Slashdot Mirror


More Analysis Of Pentium M Desktops

Hack Jandy writes "The Pentium 4 has gotten enough attention lately as a slow, over heated monstrosity; but does Intel's Pentium M fare any better? Intel's decision to introduce the Pentium M as a desktop processor (East Fork) may not be all it's cracked up to be. Sudhian has an in-depth article, and Anand has benchmarks (on Linux!). I will stick with my Athlon 64, thank you very much."

347 comments

  1. Big Surprise by Anonymous Coward · · Score: 0, Funny
    I will stick with my Athlon 64, thank you very much.
    Translation: I am 14 years old and I have no need for any sort of stability in my "rig". I don't have to worry about getting any sort of real work done, so I play games all day and look at porn. If I can overclock a 3% performance increase, I'll cream my virgin shorts.
    1. Re:Big Surprise by Anonymous Coward · · Score: 0

      Ziiiiiiiiing!

    2. Re:Big Surprise by bzebarth · · Score: 5, Funny

      That's so not true. I know a lot of 30 year old Athlon lovers who play games and look at porn all day.

    3. Re:Big Surprise by drsmack1 · · Score: 1

      "Four legs good, two legs bad!"
      "Four legs good, two legs bad!"
      "Four legs good, two legs bad!"
      You work for Dell; right?

    4. Re:Big Surprise by wvitXpert · · Score: 2, Insightful

      For what reason are Athlon 64 processors not "stable"?

    5. Re:Big Surprise by Zathras11 · · Score: 1

      Right, like Pentium users don't look at porn,
      play games or overclock...

    6. Re:Big Surprise by hendridm · · Score: 1, Funny
      For what reason are Athlon 64 processors not "stable"?
      You get what you pay for?
    7. Re:Big Surprise by BW_Nuprin · · Score: 1

      My primary work rig is an Athlon 3000+ laptop, and I haven't had an ounce of trouble with it. I've only used it to play HL2 once, other than that its been my trusting code crunching compiling companion - without crashes.

    8. Re:Big Surprise by lsmeg · · Score: 3, Funny
      For what reason are Athlon 64 processors not "stable"?

      You get what you pay for?

      I hope you don't use linux then... :P

      --
      It's OK! I'm a limo driver!
    9. Re:Big Surprise by Barto · · Score: 1

      Translation: I've got nothing better to do than troll slashdot with almost self-parodying fallacies ("Athlons crash more", "Windows Steals Your Identity", "All Slashdotters Are Virgins"), some I'm probably a 34 year old who'll cream his virgin shorts when someone moderates me up.

    10. Re:Big Surprise by lacheur · · Score: 3, Insightful
      Translation: I am 14 years old and I have no need for any sort of stability in my "rig". I don't have to worry about getting any sort of real work done, so I play games all day and look at porn. If I can overclock a 3% performance increase, I'll cream my virgin shorts.
      Translation: I am an out-of-touch Intel fanboy who enjoys hemorrhaging money for equivalent performance because of a delusion that that Intel products are somehow more "stable".
    11. Re:Big Surprise by AttillaTheNun · · Score: 5, Informative

      My Athlon64 rig is solid as a rock and the more articles I encounter like this, the more I believe I switched over from Intel at the right time. I think AMD has pulled ahead permanently this time. Like most ./ers, I put together my own 'rig', allowing me to select the most highly rated, well-known and supported components at an affordable price. I have no need for underpowered systems with a bunch of useless software bundles - I prefer to put my money where I know it counts.
      I'm not a 14 yr old gamer, either. I earn a living designing software. The Athlon64 is about the best price/performaner (esp. considering the 'free' upgrade when moving to a 64-bit OS) that's come along in a very long time.

    12. Re:Big Surprise by Anonymous Coward · · Score: 1, Interesting

      I own an amd64 and I'm a independent programmer/security expert. I run only Linux (Gentoo if you must know, troll what you will it gets the job done) and I have tons of stability. The lovely gigantic GPR and extra XMM regs make the entire system so much more powerful. Recompile glibc while rendering video and running john-the-ripper nooo problem, bring it on! And the people who OC on amd64 are just tards, isn't the thing strong enough for you, do you really have to risk your hardware because you think your l337 wind0wz OCer. Also I don't wear shorts, I wear pants like a real coder.

    13. Re:Big Surprise by N5 · · Score: 5, Informative

      I don't know where you get your info, but i've had a 64 since march and it's been rock solid. The major problems with stability people have with a 64 is:

      1. Amperage too low on 12V rail. This is becomming more common with all systems. A number of Mobo makers are now taking the CPUs power out of the 12V rail, and since the memory controller is integrated into the CPU, well you see the issue. It's a simple fix, get a decent PSU with at least 20+ amps on the 12V rail.

      2. Memory. A common problem with ALL machines. This is also an easy fix: BUY DECENT MEMORY. You don't need corsair, but get a respected name. People also claim that you can't run at any timings faster than 3-3-3, but that ISN'T the case. I'm typing this with a Gig of PC 3200 running 2-3-3. Haven't tried to oc it because I have no need, but tighter timings might be possible.

      As far as overclocking goes, no you won't get massive numbers, but with the basic understanding that you're dealing with more than just FSB and multiplyer, you can get a decent overclock.

      Intel fan boys need to calm down. Netburst hasn't been what intel needed. AMD gave them plenty of warning that they were going to release a pretty advanced chip. Intel decided that GHZ are what matter, and that everyone would want an Itanium for 64 bit computing. wrong on both accounts. It's good to know that they are going back to more reliable tech, but when I have a cool running, stable machine that can go toe to toe with an Intel Extreme Edition (I hate extreme marketing) and cost me a fraction of the price, I'm happy.

      --
      John 3:16 - The easiest way to a BETTER YOU.
    14. Re:Big Surprise by Anonymous Coward · · Score: 0

      I dont know, but the only instability I ever had on my Dual Athlon rig was the Intel Network card.
      Replacing it fixed it all.

    15. Re:Big Surprise by xanadu-xtroot.com · · Score: 1
      Well, I don't overclock, anyway...

      1. $ uname -a
        Linux aragorn 2.4.28-gentoo-r3 #2 Sun Dec 26 19:42:15 EST 2004 i686 Intel(R) Pentium(R) 4 CPU 2.40GHz GenuineIntel GNU/Linux

      --
      I'm not a prophet or a stone-age man,
      I'm just a mortal with potential of a super man.
    16. Re:Big Surprise by Phil+Urich · · Score: 1

      Hmm. Odd how I can leave my family's A64 on for seemingly unlimited periods of time, showing almost constant use, and it's never crashed. Yup, it's sure unstable.....

      --
      I remember sigs. Oh, a simpler time!
    17. Re:Big Surprise by Anonymous Coward · · Score: 1, Funny

      Also I don't wear shorts, I wear pants like a real coder.

      So you're not a REAL coder: I know a guy who has the shorts, the hawaiian shirt AND the beard! He's powered by coffee and cigarettes.

    18. Re:Big Surprise by Anonymous Coward · · Score: 0

      No but you're a gentoo user who named their machine "aragorn"

      That says a whole lot of _somthing_

      I'm just not sure what...

      -funroll-the-one-ring?

    19. Re:Big Surprise by Teckla · · Score: 2, Insightful

      Translation: I am 14 years old and I have no need for any sort of stability in my "rig". I don't have to worry about getting any sort of real work done, so I play games all day and look at porn. If I can overclock a 3% performance increase, I'll cream my virgin shorts.

      Better translation: I prefer superior hardware that also happens to cost less.

      You're a troll, and should be modded as such.

    20. Re:Big Surprise by be-fan · · Score: 1

      Um, you're about 4 years late. Sun is shipping servers based on the Athlon64 architecture now. AMD is no longer for 14 year olds.

      --
      A deep unwavering belief is a sure sign you're missing something...
    21. Re:Big Surprise by fnord242 · · Score: 1

      Bah! No, keep your Athlon64. Us 30 somethings use Alphas and have done so for the last 10+ years. We have rock stable 64 bit systems, get our work done, and still have CPU cycles left over to look at porn. We overclock the x86 systems for fun to see if they will catch up to the Alpha without burning out.

    22. Re:Big Surprise by pipsey · · Score: 1

      That's odd... please, tell my A64 server to go crash a couple of times for me then. (It hasn't seen a reboot in about 3 months)

      This isn't 1997, and AMD isn't the new kid on the block anymore. When it comes down to it, Intel and AMD perform about equally on the top end, with AMD coming out just in front, and there's no perceptable difference in "stability." The real difference between the two (excluding, for the moment, the "bonus" of future 64-bit computing, unless you're using unix) is price, and guess who wins that one?

      On the topic of the Pentium-M, it is, in my opinion, the best showing anyone has right now for mobile computing, but it's no match for an A64 on the desktop. I'm looking to replace my laptop later this year, and I'll be going P-M.

    23. Re:Big Surprise by Anonymous Coward · · Score: 1, Funny

      Pentium users mostly fish for young victims in AOL chat rooms.

      It's a little known fact, but convicted child molestors almost execlusively use Windows. Does this mean the pedophiles that run Linux don't get caught as easily?

    24. Re:Big Surprise by arose · · Score: 1

      You could pay foe a SCO license I guess...

      --
      Analogies don't equal equalities, they are merely somewhat analogous.
    25. Re:Big Surprise by geekoid · · Score: 1

      " is price, and guess who wins that one?"
      When you calculate TCO? Intel.
      You don't get that extra heat for free, you know.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    26. Re:Big Surprise by pipsey · · Score: 1

      Sounds about right. If I buy a Prescott, I can lower my gas bill in the winter!

      Athlons being hotter than Intel chips is nothing more than an urban legend. Upper-end Athlon64s and Prescotts BOTH put out a lot of heat.

    27. Re:Big Surprise by Anonymous Coward · · Score: 0

      (esp. considering the 'free' upgrade when moving to a 64-bit OS)

      i though't that linux allready supported Athlon64 and allready was 64 bit OS... am i wrong?

      or are you refferring to 64 bit Windows? nah, couldn't be, because you mentioned free upgrade...

    28. Re:Big Surprise by (H)olyGeekboy · · Score: 1

      Ain't that the truth. I run into these people all day. Still tell me "AMD is not stable at all. When I had an AMD 5x86 133, it used to crash all the time. It is worth it for me to pay $250 more for my processors because AMD sucks so bad. You'll regret it when your AMD crashes on you!"

      Right. I've been building Athlon XP systems for three years now... for myself, wife, friends, parents... and have never had stability issues except in the case of failed DIMMs.

      Just because Toyota cars may have sucked in the 70s is no reason to prance around like a nancy boy today telling everyone how Toyotas suck. :)

    29. Re:Big Surprise by xanadu-xtroot.com · · Score: 1

      No but you're a gentoo user who named their machine "aragorn"
      That says a whole lot of _somthing_
      I'm just not sure what...


      Actually, I named him far before the movies cam out. I've read the trilogy twice prior to buying the movies. Since you may know the story of the Fellowship of the Ring, there is a reason this machine is named Aragorn.
      Gandalf,
      WAS the leader. In his absence, Aragorn was tapped. Thus... (Gandalf was a dual MMX-233 machine; he is no longer active, stuck in my closet, but not active.)

      --
      I'm not a prophet or a stone-age man,
      I'm just a mortal with potential of a super man.
  2. The answer by Anonymous Coward · · Score: 1, Funny

    I will stick with my Athlon 64, thank you very much.

    You're welcome.

  3. Power Post? by Anonymous Coward · · Score: 0

    PowerPC?
    First Post Bios check?

    WhatMeWorry?

  4. Failed Pentium by superpulpsicle · · Score: 5, Funny

    So is Pentium M just a bunch of failed Pentium processors that didn't fare well in the assembly line? Sort of like Celeron... cough cough.

    1. Re:Failed Pentium by KarmaMB84 · · Score: 1

      No, it's a "new" (sorta) architecture apparently based on the Pentium III's architecture. It lacks some of the more advanced features of the P4 and even Celeron lines but it's lower power and will actually beat a Pentium 4 clock for clock in some applications.

    2. Re:Failed Pentium by 0111+1110 · · Score: 2, Interesting

      and will actually beat a Pentium 4 clock for clock in some applications.

      Actually make that all applications. At least clock for clock. It also tends to beat the Athlon64 clock for clock but that's a much closer race. The P4 is such a marketing-driven dog of a processor. Thank god I will never have to own one.

      --
      Quite an experience to live in fear, isn't it? That's what it is to be a slave.
    3. Re:Failed Pentium by Thnikkaman · · Score: 2, Funny

      Many Dothans died to bring us this information

    4. Re:Failed Pentium by Krach42 · · Score: 1

      Not really. The Pentium M has a branch penalty for all branches, where as the Pentium 4 only has a branch penalty for mispredicted branches.

      This accute difference would point to them having different cores for certain.

      --

      I am unamerican, and proud of it!
    5. Re:Failed Pentium by Krach42 · · Score: 1

      clock for clock, it's almost impossible to *not* beat the Pentium 4. I mean, it does so little on each individual clock-cycle in its pipeline, that it's ridiculous.

      The Alpha beats just about everything I've heard of clock-for-clock, same with the Itanium. But when's the last time people cared about clock-for-clock performance rather than MHz.

      If you want something that beats the P4 clock-for-clock, start looking at non-x86 architectures. PowerPC, Alpha, Itanium, they're all better architectures, because they were designed with an air of sanity.

      --

      I am unamerican, and proud of it!
    6. Re:Failed Pentium by Rosonowski · · Score: 1

      Those must be cousins of the bothans? ;-)

      --
      01101001 01100001 01101101 01101110 01101111 01110100 01100001 01101100 01100001 01110111 01111001 01100101 01110010
    7. Re:Failed Pentium by Anonymous Coward · · Score: 0

      One Dothan, Two Bothans.

  5. Pentium M clocks down too much by dotslashdot · · Score: 4, Interesting

    My experience with Pentium M is that it clocks down BIG time if you don't plug in the power cord. So much so that the laptop is virtually useless. YMMV.

    1. Re:Pentium M clocks down too much by Anonymous Coward · · Score: 1, Informative

      Meanwhile, my brand new Athlon 64 laptop:

      a) Doesn't have any drivers available online
      b) Refuses to install both 32- and 64-bit versions of Fedora 1-2, SuSe, and Mandrake (locks up before setup completes)
      c) Locks up on Windows 'shutdown'

      Better slow that retarded...

    2. Re:Pentium M clocks down too much by the+angry+liberal · · Score: 2, Informative

      Then turn it off in the BIOS and quit complaining?

    3. Re:Pentium M clocks down too much by Momoru · · Score: 3, Informative

      I had this same problem too, and I could not turn it off in the BIOS. My 1 Ghz chip ran at 500Mhz most of the time, even when plugged in.

    4. Re:Pentium M clocks down too much by dotslashdot · · Score: 1

      I was not complaining--I was pointing out one of the "features" of the Pentium M. If you turn off the clock-down "feature," then your battery only lasts for 1 hour because the Pentium M burns through more power than a Dick Cheney luncheon. And FYI: complaining about my informative comment as just "complaining" is offtopic and misplaced. Or you're new to the internet, where posting ontopic responses to articles is not considered "complaining" but is actually the whole point of a discussion forum. Maybe you should become an edumacaded liberal?

    5. Re:Pentium M clocks down too much by Xabraxas · · Score: 1

      Was it a Pentium III or a Pentium III-m? Are you using linux? Linux doesn't support the earliest version of speedstep on the Pentium III. If I start my laptop up without ac power then it will remain on 500Mhz no matter what, and it will stay on 700Mhz if I booted the computer with ac power. The speedstep options in the kernel specifically state that this version is not supported.

      --
      Time makes more converts than reason
    6. Re:Pentium M clocks down too much by Bloater · · Score: 5, Funny

      > My experience with Pentium M is that it clocks down BIG time if you don't plug in the power cord. So much so that the laptop is virtually useless.

      Funny, my desktop does the same...

    7. Re:Pentium M clocks down too much by n1ywb · · Score: 3, Interesting

      I've been using my Fujitsu Lifebook P5000 1GHz P-M for a few months now, and I have to say I don't really notice the performance difference when it clocks down. It's still perfectly useable. The only time I really notice it is when I'm compiling or something, and even then it's pretty fricking fast IMO. People are spoiled by fast CPUs nowadays, consarnit.

      --
      -73, de n1ywb
      www.n1ywb.com
    8. Re:Pentium M clocks down too much by TiggertheMad · · Score: 1

      People are spoiled by fast CPUs nowadays, consarnit.

      Yes, yes, procs are very fast these days, and all these young whippursnappurs are impatient young wags. However, wouldn't you be pissed off if you shelled out several hundred dollars extra for a fast proc that always runs at half speed?

      --

      HA! I just wasted some of your bandwidth with a frivolous sig!
    9. Re:Pentium M clocks down too much by Covener · · Score: 3, Informative

      I had this same problem too, and I could not turn it off in the BIOS. My 1 Ghz chip ran at 500Mhz most of the time, even when plugged in.

      The info returned by /proc/cpuinfo will only reflect how things looked when you booted, but it doesn't mean your CPU speed isn't changing. see cpufreq (/proc/cpurfreq)

    10. Re:Pentium M clocks down too much by cnettel · · Score: 1

      Not if that keeps the fan quiet until I run a 3 minute build in JavaCC, when it will clock up again. If I force it down, the same build will take 10 minutes. And, of course, if I check the clock speed in some software that's not designed to properly react to SpeedStep, I can get values like 209 MHz.

      It's like the belief that a lot of free memory is a good thing. A lot of memory that doesn't have to be paged out is a good thing, but "free" memory is just as useful as idle cycles due to a high frequency -- it shows that you could do better in some way.

    11. Re:Pentium M clocks down too much by darkwhite · · Score: 1

      Only if you set the SpeedStep setting in the BIOS to "maximum battery life" or some such. Mine ramps up perfectly well from 500 to 1700 MHz when crunching, whether plugged in or not.

      --

      [an error occurred while processing this directive]
    12. Re:Pentium M clocks down too much by the+angry+liberal · · Score: 1

      Or you're new to the internet, where posting ontopic responses to articles is not considered "complaining" but is actually the whole point of a discussion forum. Maybe you should become an edumacaded liberal?

      I see I left myself open to some asinine attempts at insults when I responded with a logical solution to, what appeared to be, a trivial issue.

      My apologies your laptop has a bad battery. Mine runs for a couple of hours with these features off. It really isn't a big deal, also, if you get yourself a gig of RAM. The CPU slowdown seems tripled if you are having to scoop up bits of a fragmented swap file at 1/2 speed.

      Anyway, quit complaining already.

    13. Re:Pentium M clocks down too much by kurzweilfreak · · Score: 0, Funny

      Funny, so does my blender, my TV, my refrigerator...

      --

      kurzweil_freak

      5th Kyu Genbukan Ninpo/KJJR student

      Be the darkness that allows the light to shine.

    14. Re:Pentium M clocks down too much by MooseGuy529 · · Score: 3, Interesting

      That's totally bogus.

      If you are running a modern, ACPI-enabled OS, processor speed is fully controllable by the OS. My Pentium M sits at 600 MHz all the time, unless I need it, and then it throttles up to 1700 MHz as needed. My guess is that you are running Windows, since Linux uses the highest clock speed unless you install a throttling daemon (I use speedfreqd.)

      I do know, however, that the Pentium 4-M throttles down a ton, because its power management features are less efficient and the battery life would be less than an hour. As it is, most only get 1 to 2 hours.

      What OS are you running, anyway?

      --

      Tired of free iPod sigs? Subscribe to my blacklist

    15. Re:Pentium M clocks down too much by iocat · · Score: 1

      My Pentium M (IBM ThinkPad R40 1.3Ghz) seems pretty speedy for normal apps when clocked down, and also lasts fairly long when not clocked down. But I don't know what your applications are (I'm just using it for word processing, image editing, email, and watching DVDs -- nothing crazy processor intensive).

      --

      Dude, I think I can see my house from here.

    16. Re:Pentium M clocks down too much by brsmith4 · · Score: 1

      Send it back. It's not AMD's fault. Next time, don't buy your shit from some company no one has ever heard of just to save a buck.

    17. Re:Pentium M clocks down too much by Methuseus · · Score: 1

      My Athlon-M laptop shows the change in /proc/cpuinfo, and has for a while. I haven't used anything older than a 2.6 kernel with it, though.

      --
      Two things are infinite: the universe and human stupidity, though I'm not yet sure about the universe. - A Einstein
    18. Re:Pentium M clocks down too much by darc · · Score: 1

      On my Dell 600m laptop, it doesn't permanently clock down, but instead adjusts the clock speed down until you're using an application, upon which it'll clock up to maximum speed. Pentium M chips have multiple behaviours depending on what you set it to do.

      This application lets you switch them if you desire.
      http://www.diefer.de/speedswitchxp/

      It depends on how you set your laptop 'power saving mode' in the control panel, normally, but this makes it explicit. The "Dynamic switching" function will allow it to auto throttle. You might have it set to "maximum battery life".

      You can watch it clock up and down with cpu utilization, and that's how it should work!

      --
      Tired of legitimate data sources? Try UNCYCLOPEDIA
    19. Re:Pentium M clocks down too much by Goonie · · Score: 2, Informative
      That's entirely configurable. Under Linux, on my Pentium-M laptop I run cpufreqd, which allows you to configure exactly how you want the clock speed to change under power conditions and cpu load. If the CPU load goes up, mine cranks all the way up to full speed again.

      I believe that the Windows drivers allow you to do the same thing, if you want.

      --

      Any sufficiently advanced technology is indistinguishable from a rigged demo
      --Andy Finkel (J. Klass?)
    20. Re:Pentium M clocks down too much by Sivar · · Score: 2, Informative
      Meanwhile, my brand new Athlon 64 laptop:

      a) Doesn't have any drivers available online
      b) Refuses to install both 32- and 64-bit versions of Fedora 1-2, SuSe, and Mandrake (locks up before setup completes)
      c) Locks up on Windows 'shutdown'
      Funny, my Athlon64 laptop which I purchased over a year ago:
      a) Uses standard hardware so drivers are no problem to find online (or from the reseller or on the included install discs). One exception being the Radeon 9600m, which I can still easily find drivers for.
      b) Has compiled both 32-bit and 64-bit Gentoo from stage 1 and never crashed (though even with an Athlon64 it takes forever)
      c) Doesn't lock up on Windows shutdown, suspend, hibernate, or when transcoding xvid three days in a row.

      Have you considered that maybe--possibly, your laptop might be damaged, or that the one you chose was horribly crap and that it is your fault, not AMDs? I mean, my god, even eMachine's Athlon64 laptops work fine in my experience!
      --
      Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
    21. Re:Pentium M clocks down too much by Jeff+DeMaagd · · Score: 1

      It ashould be completely adjustable. If you pick the most battery conserving mode, it probably wouldn't up the clock for the instantaneous bursts of CPU power needed for many tasks. If you were so inclined, you could set it for maximum clock even without the power cord, but usually somewhere inbetween is good for most people.

      For Windows XP, a utility called sswitchxp (short for speed step switch XP) allows the user to tweak a lot of these modes. I can't say much about Linux, I don't use Linux on my laptop.

    22. Re:Pentium M clocks down too much by dotslashdot · · Score: 1

      I WAS running XP but then switched to Linux, which didn't slow things down noticeably as much (I was also just word processing & surfing, posting on slashdot, etc.) For other reasons, I ditched the laptop for another without really looking into the issue extensively.

    23. Re:Pentium M clocks down too much by serbanp · · Score: 1

      What OS are you running on your PentiumM?

      When I got my Inspiron 600m one year ago, I installed RH9 on it, just to find out that the 2.4 kernel was not supporting Banias' SpeedStep technology. At that time, Intel was promising to release a binary driver for Linux, but nothing came out in the following few months.

      I would like to revisit the issue, as in WinXP I really appreciate the 8.5 hours of battery life (using also the second battery pack), but in Linux the juice dries off after less than 4 hours.

      Could you share some info about what worked for you?

      Thank you,

      Serban

    24. Re:Pentium M clocks down too much by angulion · · Score: 1

      Same with my desktop..
      But at least it's good as a paperweight.. ;)

    25. Re:Pentium M clocks down too much by LucidBeast · · Score: 1
      Actually, they don't know that quality of life was better back when processors were slower. One application that I first coded in 1994 used to take 45 minutes to compile. How did I use that time? I learned to drink coffee, read newspaper, unfortunately internet wasn't available to me nor did it have the same entertainment value back then.

      Now the same application takes 1 - 2 minutes to recompile. Not enough time even for grammar and spell checking of ./ posts.

    26. Re:Pentium M clocks down too much by MemoryDragon · · Score: 1

      That is controllable... even a year ago I used (funnily) in Linux PowerNowD to control that thing. And even in 600MHz the thing was perfectly usable, but once you need raw processing power, given the correct settings it scales up to whatever frequency is allowed.

    27. Re:Pentium M clocks down too much by Anonymous Coward · · Score: 1, Informative

      yeah, virtuallky useless.

      you want to make other shit up?

      I have 35 pentium M based dells right here and when on battery they are a tad slower but are still extremely useful and most users do not even notice it.

      If you wanted performance, why did you buy a notebook with an el-cheapo processor? I got the dells for less than $525.00 each, pentium M processors are the value line and are only in the cheap laptops.

      Wah, my laptop does not have a geforce FX 6800 video card with 512 meg of ram.... wah.....

      get a clue and stop makeing up lies.

    28. Re:Pentium M clocks down too much by araemo · · Score: 2, Informative

      My experience with Pentium M is that it clocks down BIG time if you don't plug in the power cord. So much so that the laptop is virtually useless. YMMV.

      Luckily, this is 100% configurable.

      Even if you don't want to turn it off completely, you can set up your system to be less aggressive in down-clocking your CPU.

      In windows 2k/XP, open the 'power' control panel, and change the setting from whatever it's set at(I'm guessing "Max Battery" to something less drastic, like Portable/Laptop, or Home/Office Desk(Yes, home/office desk will still allow it to downlock, but it will clock it back up when you start using it more.). "Always on" and "Minimal power management" both disable down-clocking, I believe. And the on-demand CPU frequency scaling does wonders for battery life. I've never felt that my laptop was hampered by being unplugged, but any time I stop using a lot cpu, it down-clocks to save battery.

      Under linux, you just need to have a decent kernel and cpufreqd or speedfreqd(And I was just compiling 2.6.9 last night, it seems that it includes a kernel driver that will dynamically change the clock speed dependant on CPU load, so cpufreqd/speedfreqd are not strictly necessary if that is enabled.

      I've also done some benchmarks/useage comparisons, and my Pentium M laptop is signifigantly faster than my desktop at several things(Nothing disk-bound though. ;P), like compiling a lot of software, and working with some encryption formats. All-in-all, my 1.5 year old Penium M is still a fairly usefull system, that even plays most games competantly, if not with all the pretty features turned on. :(

    29. Re:Pentium M clocks down too much by Fred_A · · Score: 1

      Most if not all the problems I've seen on amd64 machines come from the motherboard chipset implementation.

      I know the nForce 3 on the chaintech motherboard I use is fairly flaky. I'll probably replace the motherboard eventually for something more stable. The problem is apparently really with Chaintech here, not with the nForce3 which seems to work fine on a number of other brands.

      This is on a desktop box (which now runs fine after a lot of tweaking) but I suppose similar problems can happen with laptops.

      OTOH it never crashed.

      --

      May contain traces of nut.
      Made from the freshest electrons.
    30. Re:Pentium M clocks down too much by Anonymous Coward · · Score: 0

      at the risk of being redundant, have you tried SpeedSwitchXP? http://www.diefer.de/speedswitchxp/

      It works only on windows Xp, but there might be other tools that do the same on other OSs. The point being, that your complaint of having your laptop refusing to run at full speed when not plugged in is not a problem without solution.

    31. Re:Pentium M clocks down too much by dotslashdot · · Score: 1

      Hi Serban, Unfortunately I just ended up using the plug all the time because under Linux I too did not get much battery life. Because I gave the laptop away before investigating further, I never figured out how to get around the problem. Sorry dude/dudette.

    32. Re:Pentium M clocks down too much by MooseGuy529 · · Score: 1

      I'm running Gentoo Linux with a (unrelatedly) patched kernel. There are drivers for SpeedStep in the kernel; they're open-source.

      If you go into Power Management options > CPU Frequency scaling, you will find lotsa options. You need at least one governor. I would compile all of them as modules. Then you pick "Intel Enhanced SpeedStep", and "Use ACPI tables..." beneath it. Then do "make" and "make modules_install", and try loading the new modules. You then need either cpufreqd or speedfreqd (or another similar utility) to monitor CPU usage and change speed. The kernel also has a module called "ondemand" that does what speedfreqd does (it is equivalent to the "dynamic" policy). I haven't tried it, but it's probably more responsive and efficient since it's in kernel-space.

      If you have any more questions, email me at slashdot.thinkinginbinary@spamgourmet.org or IM me (AIM: thinkinginbinary/Jabber: name is thinkinginbinary, domain is jabber.org, with the usual character in between)

      --

      Tired of free iPod sigs? Subscribe to my blacklist

    33. Re:Pentium M clocks down too much by MooseGuy529 · · Score: 1

      Sorry for the second message.

      I have a friend who has an Inspiron 600m with both battery packs, and he runs Gentoo as well. If you give me your email address or IM name(s) I will give it to him to see if he can help.

      So you really get 8.5 hours of battery life? That's amazing! Once you get SpeedStep working in Linux, it really works like a charm... my laptop (Asus M2400Ne, only supports one battery) got 1.5-2 hours of battery life before I set up speedstep, and it now gets 4-5 with it.

      --

      Tired of free iPod sigs? Subscribe to my blacklist

  6. AMD Is Gaining Speed by sammykrupa · · Score: 0

    Yea, AMD Pulled ahead really fast.

    1. Re:AMD Is Gaining Speed by Anonymous Coward · · Score: 0

      Thank you for pointing that out captain obvious.

      Soon AMD will realise that it will have no fair competition, raise its prices knowing that people will still pay and then the I'l be able to play "Diner With Andre" emulated on my A64.

  7. It's funny because it's true. by Anonymous Coward · · Score: 0

    n/t

  8. Best place for AMD systems by augustz · · Score: 5, Interesting

    I've gotten old enough that I no longer thrill at the idea of building my own system. I'm looking for something quiet, very reliable, and inexpensive. Performance comes behind these critiera.

    Basically I'm looking for the Dell equivelant in the AMD world, someone who cranks them out in great quantities. I checked out HP etc, wasn't blown away. Also open to a smaller shop if they come with a good recommendation (and without the insanely gaudy cases, no rounded plastic please).

    1. Re:Best place for AMD systems by MrRuslan · · Score: 1

      Here in NYC you can go to any local computer store and grab a custom system built the way u want for a good price. In my opinion it is better to buy from little computer stores than from the big boys because you get better quality and service.

    2. Re:Best place for AMD systems by FuzzzyLogik · · Score: 1

      Go build a nice system from Falcon Northwest or Alienware or Voodoo PC.. you get choices and they build very nice systems.. You don't have to get the crazy cases most of the time either.. give it a shot

    3. Re:Best place for AMD systems by shellbeach · · Score: 2, Interesting

      Basically I'm looking for the Dell equivelant in the AMD world, someone who cranks them out in great quantities. I checked out HP etc, wasn't blown away. Also open to a smaller shop if they come with a good recommendation (and without the insanely gaudy cases, no rounded plastic please).

      Just because you dislike the idea of building your own system doesn't mean you should ignore white boxes from the dodgy-bros. local store. IME, you get exactly the parts you want, there's no proprietary crap and you get it all at the cheapest price (you generally get a discount for buying a system, so getting someone to put it together for you works out cheaper than doing it yourself) You don't need to pay for a useless OS, either. And if something does go wrong with a part, it's no sweat to replace it yourself even in the unlikely event that the shop won't - you save so much you'll still be ahead on the deal.

    4. Re:Best place for AMD systems by Anonymous Coward · · Score: 0

      For "real" work use, the chipset matters as much, if not much more than, the processor. You likely did take the cheap route, With the A64-939, for example, use the GA-K8NS Ultra-939; that is stability. Instability is genuinely your own fault, inability to research and locate stable chipset for use in "real" work use.

    5. Re:Best place for AMD systems by mollymoo · · Score: 5, Funny
      I've gotten old enough that I no longer thrill at the idea of building my own system. I'm looking for something quiet, very reliable,

      So buy a Mac.

      and inexpensive.

      Oh, OK, don't buy a Mac then :)

      --
      Chernobyl 'not a wildlife haven' - BBC News
    6. Re:Best place for AMD systems by jm92956n · · Score: 1

      Here in NYC you can go to any local computer store and grab a custom system built the way u want for a good price.

      OT: Might you be able to recommend a reputable dealer? I'm in NYC and I might soon be ready for a new PC (my aging celerson-500 is, I think, about ready to go). I've seen numerous ads in the Voice and N.Y. Press for places that computers at insanely low prices, but I admit I've never actually gone to one--I've always assumed they were the bait-and-switch types.

      --
      An effective signature identifies a particular user amongst a base of thousands.
    7. Re:Best place for AMD systems by augustz · · Score: 1

      I did this with my last system. I even had them test boot it in the store. It turns out though that that about 80% of the time the system has trouble with BIOS, a known issue with the motherboard being finnicky it turns out.

      The small place doesn't have a great return policy. Basically, once you've purchased the CPU / Memory / MB etc you own them.

      I've put together a TON of systems myself, and I'll be honest, until you've tested a combo of MB/CPU/Mem a significant number of times it is hard to guarantee reliability, there are a lot of little edge cases that you can spend a lot of time working on. And the focus on many review sites is speed.

      I guess I havn't run into many shops that do a super easy return policy and reasonable prices.

    8. Re:Best place for AMD systems by shellbeach · · Score: 1

      I've put together a TON of systems myself, and I'll be honest, until you've tested a combo of MB/CPU/Mem a significant number of times it is hard to guarantee reliability, there are a lot of little edge cases that you can spend a lot of time working on.

      Never had this problem - but then I generally go for a standard big-brand MB and RAM: it doesn't make much of a difference in cost. Maybe I've just been lucky, but I would have thought that if you're not overclocking things and not playing around with dodgy generic brands you'd be pretty safe. The nice thing with white-boxes is that you can specify exactly what you want.

      I guess I havn't run into many shops that do a super easy return policy and reasonable prices.

      Here in Melbourne, Australia, the place that I consider to be the cheapest I know from experience has a very easy return policy. They seem to make their profits through high turn-over and are rapidly turning into a major chain, so I guess they're doing something right!

    9. Re:Best place for AMD systems by carl0ski · · Score: 1

      lol MSY :P, ever had them helping you then they answer the phone and start chatting :P i'm too lazy to go to Malvern hope Mulgrave opens very soon :)

    10. Re:Best place for AMD systems by St.+Arbirix · · Score: 3, Interesting

      I would have said Netlux but they recently pulled a fast one on me. My laptop from them is an oversized clunky beast... but it's a 2.00Ghz machine from the summer of 2002 that cost me $1100, an amazing deal at the time (Clemson school laptops were $1500 1.5Ghz IBMs).

      And talk about dependable... I've taken it apart about 5 times, once to paint the exterior with my own designs, cut holes in the casing, etc and it still works fine. Occasionally the flourescent light for the LCD would flicker out but that was just a matter of opening her up, unplugging the screen, and plugging it back in a couple times. The parts are all Sony, Fujitsu, Toshiba, and other brand names... the laptop runs really hot and sometimes the harddrive (Toshiba) will click a few times and stop working on me which has been happening for a year and just a matter of letting it cool down. Since I never actually move the laptop (there are 6 firewire drives daisy-chained off of a poorly placed 4-pin port in the front center of the laptop) I plan on shelling out the slot fan and copper radiator in favor of some cheap water-cooling experiment (hopefully involving a decorative waterfall).

      My basic point is that a well rated off-brand computer store from pricewatch.com will land you with a Volvo among computers that outruns Miatas, isn't winning design awards, and despite the fact that sometimes it shuts off by itself or won't start immediately it can always be depended on to come up with a few retries and not get any worse with age (my girlfriend's 1984 tank/Volvo is just like this).

      --
      Direct away from face when opening.
    11. Re:Best place for AMD systems by lachlan76 · · Score: 1

      Really? Which place would that be?

    12. Re:Best place for AMD systems by FuzzyBad-Mofo · · Score: 1, Offtopic

      I have a faster computer, a larger penis and better hair than you.

      Oh yeah? Well I have a faster penis, larger hair, and a better computer.

      But maybe I've said too much..

    13. Re:Best place for AMD systems by shellbeach · · Score: 2, Informative

      That'd be CentreCom. I know that CPL isn't bad either. But check The Age's Green Guide - there's heaps of other cheap stores out there (though can't comment on their service).

    14. Re:Best place for AMD systems by Lehk228 · · Score: 1

      Have you looked at eMachines? no really don't laugh, Mine(T2200SE) has been great to me (the motherboard was replaced, but it wasn't their fault I screwed up when I was trying to update the BIOS) My old one (533id2) was great for years, and will probably soon be given to my brother's friend who needs a computer, My brother has one, my girlfriend has one, one of my friends has two, another friend has 3 or 4 of them.

      The design of their recent and semi recent ones is compact and, while looking nice on the front doesn't scream for attention from across the room like an alienware. Also they come with Nero rather than depending on windows XP (shitty) CD and DVD writing

      --
      Snowden and Manning are heroes.
    15. Re:Best place for AMD systems by spauldo · · Score: 1

      E-Machines power supplies are flaky, though - back when I used to run a computer shop, I must have had a dozen of 'em before I junked 'em.

      In general, you won't see a problem with them - the failure rate is a tad higher than normal, but the chances are still in your favor. Just don't add any electric-happy devices or you're asking for trouble.

      Depending on the model, make sure you keep it somewhere cool (i.e. away from heat vents) and where it can get airflow. Some of them have a real heat dissipation problem - HP pavillions and compaqs have the same thing. If you open the case and it's a jumbled mess, then you have a potential meltdown on your hands.

      --
      Those who can't do, teach. Those who can't teach either, do tech support.
    16. Re:Best place for AMD systems by lachlan76 · · Score: 1

      I live in Adelaide, and in the city, there's a computer store about every 40m along some streets ;)

      I get my stuff from some cheap place, good place to buy from, if you can understand people that have a really strong Vietnamese accent.

      And who may or may not have ever heard of Linux.

    17. Re:Best place for AMD systems by fbg111 · · Score: 1

      Try Shuttle XPC G5 9500g or 9500m:

      http://sys.us.shuttle.com/BuildXPC.aspx?id=1089

      Athlon64, Nforce3 Ultra, built-in heat-pipe CPU cooler, etc. etc. I just a built a barebones version, and it's well-layed-out, small, quiet, performs well, and is so far reliable.

      --
      Flying is easy, just throw yourself at the ground and miss. -Douglas Adams
    18. Re:Best place for AMD systems by HalfFlat · · Score: 1

      For Adelaide, I've had good experiences with Adrenalin on Hindley Street for the most part. Linux isn't their specialty, but they're not ignorant of it and they will research hardware compatibility.

      They do burn-in tests on the systems they assemble, and try to cater to the high-performace and (to a lesser extent) the low-noise crowds.

    19. Re:Best place for AMD systems by lachlan76 · · Score: 1

      I usually go to GetRight, their prices are quite good, and there is one person who doesn't tell me to reformat my hard drive and come back when the problem happens under knoppix.

    20. Re:Best place for AMD systems by geekoid · · Score: 3, Funny

      "I have a faster computer, a larger penis and better hair than you."

      Mom?

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    21. Re:Best place for AMD systems by Anonymous Coward · · Score: 0

      I've gotten old enough that I no longer thrill at the idea of building my own system. I'm looking for something quiet, very reliable, and inexpensive. Performance comes behind these critiera.

      That's exactly how I felt when I bought a Mac.

    22. Re:Best place for AMD systems by Anonymous Coward · · Score: 0

      try acer, and keep an eye on Aldi and Lidl if any near you

  9. A very neat processor indeed by elh_inny · · Score: 5, Interesting

    When I browsed through the test, I headed directly to the database section and I was positively surprised, P4M excels in this area.
    In my computing I actually find hard disks to be a bottleneck. I use databases all the time and any improvement in that area is a plus.
    I bet Gentoo fanboys will lament on processor's performance while compiling, I think it has more to do with the lack of the optimisations yet and what's even more important I don't compile much, I just use the computer.
    Overall I find this processor to be a very attractive solution for a typical desktop computer.
    It's a great base for a SFF or even smaller computer with more than adequate computing power.

    1. Re:A very neat processor indeed by Lisandro · · Score: 1

      It would be much appealing if it wasn't $500 (sans motherboard). The performance it's impresive indeed, but AMD still runs faster (and all of this in 32-bit software, which is half of what the Athlon-64 CPUs can do). For desktops, it doesn't make much sense, even considering the (marginal) power savings - i want to see what Intel comes up with next, their "next-gen" CPU architecture. Things will get interesting then.

      As it is, it's a terrific processor for portables, and maybe blade servers. There're better offerings for a desktop, IMHO.

    2. Re:A very neat processor indeed by Jeff+DeMaagd · · Score: 2, Informative

      I was positively surprised, P4M excels in this area.

      It is not P4M. Aside from the FSB and the latest SSE version, Pentium M is not even remotely based on the Pentium 4 core.

    3. Re:A very neat processor indeed by stupidfoo · · Score: 1

      Well, it's not a P4M (that's a Pentium 4 Mobile, which is different), but anyways...

      We're deploying some new Linux based embedded systems and we're moving towards a small form factor system with a Pentium M board. We've been testing it for about 2-3 weeks now it's exceeded our expectations. We currently use Pentiums, Pentium 2s, and Pentium 3s for some of our other embedded systems, but none of these offered the power we needed (as a side note, we also use FreeDOS and WinXP Embedded, our linux distribution of choice is currently Redhat Enterprise). We're a company that's going to be paying for Linux, so three cheers for us... or something.

      Due to the size of the unit and the installation location where much of the hardware will be located the Pentium 4 and AMD/AMD 64 chips were just too hot.

    4. Re:A very neat processor indeed by IncohereD · · Score: 1

      even considering the (marginal) power savings

      Ummm..since when is more than 50% marginal?? They're saying it's a maximum of 30W, whereas AFAIK even my 2.53 P4 is somewhere around 70W.

      Not to mention the ability to throttle back to 5W at low load. Get a grip.

    5. Re:A very neat processor indeed by Lisandro · · Score: 1

      It's marginal when compared to the latest AMD offerings, whose power consumption varies between 35W and 70W. The 90nm versions even tops at 50W! And this is the desktop versions, the mobile versions use even less power and have the option to throttle back the clock automatically when idle, just like the P-M.

      The P4 was always a power hungry processor. Again, for a modern desktop, there're comparable alternatives to the P-M for much less money.

  10. Review focused on gaming by Man+in+Spandex · · Score: 5, Informative

    http://www.gamepc.com/labs/view_content.asp?id=dot hangaming&page=1

    and it shows how differently it performs compared to things like compiling a kernel in linux. According to the review, it competes almost as fast or sometimes as fast as the A64 in some games.

    It's still an impressive cpu and better than tha bacon-cooker (prescott).

    1. Re:Review focused on gaming by Man+in+Spandex · · Score: 1

      Weird. /. added a space in that url after "dot". Not sure why that happended.

    2. Re:Review focused on gaming by WalksOnDirt · · Score: 1

      It's not a bug, it's a feature!

      Slashdot doesn't let you put in long string without spaces, because some idiots were messing up the page formatting with long strings for fun.

      --
      a,e,i,o,u and sometimes w and y (at be if of up cwm by)
    3. Re:Review focused on gaming by Zorilla · · Score: 1, Redundant

      Why wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwhatever do you mean?

      --

      It would be cool if it didn't suck.
    4. Re:Review focused on gaming by katorga · · Score: 2, Interesting

      I'm seriously thinking about a Pentium-M for my next homebuild for one reason: QUIET. I've had it with multiple fans, massive heatsinks and the noise from top of the line amd/intel chips.

      The P-M would be a perfect chip for a media center or SFF gaming rig. 95% of the performance of an AMD 55FX or P4 3.6 at half the cost and 20% the heat is nothing to sneeze at. And that running with slow ram and agp 4x. Last time I checked a 55FX as $800ish, a P4 3.4EE or 3.8 was $700ish, and the 2Ghz P-M is only $400ish.

      And Anand is missing one fact on the compiler front. GCC may not have great P-M support, but Intel gives away their optimizing compiler for Linux.

      Given the heat characteristics of the Pentium M, Intel could likely put 4 on a die for the same heat cost as a dual core Prescott or a dual core AMD64. It remains to be seen if they can overcome their Hubris and give the Isreali chip team the ball.

    5. Re:Review focused on gaming by Anonymous Coward · · Score: 0

      Redundant? Idiot mods...

  11. Pentium M by MrRuslan · · Score: 4, Interesting

    I have no clue why would anyone buy this. I mean Pentium M is great for laptops because of the lower power consumption but there is very little to gain from it on the desktop. It is very overpriced for a standard workstation onfiguration where somone dosent need power. I mean it saves power but not enogh to make it worth the trouble.

    1. Re:Pentium M by Anonymous Coward · · Score: 0

      The first four words in your post sums it up. Read the article and do some research. Only then will you understand why they want to put it in racks.

    2. Re:Pentium M by MrRuslan · · Score: 1

      I didn't say anything about racks.I was talking about desktop systems. In racks this could be usefull.

    3. Re:Pentium M by mollymoo · · Score: 4, Insightful
      I have no clue why would anyone buy this. I mean Pentium M is great for laptops because of the lower power consumption but there is very little to gain from it on the desktop.

      Noise. If you produce less heat you don't need as much cooling, so you don't need to shift as much air. Moving air through a PC makes noise.

      Other than noise, the lower power consumption may not help much for a single PC, but saving 40W per PC when you have 200 or 2000 can add up. Remember you often pay twice for you PC's power consumption - once to heat the air and once for air-con to cool it again.

      --
      Chernobyl 'not a wildlife haven' - BBC News
    4. Re:Pentium M by afidel · · Score: 4, Insightful

      electricity is expensive and certainly isn't getting cheaper. If you can get a new box with a Pentium M and a LCD then you are going to use less power even for a more capable rig. This might not matter for your home PC, but it matters if you are running an office with a couple hundred PC's or a business with thousands of them. It also matters if you want a box capable of decoding video but don't want a bunch of noisy fans, or if you run on alternative power. A capable, low power cpu is definitly going to find a niche for itself. Oh yeah and throw 4 of them on a die with a couple megs shared cache and you now have a chip that draws about the same amount of power as a P4 but kicks its butt all over the place on multithreaded code.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    5. Re:Pentium M by Anonymous Coward · · Score: 0

      Electricity here costs 6 US cents per kilowatt hour.

      Dirt cheap.

    6. Re:Pentium M by Stevyn · · Score: 1

      The Pentium M is a combination of the P3 and P4. The P4 is still slow at very high clock speeds compared to Athlons because of the long pipes. The doesn't. So a P3 running at speeds of a P4 makes for a faster processor, if I've read correctly.

      Remember how the P4 was slower than the P3 when it was introduced? They assumed a high clock speed would have fixed this problem, instead it just made a lot more heat. I've seen a 3.4 ghz P4 get hot enough to burn skin.

    7. Re:Pentium M by GreyWolf3000 · · Score: 1

      Reductions in air conditioning needs actually far outweigh reductions in power. Most of people's energy bills come from A/C.

      --
      Slashdot: Where people pretend to be twice as smart as they really are by behaving like children.
    8. Re:Pentium M by Spellbinder · · Score: 1

      performance, low noise level and low power consumption
      take a 2 ghz cpu overclock it to 2.5 with stock cooling
      you get a cpu that rocks away everything for games with peak temperature around 50 C with the small stock cooler

      --


      stop supporting microsoft with pirating their software!!!!!
    9. Re:Pentium M by FuzzyBad-Mofo · · Score: 1

      I recently had the privilege of witnessing the voltage regulator on a friend's mobo do the same thing. (Fire... bad!)

    10. Re:Pentium M by Anonymous Coward · · Score: 0

      I have no clue

      Correct.

    11. Re:Pentium M by JW+Troll · · Score: 0

      It's quiet. Some people prize silence over an extra 3fps in Quake 3.

      --
      just like the humble blood clot... turboporsche@telus.net
    12. Re:Pentium M by Jeff+DeMaagd · · Score: 2, Insightful

      Reductions in air conditioning needs actually far outweigh reductions in power.

      The energy cost of cooling is usually less than the energy cost of making heat. Usually it is about 10:1 on a decent A/C system, it takes 10W of electrical power to remove 100W of heat.

    13. Re:Pentium M by iamnotacrook · · Score: 1

      How is it that a post beginning "I have no clue" can be insightful?

    14. Re:Pentium M by Jeff+DeMaagd · · Score: 1

      Not saying that it would be, but if left on continuously, a desktop consuming 50W total at idle would cost about $26 a year to operate. A desktop operating with a 200W idle power would cost about $105 a year to operate, a difference of about $79 a year.

      Really, most of the benefit in a desktop system would be in smaller, cooler and quieter operation.

    15. Re:Pentium M by afidel · · Score: 1

      $79*200~=$16,000 or enough for a few extra servers per year =) Most IT directors will take that anytime! Not only that but it also means a few tons less sulphur dioxide and quite a few tons less CO2 produced, which if you're even a little interested in the environment is a good thing. All of that benifit at basically zero opertunity cost, why not?!?!

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    16. Re:Pentium M by Anonymous Coward · · Score: 0

      Moderators have no clue either?

      More seriously, this is a standard "Why" post. They are always present if there are some new or cool technology making a headline at slashdot. They also always are insightfull for some odd reason.

    17. Re:Pentium M by joib · · Score: 1


      $79*200~=$16,000 or enough for a few extra servers per year =) Most IT directors will take that anytime! Not only that but it also means a few tons less sulphur dioxide and quite a few tons less CO2 produced, which if you're even a little interested in the environment is a good thing. All of that benifit at basically zero opertunity cost, why not?!?!


      Probably because the Pentium M and its motherboard is significantly more expensive than a comparable A64/P4 setup. Companies are interested in minimizing total cost over the lifetime of the computer, of which yearly electricity costs are only a part.

    18. Re:Pentium M by MemoryDragon · · Score: 2, Informative

      The Pentium-M is a kickass desktop processor, it just has not the brand GHz on its package. Still one of the fastest there is, excellent power management (which is a big plus on a desktop where you dont want to hear fans all the time and where you idle 99% of the time anyway)

      The main problem is the price, it is just much to pricey compared to an AMD64, I probably will go the AMD 64 route in the middle of next year because they also have a good power management. The main selling point for AMD64 for me is its Coool&Quiet technology, I could not care less about the 64bit features, which are totally useless currently for the stuff I do. (wake me up once the emulators start to use the higher number of general purpose registers)

      So there is this decision between a Pentium-M and AMD 64 bot excellent desktop processors and in the end AMD will win again due to its lower price tag.

    19. Re:Pentium M by MemoryDragon · · Score: 1

      The P-M is in general purpose computing around 1.5x times faster than a comparable P4, it is an excellent processor and definitely the only good one Intel has produced the last 10 years. It took Intel ages to recognize that this thing is heavens sent and should not only be sold over laptops. If they now would lower the price for the processor and board towards AMD64 it could give Intel the needed time...

    20. Re:Pentium M by Anonymous Coward · · Score: 0
      once to heat the air and once for air-con to cool it again.

      This must be true, since 100% of the world lives in Southern California.

      Otherwise, you might say that having a floor full of PCs warms up the building, reducing heating bills in the winter.

      But of course, we all live in the Golden State so that can't be true. As you so sagely and cleverly put it, we're all paying twice.

    21. Re:Pentium M by dusty123 · · Score: 1

      That's not true: The cooling does not cost you much energy. All fans together consume ~ 10W - this is few compared to 200W power consumption.

    22. Re:Pentium M by Anonymous Coward · · Score: 1, Informative

      I have two systems at work. Both Dells. One is a desktop 2.6GHz P4 machine, the other is a 1.7GHz Pentium M laptop. Memory and hard drive the same. I do hardware design and spend a lot of time compiling RTL designs for FPGAs (which can take hours, even on the fastest machines). I find my Pentium M laptop is roughly 20% faster than the P4 desktop... I've always attributed this to the fact that the Pentium M comes with more cache than most P4 systems (1M/2M instead of a typical 512K). That's just my experience with it. I've had no problem with the freqency throttling, it always speeds back up without any noticable lag.

    23. Re:Pentium M by DGregory · · Score: 1

      They'd still turn the AC on so high that you have to take a thick sweater into work. If they were really concerned about energy costs, they'd at least make the office comfortable in short sleeve shirts!

    24. Re:Pentium M by evilviper · · Score: 1
      [...] the lower power consumption but there is very little to gain from it on the desktop.

      You know something... I'm SURE Intel thought the same thing when they decided to go ahead with their Pentium 4. Unfortunately, they were very, very, very wrong.

      Heat really is the single reason why the P4's death has been such a long-time comming. Heat has become a huge problem with P4s.
      --
      Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
    25. Re:Pentium M by IncohereD · · Score: 1

      But of course, we all live in the Golden State so that can't be true. As you so sagely and cleverly put it, we're all paying twice.

      He said _often_. In Ottawa (i.e. 'Silicon Valley North'...blargh), for instance, you pay twice for maybe half the year, and pay less than once the other half.

  12. But the article says... by obeythefist · · Score: 4, Informative

    That the Pentium-M isn't optimised at all for what they were benchmarking (apart from some stuff compiled with a non-commercial intel C compiler).

    While I'll be one of the first to put my boot into intel and their behind-the-market sloppy overpriced inefficient CPUs, it would be at least fair to do it on a reasonably even playing field.

    --
    I am government man, come from the government. The government has sent me. -- G.I.R.
    1. Re:But the article says... by AKnightCowboy · · Score: 2, Interesting
      That the Pentium-M isn't optimised at all for what they were benchmarking (apart from some stuff compiled with a non-commercial intel C compiler).

      Because God knows that all the software in the world is compiled with a highly optimized commercial Intel C compiler right? Come on guys, why do you expect them to use some crazy expensive compiler when 99% of the software on the shelf will never use it? It's just a marketing gimmick to boost their performance. If nobody uses it then it might as well not exist.

    2. Re:But the article says... by be-fan · · Score: 1

      The thing is, that nobody really compiles with the Intel C++ compiler. Certainly, nearly all Linux software is compiled with GCC, while nearly all Windows software is compiled with Visual C++. In practice, not using Intel C++ *does* level the playing field.

      --
      A deep unwavering belief is a sure sign you're missing something...
    3. Re:But the article says... by EspressoMachine · · Score: 1

      It's just a marketing gimmick to boost their performance. If nobody uses it then it might as well not exist.

      I couldn't agree more...buuuuut - they do exist, and while most end users will never have access to them (let alone use them), most companies use them (not this particular product, but similar, specially tweaked tools and drivers) to boost test results. Graphics chip companies do it, and so does everyone else. So, while it's not a particularly honest practice, not using one or more of these methods of tuning for specific benchmarks can make one company look much worse than another (more so than they may really be).

      On top of that, some companies force publishers into an agreement whereby to get review copies (pre-release) of software, the review has to be approved by that software company. This doesn't sound like a tempting situation, but no publisher wants to be the last one to publish a review, especially when consumers have been using it for just as long as the reviewers. Being able to offer a "first look" is a pretty big deal, and a big bargaining chip for manufacturers.

      And always remember the saying: "There's lies, damnable lies, and then there's statistics"

      --
      Despite conventional wisdom, I've discovered you can blame a guy for trying. It's called "attempted murder".
  13. I wanna do that when I grow up.. by MrElcee · · Score: 1

    Playing games and looking at pr0n would be my dream job.

    Now might be my chance to break into the lucrative pr0n game beta tester job market.

  14. Surely PowerPC G5 is the way forward? by Anonymous Coward · · Score: 1, Funny
    Why would anyone want to use an Intel chip or an AMD chip is beyond me. All that backwards-compatible legacy circuitry eating up real-estate on the die.


    With the advent of 64-bit OS-X, the Mac G5 platform is the only serious contender for my desktop. And I would have thought that the tech-savvy readership of this website would feel the same.

    1. Re:Surely PowerPC G5 is the way forward? by MrElcee · · Score: 0, Offtopic


      OS X is throwing itself against the walls of GNU/LINSUX and will be greeted, god willing by the shoe of Linus!

      (written on an iMac)

    2. Re:Surely PowerPC G5 is the way forward? by Anonymous Coward · · Score: 0

      I'm with you on that. AMD is an Intel knockoff which is a Zenith knockoff, and we all know what Cyrix was. Motorola(now IBM's chip?) always had a superior product, and I sure wish that more hardware maufacturers used them. This whole "economy of scale" thing is throwing us for a loop, as well as IP law.

    3. Re:Surely PowerPC G5 is the way forward? by Waffle+Iron · · Score: 1
      Why would anyone want to use an Intel chip or an AMD chip is beyond me. All that backwards-compatible legacy circuitry eating up real-estate on the die.

      The legacy circuitry allows the chip to use an old-school compact instruction format, thus saving die real-estate in the instruction cache. The caches on modern chips are much bigger than the logic core, with or without legacy circuits.

    4. Re:Surely PowerPC G5 is the way forward? by nxtr · · Score: 1

      >Why would anyone want to use an Intel chip or an AMD chip is beyond me. All that backwards-compatible legacy circuitry eating up real-estate on the die.

      You gotta always be prepared just in case you find yourself stuck with your computer in 1981, without software or the hard drives.

    5. Re:Surely PowerPC G5 is the way forward? by Methuseus · · Score: 1

      Are you saying there is no legacy circuitry in a G5 chip? I would have to disagree with you on that unless you could cite a specific article. And believe me, I'll look for my own article to prove my point, because I believe I have seen it.

      Also, look at the Athlon 64. They threw out a lot of the legacy circuits that have barely been used for years. Yes, there is yet to be a large commercial OS that is 64-bit (unfortunately Mandrake and Red Hat and the like don't support 64-bit as much as they do 32), but once it is there, is there really any point to using a G5 chip?

      You can make the argument go either way, so your point is really moot.

      --
      Two things are infinite: the universe and human stupidity, though I'm not yet sure about the universe. - A Einstein
    6. Re:Surely PowerPC G5 is the way forward? by Anonymous Coward · · Score: 0

      Surely, troll, you need to actually back up what you are saying with some facts. Oh, and by the way, port all the software that runs on x86 hardware only over to G5 and then come back and show us how well your new software compares the the athlon 64 specs. We'll be waiting fanboy.

  15. Every geek... by Doolspin · · Score: 3, Insightful

    Every real geek has owned a dual proc. Intel machine.

    1. Re:Every geek... by matts-reign · · Score: 0

      You mean i'm not a geek..... *sniff* /snuggles with his cray XT3....

      --
      Waffles rock.
    2. Re:Every geek... by mr_z_beeblebrox · · Score: 2, Funny

      Every real geek has owned a dual proc. Intel machine.

      No...every REAL geek can remember saving up for their first math co-processor.

    3. Re:Every geek... by Anonymous Coward · · Score: 0

      amd dude, amd!

    4. Re:Every geek... by FFFish · · Score: 2, Funny

      Every real geek can remember the first time they took a soldering iron to the motherboard to piggyback 4k RAM chips in a hack to double the memory of their machine.

      --

      --
      Don't like it? Respond with words, not karma.
    5. Re:Every geek... by Anonymous Coward · · Score: 0

      I owned a dual Alpha CPU machine, not an intel machine. NT4.0 rocked on that.

    6. Re:Every geek... by GreyWolf3000 · · Score: 1
      I built myself a dual P3-1ghz machine about 2 years ago and haven't looked back since. The only problem is that I bought a cheap motherboard based on the i810 chipset--I'm stuck at 512 megs of RAM and a 133mhz fsb.

      It still runs Linux beautifully.

      --
      Slashdot: Where people pretend to be twice as smart as they really are by behaving like children.
    7. Re:Every geek... by FuzzyBad-Mofo · · Score: 1

      Don't forget notching the opposite side of the 5.25" disk to double the capacity. ;)

    8. Re:Every geek... by ldspartan · · Score: 1

      pfffffffft.

      My mail flies through a 4-way P233 Compaq in the basement :).

      --
      lds

    9. Re:Every geek... by Anonymous Coward · · Score: 0

      Every REAL geek has made up their own motherboard with some 74x series TTL gates and a bunch of wire-wrap.

      Oh wait, I guess that was just me.

    10. Re:Every geek... by mr_z_beeblebrox · · Score: 1

      Every REAL geek has made up their own motherboard with some 74x series TTL gates and a bunch of wire-wrap.

      Wow! Now I feel like just a PC user. Though I did install a bunch of 4k ram chips on a RISC box, but never didI even KNOW someone who made there own MB.

    11. Re:Every geek... by crisco · · Score: 1

      ha! I have the same board running Linux. Or at least a board using the same hack on the 81x chipset (I thought mine was an 815). Picked it all up cheap, too. Now I just need faster hard drives.

      --

      Bleh!

    12. Re:Every geek... by Anonymous Coward · · Score: 0

      Ahh the Abit bp6.

      Mines still going strong as a file server.

      Those were the days...

    13. Re:Every geek... by Anonymous Coward · · Score: 0

      I suppose you could call a 486sx+487 combination a dual proc. machine.. it's not like you didn't really have 2 alus + 2 fpus.

  16. 64-bit goodness by pp · · Score: 5, Interesting

    Was unfortunately left out. I mean, Athlon 64 makes a fine Pentium 4 competitor when running a legacy 32-bit operating system, but it's so much more. Those cool extra registers you get in 64-bit mode make the thing just scream!

    And no, the intel EM64T stuff isn't even competing in the same league, 40-45% slower with 40% more GHz is what I've seen in real-life workloads (heavy numbercrunching). For some other types of loads it does just about as well as the a64/opteron, though.

    Revised x86_64 support (possibly in the pentium m core and in the same price range as the new 90nm a64's) and Intel has a chance. That and Microsoft delaying 64-bit Windows for a couple more years.

    1. Re:64-bit goodness by FuzzzyLogik · · Score: 1

      Nothing real world (for normal users anyway) exists for it yet. I know my parents don't care if it runs a database faster, can address more than 4gigs of ram or what have you. Nor do my brothers care about this, they want games to run as fast as possible. Those are all 32 bit games, and there's no real 64 bit windows yet.. sure the beta, but good luck getting it to work as an everyday system. They don't care about linux either so it doesn't make sense to judge the 64 bit part of this for most normal users. besides, you'd be comparing apples to oranges... the Athlon64 vs the Pentium 4 in 64 bitness? Obviously the Athlon64 will win.. it's the only one to have 64-bit registers. There's really nothing to compare it to on the desktop side of things either other than a mac. and even then you aren't comparing things fairly.

    2. Re:64-bit goodness by Anonymous Coward · · Score: 1, Informative

      he's probably talking about a64/opteron versus Nocona (EM64T, meaning Intel's equivalent of amd64) so that would be apples to apples. Dell had been selling Noconas for quite a while now, so it's fairly "real world". And x86-64 is not just about more that 4G addressable space - read the OP again.

      yeah, and for number crunching windows (64bit or not) is irrelevant, so again, what was it you were addressing from the parent post?

    3. Re:64-bit goodness by louarnkoz · · Score: 1
      pp wrote:

      Revised x86_64 support (possibly in the pentium m core and in the same price range as the new 90nm a64's) and Intel has a chance. That and Microsoft delaying 64-bit Windows for a couple more years.

      Actually, the beta version of 64 bit Windows XP is already available on the microsoft web site. I would not bet that it will be delayed "for a couple more year".

      Who wants to bet that two years from now all desktops will be 64 bit?

    4. Re:64-bit goodness by Anonymous Coward · · Score: 0

      your brothers can lick my hairy nuts. gaymers are only a small percentage of computer users.

    5. Re:64-bit goodness by Jeff+DeMaagd · · Score: 1

      Who wants to bet that two years from now all desktops will be 64 bit?

      All desktops, no, I doubt it, the average lifecycle is maybe three or so years and many 32 bit systems are still being sold, 64 bit is still an exception.

      But if you mean all new desktops, I can definitely see this happening.

    6. Re:64-bit goodness by pp · · Score: 1

      Should have been "That or Microsoft..." really.
      Once those 64bit FPSs start showing up for Windows, people will run whatever gives them the best framerate. Until then, whatever happens in the 64bit mode is irrelevant apart from the Linux market.

    7. Re:64-bit goodness by TheGavster · · Score: 1

      Both the Athlon64 and the Pentium use registers rather wider than 32-bit when handling massive calculations. The ALUs in use are something like 128 bits wide, though usually that's partitioned down to several 32 or a pair of 64 bit operations, simply because the software is only asking for 32 bits worth of range. I'm so far unconvinced that there are all that many benefits to running a 64-bit processor; most of the benefits of wide registers have been exploited in MMX and SIMD, and larger memory spaces are handled through segmentation (not that I have seen a machine with more than 2GB of memory, let alone anything close to that 4GB ceiling you hear so much about).

      --
      "Because Science" is one step from "Because old book". Try "Because of my experiment testing my falsifiable assertion".
  17. How slow do you want to go today? by Anonymous Coward · · Score: 0

    Unsafe at any speed?
    Great quote
    "Frankly, Intel, this is getting pathetic"

  18. Bus speed is the big issue by StandardCell · · Score: 4, Informative

    If you look at the front side bus speeds of the Pentium M, they're low. Very low, in fact, at 400MHz. Certainly not in the 800MHz -1066MHz range that's required for a lot of operations. A 1.5GHz P-M is about the equivalent of a 2.4GHz P4 Northwood UNLESS it comes to data-intensive operations requiring FSB access, and then it gets constricted.

    Let's reserve judgment on the P-M's future unless and until Intel builds a higher FSB speed or unless the biggest priority is low overall system power.

    1. Re:Bus speed is the big issue by Anonymous Coward · · Score: 1, Informative

      I don't disagree that FSB is important for certain type of algorithms. However, there are other things that impact performance such as pipeline setup (length, qantity, etc) and cache (size and other attributes).

      My feeling is that having a faster FSB might improve operations involving long arrays, but apart from that, it won't help things much. Remember that the P-M already comes with a fair bit of cache (for a desktop CPU).

    2. Re:Bus speed is the big issue by Jeff+DeMaagd · · Score: 2, Informative

      If you really looked, you'd find that the newest one comes in 533MHz FSB.

      I really don't know what operations require 800MHz and up for best performance (I/O intensive vs. compute intensive), but if East Fork is a reality, I bet that faster FSB speeds are in the works for a desktop version. The lower FSB was intended to minimize power consumption, which is a major priority on power sensitive laptops, but even if a higher bus speed means a couple extra watts, I doubt it would make a difference even in the fanless systems.

    3. Re:Bus speed is the big issue by ErikZ · · Score: 1

      (blink)

      What operation *requires* a 800-1066MHz bus speed?

      --
      Democrats or Republicans. They are both taking us to the same place and they are not afraid of us anymore.
    4. Re:Bus speed is the big issue by noewun · · Score: 1
      Requires? Dunno.

      However, moving around huge amounts of data gets easier the faster the bus -2 gigabyte Photoshop files are an example.

      --
      I am a believer of momentum and curves.
    5. Re:Bus speed is the big issue by Anonymous Coward · · Score: 1, Informative

      the 800 and 1066 MHz "FSB" speeds used in the Pentium 4 are bullshit numbers.

      The bus runs at 200 and 266MHZ
      The Pentium M bus runs at 200Mhz, with 266 coming very soon.

      The pentium 4 uses a quad pumped bus, that samples 4 times per cylcle hence 800 and 1066, however the memory it's talking to only does 400 or 533Mhz (that's 200 and 266DDR), What good is 800Mhz if you only use every other one?
      Remember the Pentium 4 was originally designed to use Rambus RDRAM that worked quite differently to SDRAM

      The Pentium M - and Athlon for that matter - are double pumped, hence 400 and 533Mhz, and match up perfectly to the 400Mhz (200Mhz, DDR) and 533Mhz (266Mhz, DDR) memory being used...
      The athlon has never shown itself to suffer from the "lower" speed bus, so why should the Pentium M?

    6. Re:Bus speed is the big issue by 1000StonedMonkeys · · Score: 2, Informative

      That's why the 800MHz bus pentiums use dual channel DDR. Two 400MHz channels sync up with an 800MHz channel pretty well.

    7. Re:Bus speed is the big issue by Anonymous Coward · · Score: 0

      Well... (correct me if i'm wrong, like i know you will, but) the high bus speed on the P4 is to provided lots of bandwidth to the deep pipeline design of the P4. The PM does not have nearly as a deep as a pipeline (it has a higher IPC), so it would not need as much memory badwidth as the P4 does.

    8. Re:Bus speed is the big issue by geekoid · · Score: 1

      you know how it is, a geeks first beer, and they get all crazy with the numbers.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    9. Re:Bus speed is the big issue by Alioth · · Score: 1

      A big issue to who, though?

      Maybe to me, since I play games on my PC.

      But to the vast majority of people out there who spend all day in something like MS Word or a web browser - machines basically waiting for I/O from a human - it's not even an issue. Given adequate amounts of RAM, most of them would be fine with a 500MHz PIII.

  19. Anyone know what Gentoo stage to use? by (H)elix1 · · Score: 1, Insightful

    Never have figured out if the Pentium 3 or 4 flags are what you are supposed to use. Forms seem to be split down the middle of calling it one or the other. I used the Pentium 3 config without any issue, but I know there is someone out there who knows....

    1. Re:Anyone know what Gentoo stage to use? by Anonymous Coward · · Score: 0

      use the Type-R Flag

    2. Re:Anyone know what Gentoo stage to use? by ultraslacker · · Score: 2, Informative

      gcc 3.3.x? Then pentium3.

      More recent gcc (3.4.x) has explicit support for centrino, with cpu-type pentium-m

    3. Re:Anyone know what Gentoo stage to use? by MarcQuadra · · Score: 5, Informative

      The pentium M is NOT a P4 at heart, it's an intel i686 core (same as PIII) with added instructions and a different bus to the northbridge. It has the same feature set of a P4 but that doesn't make it a P4.

      GCC-3.4.3 has a "-march=pentium-m" option, btw.

      If you're stuck with an older gcc, try:

      "march=pentium3 -msse2"

      which should get you as close as can be to optimal scheduling.

      Using "march=pentium4" will probably yield slower code than using just "pentium3" because the scheduling for these CPUs is so different.

      I spend too much time doing this shit.

      --
      "Sometimes, I think Trent just needs a cup of hot chocolate and a blankie." -Tori Amos on Nine Inch Nails
    4. Re:Anyone know what Gentoo stage to use? by Florian+Weimer · · Score: 1

      Using "march=pentium4" will probably yield slower code than using just "pentium3" because the scheduling for these CPUs is so different.

      GCC hasn't got a scheduler for the Pentium 4 anyway.

    5. Re:Anyone know what Gentoo stage to use? by MarcQuadra · · Score: 1

      GCC hasn't got a scheduler for the Pentium 4 anyway.

      I think it does have at least something rudimentary in that department, the VIA C3 chips are unpopular and largely undocumented, and the GCC notes mention that they are 'unscheduled', GCC treates them like a generic i586 or i686 with extensions (depending on the C3 or C3-2). I would expect there to be a note similar to the ones attached to the C3 series if the P4 option was unscheduled.

      I'm working off notes for GCC-3.4.3, which is only a few weeks old.

      --
      "Sometimes, I think Trent just needs a cup of hot chocolate and a blankie." -Tori Amos on Nine Inch Nails
    6. Re:Anyone know what Gentoo stage to use? by Florian+Weimer · · Score: 2, Informative

      [Pentium 4 scheduling in GCC]

      I think it does have at least something rudimentary in that department

      This was a bug.

  20. cat 600000 /sys/devices/system/cpu/cpu0/cpufreq/ by n1ywb · · Score: 1

    Shouldn't that be "echo 600000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_setsp eed"?

    --
    -73, de n1ywb
    www.n1ywb.com
  21. Intel still holds the major share in market by adeydas · · Score: 5, Insightful

    Though AMD Atlon may be rising as a global giant, Intel's Pentium processors still holds a major share of the market primarily because the large choice of processors it offers, the number of years it has been in business and compatibility with a large number of OS and softwares not to mention hardwares. I believe that Intel is going to stay in business atleast for some more years. What do you think?

    1. Re:Intel still holds the major share in market by Mindmatter · · Score: 1

      "compatibility with a large number of OS" All x86 compatible OS's run on x86 chips. Both intel and AMD chips are full x86 compatible chips. So how would intel have more compatible OS's than AMD? "not to mention hardwares" Again, if an intel motherboard has a PCI slot and an AMD motherboard has a PCI slot, how would the intel board be "more" compatible with a standard PCI card. I hear people say this compatibility stuff a lot. Maybe you could explain what you mean. But from what I can see this is just marketing getting the best of people. BTW AMD has been in business since 1969 and has many processors to choose from. And intel is not going anywhere.AMD and intel will be neck and neck for a long as I can imagine. Who knows maybe VIA or Transmeta will catch up someday and make some crazy competition.

  22. Is AMD really that much better for games? by Saint+Stephen · · Score: 4, Informative

    It seems to me that midrange AMD is far better than midrange Intel for games -- so this is probably why the impression "AMD rules for games" is out there. Buncha kids with no money think it is.

    But my 3.2 P4 Northwood running at 3.52 with 6800GT seems plenty competitive -- with everything except the FX 55, which is *extraordinarily expensive*.

    It seems that AMD is better at the low end and the extreme high end, but the "ordinary" high end (3500+ and 3.2 P4), Intel and AMD are about the same. Plus with things like MPEG encoding and compiling, which is also important to me, P4 beats even the AMD FX.

    So AMD is only better than Intel at the extreme high end and the low end. But the low end isn't worth playing at, unless you ain't got no money.

    So in short it seems to me that in the real world a 3.2@3.52 P4 is plenty great for games.

    Or would an AMD 3500+ give me a "smoother feeling" experience?

    1. Re:Is AMD really that much better for games? by lewp · · Score: 1

      Doubtful. Your box is fast enough that you'll be able to play most games at whatever resolution you want with all the visuals maxed out. Even if the A64 gives you 5% higher performance, as long as you're happy with the performance of the P4, who gives a fuck?

      I've got a 3500+ and a 6800GT, though. If you really want to do the penis measuring thing run a few benchmarks and I'll do the same.

      I bought AMD because the last 3 chips I've bought have been AMD. I knew the memory I had worked well with AMD processors, and I like the nForce chipsets. But I think most people know if you're willing to spend a decent bit of money on the chip you can get a good CPU from either manufacturer.

      (OT: You wouldn't happen to be the same Saint [Stephen] that runs nachos.net, would you? Used to post there a long time ago.)

      --
      Game... blouses.
    2. Re:Is AMD really that much better for games? by Anonymous Coward · · Score: 1, Insightful

      By your own logic, if AMD is better than Intel at the high and low ends, and in the ordinary they're "about the same", then it stands to reason that Intel isn't worth buying because AMD beats Intel in most areas.

      Overclocking a $213 processor to compete with a $260 processor and calling the results close and the kids "cheap" seems a bit ridiculous. The price difference is $40 and overclocking is a pretty "cheap" way of getting more speed.

      Who's being "cheap" here?

    3. Re:Is AMD really that much better for games? by Anonymous Coward · · Score: 0

      don't forget that the extra $40 pays for something that Intel doesn't have: 64-bit processing

    4. Re:Is AMD really that much better for games? by zx75 · · Score: 3, Insightful

      Yesterday's 'extreme' high-end is today's 'ordinary' high-end. AMD has been putting out some amazing processors for a couple years now, so if you're looking for an 'ordinary' high-end CPU check out some of their previous offerings. I've been running a 2800+ for over a year now under a lot of stress (I'm a programmer, gamer, and play around with a number of CG renderers) and it still holds up admirably.

      --
      This is not a sig.
    5. Re:Is AMD really that much better for games? by RoboProg · · Score: 1

      I don't know about games, but I often use AMD chips for "nominal" upgrades to existing machines at home when they break or start to get too weak for the task-of-the-day.

      They are cheap, and they work well enough. By the time I buy a board, chip and RAM, I'm only paying about 200 to 300 dollars, and I don't want to add another $100 to that for a chip that is perhaps 20 to 50 percent faster. I'll just upgrade another one of my computers next year.

      --
      Yow! I'm supposed to have a plan?
    6. Re:Is AMD really that much better for games? by PhrostyMcByte · · Score: 1

      Plenty fast? Yes. Fastest? No.

      The major difference is floating point speeds - AMD chips tend to excel far past P4s in this area. Hence why they are usually best for games no matter what "end" you are at.

      Not to mention AMD chips have a much better price/performance ratio. Though their expensive A64 line is quickly killing this trend.

    7. Re:Is AMD really that much better for games? by Anonymous Coward · · Score: 1, Informative
      Plus with things like MPEG encoding and compiling, which is also important to me, P4 beats even the AMD FX.

      With regards to compiling, it is absolutely not true that P4 beats the AMD FX. In fact Athlon 64 wins all the compiling benchmarks.

      As for video encoding, if encoding performance is seriously important to you, then you should be using a 64-bit OS for video encoding. The use of 64-bit code gives a significant performance boost compared to 32-bit code, enough to surpass the Intel offerings.

    8. Re:Is AMD really that much better for games? by Anonymous Coward · · Score: 0

      Check the benchmark links given in other replies.

      Yes, a 3.2 P4 runs just about every game out there just fine. So does a high end Athlon. But what you're missing is the fact that the Athlon delivers a signifigant amount of added performance for games that you're just not seeing yet. This means that when you start player more taxing games, your 3.2 P4 is going to start slowing where an equivalent Athlon might not. Alternately, it means that you could get a lower level Athlon to get the same gaming performance and save a large chunk of cash.

      Saying that your Dual Xeon system gives you decent frame rates in Doom 3 doesn't mean that you wouldn't be just as well off with an Athlon 64 3500+ for that application, it just means that you spent more than you needed to. Intel processors aren't bad for games if you're at the high end. Athlon 64 processors aren't bad for games... period. (ok, ok, slightly overstating the case, but thats what it boils down to)

      And double check the benchmarks for video encoding. Some of the ones that the Athlon 64 did well in suprised me. It looks like it depends a lot on whether the program you're using is designed and compiled to take the huge pipeline of the P4 into account. If it isn't (and not all the major ones are), an Athlon might actually end up better.

    9. Re:Is AMD really that much better for games? by Saint+Stephen · · Score: 0, Troll

      I'm not trolling. Stop modding me down. Jesus, can't you have an opinion on Slashdot anymore? Modding is so formulaic now, a robot could do it.

    10. Re:Is AMD really that much better for games? by servognome · · Score: 0, Offtopic

      The /. mod bot results of your post:
      I'm not trolling.: +1 Insightful
      Stop modding me down: +1 Interesting
      Jesus: -1 troll
      can't you have an opinion on Slashdot anymore?: +1 Insightful, -2 flamebait
      Modding is so formulaic now: +1 interesting
      a robot could do it.: +1 funny
      SCO/Microsoft/RIAA Modifier: 0
      Total score: +2

      --
      D6 63 0D 70 89 81 BB 8E 7B 7C 5F 5D 54 EA AB 73
  23. Has Intel Peaked? by upsidedown_duck · · Score: 3, Insightful


    Opteron is better than Xeon in most ways that matter. Itanium, even with all its FP muscle, has to be given away. Has Intel peaked?

    --
    -- "Makes Little Debbie look like a pile of puke!" - Moe Szyslak
    1. Re:Has Intel Peaked? by the_2nd_coming · · Score: 1

      umm... intel PEAKED back in 1998 with the P3. they have been in decline ever since. the question is "Has intel hit the bottom".

      --



      I am the Alpha and the Omega-3
    2. Re:Has Intel Peaked? by Kehvarl · · Score: 1

      Intel. Shark.

      There may have been jumping.

      Then again, what if it was a mock shark and not the real thing?

  24. fscommand protocol? by Bloater · · Score: 1

    I wonder if that sudhian is carring a 'sploit. I went with javascript enabled (loaded both at the same time though, so it *could* be anandtech). Fortunately, running Linux, there is no fscommand protocol, but if that fs stands for filesystem - then boy am I glad my browser doesn't support it.

    Doesn't seem to be happening with javascript disabled.

    1. Re:fscommand protocol? by BoneMarrow · · Score: 2, Informative

      From www.macromedia.com: Macromedia Flash can use the fscommand action to control the playback and appearance of standalone projectors, as well as launch external applications. The fscommand action takes two parameters: a command and an argument. In some cases, an argument is not required.

      --
      Unfortunately, no one can be told what my sig is...
    2. Re:fscommand protocol? by Bloater · · Score: 1

      Great, so the site is using macromedia flash to run a local binary...

    3. Re:fscommand protocol? by Anonymous Coward · · Score: 0

      It does nothing of the sort. Its just used to display the ads. Shees not everyone is evil

    4. Re:fscommand protocol? by Bloater · · Score: 1

      I feel I should make the context of my comment more clear:

      > Macromedia Flash can use the fscommand action to control the playback and appearance of standalone projectors, as well as launch external applications

      Controlling a projector and launching external applications sounds like it has the run of the local machine. If Macromedia have chosen to use overloaded market-speak to give the appearance of explaining a technical mechanism then I retract my concerns.

  25. for the record by Anonymous Coward · · Score: 0

    there is no a64 architecture, you must mean amd64.

  26. Intel is trying to shift the battle, not catch up by Anonymous Coward · · Score: 4, Interesting

    My parents and the bulk of the people out there do not need a 64bit 5ghz monster under their desk. And honestly most of thosethat have them probably only use the power 5-10% of the time, if that.

    Intel could care less about us, they care about Fortune 500 companies that buy computers by the truck load... and what those companies care about is saving money. 5-20W here and there don't really mean much to you and I, but when you're footing the electric bill for several hundreds or thousands of people then giving everyone barn burners to run Excel starts to look pretty foolish.

    You might as well be comparing a Prius and a Ferrari or a jumbo jet and an SR-71.

    Use the right tool for the job folks.

  27. Nothing wrong with a PM by AbRASiON · · Score: 2, Informative

    I've been using a Pentium M Dell Inspiron 8600 for the past week (configuring it for a friend)

    I've had nothing but luck with it, it's warm at worst and the fan comes on for 90 seconds every 25 minutes when it's sitting on a soft pillow (practically covering ALL vents in the machine)

    It browses very fast, it's responsive and it plays back movies fine

    Absoloutely no qualms here, an Athlon 64 would be far hotter, far noiser and (potentially) less stable.
    (intel chipset / cpu in a laptop is just the only way to go... I wish it was different but it's not)

    1. Re:Nothing wrong with a PM by michaeldot · · Score: 1

      It browses very fast, it's responsive and it plays back movies fine

      No argument that you're happy, but tasks like these require at most a 768 Mhz machine, which is probably what it's stepping down to when the power cord's unplugged.

      The article poster is talking about more computationally intensive tasks.

      (Browsing these days on almost every computer is limited by the speed of the internet connection, not the computer's CPU performance.)

    2. Re:Nothing wrong with a PM by AbRASiON · · Score: 2, Interesting

      Actually I find browsing nowadays is limited to the shitty browser and the way windows works with files on the hard disks etc.

      If you pump up the threads in firefox and tweak it a little bit plus defrag you can get pages to load fast.

      I'm betting under 98 it could be even quicker as smartdrv was an awesomely fast caching tool whereas XP's is a little safer but ass slow in comparison

  28. Intel CPU sticking to the heatsink... by __aaclcg7560 · · Score: 0, Offtopic

    My indirect experience with a Intel P4 was when my roommate upgraded his computer. Since it didn't boot, he was packing the parts away to take back to the store. He came to me asking if it was normal for the CPU to be sticking to the heatsink, and I died laughing when I saw it. Turns out my roommate didn't know about the little lever thingie on the CPU socket. :)

    1. Re:Intel CPU sticking to the heatsink... by Anonymous Coward · · Score: 0

      If you have ever done a lot of removing of Intel CPU's you will get a few that stick to the heatsink and pull up without the leaver being used. (fortunately enough the pins have always been ok)

  29. Stability, what's that? by Anonymous Coward · · Score: 0

    I've worked on most operating systems out there and a large number of different hardware configurations and I've had everyone of them crash just some more than others. Stability is like Bigfoot, a lot of claims by people but still no physical evidence. Do what I do, Back up often, say the occational prayer and once a day face Seattle Washington and flip Bill Gates the bird. Works so far.

    1. Re:Stability, what's that? by Anonymous Coward · · Score: 0

      Renton, WA Us Seattle people don't like being associated with them Renton folk.

  30. Re:Intel is trying to shift the battle, not catch by lewiz · · Score: 1
    5-20W here and there don't really mean much to you and I, but when you're footing the electric bill for several hundreds or thousands of people then giving everyone barn burners to run Excel starts to look pretty foolish.

    You really think so? Where I work monitors (big 21" CRTs) are left on 24/7. In fact, I don't think I've ever seen anybody sit down at them. It's always roasting and the lights never go off.
  31. Re:cat 600000 /sys/devices/system/cpu/cpu0/cpufreq by Jimmy+The+Leper · · Score: 1

    No, they obviously had a file called 600000 which contained the string "600000\n"

    --
    -You're only as clean as your towel.
  32. Re:Intel is trying to shift the battle, not catch by Anonymous Coward · · Score: 0

    Intention and implementation are two entirely different things :)

    I've noticed the same thing at the office I work at, they go through great lengths to make sure that the overhead lights and HVAC are turned-down/shut-off during off hours but all the computers are left on and none of them are set to auto-sleep/hibernate.

    The sad thing is everyone will complain about how hot the office is over the weekend when it's 30 degrees outside. It should be getting *colder* (no one is in the office, everything else is shut off and the window to floor space ratio is pretty high). Apparently it hasn't occurred to anyone important that it's the computers that are heating it up.

  33. Last I heard.. by BalkanBoy · · Score: 1

    wasn't it AMD's CPUs that you could literally fry an egg on? How did this come about? Sorry if it's redundant, but that was my impression for the longest time - AMD runs hotter...

    --
    'A lie if repeated often enough, becomes the truth.' - Goebbels
    1. Re:Last I heard.. by lachlan76 · · Score: 1

      P4s do run really hot, but when they start to overheat (ie. you take off the heatsink) they slow down, and this lowers the amount of heat output.

    2. Re:Last I heard.. by Anonymous Coward · · Score: 0


      times change my friend...

    3. Re:Last I heard.. by Anonymous Coward · · Score: 0

      They were like that in the K6-2 era...then they were even, and ever since the P4 Intel has been way way higher - it's an ongoing myth perpetrated by AMD haters (and why you would hate a chip company just putting out products I don't know)

    4. Re:Last I heard.. by Mindmatter · · Score: 1

      LOL don't take off an Athlons heatsink and you will be fine.

    5. Re:Last I heard.. by Anonymous Coward · · Score: 0

      That was true in the times of PIII and plain Athlon. Since Athlon XP and P4, Intels run hotter.

    6. Re:Last I heard.. by Shokac · · Score: 1
      wasn't it AMD's CPUs that you could literally fry an egg on? How did this come about? Sorry if it's redundant, but that was my impression for the longest time - AMD runs hotter...

      Time changes, my friend. before,Intel CPUs was faster on same clock speed than AMD, and look at situation now.

    7. Re:Last I heard.. by lachlan76 · · Score: 1

      I take it you've seen the video too? ;)

    8. Re:Last I heard.. by Anonymous Coward · · Score: 0

      My Athlon 64 has been running at 100% cpu for 3 hours now, and the cpu temperature is 44 degree celsius.
      When idle it scales down to 1000 Mhz, and the temperaure is around 30 degree celsius.

    9. Re:Last I heard.. by Mindmatter · · Score: 1

      I've seen the video. But I'm still a die hard AMD fan and have never seen an Athlon overheat with a heatsink on, and as long as it doesn't cost anymore to cool an Athlon than it does a P4 then I see this as a non issue. A good rule of thumb for any processor simply is don't run it without proper cooling. BTW last I checked, the Athlon XP puts out about 70 Watts of heat vs. the, latest at the time, P4 which was about 60 Watts of heat. That basically makes them both good space heaters ;) . Here is an interesting review I just found. http://www.hardwareanalysis.com/content/popular_ar ticles/article/1758/

    10. Re:Last I heard.. by lachlan76 · · Score: 1

      I agree, I haven't bought any Intel procs since '97.

      However, I did have a problem with overheating on my current box, but it wasn't a processor problem, it was actually a fan failure.

      However, I *DO* like to use my many computers to keep my bedroom warm, so there is a fair amount of heat output ;)

  34. I'm so glad I have a Pentium IV Mobile by digitalgimpus · · Score: 3, Informative

    My ideal laptop for college was an Apple G4, but school insisted on Windows for it's 'stability, reliability, and security'.... yea, that's a joke in itself.

    Decided to get an IBM Thinkpad with a Pentium IV Mobile.

    Everyone with laptops running P4's seen to have issues with heat, and power consumption. Despite my oversided screen, dual HD's, and CD-RW... I'm actually doing all right.

    It cost more to get a laptop with a real mobile chip, rather than just a P4 as some companies offer... but I think it saved me a lot of trouble.

    1. Re:I'm so glad I have a Pentium IV Mobile by lachlan76 · · Score: 1

      I'd still get a Powerbook and run Virtual PC, if the performance is any good.

      What is the performance of VirtualPC like on PPC? Anyone know? Emulating a PPC would be slow on x86, being CISC, but how fast is it to emulate a CISC on a RISC processor?

    2. Re:I'm so glad I have a Pentium IV Mobile by Anonymous Coward · · Score: 1, Interesting

      I run VPC on a 1.5GHZ G4 with WindowsXP pro.

      It is useable, but only just. It works fine for what I want (some netmeeting, visio and IE testing) but if you were using programs in it all day I think you'd get frustrated.

      Also, based on my extensive testing (fiddled around with an imac for 5 mins in a shop) VPC7 on a 1.8ghz G5 doesn't seem to be any faster than my 1.5ghz G4.

    3. Re:I'm so glad I have a Pentium IV Mobile by Dwedit · · Score: 1

      Cisc and Risc don't matter much anymore, since everything since the Pentium 2 has internally converted CISC instructions to RISC style instructions.

    4. Re:I'm so glad I have a Pentium IV Mobile by burns210 · · Score: 1

      1ghz 12" powerbook, 768MB of RAM. Nice little laptop, but I should have waited for the next revision and got the better video card.

      VPC6 runs just a hair under acceptable(with mild, realistic standards.) It is usable, but not for long term, more for testing or for fulfilling the nitch of a program or 2 (IE web-dev testing, etc). With VPC7, it isn't(according to the macworld review) much better.

      Microsoft was suppose to improve 3d acceleration(add it, actually) but hasn't I do not believe. If you set aside adequite RAM, and strip down XP (turn off ANYTHING graphical that isn't needed, no eye candy!) it would be usable.

      Overall. Yea, it is usable, with some patience and configuration(and a spare 512meg of ram in your computer), but every task and every type of program I have sought out, had a good(if not better) mac equivalent. (Note: Linux/FreeBSD software is included in the pool of mac software, IMHO).

      A Mac, IMO, is a good purchase, but not for a Windows box. It would be better to buy a crap p2 or p3 mini-tower for a hundred bucks(if you don't already have one around) and run remote desktop(free, decent quality) to your mac.

    5. Re:I'm so glad I have a Pentium IV Mobile by lachlan76 · · Score: 1

      But I take it there is no way to use them in applications, and you have to use the CISC instructions?

    6. Re:I'm so glad I have a Pentium IV Mobile by lachlan76 · · Score: 2, Interesting

      I'm a Linux user, and use x86 assembly enough that I don't expect to switch in the near future.

      So I just use VMWare, which is IMHO one of the greatest closed source programs I have ever used.

    7. Re:I'm so glad I have a Pentium IV Mobile by burns210 · · Score: 1

      Absolutely. I haven't played much with VMWare, only heard good(great) things about it.

      Mac OS X does have some, in my opinion, distinct advantages (and in some ways, disadvantages) and is a nice complete system. With that said, Linux is great and the ability to run VMWare would make Windows without Windows as it were very nice.

      Personally, I can't WAIT for the multi-tasking whatever IBM Power5 (G5++) processor, So I can have Mac OS X (10.5, or .6 by then) and Yellow Dog Linux and Solaris 10 interchangeably.

  35. Comment removed by account_deleted · · Score: 1

    Comment removed based on user account deletion

  36. always the lame excuse. by Anonymous Coward · · Score: 0

    People who are clueless and overpay for items use the lame phrase "you get what you pay for" to make themselves feel better.

    I had a "friend" even try to use that line on me when I bought the same exact item online for much less than he got it for at the store. I guess he thinks that somehow paying more money for something must mean it's more valuable.

    1. Re:always the lame excuse. by BeaverCleaver · · Score: 1

      Well if something does go wrong with his hardware your friend might get better service from the store than trying to get something replaced online. Just a thought, I'd go for the lower price too...

  37. Re:Intel is trying to shift the battle, not catch by statusbar · · Score: 4, Funny
    My parents and the bulk of the people out there do not need a 64bit 5ghz monster under their desk. And honestly most of thosethat have them probably only use the power 5-10% of the time, if that.

    You know, 10 years ago I was saying the same thing about the bulk of the people not needing a 32bit 1 gHz monster under their desk. Now this is the minimum that they need.

    Don't underestimate Microsoft's ability to make software that brings yesterday's supercomputers to their knees today.

    --jeff++

    --
    ipv6 is my vpn
  38. Comment removed by account_deleted · · Score: 1

    Comment removed based on user account deletion

  39. Bigger surprise by Anonymous Coward · · Score: 0

    I've built a pentium-M desktop right after reading /.'s article. I dunno bout the rest of you, but I enjoy the fact it kicks most competitor's asses in gaming when it's similarly equppied. This thing screams playing games, regardless of it's limited architecture. 2.0 ghz and kicking the crap out of my 1.8 p4 desktop by MILES.

  40. You don't appear to be very smart by melted · · Score: 2, Informative

    First of all, you will only find P4 "Mobile" in P4 laptops. Non mobile versions draw too much juice and have heat dissipation of at least 100W.

    Second of all, Pentium-M blows the doors off P4 "Mobile" while at the same time running much cooler and consuming much less power.

    Third, I type this from a 20" iMac G5. Envy me. :0)

    1. Re:You don't appear to be very smart by bStrom · · Score: 3, Informative

      That's not true. Some laptops use a Mobile Pentium IV (upt to 3.2 Ghz w/533Mhz FSB), while others use a plain Pentium IV (up to 3.4 Ghz w/800 Mhz FSB). See Pentium IV vs. Mobile Pentium IV.

      I do agree with you about the Pentium-M, though.

      --
      Try eMusic. DRM free, legal, MP3 downloads.
  41. Re:Intel is trying to shift the battle, not catch by Anonymous Coward · · Score: 0

    Your remark about Microsoft could apply to Linux as well. A login to KDE or GNOME on my computer takes much longer than in Windows. So much for the days when small light and fast Linux could be used to rejuvinate old systems.

  42. I actually read the review, and... by Dr.+Spork · · Score: 4, Insightful
    I think the blurb on the front page was quite misleading. I was pretty blown away by the performance data of the Pentium M. It was on a crappy board with prehistoric features, and it was still kicking ass. It overclocked perfectly to 2.1 Ghz and beat a standard-clocked A64 3500+ on almost every benchmark, and never broke 40C with a tiny little heatsink and a tiny little fan.

    If Intel were serious, they could be making these right now at 2.4 GHz (I'm sure they'd run fine, and still quite cool) at which point it would be beating every desktop processor in the world. I say that's a hell of a start for an Intel processor line. The most important thing is that with such a low heat output, Intel can eventually clock these things pretty high. The Athlon64 seems to have less headroom.

    One clear lesson is that the Pentium4 and everything based on it is done. The P4 gets creamed by the M, it's quite embarassing. I think Intel will just ride out the P4 advertising investment, but we know that their next big thing involves the M cores. And they will be quite fearsome once they start putting multiple M cores into desktop chips, and putting their marketing muscle behind the result.

    I'm a huge AMD fan and will remain loyal, but... I think AMD is in a good place now only because they've consistently out-engineered Intel since the first Athlon. Now I'm scared that they won't pull it off in the next generation. Intel seems to have a really promising starting point.

    1. Re:I actually read the review, and... by HalfFlat · · Score: 1

      The Pentium M is a nice chip, but as far as Linux comparisons go, comparing it to 32-bit compiled applications on the A64 is also misleading.

      With Linux on the Athlon64 and sources for your programs, there is just no reason not to compare against x86-64 versions of the code. This would offer a much more realistic comparison between a Linux Pentium-M machine and a Linux Athlon64 machine.

      Given that performance generally improves greatly on the Athlons when everything is compiled for and run in 64-bit mode, it seems premature to state that the Pentium M and the A64 are on par for a system running a recent, x86-64 capable version of Linux.

    2. Re:I actually read the review, and... by mpsmps · · Score: 1
      Just in case you didn't RTFR. Here are some choice quotes:

      Right now, if you wanted the absolute best gaming performance from an Intel chip, you wouldn't buy a Prescott, a Northwood, or an EE. You wouldn't buy Grantsdale or Canterwood, or PCI-Express.

      You'd buy a DFI board based on Intel's Montara chipset, you'd use single-channel DDR333, AGP 4x, and a regular ATA hard drive--and your frame rates would rock in silence.

      And my favorite:
      In three words: This chip rocks.

      It looks like a pretty impressive platform. I hope Intel ditches the P4 and moves their efforts towards further developing the Pentium M.
    3. Re:I actually read the review, and... by klui · · Score: 1

      It kind of gives you a new perspective on where the "state-of-the-art" in the current 32-bit Wintel platform is. PCIe/1066MHz FSB/Prescott really don't buy you a whole lot except nice specs on a piece of paper (and a lot of excess heat). I really hope in the future, the mobo manufacturers will make a full-size ATX board with 2 or 3 PCI-X slots along with a standard CPU socket.

    4. Re:I actually read the review, and... by MemoryDragon · · Score: 1

      I have been telling that for years (onwning one) the main problem I have faced is that the GHz is so hammered into peopls minds that even people who were knowledgable, were totally clueless, once I told them that this thing runs cicrcles around a much higher clocked P4 or Athlon. But they did not want to believe that until I showed them my laptop.

    5. Re:I actually read the review, and... by Anonymous Coward · · Score: 0

      What this reveals is that the P4 arch is a failure. Outperformed by an updated PPro at the same clockspeeds.

  43. Re:Intel is trying to shift the battle, not catch by Anonymous Coward · · Score: 0

    yeap, 640KB of memory should be enought for everybody...

  44. Re:2 fps in the same day! by Anonymous Coward · · Score: 0

    Gah, I have Slashdot set up to show that whole damned comment. Now I know why people don't do that.

  45. I guess I'm a "Real Geek" by Nick+Driver · · Score: 1

    Every real geek has owned a dual proc. Intel machine.

    Only problem is that my dual processor Intel machine is a dual slot-1 P3 450MHz box... built back when this machine was state of the art. But the processor speeds skyrocketed so rapidly that this machine became utterly worthless, virtually overnight. It still runs fine, and runs Linux pretty well, but still rather slowly by today's standards. The motherboard won't support any better processors without using PowerLeap adapters, and the cost of a pair of PowerLeaps with 1.4GHz Tualatin P3 cpu's just isn't worthwhile. The money would be better spent towards something else.

  46. Re:Intel is trying to shift the battle, not catch by Teckla · · Score: 1

    Use the right tool for the job folks.

    Most companies should be using low cost, diskless workstations whose only job is to connect to powerful servers where the real work is done, that can be easily managed by a small staff, instead of giving everyone a full fledged PC, which for most business users, is nothing more than fertile soil for viruses, trojans, and worms, and is a maintenance nightmare for I.T. staffs.

    In my opinion... :)

  47. Every *real* geek... by rarose · · Score: 1

    Every real geek can remember the first time they took a soldering iron to the motherboard to secretly modify their father's computer...

    --
    --Rob
  48. Not "what you pay for" but "effort that goes in" by AHumbleOpinion · · Score: 1

    For what reason are Athlon 64 processors not "stable"?

    Usually its the user. Someone will upgrade a system by replacing the motherboard but that old power supply is not Athlon rated. Or they will buy brand new crappy parts. Or they will have a rats nest of ribbon cables block airflow through the case. Or they will overclock and think because the systems seems to be running fine everything is OK, being totally oblivious to the fact that not all overclocking induced errors are obvious - some are as subtle as an instruction giving the wrong answer. The build-your-own folks who buy high quality parts after doing their homework, who properly route cables to permit airflow, and who avoid risky behaviors like overclocking are not heard from as often as the former.

    Someone else posted "you get what you pay for", well yes and no. You get back the effort that went in, now you could pay for someone else's effort, or you could invest your own time. Either works.

  49. www.monarchcomputer.com by Anonymous Coward · · Score: 0

    I've purchased a custom-built Athlon 64 (3500+ 939) system from them. I've also purchased additional RAM from them for that same system.

    They have good standing on www.resellerratings.com:
    http://www.resellerratin gs.com/seller2079.html

    In addition to their good ranking here, my top reason for going with them was that for custom builds, for every component available they list full specifications and provide an image (in a popup window).

    Many of the sites I checked out before making my decision just give you a drop down box for each component category, and leave it up to you to determine what the item is, and why stated specs are sometimes inconsistent.

    They will pre-install Windows, Linux, or ship without an OS.

    For a quality case (without plastic and neon) I recommend the Lian Li PC-60.

    And in my opinion, the 3500+ 939 is still the best "bang for your buck". A 3800+, FX, or dual-Opteron system going to be expensive wherever you go. Since its socket 939, you can be pretty well assured you will have an upgrade path at least late into next year.

    The retailer is a bit more expensive than others.

    dircha AT dircha DOT com

  50. This form factor by ubercombatwombat · · Score: 1

    can be a fair gaming machine. After reading: http://slashdot.org/article.pl?sid=04/11/20/205821 2&tid=137 I looked into buying the pieces from TD Curran for a very small quiet gamer that would play anything on the market. It came in at about $1200.00 without a monitor.

  51. Re:Intel is trying to shift the battle, not catch by Lehk228 · · Score: 1

    I agree the times are ripe for the reintroduction of Thin client deployment, obviously a UNIX varient would be the right tool, perhapse even thin clients which pooled resources, data storage taken care of by a NAS and an image server for the initial loading of each machine. It would be weird having a few thousand 3-400 MHz Geode CPU based machines logging on to a distributed cluster which was running on each machine but it would do several things.

    1. it would be Frickin' cool
    2. it would be cheap compared to either terminal + mainframe or thousands of desktop machines
    3. As long as the cluster software was done right and deployed properly it would scale very well (hopefully automatically)
    4. If you needed a large amount of power added to the system for a short time you could rent a few beefy machines to hook up to the cluster temporarily.
    5. In the future if your computing needs diminish rather than being stuck with semi out dated PC's or more Iron than you need just sell some of the cluster clients to someone who is using a similar system and is in need of additional machines.

    --
    Snowden and Manning are heroes.
  52. the pentium M is amazing by 0111+1110 · · Score: 4, Insightful

    A couple of more links
    here and here.

    At the moment AMD is kicking Intel's arse in the performance sector. The pentium M (Banias) is the only remaining tech that Intel really has. Lots of chickens have come home to roost now that Intel's super-ultra-mega clockspeed boosted chip has reached the end of the line.

    For the sake of a continuing healthy, competive market even the most die hard AMD fans had better hope that Intel gets back on track and allows some engineers to actually make some product decisions for a change. The Banias core seems to be their only hope.

    I have found all of these recent benchmarks to be rather amazing. It's tough for anything to beat an overclocked Pentium M in games even with the huge disadvantages of an aging platform without all the latest goodies. Intel should be embarrassed. Deeply. Their Pentium 4 is a disgrace.

    It is clear that for anyone who cares at all about power consumption, heat, or noise, nothing can touch a Pentium M, not even a Cool n' Quiet enabled 90nm Winchester Athlon64. If Aopen releases a desktop motherboard with the upcoming alviso (PCI-E, DDR2 etc) chipset, things could get very interesting indeed.

    --
    Quite an experience to live in fear, isn't it? That's what it is to be a slave.
    1. Re:the pentium M is amazing by rugger · · Score: 1

      To be fair though, without the Pentium 4, the pentium-m would never have been this good.

      With the pentium 4, Intel was forced to face the problems that plauge pipelined chips and create solutions that worked with excessively long pinelines. The experience gained by intel engineers in tackling these problems has been critical during the redesign of the P3 core into the P-M core.

    2. Re:the pentium M is amazing by C_Kode · · Score: 1

      You do not learn from success. You learn from failure.

  53. Air Conditioning? In Northern California?? by billstewart · · Score: 1
    Sure, my office has air conditioning, and PC power consumption really does make a difference there, but my home doesn't need it except maybe 1-2 weeks a year (and doesn't get it then.) Fans and high ceilings do the job just fine (or if they don't, I _could_ go to the office instead of telecommuting.) And I've already got electric heat, so my heat bills don't go up significantly just because the heat's coming from a computer instead of a baseboard heater. (

    Noise, on the other hand, is a real issue, and having a silent or very quiet machine would be a pleasant change.

    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
  54. Jesus Tapdancing Christ... by Anonymous Coward · · Score: 0

    I realize no one reads TFM, but is it too much to ask that you actually RTFPost you're commenting on?

    And WHO in god's name gave a +4, interesting to an obviously smartassed comment about a laptop clocking down WITH THE POWER CORD UNPLUGGED?!?!

  55. Except you forgot that this thing comes with a 2M by melted · · Score: 1

    of on-die full-speed L2 cache. The same amount as they have on Xeons. So if by "data intensive" you mean not just copying large blocks of RAM, but actually doing something with the data, I think P-M should be fairly competitive despite its lowish FSB speed.

  56. Dual-CPU? Either overkill or small and lame by billstewart · · Score: 1
    Back when I had a _big_ database machine, I think we'd installed about 4 or 8 of the 16 CPUs it could hold, because that was enough for most of the customer benchmarks we wanted to do. This was at NCR in ~1994-95, so they were starting to be Pentiums, and I think we also made 32-CPU machines. (That doesn't count the Teradatas which had several hundred CPUs, one per disk drive plus a few master CPUs :-) We were using the Unix System V Release 4 multiprocessor support, which did a decent job of balancing loads on them.

    But for home, since I'm not a gamer, I've never had a good reason to buy the current year's CPU. Sure, I recently upgraded from 233 MHz to 2.4 GHz, but that was mostly because my old motherboard fried itself and 2.4 GHz was $10 more than 2 GHz. It's an embarrasingly large amount of CPU, and the only reasons that I perceive it as slow are that Windows XP isn't very good at switching users (probably more RAM would help), plus I'm usually burning CPU with Folding@Home and running BitTorrent uploads in the background. When I get to use it as a geek machine and boot it into Linux, it's far faster than it needs to be, even if I'm running bloatware window managers.

    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
  57. Bits and Bloatware by billstewart · · Score: 1
    32 bits? 10 Years ago? I've been using 32-bit machines for 25 years, though the Vaxen did hide some memory limitations inside that made it hard to do more than about 26 bits of address space some years (or was it 22?) and some various Motorola things and DSPs tended to be 24 bits of usable space.

    But yes, yesterday's supercomputer, the Cray-1, was about the speed of a Pentium-133. Too slow to run Microsoftware on these days, though if you ran Win95 and Netscape 2 and Excel 5 you'd probably think it was pretty responsive. And X Windows ran ok on my 33 MHz 386 box, though Open Look was a bit slow, so if you want a modern Linux on a P133 (which you do) you probably need a bunch of extra RAM and a lightweight window manager - fvtwm or blackbox or something is a lot friendlier than KDE/Gnome.

    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
  58. whew! by geekoid · · Score: 1

    "slow, over heated monstrosity;"

    thank goodnes AMD isn't known for that!

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
  59. And be at the mercy of Apple? by cbreaker · · Score: 1

    Naa.

    With x86, you have the choice to go AMD or Intel (or even one of the others) with the CPU, about a dozen or so motherboard makers (and a handful of chipset makers) the best compatibility with hardware, software, etc.

    You're not at the mercy of one single company. Yes, the Microsoft point can be made, but you CAN run other OS's and moving forward all these Linux distributions just keep on getting better..

    These Athlons are no slouch, either. They're quick chips and they shine when they run in 64-bit mode. While the OS support is still tentative, it will come, and then I see no reason to use a Mac at all unless you really like MacOS. It's not like this is the last batch of CPU's from AMD and Intel; they'll both probably have really great second generation 64-bit performance.

    I like the Macs. I do find the MacOSX to be annoying to use just as any MacOS before it, but it works well and the machines are fast. Unfortunately the whole single-company thing just scares me. You never know, Apple could completely blow it and decide their desktop computer division is too much work and decide to only make iPods; then we'd be down shit's creek.

    Plus. Macs are kinda expensive.

    --
    - It's not the Macs I hate. It's Digg users. -
    1. Re:And be at the mercy of Apple? by geekoid · · Score: 1

      "You never know, Apple could completely blow it and decide their desktop computer division is too much work and decide to only make iPods; then we'd be down shit's creek."
      not likley, but Intel could do the same thing. And its as likly as Apple doing it.
      iTunes is used to sell iPods, iPods are used to sell Macs.

      Macs are a high end systems, and so when you talk about price please keep that in mind.

      I understand the trepidation about single computers maker, but Apple has a decent history. Plus if you did have your nice Mac, and apple stopped selling them they won't come to your door and take your computer away.

      Apple has 10% or so of a multi-billion dollar market. So they are a big company.

      No, I don't own a Mac, but have been looking into buying one next year.

      P.S. I was looking at Mac games, and they had every game available that I am interested in playing, plus the stuff my young kids want.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    2. Re:And be at the mercy of Apple? by cbreaker · · Score: 1

      But alas, as I stated in my post, there's multiple companies making the CPU. If Intel called it quits (which is unlikly with it's significantly more then 10% market share) we'd still have AMD, or Transmeta, or others.

      The only reliable history Apple has with the Macintosh is the fact that they've continued to make them.

      Not all Macintoshes are high end, either. But they are almost always more expensive then an equivelant x86 machine.

      Maybe you just want something different, or something trendy, or something Not Microsoft. Well, more power to you, really. Go nuts. I'm sure you'll enjoy using your Mac.

      --
      - It's not the Macs I hate. It's Digg users. -
  60. Re:Intel is trying to shift the battle, not catch by Fancia · · Score: 1

    Try Xfce. I have Xfce 4 on an old PII of mine (the computer I'm typing this on, actually), and with a light GTK theme (Bluecurve), it runs much faster than even Win98 did on the same hardware.

    --

    Bít, zabít, jen proto, ze su liska!
  61. 20" huh? by geekoid · · Score: 1

    I don't fold mine in half for anything.

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
  62. Re:Intel is trying to shift the battle, not catch by Anonymous Coward · · Score: 0

    Thanks! Just took a look and will try it out.

    I wish distributions would take speed a little more seriously though. A coworker or relative installing Linux to try it out typically chooses the defaults (KDE? GNOME? huh?) and goes by first (bad) impression.

  63. Re:Intel is trying to shift the battle, not catch by geekoid · · Score: 1

    1 gig? 10 years ago? I think not.

    A 1Gig machine with 512 MEGS of memory can run any application needed for standard office work... with Windows 2000.

    Guess what? a lot of companies have figured that out as well. The seldom buy cutting edge desktop computers any more, and the corporate uptaks on XP has been slow. It may be the first MS OS that home users pick up before a lot of corporations. MS is not happy about XP or Office sales.

    I cimpile a decent size .net application on a 2 gig machine(200 bucks at frys) + 84 bucks for 512 MRam, and it cimpiles the application in about 45 seconads.
    My 1.7 Gig machine compiles it in 35 seconds. Higher end chip.

    my partende 1.2 gig lap top compiles it in 60 seconds.

    Would it be worth spending 1500 bucks to shave 25% of the time? not any more. 10 years ago, sure becasue 20% could be half a day.

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
  64. Stability issues with AMD 64 by Anonymous Coward · · Score: 0

    I wish I could say Athlon rocks, but I just can't. I've had some major stability issues with AMD 64 processors. That's why my next computer will run mission critical applications on Intel hardware.

    If I was just playing games with my box, then it naturally wouldn't matter.

  65. I disagree by geekoid · · Score: 1

    I am running an Intel celeron 1.7 gig, 3/4 Gig of RAM. I can run games as well as my friends system. HIs system was what ever was top of the line AMD 4 months ago. We both use the same video card, but he has a Gig of memory.

    I think the "AMD rules for games" impression is crated by a bunch of gamers who don't know processors, but do know that intel is big, and therefore evil, and AMD is the 'underdog'.

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
  66. which is where by geekoid · · Score: 1

    I believe Intel will strive for the next step.
    Face it, the speed wars are over*.
    Next up, quite computers. Intel is working on a number of different angles on this.
    It probably wont be too long before you start seeing commercials that show two computer running the same application and it will look just as fast on both machine, then some yopung ladies hand willr each out and turn of one system and that back ground hum with disapear. It will be replaced with a bird chirpping, the sound of kids playing.
    "Listen to what you have been missing" will be the tag line. The Intel Hush.

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
  67. serious error in TFA by Anonymous Coward · · Score: 0

    there is a serious error in TFA.

    For Pentium M compiling benchmarks, look carefully HERE

  68. Re:Intel is trying to shift the battle, not catch by Chanc_Gorkon · · Score: 1

    You know what? A 1 Gig machine with 512 MB runs XP just fine too. We have alot of them at work. In fact, we have a lot of 800 MHz machines that run XP. XP does not have that much bloat when compared to 2000.

    --

    Gorkman

  69. Re:Air Conditioning? In Northern California?? by Anonymous Coward · · Score: 0

    ELECTRIC HEAT! IN CALIFORNIA? I realize you don't use it as heavily as someone in Wisconsin, but jeez, thats just stupid. On one level is the very high price your paying for the heat, and two is the utter sillines of burning fuel, converting it into heat, converting the temperature gradient into electricity at maybe 1/3 to 1/2 efficiency, and then sending that juice down wires at a significant loss of 3 to ten percent and then converting BACK to heat rather than just burning it in the first place means that you're using about 4 to 5 times as much fuel. And in California with the most expensive electricity in the country. You might as well just burn dollar bills for heat.

  70. Or by Mark_MF-WN · · Score: 1

    Or, for the more evolved, Gnu/Linux running on a G5. OS-X is nice for novices, but a Gnu OS is the only real operating system for pros.

  71. Dual P90 by Zo0ok · · Score: 1

    I run my firewall on a Dual Pentium 90MHz.

    Does that mean I am a geek?

    1. Re:Dual P90 by paranoidgeek · · Score: 1

      If you have a dedicated firewall that you set up yourself , then yes you are a geek.

      --
      Lima India November Uniform X-ray
  72. Making the PowerPC look good by Nice2Cats · · Score: 1
    One of the reasons I am leaning towards a Mac as my next machine despite the cost of buying a whole computer instead of just upgrading components is the PowerPC CPU. I have a G4 iBook and the thing is just completely silent -- I have managed to push it hard enough to make it turn on the fan, but you really have to work on it. Heat doesn't seem to be a problem at all. You can tell the difference when you sit next to some guy with his Dell when it starts making those airplane noises with the fan. Apple has some pretty cool hardware going here.

    Now if only Yellow Dog Linux would cleanly support the iBooks's sleep mode...

  73. Much better read by Anonymous Coward · · Score: 1, Informative

    http://www.anandtech.com/mobile/showdoc.aspx?i=180 0

    Gives more detail on the why & how ...also notice how a P4M gives about the same type of performance of a normal P4: actually it's pretty on par with equivalent speed athlon cpu's.

    1. Re:Much better read by Anonymous Coward · · Score: 0

      Meant: P4M gives about the same type of performance of a normal P4 (clocked 1ghz faster)

  74. Eww... by Anonymous Coward · · Score: 0

    When looking at my XP taskbar, the title in Firefox says "Slashdot: More Anal..."

  75. Re:2 fps in the same day! by Anonymous Coward · · Score: 0

    You made a mistake, the 1283 number should be a 7 you sux0r!

  76. Neat yes, but mobo prices are insane by Sumocide · · Score: 1

    Prices for the DFI 855GME are around euro 270 ($360). For the AOpen i855GMEm-LFS euro 205 ($270).

    That's just crazy. For 270 you can get an Asus A8V Deluxe AND an Athlon 64 3200+ Winchester.

  77. Inspiron i8600 smokes the G4 PowerBook. by Luke+Mewburn · · Score: 1
    "which Dell"?

    The i8600 has the following benefits over an Apple laptop:

    • a higher quality and higher resolution screen (1920x1200 vs 1280x854); this allows ClearType in XP or the equivalent in XFree86 to render text far superior to the PowerBook (even with OS X.3).
    • far superior warranty; 3 year onsite next business day with accident & theft protection is A$400. The PowerBook only gets 3 year "return to base and wait 1 week to fix" without the accident & theft protection for A$600. Apple.au need to lift there game here; as a professional I can't do without my laptop for 1-2 weeks whilst the reseller fixes it.
    • faster; the Pentium M @ 2.0GHz is sweet. In fact, it compiles kernels at speeds equivalent to a P4/Xeon 3.4-3.6 and an AMD64 3200 (all of these were workstations, compared to my laptop with a slow laptop disk.)
    • is at least 20% cheaper
    • runs NetBSD and Linux better than the PowerBooks. (OTOH, I'm told NetBSD & Linux run faster on PowerBooks than OS X)
    • runs XP vs OS X; there are applications that only work in W2K/XP, and emulating a PC at 1/10 the speed for more money is not acceptable.
    • comparable or better battery life (IIRC)
    • comparably quiet
    • isn't purchased to get a woody about being an elitist brand snob (AKA "I paid more, I have to justify why it's better!". Same goes for .eu cars for .jp cars -- I'm an apostate of the former :)

    Where the i8600 lacks compared to the PowerBook:

    • only has 100Mb ethernet instead of Gig-E. The Dell Latitude D800 (equivalent to the i8600) does have Gig-E but only ships with nVidia graphics, which I don't want to use because of inferior support in XFree86 for nVidia (vs the ATI 9600)
    • only has 400Mb firewire instead of 800Mb firewire
    • only has VGA out and not DVI out
    • subjectively it is not as well built. However, see the comment above about "better after-sales support". All laptops break, even Apple, as I have seen with iBooks and TiBooks.
    • a warm inner glow for being one of the "skiv" Elites who've seen the light and Converted to the Church of Apple.

    A while ago I compared a 15" G4 PowerBook and a Dell Inspiron i8600 "side by side" at my home office for a few weeks. I have two i8600's and one i8500 (about to sell the latter), and had a Inspiron 5000e for over 3 years. I have co-workers and collegues who are very happy with the i8600's. I also have co-workers and collegues who are happy with the G4 PowerBooks. On the other hand, one co-worker had a very unreliable 800Mhz iBook (and the story behind the crap support he got from the local reseller in Australia is a separate rant), and another friend had a similar "bad support" story for his 15" TiBook... I have a friend who swears Dell are unreliable and raves about his PowerBook. What widely divergant views!

    In short, use whatever you like for your own reasons. I really appreciate the high quality high resolution screen and far superior after-sale warranty of the i8600 over the G4 PowerBook. If Apple resolved these issues in future models, I may reevaluate.

    1. Re:Inspiron i8600 smokes the G4 PowerBook. by Mistah+Blue · · Score: 1

      With regards to DVI... I have a work supplied Dell D800 (very nice even with 1.4GHz Pentium-M it smokes 2.8GHz P4's on certain apps). The Advanced Port Replicator does have a DVI output.

  78. www.ibuypower.com by Foo2rama · · Score: 1

    check these guys out great prices and decent service beleive it or not. I have been recommending these guys around for awhile. I will warn you you might have to go out and buy some stealth fans to replace the ones they use, to quiet down the system.

    --


    ---In a time of Chimpanzees I was a Monkey.
    1. Re:www.ibuypower.com by aeroelastic · · Score: 1

      A co-worker of mine recommended this site to me, he says his system from there is a dream. Plus it would cost less than to get a comparable system (even with a govt. discount). As soon as I can afford it I'm buying there.

      --
      "It doesn't take a rocket scientist" -I guess I should leave then
  79. Re:Intel is trying to shift the battle, not catch by statusbar · · Score: 1

    10 years ago 1 ghz was a FANTASY. Just like 100 mhz was a fantasy in 1987. It was of course inconceivable that anyone would be EVER needing such power, and now you are saying this is minimum requirements.

    Anyways, people are buying off the shelf computers with XP and >2ghz. Do they need them? They are not just running office, they are running Doom and Half Life 2 and Halo 2, etc. etc.

    Jeff

    --
    ipv6 is my vpn
  80. Re:Intel is trying to shift the battle, not catch by johannesg · · Score: 1

    If those companies really cared about power they would be buying Arm or Transmeta, not Intel. But in reality, what they care about is using a vendor whose name they've heard about, and that is Intel, despite the fact that they make the hottest, most power hungry CPU's available on the desktop today.

  81. Re:Intel is trying to shift the battle, not catch by Shokac · · Score: 1
    My parents and the bulk of the people out there do not need a 64bit 5ghz monster under their desk. And honestly most of thosethat have them probably only use the power 5-10% of the time, if that.

    You know, 10 years ago I was saying the same thing about the bulk of the people not needing a 32bit 1 gHz monster under their desk. Now this is the minimum that they need.

    Don't underestimate Microsoft's ability to make software that brings yesterday's supercomputers to their knees today.

    Nah, my Win 95 realy runs FAST on my new XP2500+ :)



    anyway, there is other operating systems out there, not just M$ stuff, like LINUX
  82. AMD 64's stability?? by Anonymous Coward · · Score: 0

    Hello to all AMD 64 owners out there. What I would like to know is - have you got any stability problems with your computers? I administrate 7 recently built AMD 64 computers and I have faced weird stability problems on daily basis. Well, it could be the VIA chipset which is notorious for being unstable, but I'm not sure. I'm afraid these first incarnations of 64 bit processors made by AMD are just no there yet. Like most new products, these seem to have flaws. Probably should have waited a year and then get me 64 bit processors. On the other hand I must say that I've been impressed with the stability of Intel hardware.

    Now, I know there are going to be people out there who disagree with me on this. They're going to say that Intel sucks, and that they know a dozen reasons why AMD processors are better/faster/cooler. Take my advice and don't bother listening to their misinformed chatter. Instead, do what I do: drop into fifth and accelerate past them while they're still talking. Very few things are as satisfying as the stability of Intel hardware.

    1. Re:AMD 64's stability?? by jazzmans · · Score: 1

      My amd64 3000+ runs flawlessly, I'm currently at 19 days uptime, but that's only because of a power outage here.

      It's never crashed on me yet, running 64 bit debian.

      I've also got win 2k as a dual boot, and last time I ran windows for doom 3, it worked perfectly.

      jaz

      --
      Life is what happens to you while you are busy making other plans. No-one sees motorcycles
  83. Don't believe the hype! Use your brains! by Anonymous Coward · · Score: 0

    Some people who have bought 64 bit processors are eager to brag about it when in reality their 64 bit processors are not necessarily any faster than your 32 bit processor. It's just marketing hype! It's the same marketing bullshit we've seen with 192kHz / 24 bit audio cards. The average range of hearing for children is from a low of 20 Hertz (20 Hz) to a high of 20,000 Hertz (20 kHz). By the time a person hits age 20, years of loud listening to rock music has already killed off the 'high tones'. The high range for a young adult has dropped to 16,000 Hertz (16 kHz). Human ear can not even hear those frequencies! So really, any music that plays outside those ranges won't be heard by a human ear and is worthless as far as humans are concerned. Similar marketing hype is happening with Athlon 64 bit processors.

  84. Did AMD run on a 64-bit OS? by Jeppe+Salvesen · · Score: 2, Informative

    It seems to me that Anandtech didn't run the Linux benchmarks in 64-bit mode for the AMD processors. Given that they are giving an indication of processor performance, they should allow AMD to use that extra gear. It's there to be used!

    --

    Stop the brainwash

  85. On Intel's stability. by thegnu · · Score: 0

    I think one of the biggest reasons Intel has classically been seen as stable and AMD hasn't is because of the motherboard chipset. I'm willing to bet you've built an Intel system that was stable running alongside an instable system with the same processor and different motherboard.

    Intel processors have a higher likelyhood of resting on a decent chipset that works well in conjunction with the processor. It happens to be an Intel chipset, so I'm not debating the fact that Intel makes good hardware. I have been building AMD systems for a long time, and the stability varies with the motherboard chipset.

    After all this hot air (undoubtedly from a P4, har har), build an AMD system on an nforce chipset and you'll have the most rock-solid PC in the world.
    Love,
    Me

    --
    Please stop stalking me, bro.
  86. AMD older products by Anonymous Coward · · Score: 0

    Actually, even my K6-2 450mhz does not crash for 3 months.

    And it is not hot.

  87. The question is: can Intel beat its own marketing by kompiluj · · Score: 1

    Ok, Dothan is an excellent processor. The only question is whether Intel can beat its own marketing. When I recommend Athlon 64 over P4 Prescott to my friends they always say: P4 has HyperThreading! I will be able to run two apps a the same time! That's tremendous! I can't buy this Athlon thing, it has no HT!. And unfortunately for Intel there are some troubles with incorporating HyperThreading into Dothan's architecture due to short pipeline (HT was a hack to fix problems resulting from large wrong-branch-prediction penalty). So people not willing to buy Athlon will not want to buy Dothan (or any Pentium M) either. At least until it has HT or Intel convinces people that all their PR said was wrong.

    --
    You can defy gravity... for a short time
  88. PentiM is worst than PentiII, false performance. by Anonymous Coward · · Score: 0
    ftp://download.intel.com/design/mobile/SPECUPDT/25 266512.pdf

    35 bugs or more NOT TO BE FIXXED, a lot of awful flaws from Intel!!!
    worst performance indicator!!! worst protection!!! worst of worst!!!

    Y3. RDTSC Instruction May Report the Wrong Time Stamp Counter Value

    Problem: The Time Stamp Counter is a 64-bit counter that is read in two 32-bit chunks. The counter incorrectly advances and therefore the two chunks may go out of synchronization causing the Read Time Stamp Counter (RDTSC) instruction to report the wrong time stamp counter value

    Implication: This erratum may cause software to see the wrong representation of processor time and may result in unpredictable software operation.

    Don't buy the EXPENSIVE & AWFUL CENTRINO's PENTIUM-M !!!

    open4free ©

  89. PentiM is worst than PentiII, false performance. by Anonymous Coward · · Score: 0
    ftp://download.intel.com/design/mobile/SPECUPDT/25 266512.pdf

    35 bugs or more NOT TO BE FIXXED, a lot of awful flaws from Intel!!!
    worst performance indicator!!! worst protection!!! worst of worst!!!

    Y3. RDTSC Instruction May Report the Wrong Time Stamp Counter Value

    Problem: The Time Stamp Counter is a 64-bit counter that is read in two 32-bit chunks. The counter incorrectly advances and therefore the two chunks may go out of synchronization causing the Read Time Stamp Counter (RDTSC) instruction to report the wrong time stamp counter value

    Implication: This erratum may cause software to see the wrong representation of processor time and may result in unpredictable software operation.

    Don't buy the EXPENSIVE & AWFUL CENTRINO's PENTIUM-M !!!

    open4free ©

  90. PentiM is worst than PentiII, false performance. by Anonymous Coward · · Score: 0
    ftp://download.intel.com/design/mobile/SPECUPDT/25 266512.pdf

    35 bugs or more NOT TO BE FIXXED, a lot of awful flaws from Intel!!!
    worst performance indicator!!! worst protection!!! worst of worst!!!

    Y3. RDTSC Instruction May Report the Wrong Time Stamp Counter Value

    Problem: The Time Stamp Counter is a 64-bit counter that is read in two 32-bit chunks. The counter incorrectly advances and therefore the two chunks may go out of synchronization causing the Read Time Stamp Counter (RDTSC) instruction to report the wrong time stamp counter value

    Implication: This erratum may cause software to see the wrong representation of processor time and may result in unpredictable software operation.

    Don't buy the EXPENSIVE & AWFUL CENTRINO's PENTIUM-M !!!

    open4free ©

  91. PentiM is worst than PentiII, false performance. by Anonymous Coward · · Score: 0
    ftp://download.intel.com/design/mobile/SPECUPDT/25 266512.pdf

    35 bugs or more NOT TO BE FIXXED, a lot of awful flaws from Intel!!!
    worst performance indicator!!! worst protection!!! worst of worst!!!

    Y3. RDTSC Instruction May Report the Wrong Time Stamp Counter Value

    Problem: The Time Stamp Counter is a 64-bit counter that is read in two 32-bit chunks. The counter incorrectly advances and therefore the two chunks may go out of synchronization causing the Read Time Stamp Counter (RDTSC) instruction to report the wrong time stamp counter value

    Implication: This erratum may cause software to see the wrong representation of processor time and may result in unpredictable software operation.

    Don't buy the EXPENSIVE & AWFUL CENTRINO's PENTIUM-M !!!

    open4free ©

  92. Intel rushes to fix Alviso bug. by Anonymous Coward · · Score: 0
    http://theinquirer.net/?article=20405

    whoopty-f*i*kin-ding-dong!!!

    THERE IS A MINOR, and I mean minor bug in Alviso that won't be fixed before launch. The DDR subsystem doesn't like to power down the way it should, so if you look at the initial numbers, DRR2 will end up looking much better than it really is.

    I say this is a minor problem because you would pretty much have to me a moron to buy an Alviso machine with DDR anyway. The 15 or so minutes of battery life when you use DDR2 is well worth the price premium, especially considering it is probably not going to be in cheap laptops to begin with.

    ...

    Note: the Alviso machine is a based-Sonoma (==Dothan with 533MHz FSB instead of 400MHz FSB) notebook that Intel will start to release it in 16-January-2005.

    Don't buy the EXPENSIVE & AWFUL CENTRINO's PENTIUM-M !!!

    open4free ©

  93. Intel rushes to fix Alviso bug. by Anonymous Coward · · Score: 0
    http://theinquirer.net/?article=20405

    whoopty-f*i*kin-ding-dong!!!

    THERE IS A MINOR, and I mean minor bug in Alviso that won't be fixed before launch. The DDR subsystem doesn't like to power down the way it should, so if you look at the initial numbers, DRR2 will end up looking much better than it really is.

    I say this is a minor problem because you would pretty much have to me a moron to buy an Alviso machine with DDR anyway. The 15 or so minutes of battery life when you use DDR2 is well worth the price premium, especially considering it is probably not going to be in cheap laptops to begin with.

    ...

    Note: the Alviso machine is a based-Sonoma (==Dothan with 533MHz FSB instead of 400MHz FSB) notebook that Intel will start to release it in 16-January-2005.

    Don't buy the EXPENSIVE & AWFUL CENTRINO's PENTIUM-M !!!

    open4free ©

  94. Pentium 4 Mobile, numbnuts, is not the Pentium M by Anonymous Coward · · Score: 0

    You're not using a Pentium M, then (Dothan/Banias) - you're using a Pentium 4 Mobile.

    Get a clue.

  95. He's not using a Pentium M. by Anonymous Coward · · Score: 0

    He's using the Pentium 4 Mobile, I'd say. He's just too clueless to realize there's a difference and the P4 Mobile is more power hungry.

  96. Intel rushes to fix Alviso bug. by Anonymous Coward · · Score: 0
    http://theinquirer.net/?article=20405

    whoopty-f*i*kin-ding-dong!!!

    THERE IS A MINOR, and I mean minor bug in Alviso that won't be fixed before launch. The DDR subsystem doesn't like to power down the way it should, so if you look at the initial numbers, DRR2 will end up looking much better than it really is.

    I say this is a minor problem because you would pretty much have to me a moron to buy an Alviso machine with DDR anyway. The 15 or so minutes of battery life when you use DDR2 is well worth the price premium, especially considering it is probably not going to be in cheap laptops to begin with.

    ...

    Note: the Alviso machine is a Sonoma-based (==Dothan with 533MHz FSB instead of 400MHz FSB) notebook that Intel will start to release it in 16-January-2005.

    Don't buy the EXPENSIVE & AWFUL CENTRINO's PENTIUM-M !!!

    open4free ©

  97. Guh? by Anonymous Coward · · Score: 0

    This post got a +5 Interesting mod? Truly sad, because it's the worst sort of ignorant fanboyism.

    First, there's nothing "legacy" about a 32-bit operating system. How many people, even hard core nerds, have more than 3G of memory in their workstations/desktops?

    Second, the extra registers are somewhat helpful, but unfortunately for FanBoydom there aren't a lot of mainstream apps/games that use them.

    Windows 64-bit is pointless for the desktop, of course, I'm not sure what crack you're smoking there. I wouldn't use it for _3_ years even if it was released tomorrow. No point - immature drivers, no recompiled software, etc...

    Finally, it's fun to pull numbers out of your ass, isn't it? Your post was 40% more idiotic than most of the rest of the drooling fanboy posts.

    1. Re:Guh? by pp · · Score: 1

      Duh...

      Every application compiled with gcc for x64_64 gets to use said registers. Every application compiled with Visual C++ for x86_64/EM-64T gets to use said registers. That includes Doom 5 or whatever the application is. Applications compiled with gcc -m32 or Visual C++ for 32-bit windows do NOT get to use those registers. Once Quake XYZ gets compiled with Visual C++/64 and gets x % more FPS, people will switch to whatever performs better. That's why I said Intel has to shape up before that happens. Since there's no huge 64 bit market there _NOW_ they still have time. Apart from Linux hobbyists and scientific computing.

      40-45% is matlab simulating radio waves as well as a home-brew program for factorizing co-primes (primes I can fortunately factorize in my head!). Latter runs in 15 mins (2.4GHz Opteron), a 3.4GHz EM-64T uses half an hour and a normal 3Ghz Xeon about an hour. (4* less operations since it's really just multiplying big integers, really)

      Also even with 4GB of RAM you get some of the benefits, in fact anything over 1GB is enough
      with current Linux/Windows versions to get some benefits. Ever heard about 1:3 kernel:user VM splits (same on all x86 OS's) and TLB flushes?

      Feel free to discuss operating system or hardware architectures with me any time!

  98. Intel rushes to fix Alviso bug. by Anonymous Coward · · Score: 0
    http://theinquirer.net/?article=20405

    whoopty-f*i*kin-ding-dong!!!

    THERE IS A MINOR, and I mean minor bug in Alviso that won't be fixed before launch. The DDR subsystem doesn't like to power down the way it should, so if you look at the initial numbers, DRR2 will end up looking much better than it really is.

    I say this is a minor problem because you would pretty much have to me a moron to buy an Alviso machine with DDR anyway. The 15 or so minutes of battery life when you use DDR2 is well worth the price premium, especially considering it is probably not going to be in cheap laptops to begin with.

    ...

    Note: the Alviso machine is a Sonoma-based (==Dothan with 533MHz FSB instead of 400MHz FSB) notebook that Intel will start to release it in 16-January-2005.

    Don't buy the EXPENSIVE & AWFUL CENTRINO's PENTIUM-M !!!

    open4free ©

  99. No, they care about who gives them the most gifts by spackard · · Score: 1

    Gifts, and "executive briefs" are what execs thrive on. The ability not to look stupid among his other MBAs is important, so he wants to be able to call the company and have a brief made on a moment's notice as to why chip X beats chip Y.
    Power consumption really is in the noise with these guys.

  100. Wrong!!! by Anonymous Coward · · Score: 0
    The Pentium M - and Athlon for that matter - are double pumped, hence 400 and 533Mhz, and match up perfectly to the 400Mhz (200Mhz, DDR) and 533Mhz (266Mhz, DDR) memory being used...

    Wrong!!!

    Pentium M is not double pumped, it's quadpumped!!!.

    Pentium M (Dothan) uses PC1600 RAM SO-DIMM (it's DDR200 RAM SO-DIMM for 400 MHz FSB).
    Athlon XP 3200+ (Barton) uses PC3200 RAM DIMM (it's DDR400 RAM DIMM for 400 MHz FSB) (it's really double pumped).

    Pentium M (Sonoma) will use PC2100 RAM SO-DIMM (it's DDR266 RAM SO-DIMM for 533 MHz FSB).

    open4free ©

  101. Re:Air Conditioning? In Northern California?? by Iffy+Bonzoolie · · Score: 1

    It is my understanding that a lot of residences here in California do not have gas stoves or heat because we have earthquakes, which can rupture gas mains. I've lived in a bunch of places, and all the stoves/heating were all electric. Cooking on an electric range is pretty terrible, as well - restaurants and such have the industrial gas stoves, but they seem to be more rare in residences.

    -If

    --
    Run a pencil-and-paper RPG campaign with your far-off friends: Gametable!
  102. Re:Intel is trying to shift the battle, not catch by evilviper · · Score: 1
    5-20W here and there don't really mean much to you and I

    Speak for yourself. Every 10 watts or so that my systems put out raises the tempurature in here by another 1 degree. During the summer (which is about 9 months of the year) that's a big problem.

    With 3 computers running constantly, and a few more running intermittently, you can imagine what kind of wattage I'm using here, and just 20watts less from each one would make a very big difference to me.

    And that's not even accounting for the fact that electricity is too damn expensive here in California. 20 watts continually means a few killowatts every month, which means something like 50 cents, per machine, per month. And even more is spent during the summer on air conditioning to remove the heat they are generating. Very rough numbers, but you get the idea.
    --
    Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
  103. Re:The question is: can Intel beat its own marketi by Psykechan · · Score: 1

    But then again, neither architechture supports Blast Processing.

    This post is brought to you in 100 percent digital quality.

  104. Re:Air Conditioning? In Northern California?? by IncohereD · · Score: 1

    ELECTRIC HEAT! IN CALIFORNIA? I realize you don't use it as heavily as someone in Wisconsin, but jeez, thats just stupid.

    I live in Ottawa...and at least from a renters perspective the poeple I know with electric are much, much, much happier than people with oil heat. Judging by the other response you're implying gas is better...but I guess with older neighborhoods this may not be an option. It's certainly what my parents have in their suburb built in 1980, though.

  105. Re:Electric Heat In? Northern California?? by billstewart · · Score: 1
    Yup. That orphaned open parentheses in my article was supposed to be followed by a comment that the year I moved here I ended up spending more on electricity heating my condo in the Bay Area than I had spent on oil heating my standalone house in New Jersey, where we actually had something resembling winter, as opposed to a moderately cool rainy season. It's gotten a bit better, and my electricity costs haven't gone up as fast as oil, and for a few years we had a cheaper-at-night pricing system that cut my electric bills down. (I think that died when now-Terminated governor Grey Davis was mismangling our electric system.)

    On the other hand, I lived in California for a year in the late 70s, back when they were building those electrically-heated buildings, in a gas-heated one-bedroom apartment that never used more than $13 for gas+electricity in the worst of the winter, which was less than the electric bill alone for the gas-heated two-bedroom townhouse in New Jersey the next year.

    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks