Slashdot Mirror


Intel in the GHz Game Again - Skulltrail Hits 5 GHz

An anonymous reader writes "Intel's Skulltrail dual-socket enthusiast platform has been making the rounds on the web for half a year or so, but we haven't seen many details yet. TG Daily got a close look at an almost complete prototype, which surely sounds almost like a production ready version, judging from the article. Everything that TG Daily describes sounds like Skulltrail PCs will be very limited in availability and insanely expensive. Intel also has said it has developed 'special' Xeon processors with desktop processor attributes just for Skulltrail. These chips are currently running at a stable 5 GHz."

229 comments

  1. I guess... by cesman · · Score: 4, Funny

    I guess the skulls in it's trail are the heads of AMD execs.

    --
    When the source is open, the possibilities are endless.
    1. Re:I guess... by arivanov · · Score: 4, Funny

      No... They are the skulls of White Bears who have fallen through the ice because the water in the arctic got warmed up too much by the water cooling kit this beast requires to operate.

      --
      Baker's Law: Misery no longer loves company. Nowadays it insists on it
      http://www.sigsegv.cx/
    2. Re:I guess... by Hal_Porter · · Score: 3, Funny

      No... They are the skulls of White Bears who have fallen through the ice because the water in the arctic got warmed up too much by the water cooling kit this beast requires to operate. You could probably render photorealistic White Bears in realtime on this beasty. And do the AI and physics too. Certainly good enough to use them as enemies in a arctic themed FPS.

      So it all works out in the long run.
      --
      echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
    3. Re:I guess... by darkwhite · · Score: 1

      (Score: 0, Sad)

      --

      [an error occurred while processing this directive]
    4. Re:I guess... by moogs · · Score: 0

      arctic themed fps? Gordon Freeman vs. White Bear in Half Life 2 Episode 3... maybe use a portal gun and portal the bear's white shit on it's white head shithead bear

      --
      I have bad karma. What do I care what you think?
    5. Re:I guess... by Anonymous Coward · · Score: 0

      White Bears? What are they part of an underground White Bear supremist clan? Sounds like a really bad Native American stripper name.

    6. Re:I guess... by Hal_Porter · · Score: 1

      Oddly enough, that might actually happen

      http://www.computerandvideogames.com/article.php?id=163601

      --
      echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
  2. Also, Duncan Hill Coffee by User+956 · · Score: 4, Funny

    Everything that TG Daily describes sounds like Skulltrail PCs will be very limited in availability and insanely expensive.

    Obviously, it's the only architecture hand-designed by Dethklok.

    --
    The theory of relativity doesn't work right in Arkansas.
    1. Re:Also, Duncan Hill Coffee by dino2gnt · · Score: 2, Insightful

      Yes, but is it blacker than the blackest black times infinity?

      --
      Future events such as these may affect you in the future!
    2. Re:Also, Duncan Hill Coffee by myz24 · · Score: 1

      Coffee? No, I'm selling toner.

  3. Excessive? by Vexor · · Score: 2, Interesting
    So this is most likely targeted at gamers because we all know that games generally speaking are the most intensive software ever run on a PC. As far as I know though there is no game on the market that requires anywhere near that kind of horsepower. Not that more is a bad thing but I'm running a Intel E6750 at 3gigs and even that rates a 5.9 on the Vista-Meter (not that Vista is a "reliable" benchmark).

    On the other hand...will this be out in time for Crysis?

    --
    ~Vexed and loving it!
    1. Re:Excessive? by drix · · Score: 4, Insightful

      we all know that games generally speaking are the most intensive software ever run on a PC Not even close. Games, after all, run in realtime. There are many, many applications out there that have no problem pegging top-of-the-line hardware for hours on end: DV editing, raytracing, scientific computing. In fact, the whole reason I'm posting this is because I'm waiting for my PC to solve a big math problem :-)
      --

      I think there is a world market for maybe five personal web logs.
    2. Re:Excessive? by physicsboy500 · · Score: 0

      This isn't specifically targeted towards gamers.

      While a lot of games won't reach the point that they need a clock that fast, Applications that use thousands of arithmetic or other basic CPU-based operations over long periods of time will easily use this power.

      I'm guessing these processors will show up more frequently when associated with cad development / computer animation / data mining / compiling / ect. while there may be gamers like this that mistakenly think a high number will help them with everything, it is not the direction this processor is taking.

      --
      The original generic sig.
    3. Re:Excessive? by everphilski · · Score: 2, Informative

      I bought a top of the line processor and quadrupled my RAM the beginning of last year - not for video gaming (although it sure didn't hurt, I play occasionally not very hardcore anymore) but to do scientific computing for my thesis. I did a 6DOF model of a guided bullet, with this spiffy guidance model. 500 monte carlo runs took about 2 hours. I needed to do a ton of sets. All in all, my entire master's dissertation worth of sets took about a month worth of running 16 hours a day on a dual-core machine. And of course I had to do a lot of pre-emptive runs to determine the domain of my problem, etc.

      I still do a lot of scientific computing at home - 6DOF's, playing with CFD, etc. There are plenty of 'hobbyists' out there who can keep a CPU pegged more often than its not ...

    4. Re:Excessive? by Luyseyal · · Score: 3, Insightful

      Not to mention running something like World Community Grid. I love using my idle processor time to tackle AIDS, Cancer, Muscular Dystrophy, Dengue, etc.

      -l

      --
      Help cure AIDS, cancer, and more. Donate your unused computer time to worldcommunitygrid.org. Join Team Slashdot!
    5. Re:Excessive? by Anonymous Coward · · Score: 1, Informative

      Not that more is a bad thing but I'm running a Intel E6750 at 3gigs and even that rates a 5.9 on the Vista-Meter (not that Vista is a "reliable" benchmark). Just to let you know that you could have a 5GHz system and you still wouldn't go over 5.9. The scale only goes from 1 to 5.9. it says it on the MS site.

      http://www.microsoft.com/windows/products/windowsvista/buyorupgrade/experienceindex.mspx

    6. Re:Excessive? by Svet-Am · · Score: 1

      flight simulator x would certainly use all of that hardware. how well is a different matter, but FSX is upward scalable for memory, CPU, and GPU.

      --
      [move .sig! for great justice, take off every .sig!]
    7. Re:Excessive? by Tim+C · · Score: 1

      Not only that, but generally speaking the single most demanding pat of a game is the graphics, and we have dedicated GPUs for that.

      I agree with you; games are power-hungry, but by no means the most power-hungry things you can do with a PC. Mind you, I'm weird - I've actually done proper scientific numerical simulation work (and had to leave it running overnight to finish). I've also done video transcoding, and while that doesn't take as long it wasn't quite real-time last time I did it, so there's definitely still room for improvement.

    8. Re:Excessive? by LWATCDR · · Score: 1

      Most people will say games because they are just about the only programs they use that they have to wait on.
      But yes your correct. You did leave out data mining and a few other applications.
      What is amazing is the size of problems that we are now willing to tackle with desktop hardware.

      --
      See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
    9. Re:Excessive? by rrohbeck · · Score: 1

      Gamers are probably the only users that are more or less immune to price/performance considerations.

    10. Re:Excessive? by fitten · · Score: 1

      Yup... we've had code that ran for 7+ days on 1024+ processors to produce one dataset before. What sucks is when on the 6th day of running, you get a segmentation fault or something :(

      I think that one run used up a gamer's quota of CPU power for at least 14 days of game play ;)

    11. Re:Excessive? by p3d0 · · Score: 3, Funny

      we all know that games generally speaking are the most intensive software ever run on a PC Not even close. Games, after all, run in realtime. So? That's because they are tuned that way. I haven't played this sort of game in a while, but in the day, I remember you could tune the game for your system, and it would take 100% of your CPU, GPU, ALU, FPU, and any other U you wanted to throw at it. How long it runs is entirely irrelevant.
      --
      Patrick Doyle
      I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
    12. Re:Excessive? by Sta7ic · · Score: 1

      Yes, this will be targeted at gamers. The same gamers who spend hundreds of millions of dollars constructing supercomputers for "running games". We needn't consider molecular or weather simulations, modeling nuclear explosions, or solving Go.

      Bigger fish to fry than those in BioShock.

    13. Re:Excessive? by Vexor · · Score: 1

      I know it only goes to 5.9 that's my point! :) That's 2gigs over my CPU. Saying games are generally the most intensive is a bit inaccurate. I was trying to imply the most "common" use.

      --
      ~Vexed and loving it!
    14. Re:Excessive? by PitaBred · · Score: 2, Funny

      Get a faster machine. My laptop will transcode from a DVD to H264 at right near real-time (sans ratio changes), and it's nothing terribly special.

    15. Re:Excessive? by bronzey214 · · Score: 1

      Your CPU only hits 5.9 because that's the top of the scale right now for the Window's experience meter.

      The base score levels at a glance:

      1.0 - 1.9
        Basic performance. Productivity applications, IM, web, email, simple games - like Solitaire, educational games.
        Minimum specification needed to run Windows Vista(TM).

      2.0 - 2.9
        Improved responsiveness. Same applications as a base score of 1.0 - 1.9.
        PCs will run Windows Vista but in most cases will not be Aero capable.

      3.0-3.9
        Aero graphics, Media Center with standard definition TV, basic graphical games, basic performance while running high-end graphical games.
        Minimum specification needed to run Windows Vista Premium features, including the new Aero user interface.

      4.0 - 4.9
        Snappy performance, high definition video, high resolution monitors, dual monitors.
        Very good performing PCs.

      5.0 - 5.9
        Fast moving games with amazingly rich graphics, 3D modeling, high-end multimedia and high performance applications.
        Top end of the PC market for the Vista time frame.

      6.0
        Base scores of 6.0 and higher are not defined yet. They will be defined when the time comes and new innovations in hardware allow new capabilities. From past experience, it is expected this will happen at a rate of once every 12-18 months.

      from the Vista blog (http://windowsvistablog.com/blogs/windowsvista/pages/458117.aspx)

    16. Re:Excessive? by Anonymous Coward · · Score: 0

      I can make a simple loop that will eat 100% of my processor cycles. Gaming, unlike most other forms of use, will max out the cpu, the video card, and a few other components all at the same time. Trust me, running Crysis (or HL2 or Quake 1 with vsync turned off) involves more computations and power draw than your puny mathematics.

    17. Re:Excessive? by Anonymous Coward · · Score: 0

      FSX is also ugly as a pig and shit software. I'm not sure it really matters that it'll (easily) eat 5G. It doesn't really matter much if you wash a brick of shit.

    18. Re:Excessive? by MojoStan · · Score: 1

      we all know that games generally speaking are the most intensive software ever run on a PC Not even close. Games, after all, run in realtime. There are many, many applications out there that have no problem pegging top-of-the-line hardware for hours on end: DV editing, raytracing, scientific computing. Maybe by "PC" the GP meant "desktop PC" and not "workstation." Of course, the Skulltrail platform is just workstation hardware (dual Xeons, ECC FB-DIMMS) with some modifications for uber-gamers that have more money than common sense.

      Isn't a standard workstation (dual workstation CPUs, ECC RAM, workstation graphics card) more appropriate for DV editing, raytracing, and scientific computing? Isn't "desktop" hardware (fast single desktop CPU, faster non-ECC DDR2, "gamer" graphics card/cards) more appropriate for gaming?

      That's why I'm confused about the existence of the Skulltrail/V8 platform. Does anybody buy this shit? I thought V8 (Skulltrail's predecessor) was only created in response to AMD's ridiculous Quad FX platform (dual CPU for desktops). I thought Quad FX was only created in response to Intel's quad-core desktop CPUs, since AMD only had dual-core at the time (the only way AMD could "match" Intel's quad-core desktops was to create a dual-CPU platform for the desktop).

      Intel's fast single-CPU quad-core desktops outperformed Quad FX in most important benchmarks anyway. I thought Intel's V8 was just a demo to rub it in. Now that both Intel and AMD have quad-core CPUs, why do either of them need dual-CPU desktops that use workstation hardware? Uber-gamers are better served with single-CPU enthusiast desktop platforms (Core 2, Athlon). Other power-users are better served with dual-CPU standard workstation platforms (Xeon, Opteron).

      --
      TO START
      PRESS ANY KEY

      Where's the 'ANY' key? I see Esk, Kitarl, and Pig-Up...

    19. Re:Excessive? by scheme · · Score: 2, Insightful

      Im pretty sure games stress the whole system overall a lot more than any application im aware of. math problems and ray tracing and DV editing if im not mistaken are CPU exclusive operations. Im not an expert but high end graphics cards are more powerful than cpus, even if they are specialized. i cant think of any other application that will stress the CPU, GPU, RAM, HDD and everything else to 100% other than games.

      You are mistaken. Take particle physics simulations for example. The system might be downloading a 10GB dataset to do the next simulation while it's working on simulations of a detector which involves working with the current dataset. The download would max out your net connection while the simulation work would max your cpu and require something like 2-3GB of ram. The two activities are probably generating a decent i/o load as well.

      Same deal with audio or video processing, if you're streaming a video or audio source or two from an array, processing it and writing it back, well that's pretty much using everything. Given a raw video stream can be about 20MB/s, you're generating about 40MB/s of read/write per stream and you might be working with a few streams if you're trying to overlay two video sources or something. That's significantly more activity than a game will produce.

      --
      "When you sit with a nice girl for two hours, it seems like two minutes. When you sit on a hot stove for two minutes, it
    20. Re:Excessive? by sssssss27 · · Score: 1

      Games, after all, run in realtime.

      Which is exactly why more processing power is important. You can't wait hours for it to render the next scene. Then again it really is just a semantics issue on what he meant by intensive. You could always just throw your computer into an infinite loop and peg the processor as well.

    21. Re:Excessive? by Anonymous Coward · · Score: 0

      In fact, the whole reason I'm posting this is because I'm waiting for my PC to solve a big math problem :-)


      Well good news everyone!

      Once you have a Skulltrail system, you should be able to balance your checkbook in halfthe time!
    22. Re:Excessive? by Anonymous Coward · · Score: 0

      The 2,3 Turing Machine has already been proven Universal, dummy.

    23. Re:Excessive? by dbIII · · Score: 1
      In some cases make that weeks per processing job on quite a few nodes at once. What is nice is that gear that really spins off from the mass market for gamers gets shoved into racks at a relatively cheap price to do this sort of stuff. Sun's new multicore machines are nice but you can get what is effectively a few gaming Godboxes for less to get more done with CPU bound stuff.

      One bizzare thing is I ran effectively the same job on an Intel 8 CPU machine and an AMD 8 CPU machine of very similar specs - there was a difference of a few minutes in the six days it took to run. The margin of error due to network I/O access and the slight differences in the parameters would be a few hours - so even over that time I couldn't really tell the difference. It's all good.

    24. Re:Excessive? by Nullav · · Score: 1

      Mine goes to 11.

      --
      I just read Slashdot for the articles.
    25. Re:Excessive? by Anonymous Coward · · Score: 0

      You've forgotten GPU, but a lot of intensive computing projects are now being written to utilise the increasingly-powerful GPUs (especially distributed projects like folding@home)

    26. Re:Excessive? by afroborg · · Score: 1

      You left out the GPU, though if you're visualising the output at the same time you could probably give it a work out too. Probably talking something like a Quadro though rather than a gamer card.

      --
      my sig could kick your sig's arse...
    27. Re:Excessive? by Tom+Womack · · Score: 1

      Generally these absolute-top-of-the-range chips don't go into supercomputers; for supercomputing, you do the price-performance calculation and find that more, slower chips suffice.

      Though there are some applications where you need very large memories on the nodes, and if you're buying 32GB of RAM the difference between two 3.3GHz quad-core processors and two 2.4GHz quad-core processors is lost in the noise.

    28. Re:Excessive? by dbIII · · Score: 0, Offtopic

      I did point out they were different but I suppose nasty trolls with a logout link trick are not expected to be have high levels of reading comprehension.

    29. Re:Excessive? by TheLink · · Score: 0, Offtopic

      Oh it's you again.

      I did read your post correctly. Maybe you didn't read mine correctly.

      I suppose I have been trolled, and your replies should really be beneath my threshold.

      Oh well, have fun nonetheless :).

      --
    30. Re:Excessive? by Glock27 · · Score: 1
      Not even close. Games, after all, run in realtime. There are many, many applications out there that have no problem pegging top-of-the-line hardware for hours on end: DV editing, raytracing, scientific computing. In fact, the whole reason I'm posting this is because I'm waiting for my PC to solve a big math problem :-)

      You've conflated two different issues. Games are intended to be realtime, but often aren't. Since they're soft realtime, this is characterized by framerates dropping below some arbitrary rate be it 120/s, 60/s or 30/s. Most science/math applications don't have the realtime constraint but utilize no more system resources than the game does...both are using 100% of CPU, memory bandwidth, or both. For serious flight simulators, the frame rate is a hard realtime requirement and not meeting it is considered a fatal error.

      In fact, since games tend to use the massive compute resources on the GPU more intensively, I'd still argue they're the most intensive applications - and their soft realtime nature only adds to that opinion. Game programming issues certainly make it more difficult to achieve acceptable performance, due to that pesky realtime requirement.

      --
      Galileo: "The Earth revolves around the Sun!"
      Score: -1 100% Flamebait
  4. Traslation by king-manic · · Score: 5, Insightful

    These chips are currently running at a stable 5 GHz. A practical translation:

    It will be 20% faster, 200% hotter, needs a 300% nosier fan, consumes 500% as much power.
    --
    "There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy."
    1. Re:Traslation by Anonymous Coward · · Score: 0

      Not to mention that it spins in an empty loop at 5GHz, as soon as it needs to fetch a byte from memory, performance drops.

    2. Re:Traslation by rubycodez · · Score: 0, Troll

      and will suffer a major component failure in 20% of the time. On the plus side, this will make the startup of a jvm much faster to make java based applications nearly useable.

    3. Re:Traslation by TemporalBeing · · Score: 1

      These chips are currently running at a stable 5 GHz.
      A practical translation:
      It will be 20% faster, 200% hotter, needs a 300% nosier fan, consumes 500% as much power.
      And yet it will only deliver 33% of the performance.
      --
      Truth is like the sun. You can shut it out for a time, but it ain't goin' away. - Elvis Presley (source: imdb.com)
    4. Re:Traslation by Anonymous Coward · · Score: 1, Funny

      On the plus side, this will make the startup of a jvm much faster to make java based applications nearly useable.

      Hey rubycodez; is that funny because Java is almost as slow as ruby?

    5. Re:Traslation by Anonymous Coward · · Score: 0

      A small physics nitpick. As long as the chip is doing negligible work (and pushing electrons to other components at high velocity doesn't really rank up there that much), heat produced must be equal to power consumed.

    6. Re:Traslation by Anonymous Coward · · Score: 0

      But you'll still need 16 GB of RAM to simultaneously run without bottlenecking each of the different versions of the JRE your applications require.

    7. Re:Traslation by RzUpAnmsCwrds · · Score: 1

      It will be 20% faster, 200% hotter, needs a 300% nosier fan, consumes 500% as much power.


      Don't you mean 500% hotter? For any practical purpose, heat produced in an IC is equivalent to the amount of power it draws.
    8. Re:Traslation by horsman · · Score: 1

      actually java is about 30 times faster than ruby.

    9. Re:Traslation by Gideon+Fubar · · Score: 1

      the extra 300% will be to power the northbridge, bus, ram and the industrial cooler needed to prevent it melting.

      --
      http://www.xkcd.com/354/
    10. Re:Traslation by Sebastopol · · Score: 2, Funny

      power = heat

      if you could make something require 500% more power but convert 200% more energy to heat (ignore photonic emissions), you'd have yourself a nobel prize.

      i'm just sayin.

      --
      https://www.accountkiller.com/removal-requested
    11. Re:Traslation by dbIII · · Score: 1

      needs a 300% nosier fan

      That's pretty rough - by that stage the fans become stalkers.

    12. Re:Traslation by Mattsson · · Score: 1

      It will be 20% faster, 200% hotter, needs a 300% nosier fan, consumes 500% as much power. Doesn't the "500% as much power" kind of implicate "500% hotter", since it's simply a tweaked top of the line selection of a cpu rather than a completely separate cpu-design? =)
      --
      /.Mattsson - My native language is not English, so please don't whine over linguistic errors. (That's lame anyway...)
    13. Re:Traslation by Ddalex · · Score: 1

      but 500% heat doesn't translate to 500% hotter (heat is not temperature)

      now, give me my nobel

      PS> captcha: cannabis.. how desirable

      --
      Carefully crafted sig.
    14. Re:Traslation by FuzzyDaddy · · Score: 1

      Given the same thermal dissipation, 500% more heat means about 500% more temperature rise above ambient.

      --
      It's not wasting time, I'm educating myself.
    15. Re:Traslation by Glock27 · · Score: 1
      It will be 20% faster, 200% hotter, needs a 300% nosier fan, consumes 500% as much power.

      I know it'd require reading the article ;-) but the stated power per CPU is 100-120 W. That's a far cry from 500% over current processors, isn't it? BTW, heat scales directly with power IIRC.

      --
      Galileo: "The Earth revolves around the Sun!"
      Score: -1 100% Flamebait
    16. Re:Traslation by Bee1zebub · · Score: 1

      But the thermal dissipation will not be the same: given that as far better fan will be used, and possibly water cooling, more haet will be dissipated, so the temperature of the chipset itself will not rise by so much.

    17. Re:Traslation by rubycodez · · Score: 1

      no, it's funny because java is such a pig even though its a compiled language. Even the objective COBOLs are faster than Java when it comes to actual oo tasks. And for performance intensive tasks, Ruby deals with compiled libraries very well, thanks.

  5. Re:But... by mrbluze · · Score: 0, Flamebait

    Everything that TG Daily describes sounds like Skulltrail PCs will be very limited in availability and insanely expensive To be purchased using US taxpayer money out of the intelligence budget, no doubt. If only they used these things for good instead of evil.
    --
    Do it yourself, because no one else will do it yourself. [beta blockade 10-17 Feb]
  6. Yes by Anonymous Coward · · Score: 0

    Just won't have any cool processor extension support until 2010 when the technology is a bit more obsolete.

  7. Insanely expensive... by Anonymous Coward · · Score: 5, Insightful

    And will be obsolete in a year. Honestly, who spends thousands of dollars every year for the most advanced stuff? Even if you did have a Skulltrail, the rest of you system would bottleneck it. 3 8800GTX's would be the bottleneck, 8GB's of the fastest DDR3 ram would bottleneck, and your harddrive would bottleneck too. The only thing Skulltrail gives you is bragging rights.

    1. Re:Insanely expensive... by tilandal · · Score: 1

      Intel caters to high end users because there is no profit in low end users. The gains made in CPU power in the last 5 years have had little effect on your average user. Most of the computers sold today don't do anything that an old PIII couldn't do back in 1999. People surf the internet, read email, use productivity apps, listen to music, watch movies, maybe some photo albums. That covers the vast majority of users. The last 10% of science, engineering, gaming, and server apps is where Intel makes its bread a butter.

    2. Re:Insanely expensive... by Chandon+Seldon · · Score: 1

      This is only talking about a 20% clockspeed jump over a current overclocked workstation platform. Sure, that won't translate into a 20% performance jump overall, but it'll still be a performance jump. I'm not really up for spending an extra $1000 for an 8% framerate increase, but for those people who like that plan this is a perfectly reasonable product. Further, it sets the bar higher for future products - and as a technology enthusiast I can always get behind rasing the bar.

      --
      -- The act of censorship is always worse than whatever is being censored. Always.
    3. Re:Insanely expensive... by Kaenneth · · Score: 1

      Game Developers can would buy it.

      Insanely expensive now will be standard consumer performance over a major game platform development lifecycle.

    4. Re:Insanely expensive... by neersign · · Score: 1

      And will be obsolete in a year. Honestly, who spends thousands of dollars every year for the most advanced stuff?

      to answer your question, the people at Overclockers Forums, Xtremesystems Forums, and any other person who tweaks their computer for fun. It's a cheaper hobby than boating, for example.

  8. Imagine a.... by HalifaxRage · · Score: 1, Funny

    Just imagine a Cluster of nerds waiting in line for Beowulf!

    --
    bomb the us up set someone
  9. Yes, but by paranode · · Score: 4, Funny

    The silicon pathways are provided by Monster Cable.

    1. Re:Yes, but by Cecil · · Score: 4, Funny

      And they're made out of superconducting adamantium. As a result, all games played on these processors will have higher quality storylines. It's a little known fact that copper causes destructive interference with the story's sine wave.

    2. Re:Yes, but by flappinbooger · · Score: 1

      And all the cables are super duper oxygen free copper on individual plinths! YES!

      --
      Flappinbooger isn't my real name
    3. Re:Yes, but by Copid · · Score: 1

      And they're made out of superconducting adamantium. As a result, all games played on these processors will have higher quality storylines. It's a little known fact that copper causes destructive interference with the story's sine wave.
      The fact that this was modded "interesting" rather than "funny" should serve as a reminder to all of us that the people at Monster Cable have a very viable business model.
      --
      An interesting anagram of "BANACH TARSKI" is "BANACH TARSKI BANACH TARSKI"
  10. Hertz by themselves are useless by porkThreeWays · · Score: 5, Insightful

    Measuring computer performance in Hz is like buying a car based on red line RPMs. It only tells you one component that is meaningless by itself. Just like a car needs torque to give rpm's context, processors need how many instructions can be completed per cycle to be compared to the frequency. I've lost faith in the MHz race and generally look at benchmarks closest to the intended purpose of the processor.

    --
    If an officer ever threatens to taze you, say you have a pacemaker.
    1. Re:Hertz by themselves are useless by RightSaidFred99 · · Score: 1

      Is there anyone who doesn't know this by now? I think we all figured this out back in the P4 days. The point is the C2D has a high IPC as well as a high MHz.

    2. Re:Hertz by themselves are useless by Chris+Burke · · Score: 1

      Measuring computer performance in Hz is like buying a car based on red line RPMs. Just like a car needs torque to give rpm's context, processors need how many instructions can be completed per cycle to be compared to the frequency. I've lost faith in the MHz race and generally look at benchmarks closest to the intended purpose of the processor.

      That's exactly how I explained it to my father, back in the K7 vs Pentium 4 days, and as far as he knew MHz was how fast it went and Intel was selling chips with bigger MHz. Because it was a physical attribute of the chip it was easier to advertise with, and thus became a metric to compete on. As soon as you want to start talking seriously about IPC, like you said you have to start talking about benchmarks and that's a much, much trickier subject. It wasn't completely riduculous to sell based on MHz back when the competitors were not wildly different architectures, but it got worse over time and the P4 really threw things out of wack by being entirely designed around high clock frequencies.

      The biggest difference between now and then is that while the P4's "netburst" architecture was high frequency at the expected tradeoff of low IPC, the Core 2 is actually a well balanced machine with good IPC and frequency headroom. That it has this much headroom is thus fairly significant.

      --

      The enemies of Democracy are
    3. Re:Hertz by themselves are useless by Mattsson · · Score: 1

      This is true if comparing different architectures (like the AMD/Intel MHz race) or even the same architecture but wildly huge deltas of clockspeed.
      If you compare two identical cpus on two identical systems but one clocked at 2GHz and the other one at 3GHz, the clockspeed give a rough indication on their performance relative to each other when doing cpu-dependent stuff.
      When doing things that aren't cpu-dependent, it indicates that the two systems will perform more or less equal.

      Taking the car-analogy, it's like comparing two identical performance cars where one is RPM-capped at 6000RPM and one is capped at 10000RPM.
      On a straight asphalt-road, the higher capped one will reach a higher top-speed. On a small country road, they will perform equally.

      --
      /.Mattsson - My native language is not English, so please don't whine over linguistic errors. (That's lame anyway...)
    4. Re:Hertz by themselves are useless by Anonymous Coward · · Score: 0

      The rev limiter is not meaningless to car enthusiasts. I know of a friend who bought a Honda S2000 simply because of its insane high red line (8500 rpm I guess).

      This intel CPI is tailored to the same market as the S2000 is, hardcore enthusiasts. The Ghz plays a big part in this game.

    5. Re:Hertz by themselves are useless by cant_get_a_good_nick · · Score: 1

      Decent analogy, but I've always seen the 'Clock speed in computers is like Horsepower in Cars' analogy better.

      A lot of Horsepower, like a high clock, has some correlation to performance, but there are always confounding factors so you can not compare two machines clock for clock, HP for HP. For computers: bus speed, bus width, how much memory, etc. For cars: gearing, car weight, torque curve, transmission type and shifting speed, etc.

      Redline is more of a side effect. A high redline will possibly give higher horsepower depending on the torque curve, but why use a side effect (redline) instead of using the direct quantity (HP).

  11. Re:But... by Constantine+XVI · · Score: 1

    Somehow, I'm pretty sure the NSA uses more than just "vanilla" x86/AMD64.

    --
    "I think an etch-a-sketch with an ethernet port would beat IE7 in web standards compliance."
  12. SkunkTail? by Anonymous Coward · · Score: 0

    Could have sworn that's how it was spelled when I first looked at it.

  13. revision? by kernelphr34k · · Score: 1

    What revision of the board I wonder... hmm And yes, these boards are geared for gamers and overclockers.

  14. This is just stupid. by moogied · · Score: 0

    Clearly a publicity move to get the name out.. The systems will not be as fast as everyone thinks. I am pretty sure fasters systems already exist..

    --
    So basically, -1 troll/offtopic is really slashdots way of saying "I hate that you thought of something before me."
  15. MHz wars are over by geekoid · · Score: 2, Insightful

    Please, it's all about cores.

    Look at the history of processors speed. We've been pretty flat, and will stay that way in all practical manner for a while.
    Before someone throws the quote like they are smart, Moore's law refers to transistor not speed.

    1) Faster chips require better fabs. Fabs are having difficulty producing better platters with a few enough flaws to produce mass quantities. Strides are being made, but know massive breakthroughs.

    2) Multiples cores and real parallel processing development is just starting to become expected knowledge for the average application developer. Lets be honest, a lot of developers don't bother to understand multi-threading and avoid it like a plague. Fortunately there are some IDEs that make it easier for developers.

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    1. Re:MHz wars are over by the0 · · Score: 0

      Um... yeah...Mhz wars were over in 1999. It has been Ghz wars since then.

      Which rock have you been hiding under?

    2. Re:MHz wars are over by AuMatar · · Score: 5, Insightful

      Cores only help so much- if your problem is not paralelizable, or if it is only minimally so, a billion cores won't help. A word processor is not going to work any faster on a 1000 core machine than on a 1 core machine. Video games might see a small speed up from a multicore, but not that much of one- it doesn't break down into equally weighted threads. For the vast majority of users, 2 cores aren't even really utalized (email and web browsing doesn't use 2 cores). I doubt any home user will see much improvement beyond 2 cores, and absolutely none after 4 even for hardcore multitaskers. Business and scientific apps will see some beyond that, but memory tends to be the bottleneck there- we'd be better off increasing memory bandwidth and latency than clock speed.

      --
      I still have more fans than freaks. WTF is wrong with you people?
    3. Re:MHz wars are over by BlueParrot · · Score: 0

      Sure, but no matter how many cores you have it won't speed up an algorithm where every step depends on the previous one. Perhaps you can increase the size of the instruction set and perform a more complicated instruction in fewer clock cycles, but at the end of the day, everything else held constant the clock cycle obviously has a huge impact on performance.

      What it boils down to is "what is the cheapest way for us to crank up performance here?". If some of your engineers come up with a way to increase the clock frequency 3-4 times without too many disadvantages, you're going to go for it. It isn't a matter of weather multiple cores are better than rapid clock frequencies, it is a matter of which improvement provides the most benefit, with the fewest disadvantages, at the lowest price. Some times it will be cheaper and easier to add a core or two, some times it will be to ramp up the clock frequency.

    4. Re:MHz wars are over by paranode · · Score: 0, Troll

      Strides are being made, but know massive breakthroughs.

      But HOW do I know them?

    5. Re:MHz wars are over by Kugala · · Score: 1

      Faster clock speeds also mean that signals travel a shorter distance per clock cycle, which if your chip stays the same size starts to become more of an issue when wire delays become the dominant factor over switching delay. 5GHz would give a distance of just over an inch or so for your signal per cycle.

    6. Re:MHz wars are over by Firethorn · · Score: 4, Interesting

      This sounds a lot like a '640k' quote to me.

      A properly functioning word processor can already do pretty much everything 99.99% of what a user asks of it as fast as the user can tell it to do something, even on the bottom line processor.

      Today's video games, sure, aren't going to benefit much from multicore. But I disagree that the benefits for future games will top out at 2. I mean - you could have 1 core handling user input and processing, 1 core handling the physics enviroment, 1 core for unit AI, 1 core for graphics information. There's a quad core right there.

      Business and scientific apps will see some beyond that, but memory tends to be the bottleneck there- we'd be better off increasing memory bandwidth and latency than clock speed

      Then they can start worrying about beefing up memory bandwidth - I've read about some technologies in the pipe that will help with this. And the scientific community can always use more bandwidth - they are one of the larger users of supercomputers, and this might take a project from 'Need to rent 24hrs on the supercomputer for $$$' to 'I can run this on my work computer for a month/week to get the same results for $'.

      --
      I don't read AC A human right
    7. Re:MHz wars are over by n+dot+l · · Score: 1

      Video games might see a small speed up from a multicore, but not that much of one- it doesn't break down into equally weighted threads. Not at all true. Most of the heavy physics and graphics processing is extremely parallelizable. I've even seen AI computations parallelized. Taking advantage of multi-core is a very hot topic in the game industry right now and I assure you we're far from lacking ideas.
    8. Re:MHz wars are over by jpfed · · Score: 1

      What if "your problem" is really several problems? Are modern operating systems able to put different applications on different cores? (Serious question- I don't know) If they were, then multiple cores would certainly help me; at work, I'm never just running one application, and it's typical for me to run four.

    9. Re:MHz wars are over by Chris+Mattern · · Score: 1

      You fought in the MHz wars?

      Yes, I was once an AMD knight, the same as your father.

      Chris Mattern

    10. Re:MHz wars are over by Anpheus · · Score: 0

      Yep.

      You can run Photoshop running and have it stall a core on an image operation and have Firefox open to a flash or javascript intensive website and have it stall on another core.

      Progress is being able to stall your CPU twice!

    11. Re:MHz wars are over by qmaqdk · · Score: 1

      Score:1, Troll? Shouldn't it rather be Score:-1, Funny?

      --
      My UID is prime. Hah!
    12. Re:MHz wars are over by TheRaven64 · · Score: 3, Insightful
      We have had about 40 years of practice getting one processing unit to pretend to be n, and we're pretty good at it now. We have no good ways (even in theory) of getting n processing units to pretend to be one in the general case. If you have a 5GHz core then you can run two processes on it happily with only a small amount of overhead. If you have two 2.5GHz cores and only one process, you will end up running that process at half of the theoretical speed of your CPU.

      Fewer faster cores will always be more flexible than more slower ones. The reason we go with more slower ones is that slower cores use less power (power scales much worse than linearly with speed, so two 1GHz cores will use a lot less power than one 2GHz one). Some workloads are intrinsically parallel (e.g. web serving) and so having lots of cores using less power is a big win. Others are not and so extra cores are just a waste (although you can often consolidate multiple serial tasks onto one machine with lots of cores).

      --
      I am TheRaven on Soylent News
    13. Re:MHz wars are over by Bill+Dog · · Score: 0

      Cores only help so much-...

      Then we need to get creative.
       
      ...if your problem is not paralelizable, or if it is only minimally so,...

      What percentage of problems really aren't parallelizable?

      A word processor is not going to work any faster on a 1000 core machine...

      We don't need our word processors to work faster. So that's not one of our "problems".

      Video games might see a small speed up from a multicore,...

      Video games could see a huge speedup, or get a lot smarter -- imagine a processor for every enemy in play. Compare for a moment to real-life: Our reacting to each other sort of resembles a turn-based dynamic, but mostly we act, and decide how we're going to act, independently, and all at the same time as everyone else.

      I doubt any home user will see much improvement beyond 2 cores,...

      My folks still use AOL (Security Edition) -- you should see all the crap that runs on their system while they're trying to do their own stuff. I'd say they wouldn't see much improvement beyond 8 cores. But that is as software stands today -- when virus scanners and other background and foreground processes are written to use multiple cores, then even the average home users' needs for greater numbers of cores will escalate. Never say x should be enough.

      --
      Attention zealots and haters: 00100 00100
    14. Re:MHz wars are over by Anonymous Coward · · Score: 0

      My father's name isn't Chris Mattern, you insensitive clod!

    15. Re:MHz wars are over by Surt · · Score: 1

      Video games are very parallelizable at the render and physics layers. They'll have no problem usefully scaling to hundreds of cores.

      --
      "Who is the Journal of Quantum Physics going to believe?" --Stephen Hawking
    16. Re:MHz wars are over by Chandon+Seldon · · Score: 1

      Cores only help so much- if your problem is not paralelizable, or if it is only minimally so, a billion cores won't help.

      This is true.

      Here's the thing: Every one of the applications that people commonly run on a desktop PC can be parallelized.

      The real problem is that programmers who are used to single-thread designs cringe when they see the parallel version. Not only is it moderately more complex, but to generalize to many cores a design frequently entails a 10% to 50% performance penalty compared to the single-threaded version. That means that the code that runs four times as fast on an 8 core processor may only run half as fast on a single core, and may see no performance gain at all on a dual core.

      So the problem isn't that word processors and video games can't be parallelized, it's that there seems to be no advantage to writing solid parallel code that runs best on 8+ core machines when most people are just moving from single to dual core.

      --
      -- The act of censorship is always worse than whatever is being censored. Always.
    17. Re:MHz wars are over by Anonymous Coward · · Score: 0

      A word processor is not going to work any faster on a 1000 core machine than on a 1 core machine.
      If your word processor is doing spell checking and grammar checking to transcode your voice to text, more than one core becomes quite useful.
      Although 1000 cores is a tad extreme, a handful or

    18. Re:MHz wars are over by Chandon+Seldon · · Score: 1

      That's true, but the algorithms / program designs that work great with a hundred cores work like crap on one or two cores. Personally, I expect to see video games designed to be truly concurrent just as soon as low-end gamers have quad core machines (and high-end gamers have 32-thread systems).

      --
      -- The act of censorship is always worse than whatever is being censored. Always.
    19. Re:MHz wars are over by Chandon+Seldon · · Score: 1

      We have had about 40 years of practice getting one processing unit to pretend to be n, and we're pretty good at it now.

      We've also had decades of practice solving problems using multiple processing units. In fact, for the sort of problems that today's processors can just barely handle (i.e. those problems that processing power is still an issue on) we've had *more* practice solving them on compute arrays than we have solving them on single processors.

      --
      -- The act of censorship is always worse than whatever is being censored. Always.
    20. Re:MHz wars are over by Surt · · Score: 1

      Absolutely. My point was only that video games are not in a category of 'can't use multiple cores'. At all. They would love to have tons of cores. Whether or not their user base has multiple cores, is an entirely different question.

      --
      "Who is the Journal of Quantum Physics going to believe?" --Stephen Hawking
    21. Re:MHz wars are over by TheRaven64 · · Score: 2, Insightful
      You are missing the point. Any problem that can be solved on a parallel machine can be solved on a serial machine of the same computational power in the same time. The converse, that any problem that can be solved on a serial machine can be solved on a parallel machine of the same processing power, is not true. At the abstract level, any nondeterministic finite automaton can be reduced trivially to a deterministic equivalent, but an arbitrary degree of nondeterminism can not be trivially added to a DFA.

      If you can solve a problem on an n-core mGHz machine, you can solve it on a n/2-core, 2mGHz machine in (roughly) the same amount of time. Whether you can solve it on a 2n-core, m/2GHz machine in anything like the same amount of time depends heavily on the problem (and also the value of n - see Amdahl's law).

      It doesn't matter how much more practice we've had solving problems on parallel systems, because a serial system can trivially emulate a parallel system, while the converse is not true. This means that all of our experience solving problems on parallel systems can be directly applied to serial systems, while our experience solving problems on serial systems can not be applied to parallel systems.

      Don't get me wrong; I like designing parallel systems and did quite a bit of it for my PhD (among other things), but at the end of the day I'd rather have serial hardware than parallel, all other things being equal. Of course, in the real world, all other things are not equal. You can't get a general-purpose serial processor with the same IPC as a quad-core Opteron (for example), and individual processors typically have less (total) cache than multiprocessors.

      Some of the code I wrote was originally aimed for a 64-processor 600MHz MIPS box and ended up running on a 16-core 2.4GHz Opteron system. It was highly parallel (roughly 1000 lightweight processes, often more), but still ran faster on the 16-processor machine than the 64-processor one, because each of the 16 processors was more than four times the speed of the old MIPS chips (and that's ignoring the latency differences).

      --
      I am TheRaven on Soylent News
    22. Re:MHz wars are over by Have+Brain+Will+Rent · · Score: 1

      Multiples cores and real parallel processing development is just starting to become expected knowledge for the average application developer.

      Multiple cores is a pretty simple concept. Parallel processing is a little more complex but should have been part of any CS degree for the last 30 years at least.

      a lot of developers don't bother to understand multi-threading

      Multiple threads and processes should have been covered in any good CS degree for at least the last 20 years. You can't understand OS design without knowing this stuff.

      Fortunately there are some IDEs that make it easier for developers.

      What IDE's make it easier to develop multi-core, parallel processing applications?

      --
      The tyrant will always find a pretext for his tyranny - Aesop
    23. Re:MHz wars are over by mdwh2 · · Score: 1

      I agree that getting video games to take advantage of multiple cores isn't as easy as some people seem to think - OTOH, games already are massively parallelised. It's just that the parallelisation has already happened on dedicated hardware - a graphics card. Since the CPU tends not to be the bottleneck, it wouldn't benefit from a faster CPU anyway, whether that's higher GHz or more cores.

      But think of all the other things that take up CPU time. That's not things like word processors - it's things like DVD authoring, mp3 encoding, 3D rendering, CAD, scientific applications. All of these things are usually easily parallelised.

      It's true that (for now, at least) most home users won't need more than 2 cores most the time - but they also won't need a CPU that's twice as fast most the time.

    24. Re:MHz wars are over by drspliff · · Score: 1

      The example you've used, 16 2.4ghz cores vs 64 600mhz cores is a pretty typical example.

      If you go on mhz alone, the 16 core machine should process work units at the same speed as the 64 core machine (1/4th the number of processors, but 4x faster cpus, vs 4x more processors and 1/4th speed cpus). But that's not taking into account faster bus speeds, better architecture, improved floating point performance etc.

      However, a 96 core machine at 600mhz would process far more units of work than the opteron machine, but that's not the point. Today it's all about what's cheapest now that can do the job, the power draw per server and how many you can fit in a rack and how many units of work you can process per $$$.

      But, if you took the same 64 core server from yesteryear and re-vamped it with the latest technology, wattage goes down, chip size goes down, more chips per rack etc. It's just not possible to compare hardware from 5 years ago with what's available today will always be 2-4x as powerful as what was available then.

      The MHz wars are still here, but with an added dimension - MHz per square foot :) IMO they'll keep rising with Moore's law for atleast another decade or so before the next tech revolution happens, IBM Power6 processors have become the new Alpha and hopefully it'll spur the rest of the industry on :D

    25. Re:MHz wars are over by phillips321 · · Score: 1
      Crytek: Crysis multicore support http://www.next-gen.biz/index.php?option=com_content&task=view&id=4690&Itemid=2

      To utilize multi-core machines we identified good candidates, [such as] particle system and physics, and made them multi-threaded. It's normal development, but bugs are harder to track, performance is less predictable and there is no benefit for single core processors.
    26. Re:MHz wars are over by SuperMog2002 · · Score: 1

      You're certainly right in your video games point. A few patches ago, the Mac client for World of Warcraft went multi-threaded. If you have two cores, the game engine runs on one of them while the other assists with the graphics processing. If you have a low end GPU (as I do in my MacBook), this made a TREMENDOUS difference in performance. In early October, they also added a built in video recorder to the Mac client, and it too is multi-threaded. I use my quad-core Mac Pro to record boss fights on occasion, and as long as I'm not loading at the same time, I hardly take a performance hit at all, as the encoder is happily running on the other two cores and isn't effecting game play much.

      --
      Sunwalker Dezco for Warchief in 2016
    27. Re:MHz wars are over by darkwhite · · Score: 1

      Please, it's all about cores. I think to you it's all about throwing around bullshit unfounded opinions.

      There are many tasks, some of them on the dekstop, which will never be parallelizable. Single-core performance has been and will remain absolutely crucial, even when everyone and their mom can write code in a parallelizing toolkit.

      Faster chips require better fabs. Fabs are having difficulty producing better platters with a few enough flaws to produce mass quantities. Strides are being made, but know massive breakthroughs. My phoniness meter just exploded.
      --

      [an error occurred while processing this directive]
    28. Re:MHz wars are over by Firethorn · · Score: 1

      Right now I only have a dual core system; what I see happening is that a CPU hungry program like a game ends up on one core while the other stuff like CPU and background systems end up on the other. So a dual core does help me out a fair bit, but doesn't double performance.

      It'll get better if they improve game threading, but to truly double performance I'll probably need 4 cores.

      I figure many(IE more than four) core systems are about four years away. An eternity in computer terms, of course.

      --
      I don't read AC A human right
    29. Re:MHz wars are over by Firethorn · · Score: 1

      The 8800 GTX has 128 graphics pipelines - that is indeed a good amount of parallelization.

      I agree that getting video games to take advantage of multiple cores isn't as easy as some people seem to think - OTOH, games already are massively parallelised. It's just that the parallelisation has already happened on dedicated hardware - a graphics card. Since the CPU tends not to be the bottleneck, it wouldn't benefit from a faster CPU anyway, whether that's higher GHz or more cores.

      CPU right now might not be the bottleneck - but it is a concern. Graphics cards enable great eye candy, but just as an example, AI units* are still quite dumb. I've read some reviews where they had to scale back the AI for units, go to more scripting because machines couldn't handle the computational expense of AI. We're talking about where going to scripting to essentially fake AI can reduce CPU load for that unit ~100X. With more cores, you'd be able to make the enemies smarter. Go into more detail with the physics models, etc...

      Sure, it's going to require a shift in skill set for game programmers - but they have to deal with new engines all the time anyways. Heck, it'd only take one company making a massively parallelized engine and other companies can license the tech to make their own games. Even better if the company produces a non-parallel optimized version of the engine that can take the same inputs, so people with single or dual core systems can still play the game at a decent efficiency.

      *I'm essentially talking about NPCs here.

      --
      I don't read AC A human right
    30. Re:MHz wars are over by Distortions · · Score: 1

      Not even close. The wars have just begun.

      Wait until the raytraced games come out and more cores is all that counts.
      Or much more sophicated voice recognition, better AIs, facial recognition / lip reading, more and more security, etc etc.

      Hell, even today its not unreasonable to say: You need a core for the OS, a core for anti virus/trojen/spypware software, a core for the foreground app, a core for background tasks. 2-4 cores already reaonable.

      --
      Talent hits a target no one else can hit; Genius hits a target no one else can see.
    31. Re:MHz wars are over by dbIII · · Score: 1
      A lot of things are. Digital video where you apply the same or similar transforms to every frame is one. Image manipulatation in general is another. Sound processing is yet another once you get beyond mono.

      Word processing is a bad example because the software is so badly written and ridulously resource hungry while providing less features than a Desktop Publishing program that ran on a 286 or even one that ran on an Atari ST.

    32. Re:MHz wars are over by revengebomber · · Score: 1

      Today's video games, sure, aren't going to benefit much from multicore. But I disagree that the benefits for future games will top out at 2. I mean - you could have 1 core handling user input and processing, 1 core handling the physics enviroment, 1 core for unit AI, 1 core for graphics information. There's a quad core right there. Or you could build a game on a raytracing graphics engine, and utilize up to (1280x1024) cores.
      --
      09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
      45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
    33. Re:MHz wars are over by shutdown+-p+now · · Score: 1

      WRT games optimized for multiple cores, it is certainly possible, even though it requires a lot of rethinking. This presentation by Tim Sweeney of the Unreal fame touches on the topic.

    34. Re:MHz wars are over by Anonymous Coward · · Score: 0

      Interesting. One core for handling USER INPUT. *rolls eyes*, how fast you type, again? ;) Mouse sends probably the largest amount of data, which doesn't take even 1% of 100 MHz 486 to handle.. well, right, later.

    35. Re:MHz wars are over by ginbot462 · · Score: 1

      How did this get modded insightful; did no one read beyond the first sentence? Have you not seen how many processes run on home computers? Hell, Bonzi Buddy alone takes up like 64 Cores (measured in apple(tm) cores).

      [me] Leaves to kick off 4 builds on a dual processor & dual core machine (== CPU bound!).

      --
      Atlas Shrugged : Thematic Story :: Battlefield Earth : Organized Religion
    36. Re:MHz wars are over by Firethorn · · Score: 1

      Hell, even today its not unreasonable to say: You need a core for the OS, a core for anti virus/trojen/spypware software, a core for the foreground app, a core for background tasks. 2-4 cores already reaonable.

      I wouldn't segment that far, at least right now. A single core should be able to handle the OS, antivirus/spyware, and background tasks like handling non-video drivers like disk access, sound and network and still not be strained. My computer with a single core processer, just sitting there is 99% idle. Intensive disk browsing might get it up to 10% in spurts.

      A smart OS with even halfway decent singlethreaded programs should be able to move these tasks between CPUs as necessary for load balancing.

      With a dual core you'd be able to play a intensive single thread game and compress a DVD on the other core in the background without noticing anything but longer load times(and not even that if they're using different disks).

      A dual core is probably on average best off not worrying about multithreaded applications yet - let the OS handle it. Get above that though, and smart applications will have substantial benefits to gain by multithreading.

      Though using additional cores for AI enemies might make for interesting reviews: 'The enemies are real dumb on our ancient 5.0 Ghz Quadcore test bed, but are frighteningly intelligent on our 5.0 Ghz 64 Core GamerXtreme hyperchip!'

      --
      I don't read AC A human right
    37. Re:MHz wars are over by Chandon+Seldon · · Score: 1

      You are missing the point.

      Not at all. The point is this:

      • Fast computers today are parallel, serial systems aren't an option. Further, they're getting more parallel rather than less.
      • Contrary to popular FUD, almost all programs can be usefully parallelized.

      Sure, a serial system with the computational power would be nicer - but that's really irrelevant in a world where the fastest/cheapest serial processor available comes with three, seven, or fifteen free copies of itself attached.

      --
      -- The act of censorship is always worse than whatever is being censored. Always.
    38. Re:MHz wars are over by earthbound+kid · · Score: 1

      An exercise for those who believe more Hz is still the future:

      Open Google.com

      Type "299792458 m/s / 1cm"

      You'll get "29.9792458 gigahertz"

      That's the fastest light can get from one side of a half centimeter chip to another and back. Think about what that means.

    39. Re:MHz wars are over by Firethorn · · Score: 1

      Only for a standard LCD screen. What about those with 1600x1200 screens? And unit AI? 2 Million cores sounds about right... ;)

      --
      I don't read AC A human right
    40. Re:MHz wars are over by earthbound+kid · · Score: 1

      Faster clock speeds also mean that signals travel a shorter distance per clock cycle, which if your chip stays the same size starts to become more of an issue when wire delays become the dominant factor over switching delay. 5GHz would give a distance of just over an inch or so for your signal per cycle.


      Bingo. People can say, "Parallelization suxx0rz!" all they want, but the fact of the matter is that 299,792,458 m/s / 1 cm = 29.9792458 gigahertz. (Thanks Google calculator.) That's the absolute ceiling: the speed of light divided by chip size. People can deride multi-core all they want, but they'll never get Intel to put resources into creating a 40GHz chip for you, since, you know, it can't actually be created without FTL processors.
    41. Re:MHz wars are over by superflyguy · · Score: 1

      The point was that he prefers serial, all things being equal, and he noted that the problem is that all things are not equal. You completely missed that point. Certainly it's different than your point, but you're still missing the point he made.

  16. its by Anonymous Coward · · Score: 1, Funny

    its

    1. Re:its by Anonymous Coward · · Score: 0

      ... Monty Pythons Flying Circus?

  17. what does "desktop processor attributes" mean? by Thagg · · Score: 2, Interesting

    I've seen this before, I've never understood it. What does it mean?

    Thad

    --
    I love Mondays. On a Monday, anything is possible.
    1. Re:what does "desktop processor attributes" mean? by Kjella · · Score: 2, Insightful

      Servers are about making a lot of people happy at a reasonable speed. Desktops are about making one user happy at an extreme speed. A lot of crap is single-threaded or not suitable for parallelization, and the best solution is to push that single thread at maximum speed. That's the only desktop quality of significance I know of. With that said, I have a quad-core and my biggest annoyance right now is disk thrashing. My CPU is usually almost idle, but having a lot of tasks using the disks at the same time slows everything down. I really look forward to SSDs and near penalty-free random access.

      --
      Live today, because you never know what tomorrow brings
    2. Re:what does "desktop processor attributes" mean? by forkazoo · · Score: 3, Insightful

      I've seen this before, I've never understood it. What does it mean?

      Thad


      I'm not intimately familiar with the specifics in this case, but starting with a server chip and "adding desktop processor attributes" would typically entail:

      adding the inability to use ECC.
      adding a reduction in cache.
      adding a lack of fault tolerance or error checking capabilities.
      adding the feature of being impossible to use with > 2 sockets.
      adding a whizzy new marketing name.

      And, the enthusiast desktop parts are often easy to overclock, while server parts assume you'll just buy a faster CPU instead of wasting time fiddling with something that may catch fire.

      BTW, hey, I remember you from alt.movies.visual-effects "back in the day" before the death of Usenet. good to see you haven't fallen off the face of the planet. I'm not in the process of working on a compositing demo reel so I can try to jump from straight IT to visual effects in the near future. I blame this career change in part on all your interesting and informative posts getting stuck in my head. :)
    3. Re:what does "desktop processor attributes" mean? by Anonymous Coward · · Score: 0

      Thrashing? You need a lot more memory then! Are you sure you don't mean disk access? Thrashing occurs when memory is low so data is paged out to disk; but then the data is needed again so it's loaded back into memory causing other data to be paged out to disk. And then that other data is needed, and back and forth it goes. Progress is made very slowly because a sizeable chunk of the data needed can't all fit in memory so you are constantly writing/reading portions of it from disk. Slow disk access or slow performance due to many apps competing to read data isn't really thrashing as far as I know.

    4. Re:what does "desktop processor attributes" mean? by geekoid · · Score: 1

      "...and the best solution is to push that single thread at maximum speed"

      Many developers believe this, and in any practical application they are wrong.

      I have never seen an application where real world performance, and more importantly "perceived performance increase", isn't improved with threading. Too many times I have seen scenarios where a user is sitting waiting for the computer to finish. When they could be doing some other work.

      I have seen developers wither under the request to add some multitasking. Sad, really.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    5. Re:what does "desktop processor attributes" mean? by turing_m · · Score: 1

      "My CPU is usually almost idle, but having a lot of tasks using the disks at the same time slows everything down. I really look forward to SSDs and near penalty-free random access."

      You could always store your mp3s on one disk, another application on another disk, etc etc. It's a kluge though, because it requires more power to do so.

      --
      If I have seen further it is by stealing the Intellectual Property of giants.
    6. Re:what does "desktop processor attributes" mean? by PitaBred · · Score: 1

      Hell, try opening a large email over a non-local connection in Outlook. Even Microsoft withers in the multitasking arena. The POS locks up like a deer in headlights until it downloads the entire email.

    7. Re:what does "desktop processor attributes" mean? by Thagg · · Score: 1

      Forkazoo said > I'm not in the process of working on a compositing demo reel so I can try to jump from straight IT to visual effects in the near future. I blame this career change in part on all your interesting and informative posts getting stuck in my head.

      Sorry about that whole visual effects problem... Hope it works out.

      I thought your description of the difference between the server chips and desktop chips was right on.

      I'm going to be building a 8-core AMD machine in a few days, and I'll use the "server" barcelona chips if they don't actually release the "workstation" phenom chips on time, as I can't for the life of me see any disadvantage in the server chips. They're not even expensive!

      Thad

      --
      I love Mondays. On a Monday, anything is possible.
    8. Re:what does "desktop processor attributes" mean? by Kjella · · Score: 1

      Thrashing? You need a lot more memory then! Are you sure you don't mean disk access? Thrashing occurs when [constantly swapping] Quote WP:

      In computer science, thrash is the term used to describe a degenerate situation on a computer where increasing resources are used to do a decreasing amount of work. Usually it refers to two or more processes accessing a shared resource repeatedly such that serious system performance degradation occurs because the system is spending a disproportionate amount of time just accessing the shared resource. Resource access time may generally be considered as wasted, since it does not contribute to the advancement of any process. It's a degenerate situation because two or more processes are fighting over the read head on the HDD - it keeps going back and forth, getting very little done bcause it spends a disproportionate amount of time moving the read head so data can be accessed instead of advancing the process. I agree it's not the most common use of the term, but I think it's appropriate...
      --
      Live today, because you never know what tomorrow brings
    9. Re:what does "desktop processor attributes" mean? by Fweeky · · Score: 1

      Seagate apparantly did a good job with the firmware on their new 7200.11's with NCQ; there's at least one benchmark on Storage Review forums, showing serial transfer performance of 9 seperate reads running upwards of 80MB/s, while every other tested drive struggles to get past 20.

      Might be a useful excuse to upgrade to a 1TB drive if it helps more common use-cases.

  18. Re:But... by sm62704 · · Score: 5, Funny
    But can it run on Linux?

    Dude, you can run linux on a wristwatch. The question is, can it run Vista?

    From an old K5 diary:

    At any rate, I tell the guy about my dead Celeron and how I want to upgrade the motherboard, and get some memory, and get a video card so I can plug it into the TV. He tries to sell me a supercomputer cluster, and I say no, I'm not much into computer gaming any more so a pretty low end one would do.
    -mcgrew

    --
    mcgrew's razor: Never attribute to stupidity that which can be explained by greedy self-interest
  19. Re:But... by mrbluze · · Score: 1

    Somehow, I'm pretty sure the NSA uses more than just "vanilla" x86/AMD64. Yeah you're right. Just stirrin' ;)
    --
    Do it yourself, because no one else will do it yourself. [beta blockade 10-17 Feb]
  20. Where did my /. go? by blueZ3 · · Score: 4, Funny

    You just made an almost-sensible car analogy. I didn't think that was allowed here.

    --
    Interested in a Flash-based MAME front end? Visit mame.danzbb.com
    1. Re:Where did my /. go? by servognome · · Score: 1

      You just made an almost-sensible car analogy. I didn't think that was allowed here
      Ghz is like measuring a car's RPM, it's meaningless unless you also describe how loud the radio can go.
      --
      D6 63 0D 70 89 81 BB 8E 7B 7C 5F 5D 54 EA AB 73
  21. Stable by raddan · · Score: 4, Funny

    As long as you have an ample supply of liquid nitrogen.

  22. Re:But... by geekoid · · Score: 1

    This would be useless to intelligence. Try to use yours.

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
  23. bragging rights by DreadSpoon · · Score: 5, Funny

    To many people that's all they're looking for. It's like buying an F-350 when the most you use a car for is getting groceries, or getting the biggest house you can possibly afford even though you're a small family of three, and so on.

    Remember, it's not just the spammers that profit off of people with small penises. Auto manufacturers, TV manufacturers, home builders, and now Intel all profit off of them too. :)

    1. Re:bragging rights by Anonymous Coward · · Score: 2, Funny

      Remember, it's not just the spammers that profit off of people with small penises. Auto manufacturers, TV manufacturers, home builders, and now Intel all profit off of them too. :)

      Ha! I don't buy things from any of those people, and my penis is tiny!

    2. Re:bragging rights by BForrester · · Score: 1

      Proud '80 civic driving, 14" monitor viewing, 40 sq. ft. closet dwelling C-64 computer user here. (Wink wink nudge nudge).

    3. Re:bragging rights by Anonymous Coward · · Score: 0

      I guess it gives the poor something to smile about when they envision those who are wealthier than they are as having small genitalia?

      Just because you can't afford it doesn't mean that the potential audience are braggarts.

    4. Re:bragging rights by numbware · · Score: 1

      Wow. No car, no TV, no home, and a small penis. Your life must really suck.

      --
      I'm going to go create my own technology news site, with blackjack and hookers. You know what? Forget the news site.
    5. Re:bragging rights by Anonymous Coward · · Score: 0

      That's because yours is called a vagina. Your husband/boyfriend with the small penis is the one who buys from those people. Where do you think you got all that stuff from?

  24. This looks bad next to a amd dual quad-core system by Joe+The+Dragon · · Score: 3, Interesting

    single-slot graphics cards, FBDIMMS and you will need 4 of them to get the max system out of the memory system, SLI useing Nvidia nForce 100 chips over a pci-e x16 1.1 bus split to 2 x16 slots, dual eps power in, 3 chip sets chips that driver up cost and power use.

    The dual amd system that this will be like this will use DESKTOP RAM, have 2 or more chipset choices. Also the amd setup lets you have 2 full Northbridge chipsets for even more i/o the nForce 680a uses this and nvidia will likey have a new chipset with pci-e 2.0. The old has a x16 x8 x8 x16 pci-e with a total of 56 PCI-E lanes.

    The new amd chipet is also comeing and you may even see a board with 2 Northbridges = 82 pci-e lanes.

    790FX

            * Codenamed RD790, final name revealed to be "AMD 790FX chipset"
            * Dual-socket (Quad FX, Dual Socket Direct Connect Architecture) or single AMD processor configuration
            * Maximum four physical PCI-E x16 slots and discrete PCI-E x4 slot , the chipset provides a total of 52 PCI-E lanes, with 41 lanes in Northbridge
            * HyperTransport 3.0 with support for HTX slots and PCI Express 2.0
            * ATI CrossFire X, see below
            * AutoXpress, see below
            * Extreme overclocking, reported to have achieved about 420 MHz bus for overclocking an Athlon 64 FX FX-62 processor, from originally 200 MHz.
            * Discrete chipset cache memory of at least 16 KB to reduce the latencies and increase the bandwidth
            * Supports Dual Gigabit Ethernet, and teaming option
            * Reference board codenamed "Wahoo" for dual-processor system reference design board with three physical PCI-E x16 slots, and "HammerHead" for single-socket system reference design board with four physical PCI-E x16 slots, also notable was the reference boards includes two ATA ports and only four SATA 3.0 Gbit/s ports (as being paired with SB600 southbridge), but the final product with SB700 or SB750 southbridge (see below) should support up to six SATA ports
            * Northbridge made on 65 nm process, manufactured by TSMC, and runs at 3 W when idle, and maximum 10 W under load, nominal 8 W power consumption, the northbridge was seen on reference design with single passive cooling heatsink only instead of connecting to heat pipes which are frequently used on current mainstream motherboard offers, the combination of 790FX northbridge with SB600 southbridge consumes normally less than 15 W
            * Enthusiast discrete multi-graphics segment

    Even if the Intel system is faster the amd system with less costly MB and much cheaper ram will likely be a better buy.

  25. Not the first one at all by Anonymous Coward · · Score: 0

    Power6 does run at 5GHz and is in the market form some months already.

    1. Re:Not the first one at all by anss123 · · Score: 1

      Power6 is an in-order design so it better damn well run at 5GHz. OoO at 5 GHz is far more impressive - not that an overclocked Skulltrail truly counts.

    2. Re:Not the first one at all by treeves · · Score: 1

      Why does OpenOffice.org need 5 GHz?

      --
      ...the future crusty old bastards are already drinking the Kool-Aid.
    3. Re:Not the first one at all by GreggBz · · Score: 3, Funny

      So it's as fast as MS-Office. :=p

    4. Re:Not the first one at all by Anonymous Coward · · Score: 0

      Because on my 2.6 GHz machine it is too slow to use. As a guess, a 26 GHz (10x as fast) system is required to run OpenOffice so that you don't spend more time waiting than you do actually doing productive work. What ever happened to word processors (or even source editors in IDE's) that are fast enough to keep-up with someone that knows how to type? Applications just keep getting slower and slower. My C64 25 years ago was much faster than my new $2k system running these new crappy IDE's and bloated word processors.

  26. Re:But... by rtyhurst · · Score: 0, Troll

    It might.

    But you're going to need four of these plus a 1 terabyte hard drive plus 100 gigs of memory to run VISTA service pack 2.

    Get used to it...

  27. So... by Anonymous Coward · · Score: 0

    If I have a small penis, the best thing for me to do is buy an eMachines computer, drive a Corolla, and live in a studio apartment? Won't I then be assumed to completely LACK a penis?

    1. Re:So... by WinterSolstice · · Score: 1, Offtopic

      Actually - I got rid of my Corvette and bought a fuel efficient Scion. A female friend of mine quipped that a trade like that gained me 6" :D

      Third party observation by the secretary was that guys who drive reasonable cars are more comfortable with themselves and more stable. Better material, etc.

      Of course, I then bought an M3 and ruined it all!

      --
      An operating system should be like a light switch... simple, effective, easy to use, and designed for everyone.
    2. Re:So... by Bluesman · · Score: 5, Insightful

      My third party observation is that 99% girls will look at the comfortable and stable guys, wonder why THEY can't find a guy like that, and then hop in the M3 with the asshole.

      Confucius say, a small dick is still better than an unused one.

      --
      If moderation could change anything, it would be illegal.
    3. Re:So... by geekoid · · Score: 1

      That's nice, and tour friend is lying to herself.

      I got more action, looks, and flirting when I had my sports car. Lots more.
      Now it could be that the women I was seeing didn't want anything more then a quick weekend at the beach having sweaty sex.

      So enjoy your M3.

      Yes, I did post this to brag about my younger weekends. Considering I get paid the same to program computers as my colleagues who spent all their younger years constantly sitting in front of a computer, I win.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    4. Re:So... by Burning1 · · Score: 1, Offtopic
      Here's a thought...

      Perhaps those women are right; that they really can't find a comfortable, stable guy.

      Perhaps many of the comfortable, stable guys aren't interested in girls with the kind self confidence issues that put them in M3s with assholes.

      I'm drastically simplifying relationships when I write this and I'm also grossly generalizing, so please take what I write with a grain of salt. I'd love to hear other people's experience.

      Here's my dating experience, having gone through many stages of personal growth...

      - I've been a 'nice guy' who waited on women hand and foot.

      I ended up being friends with women who had self respect issues. In my opinion, that kind of woman enjoys being around people whom she sees as more pitiful than herself. She loves the attention, but the more attention she is given, the lower her opinion of the man due to her own self respect issues. She desires strength in a romantic partner. Unfortunately, she's often the only kind of woman a self confessed 'nice guy' will spend much time with. Most other women are put off by his clingy attitude. The nice guy then begins to believe that all women are like her.

      - I've been 'the asshole.'

      The asshole is the basic male equivalent of the women I spent time with when I was a 'nice guy.' I was more successful, but I wasn't happy. I could never respect the women I dated, because my false sense of self confidence did not allow me to respect myself. The relationships I had were frequently shallow, and subtly antagonistic.

      - I'm currently learning to relax, and to be confident.

      I'm not Casanova, but I've had some success. I'm learning when to be strong, and when to let things go. I'm learning when to express my feelings, and when to control them. It's not easy, and there aren't simple rules... But I have mutual respect, and I've had some great experiences. I believe I'd be more successful if I was less intimidated by social situations.

      My third party observation is that 99% girls will look at the comfortable and stable guys, wonder why THEY can't find a guy like that, and then hop in the M3 with the asshole.
      When I hear this, I usually think there is something deeply inexperienced about the guy saying it.

      I've known a lot of 'nice guys' and I've met a few truly confident, stable guys. Few (if any) of the confident, stable guys have this attitude. They aren't bothered by unstable women, because they aren't interested in them. They are generally too busy spending time around the confident, stable girls.
    5. Re:So... by Prien715 · · Score: 1

      My gonzo journalistic observation is that 99% of women want some guy who will listen to them and that instead 99% of guys THINK girls want the guy you describe and either (a) try to be him or (b) so flustered by the this fact that they are completely paralyzed by inaction or to lack confidence when they do act. Another result of people in (b) is to spend time place where the female % is like 5% (like /.) and then wonder why it's hard to find women. Confucius say "Man who wants fish should look in pond instead of mountain" and all that.

      --
      -- Political fascism requires a Fuhrer.
    6. Re:So... by Archades54 · · Score: 1

      Drive what you want to drive, women can think what they want as they are just insanely jealous that cars can demand as much or more attention/money/desire than they get.

      --
      If your neighbours roof is flying past your window, you know it's cyclone season.
    7. Re:So... by ksheff · · Score: 1

      They are generally too busy spending time around the confident, stable girls.
      who are usually playing with their unicorns in a land of sunshine and rainbows.

      the only stable women are dead ones.

      --
      the good ground has been paved over by suicidal maniacs
    8. Re:So... by 0111+1110 · · Score: 1

      It doesn't matter how 'nice' or confident you are if you've got the looks. That's the only value system that matters in the dating world. It will get you through the door (99% of the battle) and often times keep you in the pink for an awfully long time. Especially if you have the 'self-confidence' of knowing that most girls think you're hot. If you don't have the looks you can try to fake confidence all you want but chicks will know it's just a sham. Real confidence is based on past performance, not standing in front of a mirror reciting "I love myself" 10,000 times. The only exception to the 'looks is everything' rule is in a handful of very poor 3rd world countries where a small fraction of the women genuinely value money more than looks. But this only comes from living in a tin-roofed shack with 16 other family members and no shoes. True hunger and poverty is an excellent aphrodisiac, but you would be surprised at how seldom you can get anywhere without looks even in these countries. Usually they will just try to scam the ugly guys long enough to make their escape from poverty. Then they will find the good looking guy that they feel entitled to as a hot girl. They want someone they find physically attractive and perhaps even more importantly someone all of their friends will find attractive as well. Just reality 101. And common sense. I don't know why this is so hard for some people to accept.

      --
      Quite an experience to live in fear, isn't it? That's what it is to be a slave.
    9. Re:So... by Anonymous Coward · · Score: 0

      Actually - I got rid of my Corvette and bought a fuel efficient Scion. A female friend of mine quipped that a trade like that gained me 6" :D

      Congratulations, welcome to the 8" penis club!

    10. Re:So... by Burning1 · · Score: 1

      What has happened in your life that you have come to this conclusion?

    11. Re:So... by ConceptJunkie · · Score: 1

      I know lots of stable women, including my wife of 14 years, my sister, my mother, my mother-in-law, a bunch of sisters-in-law and lots of friends. Perhaps you hang around the wrong kind of women.

      --
      You are in a maze of twisty little passages, all alike.
    12. Re:So... by Burning1 · · Score: 1

      There's some truth in what you're saying, but I think that you are grossly simplifying things.

      In my experience, physical attractiveness does play an important role in relationships. A friend put it well: "The difference between a close friend and a lover is attraction."

      Just remember that attraction isn't only about looks.

      A long time ago, I took up weightlifting and martial arts in order to improve my physique. I've been at both for years. But IMO, the biggest successes I've had have come from improving my attitude.

      The absolute worst attitude problem is desperation. Until a geek conquers that, he is doomed to failure.

      I've also learned how fragile relationships are. I've destroyed some great ones over the course of a 15 minute phone call.

      I'd love it if a girl geek would be willing to chime in. : )

    13. Re:So... by stu9000 · · Score: 1

      You can keep 99% of girls. It's the the great 1% I go for.

    14. Re:So... by Bluesman · · Score: 1

      You're over-analyzing everything. All I was trying to say is that you can't take what women say about men's penises at face value.

      Your treatise on the ups and downs of dating with multiple personality disorder was a fun read, however.

      --
      If moderation could change anything, it would be illegal.
    15. Re:So... by Burning1 · · Score: 1

      With a razor wit like that, it's no wonder you've had a lot of bad experiences with women. I'm glad you fully appreciated my comment though.

    16. Re:So... by Ox0065 · · Score: 1

      tragically enough, not because of the M3, but rather because he's an asshole... ...if only comfortable, stable guys made women hot & sweaty... ...it'd be a different world.

      --
      thx e
    17. Re:So... by dintech · · Score: 1

      I know lots of stable women, including my wife of 14 years, my sister, my mother, my mother-in-law, a bunch of sisters-in-law and lots of friends.

      Are you offering some sort of service?

    18. Re:So... by dintech · · Score: 1

      I've destroyed some great ones over the course of a 15 minute phone call.

      Holy shit! *slaps forehead* What did you do?!

    19. Re:So... by cthulhu11 · · Score: 1

      Agreed, especially if the asshole is 6'5".

    20. Re:So... by ksheff · · Score: 1

      no, you've probably just become accustomed to their insanity.

      --
      the good ground has been paved over by suicidal maniacs
  28. Wait till they start modelling brains by Colin+Smith · · Score: 1

    You want an AI assistant? Thousands of small processors.

    --
    Deleted
  29. Re:But... by stonedcat · · Score: 0, Funny

    Why would you waste the processing power running Vista on this beast?

    You'd have about half of your system load working to prevent you from doing anything you actually want to do, a quarter running your software sound since DX is removing hardware sound support, and the rest would be used to make everything look "pretty". At that rate you could probably run solitaire with the remaining cpu cycles.

    --
    You can't take the sky from me.
  30. Progress by Duncan3 · · Score: 2, Funny

    Another huge technology gain for virgins living in their parents basements worried about their small penis.

    --
    - Adam L. Beberg - The Cosm Project - http://www.mithral.com/
    1. Re:Progress by couchslug · · Score: 1

      "Another huge technology gain for virgins living in their parents basements worried about their small penis."

      I don't care what motivates their purchases. They are paying for the hardware the rest of us will buy cheaply later in its product life cycle. The basement lifestyle must make for lots of disposable income...

      --
      "This post is an artistic work of fiction and falsehood. Only a fool would take anything posted here as fact."
  31. Obligatory question by Anonymous Coward · · Score: 0

    Is that in celsius or kelvin?

  32. Re:But... by zwizzlemydizzle · · Score: 1

    But can it run on Linux? Dude, you can run linux on a wristwatch. The question is, can it run Vista?

    LMAO. Such simple words could not be truer.

  33. Congratulations by Anonymous Coward · · Score: 0

    You managed to twist a story about overclocking of a new processor type into another 'the US government does evil' statement.

    1. Re:Congratulations by Anonymous Coward · · Score: 0

      But the astroturf, in it's various forms, will always be here..

    2. Re:Congratulations by mrbluze · · Score: 1

      For now the majority here on Slashdot, i.e. left-wingers, are letting the tinfoil hat minority take jabs at the government, but only because Bush is in office. But when Hillary is elected, they're going to come down hard on anyone with the audacity to do this. I think you underestimate the size of the said minority, its intelligence and rather apolitical stance. You point but one finger, yet three of your own fingers are pointing back at you.
      --
      Do it yourself, because no one else will do it yourself. [beta blockade 10-17 Feb]
  34. Hertz? I care about Watt much more by dallaylaen · · Score: 2, Insightful

    I have no need for a machine more powerful than mine. I would rather buy a silent one.

    --
    WYSIWIG, but what you see might not be what you need
  35. Not if it is putting out energy in other forms. by pavon · · Score: 4, Funny

    What king-matic didn't tell you is all that extra power is dissipated via X-Rays. It is called Skulltrail after all.

  36. Vapor Cooling? by ocirs · · Score: 1

    The article mentioned "Piednoel told us that one of these systems has been overclocked in his lab to 5.0 GHz stable, and 5.2 GHz nearly stable. The system uses vapor cooling.". The cooling system alone can easily cost more than a grand and it'll probably be a pain in the ass to maintain. I have a feeling that the CPUs will be more likely sold in the 4GHz range, without the vapor cooling system and that it was just a demonstration like how most car manufacturer's insane prototypes? The price performance ratio just plummets after that point with the inclusion of a thousand dollar cooling system that is most likely going to have higher maintenance fees and probability of failure. I just can't see any type of significant market for vapor cooled PCs, since the bottle neck is going to be the graphics cards rather than the CPUs at that high of a frequency for gaming, and computational intensive applications will probably be better off with cluster computing since it's more power and price efficient to have multiple machines, rather than squeeze every last drop of performance out of one.

    1. Re:Vapor Cooling? by smash · · Score: 1

      I just can't see any type of significant market for vapor cooled PCs, since the bottle neck is going to be the graphics cards rather than the CPUs at that high of a frequency for gaming, and computational intensive applications will probably be better off with cluster computing since it's more power and price efficient to have multiple machines, rather than squeeze every last drop of performance out of one.

      I *personally* can't see any significant market for Crays either, but they clearly exist and are worth big $.

      There comes a point where price/$ becomes insignificant, and you're more interested in MIPS/sqr metre. A cluster of pcs with less expensive processors will be outperformed by a cluster of vapour cooled machines - just like a cluster of Pentium 90s is outperformed by a cluster of Core2s. Technology marches on - what's expensive today is cheap tomorrow.

      Besides, the cost of the cooling system will drop. When i can buy a vapour cooling system (i.e. A/C) to fit to a room in my house for $1k there's no technical reason that the same sort of system for my PC should be exceedingly expensive.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    2. Re:Vapor Cooling? by smash · · Score: 1

      err.. by price/$ i meant mips/$.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    3. Re:Vapor Cooling? by ocirs · · Score: 1

      Vapor cooling a PC is a different matter though, you'd have to worry about condensation and liquid leakage. I thought the Crays were water cooled with anti conductive liquid in case of leakage? Also, if you were to install a cooling system that will easily cost more than a grand in mass, I'm sure the extra grand will buy you more than enough space for another server or two, or a dozen when it comes to blades. You also have to factor in the fact that the vapor cooling system itself is going to be taking up some space. With relatively high cheap performance blade servers or a couple of 2U servers, I think the Skulltrail setup top mips/m comparison either. I feel like its more about the bragging rights than anything, because all the applications that would actually be able to benefit from such power can get the same performance through a cluster at a much lower cost.

  37. Next, from AMD... by Junior+J.+Junior+III · · Score: 5, Funny

    The AMD Skullfucker-64 5300+ will 0wn this.

    --
    You see? You see? Your stupid minds! Stupid! Stupid!
  38. Re:But... by WhiteWolf666 · · Score: 3, Insightful

    I'm skeptical about that.

    I'd be fairly certain that the NSA uses some kind of off-the-shelf processors, whether that be Power, Itanium, or X86.

    What the NSA does different, most likely, is scale. You put 1,000 of these in a supercomputer? They'll put 100,000.

    Chip fabs are expensive, as is chip design. There's no reason not to leave that to the experts (AMD/Intel). It's a commodity process, and they'll do it better than the government ever can.

    Supercomputer design is something else. That's not commodity; and it's a simple scaling problem. More $$ = Bigger computer.

    Why should they bother reinventing the wheel?

    --
    WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
  39. Compiling kernels in our undies ? by unity100 · · Score: 1

    Cuz thats the only thing i can think of as a use for a 5 ghz computer in an average household.

    1. Re:Compiling kernels in our undies ? by Yosho · · Score: 1

      Cuz thats the only thing i can think of as a use for a 5 ghz computer in an average household.

      Editing home movies / photos? Playing games? Doing other things while watching hi-def movies? Or maybe just so that your computer and programs will start faster? That's not even getting into the "power user" that will use such a processor for rendering graphics, compiling code, and running other operating systems through virtualization. I think you're not thinking very hard.

      --
      Karma: Terrifying (mostly affected by atrocities you've committed)
    2. Re:Compiling kernels in our undies ? by unity100 · · Score: 1
      im not thinking very hard, but you apparently are :

      Editing home movies / photos? a 3 ghz amd with a decent (ok) graphics card is more than enough for this. unless you set up a cad studio at home.

      Playing games? again, with a 3 ghz + system with a good graphics card and mobo will be more than enough for it. and it is.

      Doing other things while watching hi-def movies? like doing computer animated cartoons at the same time and playing wow in windowed mode, and browsing the internet at the same time conversing with 4 people in 4 seperate im windows ? come on. get real.

      Or maybe just so that your computer and programs will start faster? 0.5 seconds instead of 1 seconds. worth it ? not. for computer start, 6 secs instead of 8 ? i guess not. remember that most of the program/computer start is tied to hard disk speed.

      That's not even getting into the "power user" that will use such a processor for rendering graphics, compiling code, and running other operating systems through virtualization. those "power" users already utilize many different setups that can be used for many different tasks. and whatever needed in the degree of 'power userness' you describe, is totally irrelevant for the masses.
  40. Re:But... by diamondsw · · Score: 1

    But can it run on Linux?

    No. But Linux can run on it...

    --
    I don't know what kind of crack I was on, but I suspect it was decaf.
  41. Re:But... by AeroIllini · · Score: 1

    Dude, you can run linux on a wristwatch. And IBM did, 6 years ago.

    And it might not be Vista, or even Windows, but Microsoft is also getting into the wristwatch game.
    --
    For security, the MD5 hash of this message and sig is 09f911029d74e35bd84156c5635688c0.
  42. Skulltrail? by Red_Icculus · · Score: 0

    That is one of the most awesome names for computer hardware ever. Almost as sexy as a Blade server. Whoever thought violent themes and computers would co-exist? I'm looking at you Hans Reiser.

  43. 5 GHz is not fast enough by ciaran.mchale · · Score: 1

    I'm going to wait till CPUs reach 640 GHz. That should be fast enough for anyone.

  44. ME TOO! by Anonymous Coward · · Score: 0

    Speaking as a game developer... I just got upgraded from a quad-core 3.2ghz Pentium 4 to a *2.4ghz* Core2 Quad which appears to be around 50% faster, despite the lower clock rate.

    I spend all of my time waiting for disk thrashing (e.g. doing a build involves reading tens of thousands of little source files, writing and reading many thousands of .obj files, etc).

    I would KILL for a decent solid-state hard drive of 64GB or more. I would put all my development stuff and temporary dirs on it and live happily ever after.

    (I'm also tempted to just turn Windows swapping off completely, since this box has 4 GB of RAM and so for me swapping just means doing nothing, slowly.)

  45. Re:This looks bad next to a amd dual quad-core sys by owlstead · · Score: 1

    Yeah, this sound like a well balanced system. It's not the fastest memory around, and the CPU will probably be beat just a bit by the Intel counterpart, but it is at least affordable. It also seems to use a lot less power/generate heat. As this is the enthusiast market, Intel must have a clear winner here.

  46. That's not a bottleneck by piotrr · · Score: 2, Informative

    I am annoyed by references to general "bottlenecks", because it's not just Anonymous Coward who comes up with this.

    It's just not that simple. A 5GHz CPU will be faster than a 3GHz CPU and 3 video cards will be faster than 1 video card almost regardless of other components. The only real bottlenecks you can talk about are the system busses and at the moment, that's not a problem either. HyperTransport 3.0 and intel's quad-pumped busses are still plenty wide enough for 5GHz processors, no sweat.

    I completely understand what you're trying to say, it's just that you're wrong. PCs aren't cars and processors aren't jet engines. A faster CPU will do more CPU work every second, a faster video card will give you higher framerates and more RAM will fix most of your stutters. The slowest part of your computer is the hard disk, so do whatever you can to exclude it from time-critical operations. If the rest of your system is waiting for data from that ancient storage device, THAT would be a bottleneck.

    --
    / Per
  47. Fixed that for ya... by holywarrior21c · · Score: 1

    Measuring computer performance in Hz is like buying a car based on red line RPMs. It only tells you one component that is meaningless by itself. Just like a car needs torque to give rpm's context, processors need how many instructions can be completed per cycle to be compared to the frequency. I've lost faith in the MHz race and generally look at benchmarks closest to the intended purpose of the processor.
    Measuring computer performance in Hz is like buying a car with good horsepower to attract woman. while you having succeeded to every woman with wrong Pron-ic vinyls all over your car that yells i am a frickest nerd in town. all i did was to buy a subaru and pu F1 engine that is awkwardly stickin out in your trunk.
  48. Wasting Time... by drgonzo59 · · Score: 3, Funny
    I'm posting this is because I'm waiting for my PC to solve a big math problem :-)

    You are wasting your time, the answer will always be 42....

    1. Re:Wasting Time... by swimin · · Score: 1

      Only if he asks the right question.

  49. Nothing to see here. by 0111+1110 · · Score: 1

    Phase change and LN2 cooled quads are already running at over 5GHZ. This is just intel themselves overclocking as is mentioned in the article. Is this even news? It's not like Intel is actually going to be selling a phase change or LN2 cooler to go along with their new platform. And even if they were this doesn't sound like any sort of advance in silicon as is implied by the article summary.

    --
    Quite an experience to live in fear, isn't it? That's what it is to be a slave.
  50. An addendum... by Burning1 · · Score: 1

    An addendum...

    I think it's very possible to be a comfortable, stable, and secure person around your friends, family members, and co-workers and still be insecure around women. Each are a unique experience.

  51. Oblig... by Anonymous Coward · · Score: 0

    I, for one, welcome our gigahertz-wielding overlords.

          --- Anon

  52. Then it's no longer idle. by SanityInAnarchy · · Score: 1

    I used to do this, until I realized that leaving my processor idle means less heat, less power usage, and a longer life.

    Now, maybe if it wasn't my computer -- for example, Amazon EC2, Flexiscale, even just a cheap host that's neglected to include power requirements in their spec. On our EC2 machines, I can imagine that we'd want to always have a certain amount idle, and I see no problem pegging that with distributed.net, SETI, whatever. We'd also probably leave them on slightly longer than we need them, as they take longer to start than they do to kill.

    But the essential difference there is: Amazon charges us the same rate per virtual machine whether or not we're actually using it. My power company absolutely does charge me more, so I won't be doing this in my own house. I'd rather just donate money to someone running this on a PS3.

    --
    Don't thank God, thank a doctor!
    1. Re:Then it's no longer idle. by Luyseyal · · Score: 1

      Of course you're donating power to the project. However, what's the point in owning an 8 core system at home if you're not going to use it????

      -l

      --
      Help cure AIDS, cancer, and more. Donate your unused computer time to worldcommunitygrid.org. Join Team Slashdot!
  53. 8GB RAM + SLI? by TheLink · · Score: 2, Interesting

    Just curious would SLI video cards and popular games actually work well with 64 bit Windows?

    Correct me if I'm wrong but if you're stuck with 32 bit windows there's no point having much more than 2GB RAM if you're doing SLI, given you have 4GB addressing space and the video cards would take a large chunk of that addressing space.

    --
  54. Re:But... by Nocterro · · Score: 1

    Not only that, but by buying commodity hardware and using their own software to make it scale, it'd be a helluva lot easier to make sure their purchases don't show up amongst the usual government spending. Their own custom chip-fab plant, plus all the associated requirements would be pretty noticeable if a foreign spy agency went looking, and from their it's only one step to studying the plant itself.

    --
    [clever sig]
  55. Estimating power at 5GHz by Tom+Womack · · Score: 2, Interesting

    Tom's Hardware just did a series of power-consumption tests on various overclocks of a Q9650, the first available 45nm processor.

    At 3GHz, it uses 8.79W when doing nothing, and 73W when running all four cores flat-out

    At 4GHz, it uses 16.83W to do nothing, and 135W with all four cores flat-out; on the other hand this required a voltage increase to 1.44V from the 1.25V that sufficed up to 3.33GHz.

    Fitting curves suggests that you would be using something like 350W for four cores at 5GHz, which is quite impressive.

    But a 5GHz Skulltrail would be a chip hand-picked by Intel to run at lower voltages in general, and would be running cryo-cooled which I think also allows a lower voltage to be used; probably 200W would be a better estimate for the chip power consumption, though the cooler will be using a comparable amount of power. This is, indeed, moderately crazy.

  56. Re:But... by encoderer · · Score: 1

    Just because their chips are made by AMD or Intel doesn't mean they're using off-the-shelf parts.

    I mean, the government doesn't have factories that produce fighter jets either, but when was the last time you rode in an American Airlines B-2 Stealth Bomber?

    There's no reason to believe that Intel or AMD isn't building a separate line of milspec chips. It might not be true, but it certainly wouldn't be anomalous.

  57. SkullTrail refers to dead on arrival... by PDX · · Score: 1

    When you overheat it something bad happens.

    1. Re:SkullTrail refers to dead on arrival... by Anonymous Coward · · Score: 0

      Is that as opposed to all of the other processors that have nothing bad happen to them when they're overheated?

  58. Re:But... by WhiteWolf666 · · Score: 1

    *shrug*

    Perhaps, it could be true, but I'm not certain they would gain any benefit.

    Let's say these milspec AMD/Intel chips are 4x as fast as consumer/commercial stuff. Let's say that AMD/Intel even sell them at a cheap price to the NSA (highly, highly, unlikely).

    It wouldn't matter one bit; the NSA would gain nothing. All they might have to do is reduce the "chip count" of their supercomputer by a factor of 4. All the innovative stuff their doing isn't in the processors; its how they arrange them.

    This would be different if AMD/Intel are producing milspec cryptograhpy chips which are 1-2+ orders of magnitude faster than commercial stuff, however, I think there would be more money in selling that stuff to the private sector; and given that they both have research/manufacturing facilities off-shore, I'd be very surprised if this escaped the notice of other governments (or, for that matter, other chipmakers).

    --
    WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
  59. Re:But... by sm62704 · · Score: 1

    Microsoft is also getting into the wristwatch game

    "What time is it?"

    "Dunno, my watch has a virus"

    --
    mcgrew's razor: Never attribute to stupidity that which can be explained by greedy self-interest
  60. Electricity. by SanityInAnarchy · · Score: 1

    I'm not talking about CPU power, I'm talking about electricity.

    If I had an eight-core system, and I had to leave it on for some reason -- BitTorrent, say -- it could easily drop to some 500 mhz per core, if it was properly designed, and still use even less power (electricity!) for each of those 500,000 cycles that were not used.

    Try running SETI on a modern laptop for more than about two minutes, and you'll get a very real demonstration of how this works. SETI on means the machine is not idle, means it will be scorchingly hot, and no longer a "lap" top. Leave it idle, and it might give you an extra hour (or two, or three) of battery life, while actually being cool enough to hold in your lap.

    --
    Don't thank God, thank a doctor!
    1. Re:Electricity. by Luyseyal · · Score: 1

      Yeah, I understand exactly what you're saying: I can save power/heat if I don't run these applications.

      My point is that if you have an 8 core desktop system (aka what this article is about), it is nigh pointless to use it as timidly as you suggest. As another poster said in this article, "You don't buy a Ferrari and then complain about the gas mileage". Anyone who will have the cash to shell out for one of these suckers is not going to give a crap about the electricity bill at the end of the month. In fact, they probably want to exploit the power of their new computational Ferrari.

      As such, I think it is very reasonable to use its power for the good of mankind. Mind you, your donation is not (yet) tax deductible.

      Lastly, you are a little outdated in your knowledge. World Community Grid has BOINC settings that allow you to throttle the usage as far down as you like. Only want 30% of CPU time on the project? No problem. 10%? No problem. Furthermore, it defaults to OFF when on battery.

      So, while you yourself may be so fiscally constrained that you cannot afford the electricity or you are worried about global warming more than AIDS or dengue virus, not all of the rest of us fit in those categories and we may wish to donate our Ferraris as we please.

      -l

      --
      Help cure AIDS, cancer, and more. Donate your unused computer time to worldcommunitygrid.org. Join Team Slashdot!
    2. Re:Electricity. by SanityInAnarchy · · Score: 1

      As another poster said in this article, "You don't buy a Ferrari and then complain about the gas mileage".

      They obviously don't know any rich assholes who own Ferraris.

      Anyone who will have the cash to shell out for one of these suckers is not going to give a crap about the electricity bill at the end of the month.

      Wasteful idiots, then. Yes, I would care about the electricity bill, for precisely the reason I outlined...

      In fact, they probably want to exploit the power of their new computational Ferrari.

      Presumably, if they bought this, they actually have some use for it.

      As such, I think it is very reasonable to use its power for the good of mankind. Mind you, your donation is not (yet) tax deductible.

      Which is why I think it would be more reasonable to simply donate money to a project which buys PS3s and turns them into nodes for a project like this.

      Lastly, you are a little outdated in your knowledge. World Community Grid has BOINC settings that allow you to throttle the usage as far down as you like.

      Irrelevant. I know how to throttle individual apps if I have to, whether they have such settings or not.

      Furthermore, it defaults to OFF when on battery.

      So then, if I have my laptop in my lap, but plugged in? Ouch.

      So, while you yourself may be so fiscally constrained that you cannot afford the electricity

      It's not about being fiscally constrained. It's about not being fiscally stupid.

      Now, maybe you have run the numbers, and decided that this is a lot more efficient use of your money to help these projects. If so, good for you, and I'd again suggest it really only makes sense on a desktop, or (especially) a server. (I mentioned EC2 as a possibility -- Amazon won't charge you more money if you bump their electric bill that way.)

      --
      Don't thank God, thank a doctor!
    3. Re:Electricity. by Luyseyal · · Score: 1

      They obviously don't know any rich assholes who own Ferraris.

      What is your point? I think it's fairly common knowledge you don't hear people who buy or review expensive sports cars complaining about gas mileage. Last time I read a review in Car and Driver, they mentioned the gas mileage but that was it.

      Wasteful idiots, then. Yes, I would care about the electricity bill, for precisely the reason I outlined...

      I happen to agree with you except why do you think PS3s are somehow better for this? Maybe your environmentalism has blinded you to caring about dumping more waste PS3s into our landfills after expensive manufacturing processes instead of using our existing hardware better? Furthermore, people who are reluctant to or unable to donate cash to a project but don't mind a nominally higher electricity bill can more easily contribute. Lastly, it distributes the weight on the various electrical grids rather than locally taxing specific areas.

      Irrelevant. I know how to throttle individual apps if I have to, whether they have such settings or not.

      If you know how to throttle processes, unless you use it on your lap for the vast majority of the time, then you're just a huge whiner if you won't bother throttling or shutting down a service when you're plugged in but using it on your lap. I mean, you wouldn't use a resource-intensive video game while on your lap, would you? You'd shut the game off first, let the system cool down, and then go about your poking in your email or whatever.

      It's not about being fiscally constrained. It's about not being fiscally stupid.

      Globally, it makes more sense to leverage widespread, existing resources rather than concentrated ones perhaps requiring more coal plants and manufacturing specialized hardware. Locally, it sucks (in the US/Canada anyway) that you cannot — yet — deduct your donation from your taxes.

      Now, maybe you have run the numbers, and decided that this is a lot more efficient use of your money to help these projects. If so, good for you, and I'd again suggest it really only makes sense on a desktop, or (especially) a server. (I mentioned EC2 as a possibility -- Amazon won't charge you more money if you bump their electric bill that way.)

      I'm not making anyone run this. Only you can decide for yourself whether it makes sense for you financially. However, the benefits of this research are far reaching and well worth the donation, IMNSHO, and I think for the vast majority of the targeted population (Americans and Europeans with disposable income), the difference in running it will be insubstantial.

      -l

      --
      Help cure AIDS, cancer, and more. Donate your unused computer time to worldcommunitygrid.org. Join Team Slashdot!
  61. Re:This looks bad next to a amd dual quad-core sys by Anonymous Coward · · Score: 0

    Multicore systems need increased parallelism in the OS and applications too. So Windows is out of the picture too. I doubt most of the ubercool, overclocking. watercooling. case-window-sporting. human-game-bot "jocks" want to use Linux on their PC. But that's precisely who's willing to pay lots of money for the latest Intel Hz-driven monstrosity in their PC.

    Otherwise the answer is medium to large businesses which don't have multithreaded/parallelized application workloads.