Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Intel Hardware IT Technology

Intel Dual-Core Systems Begin Shipping Monday 231

Posted by CowboyNeal from the see-pee-youse dept.
ThinSkin writes "The wait for Intel's dual-core processor is over, that is if you're willing to fork over some dough for a Dell or Alienware system bundled with the chip. Intel just announced that Monday marks the first day dual-core systems hit the market with Dell's Precision 380 workstation and its next generation Dimension XPS desktop, which start at $2,999. PC Magazine got a chance to play with the XPS system and came away quite impressed."
This discussion has been archived. No new comments can be posted.

Intel Dual-Core Systems Begin Shipping Monday More

Intel Dual-Core Systems Begin Shipping Monday

Comments Filter:

  • Perfect system for that North Pole Wifi Hotspot (Score:5, Funny)

    by Anonymous Coward on Saturday April 16, 2005 @08:43AM (#12254433)
    At least cooling won't be a problem.

  • XPS review (Score:5, Informative)

    by shreevatsa ( 845645 ) <shreevatsa DOT slashdot AT gmail DOT com> on Saturday April 16, 2005 @08:43AM (#12254440)
    Is just that -- a review of Dell's XPS Gen 5, rather than a review of Intel's dual core, actually. Still,I guess there's a bit about dualcore.

  • HP taking orders for dualcore opterons already (Score:5, Informative)

    by Ledskof ( 169553 ) on Saturday April 16, 2005 @08:44AM (#12254443)
    HP was, and still is taking orders for Dualcore Opterons systems:
    http://theinquirer.net/?article=22553 [theinquirer.net]

  • No thanks (Score:5, Insightful)

    by Quasar1999 ( 520073 ) on Saturday April 16, 2005 @08:46AM (#12254448) Journal
    I still haven't found anything that truely taxes my existing 3.2ghz P4. Games push the video card, not the CPU... I'm sure servers could benefit, but I don't see a major improvement in end user experience for these gaming systems.

    • Re:No thanks (Score:5, Insightful)

      by selectspec ( 74651 ) on Saturday April 16, 2005 @08:51AM (#12254476)
      Don't forget memory bandwidth and disk I/O as bottlenecks for most desktop apps. Obviously network bottlenecks are the most obvious, but disk I/O is a big pain (think about boot time and launching apps). For games and such, increasing memory bandwidth and system bus speeds would greatly improve performance over adding additional cores.

      Dual cores are great for servers and embedded systems, but not sure about typical desktops.
      • Oh man,

        I can't wait to get dual cores on my desktop. And to me the biggest advantages are responsiveness and better multitasking.

        I really dislike how unresponsive my computer gets when I'm doing something computationally intensive, such as maybe ripping a CD.

        I would also love it if my firewall and antivirus protection could be offloaded to another processor.
        • I seriously hope this post was intended to be "+5 Funny" but got modded wrong! Ripping a CD is CPU intensive? Crap, using error correction (slows down the pull from the cd) I can rip in real time to AAC (I use iTunes) and still have AVG running and still browse the web and still have OpenOffice Writer up and still have Gaim running too. All of this on a 1.83ghz single core Athlon XP-mobile.
          • Ripping a CD isn't overly taxing, but on my home system I often run two concurrent Windows 2000 Pro virtual machines on a Windows XP host (see also : VMware) and have one instance decrypting the files off of a DVD onto the hard drive, while the other instance compressing 4G~8G of data in DVD format down to 700M of .avi format.

            That is CPU intensive.
            Not sure the system would even be responsive enough to open a browser if it didn't have HyperThreading.
            • Wow, it takes two seperate windows virtual machines to do that?

              And people are here crying about their anti-virus and firewall sucking up all their cpu power in their windows machines?

              I use winxp to play a game here and there, but for productivity, what the hell, does windows really suck that bad?
          • It wasn't meant to be +5 funny, but maybe it deserved a +5 clueless. I was just trying to make a point about multitasking.

            A more accurate example for me would be that I enjoy playing EQ2 and at the same time I'd like to have the following programs running:

            1) Teamspeak
            2) Symantec Antivirus (not because I really need it while gaming, but because I don't want to turn it on and off).
            3) iTunes (I like music)
            4) My software firewall
            5) I've got two screens and in one screen I browse for quest stuff and read the g
        • If ripping a CD is computationally intensive, I'd suggest making sure DMA mode is on. Being serious now, I think you might be better served by reducing the priority of the encoding program.

          If firewall and antivirus protection take a noticible dent in performance, then maybe it's time to look for more efficient programs.

          I want dual core, but not for just wasting CPU on inefficient programs or to make up for a bad scheduler. I had a dual CPU system a couple years ago, it was very nice despite being five
        • >>I really dislike how unresponsive my computer gets when I'm doing something computationally intensive, such as maybe ripping a CD.

          Or compiling a huge Java or C++ codebase.......

          wbs.
        • Wouldn't fixing Windows' (and to a lesser extent Linux') time-slicing be a much less expensive solution to all of this?

          This problem isn't inherent; it's an OS artifact. FreeBSD has been able to run without X getting choppy at sustained loads of 20 on a single processor at least since the time of the K6. I had it at something like 80 (admittedly with dual Xeons with huge caches) a couple of weeks ago, and the interface was still perfectly smooth (though mozilla turned to a slug).

          Linux seems to be much be

    • Re:No thanks (Score:4, Insightful)

      by way2trivial ( 601132 ) on Saturday April 16, 2005 @08:55AM (#12254490) Homepage Journal
      burn many dvd's?

      Every try to play a game while encoding a dvd?

      watch a different video while encoding a dvd?

      I look forward to it like you can't believe.

      • Re:No thanks (Score:3, Insightful)

        by rpozz ( 249652 )
        So it's worth buying an insanely expensive CPU so you can encode DVDs while doing something else? Unless you do an awful lot of DVD encoding it seems like a waste of money to me. Better spending that money elsewhere.

        The original poster was correct. A >= 3GHz P4 or any AMD64 will be more than enough for normal games as long as you have a decent graphics card. And yes, that includes Doom 3.
        • Well, I was giving what I thought would be more common tasks..

          Yes, I'm transcoding video 99% of the time I'm not sitting in front of my computer..

          I'd prefer to transcode 99% of the time.-- i spend a lot of time in front of my computer.
          • sorry- and I don't intend to buy a firstgen cpu for this purpose, at this time. I'll wait a couple of iterations- today I would buy a 2.4 pIV instead of a 3.6 EE I don't have that kinda budget-

            but I look forward to the development, and buying a first gen dual core after a few more series are out..

        • Re:No thanks (Score:3, Insightful)

          by NutscrapeSucks ( 446616 )
          So it's worth buying an insanely expensive CPU...

          These EE boxes are only so expensive so that Intel can rape the early adopter hotrod crowd.

          When mainstream Dual-Core P4s hit, they won't be that much more expensive than regular P4s. As a longtime smp user, I certainly think it will be worth the extra $50-$100 to get one of these systems.
      • you could do that now and for a hell of alot less than these new chips will cost if your willing to use tech thats a few years old , the athlon-M is not that expensive and neither are twin socket -Socket A motherboards .

        Im rather sure twin 2000+ chips , a duel motherboard and some extra ram will still be less than half the price of one of those and would allow you to burn a dvd , and do whatever else you feel like doing .

    • Re:No thanks (Score:3, Insightful)

      by fshalor ( 133678 )
      PLay a game while burning Mp3's... That will *kill* your single core.

      Do it with the dual core 3.2's and you'll better game performance witlh encoding. Though sliughtly less than with the single core and single task.

      IMO, these aren't a sub for dual procs yet. But there' promising enough to give them a cost challange soon. And ther'e obviously better than single procs for any intensive cpu tasks that a user needs to do (not just wants for the money) while still having their computer usable.

      I just m$ doesn'

    • Don't Worry (Score:3, Funny)

      by Gr8Apes ( 679165 )
      MS already has plans for your second core.
    • I just bought an HP-4250. It's a laser printer that, when the planets align, can pump out 45 pages/minute. The driver for it can bring any machine to its knees though. I've got an A64/3200+ with half a gig of RAM and when I fire off a large print task, I may as well go do something else - there aren't any cpu cycles left to speak of.

      I suppose they gave the task of upgrading the driver to a summer intern or outsourced it to someone who didn't give a damn about HP's reputation. I've considered getting a dual

    • Re:No thanks (Score:4, Insightful)

      by YrWrstNtmr ( 564987 ) on Saturday April 16, 2005 @09:34AM (#12254702)
      I still haven't found anything that truely taxes my existing 3.2ghz P4.

      They said the same thing about the 386/25 way back when. Don't worry, the software will catch up.

    • "I still haven't found anything that truely taxes my existing 3.2ghz P4. Games push the video card, not the CPU... I'm sure servers could benefit, but I don't see a major improvement in end user experience for these gaming systems."

      Indeed. I've found that loading up on memory is far more effective at increasing system responsiveness. I've noticed that Windows is a bit worse at dealing with CPU hogs than Linux, so it probably benefits from dual cores more, but my Linux system is quite modest (P4 2.4 ghz, 1

    • Alright, so this isn't a typical "desktop use", but my 3.2 GHz NWood P4 is much much better at 3D rendering... not games, but things like Maya and Viz (3DS Max adapted for different things). Then again, I'm not the stereotypical desktop user I suppose.

      I know I'll be modded down for this, but my P4 3.2 GHz does this *much* faster than my 1.33 GHz Athlon TBird.

      And remember- you can have TWO instances of Folding @ Home going at once!!!

    • Re:No thanks (Score:3, Informative)

      by NanoGator ( 522640 )
      "I still haven't found anything that truely taxes my existing 3.2ghz P4. "

      In all seriousness, you'd probably see the difference right away when running on a dual. It's not so much about getting things done faster, but getting two done at one time. In Windows, anyway, I noticed a big difference when switching to Dual. If I start an app, for example, Explorer doesn't just sit there waiting until it's done loading. It's still there, ready to go, presumably because it still has CPU resources.

      Dual core i
    • Do you really believes that games couldn't use your CPU to improve the AI of the characters? the physics simulation?
      You know someone is trying to sell add-on cards to handle physic simulation because CPUs are too slow..

      The reason why you see games not pushing your CPU is that you have a 'top' CPU and game manufacturers will not go through the hassle of trying to improve so subbtle things as the AI or the physics for the benefit of a few people: improved visual is much more likely to give good reviews on a
    • I still haven't found anything that truely taxes my existing 3.2ghz P4. Games push the video card, not the CPU...

      Having spent 12 hour days the last week sitting in front of VTune and NvPerfHUD, I would like to strongly disagree.

      Physics, collision detection, and software vertex skinning are all amenable to parallelization.

  • So much for the AMD threats (Score:5, Insightful)

    by hazee ( 728152 ) on Saturday April 16, 2005 @08:46AM (#12254452)
    What a massive co-incidence that Dell threatened (again) to look at supplying AMD chips just days ago. Not.

    I wonder if the threats did them any good, or if Intel have now got so used to the cries of wolf that they called Dell's bluff? Intel probably told Dell to shut the hell up or miss out on the launch.

  • intresting looking on the inside.. (Score:3, Insightful)

    by gl4ss ( 559668 ) on Saturday April 16, 2005 @08:46AM (#12254454) Homepage Journal
    pic [pcmag.com].

    but 4000 bucks.. well, it SHOULD be impressing.

    but seriously though.. it seems like a "thank you mates! were so happy we actually GOT this thing and not having to just do a paper review on your paper launch".
    just check these:
    "
    Subratings (out of 100):
    Video: 100
    Gaming: 100
    Music: 100
    Photo: 100
    "
    ok.

  • 100% for Gaming? NOT! (Score:5, Insightful)

    by Brian Stretch ( 5304 ) * on Saturday April 16, 2005 @08:48AM (#12254458)
    The system doesn't quite hit the perfect 60+ frames per second score in Doom 3 at 1,600-by-1,200, but no single graphics card solution has so far, and 40 fps is still quite playable.

    Yeah, but Athlon 64 SLI graphics card solutions have. Oddly enough, PCMag only directly compares this Intel Pentium EE 840 box with an Intel Pentium 4 EE 3.73GHz box. Any hard-core gamer who buys an Intel dual-core machine to play his SINGLE-THREADED GAMES instead of an Athlon 64 dual video card SLI box is beyond hope. Torch your money responsibly, kids.

    Dell and Intel get 100% from PCMag for "Best Bribes Paid". Geeze.

  • An improvement (Score:3, Interesting)

    by Gr8Apes ( 679165 ) on Saturday April 16, 2005 @08:48AM (#12254464)
    over their regular single CPU offerings for those of us that run multiple apps, but I truly would like to see the real heat/performance numbers, and whether the rumored performance throttling is enabled on these. If not, how bad is the heat generation?

    From what I've read up to now, AMD's solution will outperform Intel's offering with significantly lower heat dissappation, making it a double winner. However, testing shipping units will finally quantify these processors. Can't wait for AMD's unit to ship and get compared.

  • Hit the Market... It's so meaningless (Score:5, Insightful)

    by SuperficialRhyme ( 731757 ) on Saturday April 16, 2005 @08:56AM (#12254493) Homepage
    These are hitting the market but won't be shipped for a few weeks - or so I gather from what I read in TFAs. By "Hit the market" they seem to mean "vendors are taking orders" which - to me - seems meaningless.

    AMD claims not to do this in one of the articles:
    ""'t is important to note that AMD only announces products when we are able to immediately begin shipping for revenue and that we have been shipping dual-core AMD Opteron processor production samples to customers and partners since January,' the statement added."

    I guess we'll just have to see if AMD actually has products available at their release or if they're just doing the same thing Intel seems to be doing here.

  • Sun's been shipping dual-cores for a while now... (Score:4, Informative)

    by Anonymous Coward on Saturday April 16, 2005 @09:17AM (#12254612)
    The UltraSparcIV is a dual-core chip. Been shipping for quite a while now - maybe even more than a year.

    And just like the first UltraSparcs from about a decade ago, it's also fully 64-bit....
    • The ironic thing is though that the Sparc IV gets destroyed by the single core AMD and Intel offerings, not to mention the IBM POWER (dual core) CPUs. For Sun, going dual core was just limitation of damage.

      • For Sun, going dual core was just limitation of damage.

        Actually, Sun is less interested in the Intel marketing game and more interested in throughput. Just wait for their 32-way Niagara next year. Their USIV should be doubling speed this year, too (USIV+).

        • IMO Sun is (also) playing the marketing game with its Niagara thingy. They didn't start to sell Opterons just because they liked to, they did because they can't compete.
          I expect Niagara to be competitive in certain task, and better than the competition in very specialiced task, the jury is just out to decide if these specialiced tasks do matter in reality. And despite what Sun wants me to believe, I don't expect Niagara to magically show superior results on Java Appservers in general.

          • They didn't start to sell Opterons just because they liked to, they did because they can't compete.

            It's more complicated than that, IMO. AMD plays the Intel Marketing Game even better than Intel does. If you look at the Opteron architecture, it's actually very similar to that of the UltraSPARC IIIi (JBUS vs. HT) and light years ahead of the Xeon architecture. For an x86 architecture, the Opteron fits really well into Sun's overall product line and is good for customers who need x86 interoperability.

  • The biggest benefit of dual core... (Score:3, Insightful)

    by JollyFinn ( 267972 ) on Saturday April 16, 2005 @09:22AM (#12254639)
    Is not the single thread performance or how fast can you finish task X.
    Its responsiveness of the system. I'm using A64 3000 and I get annoying stalls on system level.
    The CPU spends time with the backside thread, while I would love it working on UI, there is annoying stall. Multiple CPU:s according to reviews remove those issues. And don't say having 20% higher processor clock speed is going to help, its by simple fact that CPU was just doing something else at a time I would of loved it to handle UI events. Having 2 cores means, that responsiveness of a system is greatly improved, atleast until people write most of their applications to tie up more than one core ;)
    • It is responsiveness of the system, and scalability of tasks.
      Granted, a SMP box doesn't run any faster than a single CPU machine, but run two tasks in parallel and that's where the difference happens.

      Honestly I could care less about comparing this to single CPU machines (because HyperThreading already solves that issue) - the real difference I am interested in is comparing the performance of identical setups, one with a HyperThreaded CPU and the other with one of these new dual core chips.
    • "nice" was meant for this.

      With Linux you can also use "chrt" to specify that some task is "realtime": it will always get as much scheduling as it wants (make sure it will not loop endlessly though).
    • For me, the problem with one process stalling another of the same priority is only visible on Windows. Very high disk usage seems to do the same thing.

      The solution of dual core is like putting in a bigger engine into your car so your car feels less slugish, yet the real solution would be just to oil the axle and change the oil.

    • As someone who has ran dual-processor rigs for years, I can tell you that system responsiveness requires more than just a pair of speedy processors.

      For example, if your hard drive or other I/O devices are being taxed, your entire system will run slowly... which makes that second processor about worthless.

      The biggest benefit of dual processors are when you are running a single-threaded application that only taxes the CPU, or when you are running a pervasively multithreaded application (like, say, video-e
      • Gee, just after I pointed out how much of the problem is the OS, I see this :)

        I have dual Xeons with huge caches, and three of the 4 drives are 15krpm, with the other a 10krpm (all are UW, etc.)

        A few months ago, I stuck in a garden variety cheap ide drive to install for my home machine. I was stunned at the loss of responsiveness . . .

        hawk

  • So what?? (Score:3, Interesting)

    by jarich ( 733129 ) on Saturday April 16, 2005 @09:37AM (#12254717) Homepage Journal
    I've got a dual processor system. They don't cost that much more than a single... you won't pay a premium for a new "dual core" box, but you'll still get all the advantages of a fast responsive system. p. In my opinion, dual cores are for businesses whose rack space is at a premium and gamers with spare money to burn.
    • Dual core is a way for processor manufacturers to get the most performance out of a given die size and power/heat budget. The race to market with the first dual cores, OTOH, is about the hype. It really has nothing to do with rack machines and gamers. When transistor counts double in future processes, do you want 10% more computing capability or 2 processors worth of what you have now? The reason for multicore going forward is very clear. Whether it's compelling to buy one now is not.

  • I can just see the problem (Score:5, Funny)

    by stewwy ( 687854 ) on Saturday April 16, 2005 @09:40AM (#12254735)
    winxp to processor1 : please open this window.
    P1 to winxp: No I'm busy, ask P2
    winxp to P2: open this window
    P2 to winxp: ask the other lazy sod
    winxp to all: please or I'll BOS
    P1 and P2: go on then, we don't care
    user: ?

  • I don't get it (Score:2, Interesting)

    by Dolda2000 ( 759023 )
    What all this "news" about Intel suddenly doing dual-core? Isn't that what HyperThreading has been doing for, what is it, a year now?

    • Re:I don't get it (Score:4, Informative)

      by l3v1 ( 787564 ) on Saturday April 16, 2005 @11:05AM (#12255206)
      I don't get it

      Yes, you obviously don't. Hyperthreading is not in any way like a dual processor or dualcore processor. On a HT processor you still can have threads waiting and not doing anything because another thread which is using e.g. the single FPU that the system has. If two computationally heavy threads want to run, they have to wait for their turn on the single FPU. And that is just one example for HT. On dualcore and dual processor systems you have everything doubled, which is a Good Thing.

      • dual hyperthreaded dual cores :)

        A bit more seriously, a hyperthreading mechanism that allowed the virtual processors to use units from *either* core would be interesting--or just drop the distinction between cores, double the number of execution units, and have four virtual processors . . .

        hawk
        • The issue with that is the heat and die area for clock distribution over what would be simply a much larger HT CPU. The dual core approach avoids massive increases in heat dissipation and die area and essentially introduces synchronisation at the memory interface. This is the first step on the way to clockless CPUs, which is when you will see what you just wished for.

  • Comment removed (Score:5, Funny)

    by account_deleted ( 4530225 ) on Saturday April 16, 2005 @10:56AM (#12255138)
    Comment removed based on user account deletion
  • Remember, these are desktop processors that Intel is releasing. AMD is slated to release dual core OPTERONS. Those are server cpus, much more important than the puny stuff Intel is putting out.
    • Where do you draw the line between desktop and server processors?

      In my experience, servers are rarely limited by CPU power. An Opteron would be better used in a workstation.

      I've used old laptops for web servers (they come with a built-in UPS :), and I hate it when someone tries to make a hard distinction between server and client machines, because there isn't really one. I think it's usually marketing types who do this, to be able to sell twice as many machines.

  • That is, with HyperThreading, cpuid reports 2 virtual processors per package. How will a dual-core system be reported?

    I ask because our calculation-heavy product actually performs worse with multiple threads on an HT system, but speeds up by a factor of nearly 2 on a true multi-CPU system. I know how to test for HT and whether it's enabled, but how will we differentiate HT packages and dual-core packages programmatically?

  • We're Sorry We're unable to process your request. Please check back with us soon to customize this product, or click below to continue shopping Plus the customer service rep was clueless.... http://configure.us.dell.com/dellstore/nodata.aspx ?cs=19&kc=6V411&oc=XPS5PC&x=10&y=11 [dell.com]

  • Why? (Score:3, Insightful)

    by Shaheen ( 313 ) on Saturday April 16, 2005 @07:26PM (#12258264) Homepage
    Someone please answer this question for me: Why? Why does the average consumer need a dual-core processor? For that matter, why does the average consumer need a dual processor configuration at all? Are the people buying $3000 dual-proc G5s just being elitist?

    My point is this. Let's see what the average consumer does with his computer:

    - Surfs the Internet
    - Reads email
    - Watches DVDs
    - Plays music
    - Plays games
    - Maybe does some work

    Note that among all of those things, people rarely do them all at once (and I am not talking about the Slashdot population that does work, listens to music and has browser windows open all at the same time).

    Games are the only taxing item there, and most PC games are not even multithreaded. Gamers that buy the Alienware system are literally just wasting cash.

    On top of all of this, no matter how much you do, your CPU spends quite a bit of time (upwards of 85%) in the idle thread (aside from you SETI@Home people).

    When you put all that together, it makes little sense to buy a dual-core CPU for your desktop PC.

Slashdot Top Deals

Every nonzero finite dimensional inner product space has an orthonormal basis. It makes sense, when you don't think about it.

Close