13 · · Score: 1

      939 pin is for their desktop chipset, 940 is for their server chipset

    19. Re:Ugh by jericho4.0 · · Score: 1
      Because people like you can drop cash on this stuff?

      From the amount of research/impulse buying you seem to have done, it seems that these parts aren't really that 'expensive' to you. Others might have to be more selective and shop at a different price point, or would have been forced to return items if they discovered it needed more hardware. The industry counts on folks like you to cover early costs of bringing a product to market.

      Thanks for doing your part! I'm gonna spend half the cash and get 80% of the performance in a few months. :-) I'll probably get bitten by some stupid hardware thing I didn't see coming, too. It does seem to be changing faster and faster.

      --
      "A language that doesn't affect the way you think about programming, is not worth knowing" - Alan Perlis
    20. Re:Ugh by SilentChris · · Score: 1

      Stupid me... I thought "supports Athlon 64" on the new board meant, you know, actually supporting Athlon 64. I don't keep up with AMD to know that they switched pin numbers on the same chip.

    21. Re:Ugh by Anonymous Coward · · Score: 0

      "Maybe if I win the lottery some day. ;)"

      Sorry. I will be winning that one!

    22. Re:Ugh by arodland · · Score: 1

      It does, you know, actually support Athlon 64. However, it doesn't support a chip that you can't actually plug into the board, for obvious reasons. Checking for socket compatibility is pretty much the first thing you have to do, and that's been true for at least ten years.

    23. Re:Ugh by neafevoc · · Score: 1

      939: Newer Athlon 64s, most Athlon 64 FX's (53, 55, 57), and in the near future, newer Semprons and 1xx (single CPU only) Opterons

      I thought Opterons were only on 940... even the 1xx series.

    24. Re:Ugh by bhtooefr · · Score: 1

      Read the post again ;-)

      I said in the near future, there'll be some Semprons and Opteron 1xx's...

    25. Re:Ugh by neafevoc · · Score: 1

      Oops.

      You heard of selective hearing? Well, that's selective reading ;)

      Having Opteron 1xx on 939 is weird though. Having a Opteron 1xx line in general seems weird to me. I figure smaller workstations would be for 1xx... however, the speeds of the FX line would be the suited here (or even X2). Whenever I consider putting together Operton servers for clients, I always go for 2xx series at the very least.

    26. Re:Ugh by obeythefist · · Score: 1

      Doing your research is vital, as they say, a fool and his money...

      There are two major reasons for this pricing, as you suggest they are economical and technological reasons.

      Technically, it takes a while to get a new chip production line singing. I'm no lithographer, but the process is extremely complex. Initially, on any wafer, many CPU's will be faulty or fail testing. So the yield from any batch of processors can be quite low at the start. This means they spend a lot of money making very few CPU's (or GPU's or any other kind of device).

      Taking that into account, let's look at the economy of it all. We have a high initial cost per unit of production, you feel that right away. Until the process is kinked out, the new parts will cost more to produce so you will pay more to own them. Secondly, there aren't many of them out there, and people like yourself with low budget constraints and high demand will happily pay a premium for these parts. Demand and supply, at the seller-happy end of the equation.

      Now, for the rest of us without $1350 to blow every time a new GPU shows up, the trick is to state your budget, research all the components you can afford to buy at that budget, and most importantly, read the benchmarks! Understand the technology you are investing in. It will help you to understand how the components plug together, what you'll need to buy, how quickly the next part will arrive to make your new gear obselete and so on.

      Also (sigh) if the parent had done his research instead of buy first and ask questions later, AMD's socket 940 is the *first* Opteron socket available. The extra pin is used for handling registered RAM (from my understanding), hence the early Opterons and Athlon FX's on socket 940 required more expensive memory. AMD have always aimed at the server space first (they want that market!) but their yields are great so they've been able to push CPU's at the desktop market as well. There's no move to socket 940 in future, it's a move away from that.

      --
      I am government man, come from the government. The government has sent me. -- G.I.R.
    27. Re:Ugh by BagOBones · · Score: 1

      Parent should get one more Informative for actually mentioning that the difference is the dual-channel memory support.

      Going with that 939 boards support almost all the chips that need dual channel support such as the higher end AMD 64s, AMD 64 X2 (dual core) and the FX and Opteron chips.

      --
      EA David Gardner -"... but the consumers have proven that actually what they want is fun."
  19. More reviews by Thomas+DM · · Score: 1

    http://www.dvhardware.net/article5557.html

    A list of all FX-57 reviews, more will be added when I spot them.

  20. Related question by EraserMouseMan · · Score: 1

    I've got a Socket 939 motherboard. Will any AMD 939 CPU always work perfectly with my MB (after bios flash)?

    Or will there come a time when it may be a 939 cpu but won't work with my MB?

    1. Re:Related question by zachtib · · Score: 1

      afaik, yes, any 939 will work in any 939 mobo

    2. Re:Related question by DrMrLordX · · Score: 1

      Make sure the motherboard manufacturer has a BIOS that supports the newer socket-939 CPUs(notably the X2 processors). The FX-57 should work in any socket-939 motherboard.

  21. 2T memory timings by FictionPimp · · Score: 1

    I wonder if it fixes the 2T memory timings when using 4 sticks of ram. I really want 2 gigs of ram without loosing my current 1 gig.

    1. Re:2T memory timings by Slashcrap · · Score: 2, Funny

      I wonder if it fixes the 2T memory timings when using 4 sticks of ram. I really want 2 gigs of ram without loosing my current 1 gig.

      I know, it's terrible isn't it? I just can't stand knowing that my system could potentially be running nearly 0.2% faster under certain, very specific, workloads.

    2. Re:2T memory timings by Anonymous Coward · · Score: 0

      You, my friend, have discovered why you should be using more expensive registered memory with AMD. Regardless of socket, you cannot trust AMD's memory controller to work well with alot of slots.

      Not that there are many 2T registered DIMMs either but that's what you get....

    3. Re:2T memory timings by Surt · · Score: 1

      2T vs 3T can mean close to 5% in game benchmarks.

      --
      "Who is the Journal of Quantum Physics going to believe?" --Stephen Hawking
    4. Re:2T memory timings by DrMrLordX · · Score: 1

      Dude, that was fixed with the Venice and San Diego cores(for the most part). You can definitely run 4 single-sided 512 meg DIMMs on a Venice or San Diego, and I've heard reports of 4x512 working with double-sided DIMMs as well.

    5. Re:2T memory timings by FictionPimp · · Score: 1

      No, the Venice and San diego cores fixed the problem with 4 dimms dropping the bus to 333 instead of 400. But they still ran at 2T instead of 1T timings. As far as I know this is still an issue.

    6. Re:2T memory timings by Anonymous Coward · · Score: 0

      The options are 1T or 2T. 3T doesn't come into play. And yes, it's been shown that except for memory benchmarks we're talking sub-1% difference in performance.

    7. Re:2T memory timings by Babbster · · Score: 1
      2T vs 3T can mean close to 5% in game benchmarks.

      So what you're saying is that we should avoid this "Benchmarks" game?

    8. Re:2T memory timings by Surt · · Score: 1

      Nahh, it can be a lot of fun, especially if you're obsessive compulsive.

      --
      "Who is the Journal of Quantum Physics going to believe?" --Stephen Hawking
    9. Re:2T memory timings by toddestan · · Score: 1

      If you're already willing to buy the top of the line processor, you can certainly afford 2x 1GB Dimms.

  22. But... the price. by leuk_he · · Score: 1

    The price of this thing is high... what else can you buy for $1031 (when bought by 1000 pcs)

    1. Re:But... the price. by Ingolfke · · Score: 2, Funny

      what else can you buy for $1031

      Well... you've got yourself a good start on a great book collection.

    2. Re:But... the price. by IPFreely · · Score: 1
      I find it highly telling that AMD is pushing their prices up so much.

      Just a year or two ago, Intel processors were the high priced monsters, and AMD was trying to gain recognition and volume with low priced competition.

      With the introduction of the dual core processors, the tables were turned. AMD came in from $500 to $1000 on all their dual core models, while Intels entire line was way down, some aroune $200.

      It would normally be that the market leader with the best product/name recognition would have the higher price, while the wanna-bes use low prices to gain some market. By switching ends, AMD and Intel have both stated publicly who is the leader and who is the follower.

      I do like AMD, and it was great while their prices were down. But the good times are over and it's gonna be a while before I can get one of these babys though.

      --
      There is nothing so silly as other peoples traditions, and nothing so sacred as our own.
    3. Re:But... the price. by ChrisMaple · · Score: 1

      AMD has years of huge losses to make up for. Don't expect them to get a penny less than they can.

      --
      Contribute to civilization: ari.aynrand.org/donate
  23. Not quite by Cthefuture · · Score: 3, Informative

    All the FX chips are Socket 939. The only chips that are socket 940 are Opterons. The Athlon-FX chips are basically Socket 939 Opterons.

    Socket 939 is for the single CPU systems (including multiple cores).

    Socket 940 is for multiple CPU systems (ie. Opteron).

    --
    The ratio of people to cake is too big
    1. Re:Not quite by Tumbleweed · · Score: 1

      Also - single-core Opterons are coming out in Socket 939.

    2. Re:Not quite by pantherace · · Score: 1
      All the FX chips are Socket 939.

      Incorrect, the early FX chips were 940.

    3. Re:Not quite by Anonymous Coward · · Score: 0

      Yeah but you can't get those any more so... What's you're point?

    4. Re:Not quite by Anonymous Coward · · Score: 0

      the point is that he wants to be right on an internet forum :p

    5. Re:Not quite by Anonymous Coward · · Score: 0

      He must be new here.

  24. No, mebibytes by theefer · · Score: 2, Informative

    MB: Megabyte (1000*1000=10^6 bytes)
    MiB: Mebibyte (1024*1024=2^20 bytes)

    --
    theefer
  25. Amazing translation... by Anonymous Coward · · Score: 3, Funny

    The memory controller has also been tweaked to support mismatched memory module sizes - something some enthusiasts have been crying out for

    I wonder how AMD was able to translate fanboi-talk from

    Wah Wah! Make it faster! Make it faster!

    to

    Please tweak your memory controller to support mismatched memory module sizes...

    It boggles the mind!

  26. "The fastest x86 processor to date..."? by Anonymous Coward · · Score: 0

    Uh, well see there's a slight error there. There's a little clue in the name AMD 64.
    A clue that may suggest it is in fact an x64 chip and not an x86. Go on, see if you can spot the clue there.

    1. Re:"The fastest x86 processor to date..."? by Anonymous Coward · · Score: 0

      x86 is the instruction set.

      64 (not x64) is the number of bits the processor can handle at a shot.

      They are not related.

  27. Athalon to the future! by mister_llah · · Score: 1

    The only problem with it is it needs 1.21 gigawatts of electrical power to generate that kind of speed, and the only thing that can generate that much electricity... is...

    A BOLT OF LIGHTNING!

    --
    MoM++ - A Classic Expanded - [Master of Magic 1.5]
    http://mompp.sourceforge.net/
    1. Re:Athalon to the future! by ID000001 · · Score: 3, Informative

      Then Intel will need two bolt of lightning.

    2. Re:Athalon to the future! by karnal · · Score: 1

      What about Mr. Fusion?

      --
      Karnal
    3. Re:Athalon to the future! by PhaxMohdem · · Score: 1

      One bolt of lightening or two, AMD still has the integrated flux capacitor instead of having a seperate southbridge to reduce temporal latency.

      --

      The Property of One's : "The Oneitude is directly proportional to the Colditude of the one." - S.B.

    4. Re:Athalon to the future! by Anonymous Coward · · Score: 0

      I believe you mean.... 1.21 jigawatts!

    5. Re:Athalon to the future! by ID000001 · · Score: 1

      I don't get it, How is this informative and insightful?

  28. Overcloking ? by Anonymous Coward · · Score: 0

    I read fast through Tom's Hardware' and Xbitlabs' reviews, they dont say a word about overcloking.
    Having read 3 franch-language tests this morning I can tell you the results are mixed.

    1. 3050 MHz
    http://www.hardware.fr/articles/576/page3.html

    2. 3010 MHz with good air cooler
    http://www.presence-pc.com/tests/Test-AMD-Athlon-6 4-FX-57-314/3/

    3. 2991 MHz
    http://www.clubic.com/afficher-en-plein-ecran-1346 67.html

    The weather is hot here these days and they had just 2 days to do the testing under the NDA, so YMMV

    1. Re:Overcloking ? by bhtooefr · · Score: 1

      Tech Report got it to 3GHz unstable (without FSB adjustments, the next step from 2.8 is 3.0, IIRC) on a Thermaltake cooler - they suggested trying it with a different cooler, though...

      http://techreport.com/reviews/2005q2/athlon64-fx57 /index.x?pg=14

    2. Re:Overcloking ? by SebNukem · · Score: 1

      Having read 3 franch-language tests this morning I can tell you the results are mixed... Indeed!

  29. Mostly correct, except for the Socket 940 by Lonewolf666 · · Score: 1

    Socket 940 is for Opterons, not Athlon64-FX chips. It allows additional HyperTransport links (IIRC 3 more than Socket 939) for connection to other CPUs. Thus, only the Socket 940 is suitable for multiprocessor machines. This comes at the expense of requiring a more expensive six-layer mainboard, while a four-layer mainboard will do for Socket 939.

    Other differences are in supported memory. Socket 940 systems require registered RAM and will support ECC. Socket 939 systems use "normal" RAM and will not support ECC.

    Bottom line:
    Socket 940 systems are more expensive, but only Socket 940 will give you access to certain "high end" features.

    --
    C - the footgun of programming languages
    1. Re:Mostly correct, except for the Socket 940 by MechaStreisand · · Score: 1

      Mostly correct, except that 939 systems support ECC just fine. But you're right about them requiring unregistered RAM.

      --
      Disclaimer: IANAL. This post is, however, legal advice, and creates an attorney-client relationship.
    2. Re:Mostly correct, except for the Socket 940 by Lonewolf666 · · Score: 1

      Mostly correct, except that 939 systems support ECC just fine
      You are right. I must have confused something there.

      --
      C - the footgun of programming languages
  30. when do we get out damn socket 939 semprons by Anonymous Coward · · Score: 0

    grrr, thats what i'm waiting for, i may just buy a 3200+ venice instead.

  31. My head hurts by el_womble · · Score: 3, Interesting

    I'm not exactly an Apple Fan boi. I understand that they have their flaws etc but the move to x86 still doesn't fit right in my head and this news just crystalizes it for me.

    In the past, we could all group around the fact that it was impossible to really tell if PPC was actually any good when compared to x86. We knew games sucked, but were confident in the fact that the desktop apps we owned 'felt snappier' and that we were more productive as a result. Now we're being told that actually x86 is probably about as good as PPC, and in the future it will be better. Thats fine. As long as I'm running on the best hardware for my Mac and I don't have to start waiting for the x86 version of my favourite apps to become available I don't care. But the problem is that we all know that Intel isn't the best at making x86 chips. So when I buy my 'Intel inside' Mac, I'm going to know that Joe 'AMD' Linux, with his fancy new 128bit, 1024 core, $15 AMD can, and will, actually toast my Mac, no questions asked, with verifiable benchmarks to prove it and I'm powerless to do anything about it without breaking the EULA. That sucks.

    Bottom line: Apples hardware should be the best platform to run OS X. If that means using AMD, I want AMD in my Mac - and at the moment boy do I want one of those suckers in my Mac!

    --
    Scared of flying, pointy things snce 1979!
    1. Re:My head hurts by kayak334 · · Score: 1

      Yeah AMD makes faster chips than Intel. Are you actually going to notice it? No.

      It really doesn't matter.

    2. Re:My head hurts by Doctor+Faustus · · Score: 0, Troll

      I'll notice that the AMDs are cheaper. And I noticed that the 2ghz P4 I've got at work is slower than the 866mhz P3 it replaced.

    3. Re:My head hurts by kayak334 · · Score: 1

      The difference in processor price is only noticable if you are buying OEM components and building your own machine, which you won't be doing since we're talking about buying Macs. Also, some of that price difference is absorbed by the fact that Apple will get bulk discounts on the CPUs. As for the second sentence, you're wrong.

    4. Re:My head hurts by ponos · · Score: 1
      Now we're being told that actually x86 is probably about as good as PPC, and in the future it will be better.
      I can't remember a significant period of time (2 weeks between launches don't count) during which Apple had faster hardware than PCs of similar price. Not during the Motorola era, not during the PowerPC era. As a matter of fact Apple progressively moves towards cheaper PC technologies (I was shocked when Apple dropped SCSI in favor of ATA). From a business standpoint it does make sense. PC components are getting faster and better and most workstation companies were/are unable to compete with reasonable profit margins. AMD is probably unable to provide the volume that Apple requires, although it depends on which segment of the market (low end? high end? Dual core? mobile?) Apple will want to position itself.

      P.

    5. Re:My head hurts by Anonymous Coward · · Score: 0

      How about you get a fucking life? The only way you can differentiate between the high end IA64 CPU is with a benchmark, loser.

    6. Re:My head hurts by Doctor+Faustus · · Score: 1

      How the hell is this a troll?

    7. Re:My head hurts by Doctor+Faustus · · Score: 1

      The customer won't see the price difference explicitly, but that doesn't mean it's not there.

      Yes, I'm quite aware that a 2ghz P4 should be faster than a 866mhz P3, but that doesn't change the fact that this PC is noticably slower. Maybe it's because this PC is a Dell and the P3 it replaced was a no-name beige box. Both are (or were) WinXP Pro with 512MB RAM. I suppose it's possible that this one has a 5400RPM hard drive; it does have on-board video, but I think the old one did, too.

  32. There is no way we're getting an M by Anonymous Coward · · Score: 0

    Even the Apple dev kits have D's in them. M doesn't have enough processing grunt.

  33. Lies! Socket 939 propaganda! by Anonymous Coward · · Score: 0

    No no no! Nothing about socket 940 boards requires registered memory.

    Old Opterons had terrible signal quality. It's the CPU that required registered memory to work reliably. Remember, the CPU has the memory controller logic.

    AMD used this as propaganda to promote socket 939 and divide the market (to the detriment of 940 owners). The newer CPUs they put into all sockets (940, 754, 939) had better memory controllers and did not require registered memory, but AMD twisted this to present the illusion that "if you go 939 you'll not need expensive RAM".

    The reality is "if you buy a newer K8 for any socket you don't need expensive RAM, but if you plan to fill every slot you really should use it anyway because our CPUs have a history of not being able to drive that many slots reliably".

    AMD is screwing the socket 940 market left and right. You'll notice there isn't a 2.8 GHz Opteron to match this FX-57. Further, shithole companies like ASUS are further screwing 940 owners by not providing updated BIOSes to support revE/90nm chips or dual-core. ASUS hasn't updated the stable SK8V BIOS in over 1 year!

  34. Fab 36 will start volume production in 2006 by Thomas+DM · · Score: 1

    AMD Fab 36 is expected to be in volume production in 2006. The new facility is expected to require approximately 1,000 employees, most of them highly skilled engineers and technicians.

    Source: http://www.amd.com/us-en/0,,3715_10023,00.html

    1. Re:Fab 36 will start volume production in 2006 by haggar · · Score: 1

      Thanks for the info.

      Could you tell me what does the numbering mean? Does AMD have already 35 other fabs, and are they anything like the one Dresden?

      --
      Sigged!
    2. Re:Fab 36 will start volume production in 2006 by MikeBabcock · · Score: 1

      The source is always the best place to go for information. AMD's page, above, has various links and pages of information on how and why they're creating this new plant and what the details of its operation are.

      --
      - Michael T. Babcock (Yes, I blog)
  35. another review by mistermark · · Score: 2, Informative

    Well, here's another review:

    tweakers.net

    It's in Dutch but it has some nice benchmarkresults which should be clear to anybody, benchmarked next to an AMD dual-core 4200+ and a AMD 3800+ (and I know for a fact those results are correct cause it's all been benchmarked on my own desk :-))

  36. wheee by DrMrLordX · · Score: 3, Informative

    Check it out. Anandtech has a review as well.

  37. MacTel by Anonymous Coward · · Score: 0

    If only Apple went AMD...

  38. Why don't we just rename Athlon to Alpha by Anonymous Coward · · Score: 0

    and be done with it.

  39. Explanation of engineering and economics by GunFodder · · Score: 2, Informative

    AMD has limited production facilities. These "fabs" produce all of their semiconductor products. The FX57 is currently the fastest chip they produce, and typically yields on faster processors are lower than slower chips. AMD already has commitments to clients to supply certain quantities of slower chips as well. Additionally AMD doesn't want to devalue their slower processors by flooding the market with the fastest chips. These factors all boil down to the single fact that AMD can only produce a limited number of FX57 processors.

    The basic laws of supply and demand dictate that there is an inverse relationship between price and demand. If we assume that AMD has a fixed supply then their goal is to set a price that will match demand to supply as closely as possible. If the price is too high then some of their chips will go unsold, and AMD will either have to stockpile or cut the price. If the price is too low then AMD will lose profit and have a backlog of buyers.

  40. Choice by Anonymous Coward · · Score: 0

    Bottom line: Apples hardware should be the best platform to run OS X. If that means using AMD, I want AMD in my Mac - and at the moment boy do I want one of those suckers in my Mac!

    And that's why I don't run a mac; so I can have choices.

  41. Pay more attention, your head will hurt less by frankie · · Score: 1

    Apple's transition will start the LAPTOPS, where Intel is strongest. Centrino/Pentium-M is generally equal to AMD's stuff and has lower power consumption.

    On the desktop, G5 will stick around until 2007. Let's wait and see what Intel has to offer at that point before you get so stressed about it.

  42. Intel makes chipsets. by Anonymous Coward · · Score: 0

    AMD, with such exceptions as the 750 and 760MP, does not make chipsets.

    Apple needs a chipset for their powerbook line. They need something that will run at low power, with whatever CPU they choose. Intel makes both a low-power, portable friendly CPU, and a chipset for that CPU. AMD makes the Turion, a good CPU, but they rely on VIA (and SiS, nVidia) for their chipsets. They do not produce them in-house.

    That is why Apple chose Intel.

  43. Until the Pentium M and D are out in full force. by Ayanami+Rei · · Score: 1

    Then the competetion will be tight again.

    --
    THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
  44. ridiculous by edxwelch · · Score: 1

    That's a totally ridiculous assumtion. If AMD thought for a moment that all you have to do is increase capacity to create marketshare they would have done it long ago.
    No, the parent poster was closer to the truth. People buy Intel due to ignorance of processors and slick marketing by Intel and OEMs. Also, bear in mind for light weight laptops Intel actually have a superior product.

  45. Plenty of room for an FX-59... by TwoTailedFox · · Score: 1

    Most have managed to get this to overclock stably to 3GHz, which opens up the possibility of an FX-59 later on.

    I'd really like to see the following come out for AMD:

    FX-59
    FX2-59

    FX2 would be the dual-core equivalent of the FX, hopefully giving as big a boost to multimedia, as to single-threaded applications. And I'd like to see both offer 2MB of L2 Cache per core... and if they can add another 2MB of L3 Cache (shared on the Dual-core chip), that'd be kick-ass.

    --
    ~The TwoTailedFox posts again....
  46. A thought by Cedric+Tsui · · Score: 1

    So right now, most users only have 1 processor intensive application running, so two symetric cores are not overly useful.

    I wonder if the motherboard companies can make MBs that have 2 sockets, one for the current generation of processors and one for the last generation. That way, when you upgrade PCs, you can move your old processor and ram into the computer to act as a background slave.

  47. 10% of AMD's market by alexhmit01 · · Score: 1

    This Apple is small is a bit excessive. It is 2%-3% of new computer sales worldwide. That's the eighth largest computer manufacturer. In other-words, there are only 7 accounts larger than Apple to fight over.

    AMD supplies what, 20%-30% of the CPU market? Meaning grabbing Apple's share would be a 10% increase in market-share. Given the high fixed-cost, low variable costs of chip manufacturing, that could be HUGE for AMD.

    In the OS Market, Microsoft TROUNCES Apple. In the hardware market, only real Dell and HP TROUNCE Apple, making them a viable customer.

    Alex

  48. Save yourself $900, overclock instead by PRlME · · Score: 1

    Hmmm "2.8GHz on a 90nm chip, SSE3 enabled" for... $1031!! My Venice 3000+ runs at 2.8GHz on a 90nm chip, SSE3 enabled for....$147. You do the math. Thanks AMD! :) On stock retail cooling too, absolutely beautiful..

  49. Re:Until the Pentium M and D are out in full force by whorush · · Score: 1

    the PM is out in full force. its doing quite well. although, i still think the turion (25W/35W) is better. remember that turion integrates the memory controller into the core, so that adds extra power to the core. intel still has a memory controller but its in the chipset, so they get away without using its power in their wattage calculations. to be fair, you should compare the wattage of a turion to teh wattage of the PM and its memory controller on the chipest. the turion also has some other features, blocks buffer overflow viruses, and 64 bit. the pentium D sucks bad compared to AMD's dual core chip. AMD's dual core lead is much bigger than its single core lead. i hesitate to even call teh PD dual core, its two cores, but all they do is share a socket and packaging. amd has real dual core, all in one piece of silicon. much more elegant solution.

  50. We care for this reason: by MtViewGuy · · Score: 1

    Multimedia editing.

    With the proliferation of cheap digital still cameras and MiniDV/MicroDV camcorders, people are now increasingly editing their still pictures and home movies on a desktop computer. Problem is, both of these require substantial increases in computing power with lots of RAM available, so getting a faster computer is a MUST.

    Programs like Corel's Paint Shop Pro and Adobe Premiere take advantage of the multimedia extensions provided by today's latest CPU's, so using the latest CPU's means less time spending waiting for the computer to do your image/movie edits.

  51. While we are at it! by The+Creator · · Score: 1

    The 'k' in "kilo" is still not capitalized!

    --

    FRA: STFU GTFO
    1. Re:While we are at it! by Stalin · · Score: 1

      http://en.wikipedia.org/wiki/Kilobyte

      "A kilobyte (derived from the SI prefix kilo-) is a unit of information or computer storage equal to either 1024 or 1000 bytes. It is commonly abbreviated KB, or informally K."

      I believe it can be written either way. However, let's take a look at some numbers.

      1024kb: Do you agree that would be "1024 kilobits"?
      1024kB: Do you agree that would be "1024 kilobytes"?

      Okay, so we agree that the capitalization of the 'b' is significant. So wouldn't it make since to capitalize the 'k' when the 'b' is capitalized so that it doesn't look re-god-damn-tarded?

    2. Re:While we are at it! by The+Creator · · Score: 1

      No, the k is not capitalized EXLAMATION POINT

      --

      FRA: STFU GTFO