Slashdot Mirror

← Back to Stories (view on slashdot.org)

Intel Releases New Pentium M Processors

Posted by timothy on from the leapfrog dept.

doormat writes "Its been known for a while, but now it's official, as Intel releases Dothan, the 90nm version of Banias, aka the Pentium M processor. It also debuts Intel's new numbering scheme. The fastest new part is a Pentium M 755 2GHz w/ a 100MHz FSB, and 2MB of L2 on die cache. Reviews are starting to tip up as the NDA expires. One is at Tom's."

131 of 190 comments (clear)

  1. Hmm by odano · · Score: 5, Funny

    For some reason I don't think it is a coincidence that intel basically stole BMWs numbering scheme...

    1. Re:Hmm by Kegster · · Score: 1

      Well since they have ceded the cutting edge and speed advantage to AMD's 64-bit CPUs they have probably been finding it harder to sell their chips
      on clock cycles alone.

    2. Re:Hmm by Anonymous Coward · · Score: 3, Informative

      Couple of more reviews of the Dothan I came across around the web as Tom's isn't the only site reviewing new kit.

      TrustedReviews - http://www.trustedreviews.com/article.aspx?art=428

      Digit-Life - http://www.digit-life.com/articles2/asus-m6000/asu s-m6.html

      PC Mag - http://www.digit-life.com/articles2/asus-m6000/asu s-m6.html

    3. Re:Hmm by The_Mr_Flibble · · Score: 1

      Does this mean some idiot with an intel based machine is going to cut me up now ?

  2. Naming Scheme by Anonymous Coward · · Score: 2, Funny

    I'm more excited about the change of naming scheme than anything else. I was afraid they were going to call it the Super-ultra-mega-fantastic PentiumEXTREME X32000. "Pentium M 755" sounds like a car.

  3. Intel fanboys around the world do a 180... by rokzy · · Score: 5, Funny

    ...over an AMD-style rating system instead of GHz.

    the resulting transfer of angular momentum changes the Earth's orbit moving it slightly further away from the Sun.

    the increased distance and lower temperature makes cooling easier. AMD stock set to skyrocket.

    1. Re:Intel fanboys around the world do a 180... by tobirius · · Score: 2, Insightful

      I think more important than the naming scheme is, that these cpus are faster while producing less heat. I don't like my computer being louder than my hairdryer.

    2. Re:Intel fanboys around the world do a 180... by The_Mr_Flibble · · Score: 2, Funny

      You have a hairdryer ??????? I thought this was slashdot

    3. Re:Intel fanboys around the world do a 180... by sotonboy · · Score: 4, Funny

      I have solved this, my PC _IS_ my hairdryer. I dry my hair and check my email at the same time, saving 0.05 hours per day.

    4. Re:Intel fanboys around the world do a 180... by mirko · · Score: 1

      With such temperatures, I'd suggest you brought your pc closer to your hair, you might look like Freddy Krueger afterwards, but you'll definitely save your remaining 0.05 hours.

      --
      Trolling using another account since 2005.
    5. Re:Intel fanboys around the world do a 180... by Anonymous Coward · · Score: 1, Funny
      ... I don't like my computer being louder than my hairdryer.

      This isn't 1977,
      ...so I hope you're a girl.

    6. Re:Intel fanboys around the world do a 180... by hahn · · Score: 3, Interesting

      Actually...
      Didn't Cyrix first utilize this 'performance-rating' style of rating the chips back in 1995 with their 6X86 chip?

      History of Cyrix

      --
      "The only normal people are the ones you don't know very well."
    7. Re:Intel fanboys around the world do a 180... by emarkp · · Score: 1

      Alternatively, it's possible that even Intel fanboys around the world have come to realize that clock speed isn't the only (or even most) useful metric. Intel wouldn't ditch GHz if the marketing team could successfully use it.

    8. Re:Intel fanboys around the world do a 180... by bhtooefr · · Score: 1

      The P-Rating system? That was used by AMD, Cyrix, and IDT (they were in some sort of alliance against AMD), and NexGen used something similar (I don't think they used THE P-Rating system, though).

  4. The Bottom Line by haunebu · · Score: 4, Informative
    "In a direct comparison between the old Pentium M 1.7 GHz and the new Dothan with 2.0 GHz, the newcomer clearly manages to gain the upper hand. In some of the benchmarks, the mobile CPU produced with 90-nm technology is up to 22% faster. Even if you only consider the difference in clock speed between the two CPUs, Dothan still offers a 5% advantage.

    The results of the battery life benchmarks show the benefits of 90-nm process technology. The two test systems were identical, except for the CPUs, and gave nearly the same results."

    From here.

    --

    Blue skies, Barthy Burgers, girls...

  5. Some more Dothan reviews by Anonymous Coward · · Score: 5, Informative

    Couple of more reviews of the Dothan I came across around the web as Tom's isn't the only site reviewing new kit.

    TrustedReviews - http://www.trustedreviews.com/article.aspx?art=428

    Digit-Life - http://www.digit-life.com/articles2/asus-m6000/asu s-m6.html

    PC Mag - http://www.digit-life.com/articles2/asus-m6000/asu s-m6.html

  6. It's all about the Pentiums baby... by OgTheBarbarian · · Score: 1, Funny

    Technology Today 2025, Intel releases ANOTHER Pentium CPU, the ZQX4055 with a slightly smaller die size, fibre obtic AI managed L1 cache and New Car Smell! NEXT!!!

    1. Re:It's all about the Pentiums baby... by stephenisu · · Score: 1

      Where did they get that chip? a bag of Doritos?

      --
      Sigs? We don't need no stinking sigs!
  7. In other news by Anonymous Coward · · Score: 3, Funny

    Intel now has more chip models then users. In their quest to sell to all markets at the same time, they now have have more then 6 billion choises, where the current planet population is about 5.2 billion.

    1. Re:In other news by zhenlin · · Score: 2, Funny

      Forsooth, in what yeare of our Lorde didst thou cometh across this olde giokke? Methinks the population of this worlde is at six thousande million now, 'twas at five thousande million when I was a wee childe.

    2. Re:In other news by the+chao+goes+mu · · Score: 1

      Unless the original post was using the definition of billion as one million squared. In which case the population is over six thousand million, while Intel has six million million chip models.
      Ok, I doubt this was the case, but it is possible.

      --
      Boys from the City. Not yet caught by the Whirlwind of Progress. Feed soda pop to the thirsty pigs.
  8. This stuff seems to overclock nicely by Chep · · Score: 5, Informative
    ... and with quite decent per-clock performance, to boot:

    Here

    (yeah, yeah, it's in French. Machine translate it for the text, and after all the pictures and chart don't need much of an explanation, do they?)

  9. In other news ... by psergiu · · Score: 3, Funny

    AMD is switching to the Mercedes numbering scheme while releasing the two new processors: Duron A 160 and Athlon64 SLK 600 KOMPRESSOR

    --
    1% APY, No fees, Online Bank https://captl1.co/2uIErYq Don't let your $$$ sit in a no-interest acct.
    1. Re:In other news ... by Anonymous Coward · · Score: 1, Funny

      Intel is expected to counter AMD's latest efforts with the release of a chip simply called the "Pentium Type R" this fall, as reported by The Register. The stripes alone give it a 10% boost.

    2. Re:In other news ... by mav[LAG] · · Score: 1

      Good choice because as we know, KOMPRESSOR stronger than all!

      --
      --- Hot Shot City is particularly good.
    3. Re:In other news ... by dago · · Score: 1

      .. and Via will feature the new all-integrated Via Panda CPU and stripped down Via Seicento CPU ?

      --
      #include "coucou.h"
    4. Re:In other news ... by TeknoHog · · Score: 2, Funny
      Athlon64 SLK 600 KOMPRESSOR

      Does it have hardware bzip2?

      --
      Escher was the first MC and Giger invented the HR department.
  10. Don't BMW on-board computers use PowerPCs? by michaeldot · · Score: 2, Interesting

    Motorola embedded PowerPCs power those German luxury cars don't they?

    Just a tad ironic.

    1. Re:Don't BMW on-board computers use PowerPCs? by CptChipJew · · Score: 1

      I think they do, but I couldn't find anything with Google. I do know from this however that the PowerPC controls the walking mechanism in Honda's Asmino, so I wouldn't be surprised if it was in their cars as well. I remember some silly rumor a few years ago that OS X was going to power the current Accords. I think it stemmed from that.

      --
      Vonal Declosion
    2. Re:Don't BMW on-board computers use PowerPCs? by dnoyeb · · Score: 2, Interesting

      As an automotive supplier, we use lots of motorola CPUs in cars, but not quite to the power PC level. HC08 and the like. Automotive manufacturers are excessively cheap and will not pay for such a thing. Not to mention its power is wasted.

      One thing to note about the new intel numbering scheme is it directly removes revelance from AMD's numbering scheme. Intel is adding a meaningless first number as if to say "the first number is just for our purposes, you pay attemtion to the MHz." If people buy into that, it will kill AMDs scheme as people look for the MHz on the AMD chip instead of assuming its equivalent number has any value. Interesting.

    3. Re:Don't BMW on-board computers use PowerPCs? by Kiryat+Malachi · · Score: 1

      There are a few automotive systems that use PPC, but for the most part that's only in the telematics side (OnStar and the like) - I see a ton of HCs, even some 68k stuff.

      Excessively cheap - too true. The excitement people have towards saving a penny per board scares me some times. I understand it, but it scares me.

      --

      ---
      Mod me down, you fucking twits. Go ahead. I dare you.
      (I read with sigs off.)
    4. Re:Don't BMW on-board computers use PowerPCs? by jsweval · · Score: 1

      The new Audi A8L uses some kind of PowerPC to run its MMI interface and software on the QNX RTOS. I've often thought of the hacking possibilities...

    5. Re:Don't BMW on-board computers use PowerPCs? by dnoyeb · · Score: 1

      Probably why OnStar is so expensive. I also heard it is yet to be profitable. But it works doesen't it :)

      We too are looking at other CPUs since Mot is still a little pricy and they really havent thrown themselves into the automotive thing full swing yet. Some of their chips could be more useful.

    6. Re:Don't BMW on-board computers use PowerPCs? by Kiryat+Malachi · · Score: 1

      Mmm.

      I'm fairly sure that OnStar is slightly profitable for the manufacturer of the modules. I have no idea if OnStar/GM is making any money off of it.

      I also have no idea if Freescale is going to go after automotive business full-on once they seperate from Motorola; part of the problem for them has been that a lot of potential customers don't like buying from them, since they're the same company as one of their competitors (Motorola's automotive group). Motorola automotive gets the same problem, the other way; they're heavily encouraged to use Motorola chips, which makes it hard to even get bids from outside suppliers.

      Then again, the HC series micros are still great for doing control, as are the various Motorola DSPs, so even if PPC remains a non-automotive line for them, Freescale will do okay in automotive for the foreseeable future, I think.

      --

      ---
      Mod me down, you fucking twits. Go ahead. I dare you.
      (I read with sigs off.)
    7. Re:Don't BMW on-board computers use PowerPCs? by Kiryat+Malachi · · Score: 1

      PPC doesn't power BMW's braking, engine, etc. systems, but if you get the fancy map/telematics packages, I know at least some of them do use Motorola embedded PPC processors.

      --

      ---
      Mod me down, you fucking twits. Go ahead. I dare you.
      (I read with sigs off.)
  11. Welcome to the silly numbers by DrYak · · Score: 3, Funny

    I just can't wait until they start using really stupid and cryptic numbers, like the GPU companies :

    Will the Pentium 5 X159-XL-SE be more power full than a Pentium 5 X150-Pro-Ultra ? Or less powerfull than a X160-LE ?
    Does it compare to an Athlon 64 Dual FX-95e 4699+++ ? ...I think websites with benchmarks like toms' will get more popular...

    --
    "Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
    1. Re:Welcome to the silly numbers by martingunnarsson · · Score: 5, Insightful

      Those sites getting more popular is a good thing, right? Instead of just looking at the clockrate, people will actually compare performance. The average Joe has no idea what makes a P4 2.0 GHz better than a Celeron 2.0 GHz. They're the same speed for crying out loud! Yeah, you get the point.

      --
      Martin
    2. Re:Welcome to the silly numbers by PhrostyMcByte · · Score: 5, Insightful

      Someone needs to make an open source benchmark on a bootable cd so OS doesn't matter, and no background apps can cause harm to it. Moving from MHz/FSB/Cache/etc to a single common rating # would make things a lot easier for the consumer. This would also spur more competition between the CPU companies, as they couldn't so easily obfuscate the true speed from their users.

    3. Re:Welcome to the silly numbers by kryptkpr · · Score: 2, Insightful

      There is no fair way to compare CPUs of different architectures (with synthetic tests).

      If you let companies into the bechmark design process, they will cheat (see 3DMark scandal).

      If you don't let companies into the bechmark design, then your benchmarks will never be able to squeeze "the most" performance out of anything, and how much performance you do get could be determined more by how you're testing then what you're testing on.

      Comparing CPUs is a very difficult task to do.. notice the reviewer ran more then 10 "real-life" tests to compare the processors.. this is a far more useful metric to the consumer then how well it can crunch through some synthetic tests.

      They just want to know how much faster their games will go, their videos will encode, and how much quicker photoshop will render their favorite filter. Those is very difficult to represent with a single, common number .. especially across architectures.

      --
      DJ kRYPT's Free MP3s!
    4. Re:Welcome to the silly numbers by mat.h · · Score: 1
      Moving from MHz/FSB/Cache/etc to a single common rating # would make things a lot easier for the consumer.

      A single rating # would be just as useless as MHz numbers. There is no such thing as "true speed". Different users have different workloads. If you spent your time recompiling KDE, look at the gcc rating from the SPECint suite. If you want a decent framerate in Doom 3, wait for it to be released and run on different systems. Then upgrade. And if you just want Windows to boot faster -- clean up your autostart registry keys and forget about speed ratings.

    5. Re:Welcome to the silly numbers by 10101001+10101001 · · Score: 1

      It's a good thing to a point. At some point, each manufacturer needs to clarify the difference between their various products other than price or no one will want to buy their products versus someone who does do semi-sane labelling. But, I do agree that up to now it's been like AMD and Intel have each been selling the analogy of different cubical containers where only one dimension is measured and attached as a label (imagine if Pepsi and Coke were selling 2" or 1" bottles (where the 2" or 1" represents the width) -- though at least then you know what Coke and Pepsi are and you can physically try to compare the bottles to get an idea of their relative volumes..and it'd be a hell of a lot easier to use some standard measure). Now, obviously CPUs aren't pop, but there's nothing stopping AMD and Intel from measuring in something meaningful like the time to encode a DVD or the average frame rate of some CPU bound game. Or they could just tell you the Mhz and the CPU core so we could read a few benchmarks and get an idea of the scaling power of Mhz for a core.

      --
      Eurohacker European paranoia, gun rights, and h
    6. Re:Welcome to the silly numbers by EpsCylonB · · Score: 1


      They just want to know how much faster their games will go, their videos will encode, and how much quicker photoshop will render their favorite filter. Those is very difficult to represent with a single, common number .. especially across architectures.

      So why can't a benchmark CD test these things ?

  12. laptop woes by Ryan+Broomfield · · Score: 5, Insightful

    Laptops get faster but laptop users don't get any smarter. Every day I see people with a brand new processor and 128MB of memory on windows XP. They insist that their laptop is slow but refuse to spend the extra 50 bucks to get a decent amount of ram in the machine. oh well.

    --
    download games I make at: http://www.shippysite.com
    1. Re:laptop woes by Krik+Johnson · · Score: 1

      My laptop used to have 128MB ram, then I put Linux on it. KDE 3.1 runs wonders on it.

      --
      I have a fetish for traffic cones
    2. Re:laptop woes by TrancePhreak · · Score: 1

      MISTA RYANS

      I got 512MB of RAM on my Laptop and it SCREAMS. I can play the shippy at many FPS.

      --

      -]Phreak Out[-
    3. Re:laptop woes by OmniVector · · Score: 1

      man, my friend constantly complained about the speed of her laptop. i kept telling her that she can't run a modern OS with only 128mb of ram including a bunch of other apps and expect it to never page out to hard disk. *sigh* of course for $50 she could have fixed this, but constantly refused to spend even that much after spending $1000.

      --
      - tristan
    4. Re:laptop woes by tzanger · · Score: 1

      KDE works well with 128MB memory? I beg to differ. I'm running KDE3.2 (and 3.1 before that) on a P4m/2G with 512M and it's very snappy. I find that just like XP, KDE wants a lot of memory to be happy.

    5. Re:laptop woes by TheLink · · Score: 2, Insightful

      I recently used a Pentium-M notebook with 512MB RAM, Windows XP and some junk on it (instant messaging stuff).

      It was slow. In one case it did lots of paging out for dunno what reason, then when I tried to switch tasks I had to wait a long time for things to page in and for the system to regain its senses.

      A 1GHz Duron, 128MB pc with Windows 2000 was more responsive and stable in comparison - same apps (two apps).

      From my experience with other XP machines, Windows XP is a downgrade from Win2K. The only advantage XP has over Win2K pro is you get to see the various owners of processes in the task manager. For some stupid reason you have to get Windows 2000 Server with the terminal server stuff to get that feature on the task manager.

      Windows XP is the Windows ME of the NT family.

      --
    6. Re:laptop woes by sparcnut · · Score: 1
      Laptops get faster but laptop users don't get any smarter.
      That's the problem with laptops. A lot of laptops don't have a "brightness" control, unlike practically all desktops. This is why desktop users are smarter than laptop users.
      --
      perl -e 'print $i=pack(c5, (41*2), sqrt(7056), (unpack(c,H)-2), oct(115), 10);'
    7. Re:laptop woes by drinkypoo · · Score: 1
      That's true of computers in general, though. At work we'll spec someone a system to show them what we think they should buy, and then they go out and spec something else to "save money" and decrease memory from 512 to 128, then sometimes bump up their CPU, or add one of those useless Zip 750 drives to a system which comes with a flash drive and has a CDRW already. Then you have to explain the difference between physical and virtual memory to them, whee!

      I have 1GB ram in my Windows XP system, you might think that's overkill but my boot time was cut in half when I doubled my memory from 512MB, and also I have a virtual machine running most of the time to do cross-compiling via distcc for my gentoo-mips box, so that ram comes in real handy. When I see people with 128MB I just laugh. We have celeron 333 desktops everywhere and they have 64 or 128MB ram in most cases (some 256, but that's rare) and people think they're slow because they're old. No, they're slow because they don't have any ram. These people aren't doing physics simulations, they're running office which requires little to no CPU, but the programs are bloaty so when you run 'em you page page page like a madman.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    8. Re:laptop woes by skiflyer · · Score: 1

      A Pentium-M what? They range from the ultra-low power 1GhZ (or do they go all the way down to 800MhZ?) and up to 1.7GhZ (before today anyway).

      Personally, I run a 1.6 Pentium-M with 768Megs of RAM and it's perfectly fine running XP and Linux with KDE 3.2 (2.4 Kernel). The only downside I have is the harddrive is kind of slow, so file access is a bit of a bear

      Personally, I think the grandparent hits it dead on, it's important for any computer with a modern OS to have sufficient RAM, it's doubly important for us laptop users because we have slow harddrives for the most part.

    9. Re:laptop woes by drinkypoo · · Score: 1
      Windows XP also provides you remote desktop connection, the other feature found in win2k server. I suspect there is a connection here.

      In my experience Windows XP is much more stable than Windows 2000. I am aware that most people have had the opposite experience but I've bluescreened win2k repeatedly on every service pack that's been released to date. I have very rarely bluescreened XP since day one, and when I have it has generally been tracable to bad hardware (this was not true in all cases, however.)

      Windows XP does seem to be substantially more memory-hungry than Windows 2000 for no reason I can quite discern. But, to me this is not a problem, as I want large amounts of ram anyway.

      Your anecdotal evidence is just that, and it has no numbers to boot. (Nor does mine, of course.) So it's pretty meaningless. windows XP actually does have more functionality (slightly) than Windows 2000, Windows ME has less functionality than Win98SE. There's a big, big difference.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
  13. What I don't understand... by drsmithy · · Score: 4, Interesting

    ...Is why more hardware vendors aren't using these CPUs in Blade configs. They seem perfect for high density computing power.

    1. Re:What I don't understand... by leuk_he · · Score: 4, Informative

      Check again:

      HP to sharpen blade with Pentium M

      And don't forget: high end servers don't use bleeding edge processors since they need some extra time to certify the hardware.

    2. Re:What I don't understand... by CuriHP · · Score: 1

      It's off to disk you go.

      --
      If it's not on fire, it's a software problem.
  14. Re:FSB @ 100 MHz ? by Chep · · Score: 4, Insightful

    I beg to disagree, if those guys can be trusted. Sure, a better FSB would clearly help, but look at what they achieve with a single-channel FSB100.

  15. Hmmm... by Anonymous Coward · · Score: 4, Funny

    These new names seem about as useful as self confessed penis inches. Real world benchmarks?

    1. Re:Hmmm... by frozenray · · Score: 2, Funny
      > These new names seem about as useful as self confessed penis inches. Real world benchmarks?

      Your wish is my command:
      #!/bin/sh
      LC_ALL=C
      echo `uptime|grep days|sed 's/.*up \([0-9]*\) day.*/\1\/10+/'; \
      cat /proc/cpuinfo|grep MHz|awk '{print $4"/30 +";}'; free|grep '^Mem' \
      |awk '{print $3"/1024/3+"}'; df -P -k -x nfs | grep -v 1k \
      | awk '{if ($1 ~ "/dev/(scsi|sd)"){ s+= $2} s+= $2;} END \
      {print s/1024/50"/15+70";}'`|bc|sed 's/\(.$\)/.\1cm/'
      (stolen from someone's Usenet sig)

      "With a castle that big, do you think maybe Lord Farquaad is trying to compensate for something?"
      Shrek
      --
      "There are already a million monkeys on a million typewriters, and Usenet is NOTHING like Shakespeare." - Blair Houghton
    2. Re:Hmmm... by dleifelohcs · · Score: 1

      my server is 59.1cm long.

  16. Future Intel processor numbers by nacturation · · Score: 2, Funny

    If you're interested in Intel's future processors and how they'll be numbered, please look at the handy chart available here.

    --
    Want to improve your Karma? Instead of "Post Anonymously", try the "Post Humously" option.
    1. Re:Future Intel processor numbers by superbam · · Score: 1

      Dammit now I need to get a BMW 745i to go with my Pentium M 745.

      --
      We've tried nothin' and we're all out of ideas. - Ned's Mom
  17. Re:FSB @ 100 MHz ? by Anonymous Coward · · Score: 5, Informative

    Umm.. no. First its Quad-pumped, meaning that it acts like a 400MHz bus. I believe the P4 at 800MHZ quad pumped is somewhere in the 6GB/s range. So this should be sufficent since the architecture is less dependant on bandwidth.

    You could even bother to do a back of the envelope calculation.
    BW = (100*10^6)(4)(2 words)(4bytes/word)/(1024^3 GB/byte) = 2.98 GB/s

    So yeah, its sufficent.

  18. Desktop by moxruby · · Score: 4, Insightful

    My ageing duron 1.3ghz does everything I want it to.
    If someone made a reasonably priced, Pentium-M desktop using low power and heat components, I would consider buying it. Especially if it had no fan.
    The energy savings alone would make it worthwhile.

    1. Re:Desktop by drsmithy · · Score: 1
      The energy savings alone would make it worthwhile.

      Uh, over what sort of time frame ?

    2. Re:Desktop by BenjyD · · Score: 4, Interesting

      Let's see:
      Intel Pentium-M 735 $294,21W
      Comparable Athlon-XP (2600-3000?) $170, ~70W

      Price difference = ~$120
      Power difference=~50W

      Electricity cost (UK prices, don't know any others)=$0.10/Kwh

      Time to break even=~2.7 years

      Which is about the lifetime of a processor, I guess. Of course, that's assuming you use both at 100% CPU constantly for three years.

    3. Re:Desktop by tomstdenis · · Score: 1

      Athlon-XPs don't have easily accessible idle modes. Intel processors do [via HLT opcode]. So when both boxes are idle the PM is taking 6W and the Athlon-XP is basically taking the full power.

      So the diff is really 64W or 0.064Kwh so diff is around 2.2 years. ;-)

      Though the stability of Intel cpus alone is worth it. Have yet to have my P4 box shutdown because a 10 minute test run of a program [full load] over heats the core...

      Tom

      --
      Someday, I'll have a real sig.
    4. Re:Desktop by BenjyD · · Score: 1

      You must be really bad at building computers if you've had that happen to an AMD chip.

    5. Re:Desktop by Darren+Winsper · · Score: 4, Informative

      AMD chips have implemented the HLT opcode for many years.

    6. Re:Desktop by tomstdenis · · Score: 1

      Yeah, I'm the first person to notice AMD cpus produce heat.

      Except unlike most /. drool-boys I actually *used* my cpu. Under full load it doesn't take much to get an AMD cpu really hot with modest cooling.

      Tom

      --
      Someday, I'll have a real sig.
    7. Re:Desktop by hsa · · Score: 2, Insightful

      If someone made a reasonably priced, Pentium-M desktop using low power and heat components, I would consider buying it. Especially if it had no fan. The energy savings alone would make it worthwhile. Who cares about the energy savings? This baby would be *quiet* and you could really leave your machine on, while sleeping in the room.

    8. Re:Desktop by tomstdenis · · Score: 1

      Unfortunately it doesn't do anything. That's why you have to run programs like athcool or vcool to actually get it to power down.

      --
      Someday, I'll have a real sig.
    9. Re:Desktop by the+quick+brown+fox · · Score: 1

      So when both boxes are idle the PM is taking 6W and the Athlon-XP is basically taking the full power. That seems a little hard to believe. If that were the case, Athlon XP's would have the same temperature at full load as they do at idle, and that's certainly not the case.

    10. Re:Desktop by kryptkpr · · Score: 1

      I tried one of those programs once.. fried my mobo (Asus A7V133). Just shut down one day never to turn back on. YMMV, but I'm sticking to the "good thermal paste + good aftermarket cooler + not overclocking method" of cooling from now on ... no more software that does funky things with my chipset.

      --
      DJ kRYPT's Free MP3s!
    11. Re:Desktop by BenjyD · · Score: 1

      Power dissipation under load is similar between AMD and Intel chips.Anandtech shows the idle temp of the athlons a little higher, and vice versa for the load temps. There's around a 10% difference using the same heatsink.

      Why the assumption that most people don't use their cpus? Mine is at 100% a lot of the time either running tests for development work or gaming.

    12. Re:Desktop by BenjyD · · Score: 1

      No, the HLT code as in the opcode issued by the OS in its idle loop. No way that can fry your mobo, it's just a machine code instruction.
      Every OS apart from Win9X uses it (ie. Win2000+, Linux etc).

    13. Re:Desktop by BenjyD · · Score: 1

      ignore me, missed the parent post.

    14. Re:Desktop by kryptkpr · · Score: 1

      I was referring to the CPU cooling programs that reprogram your chipset, not the HLT opcode-issuing programs (which are built into modern OSs from what I understand).

      --
      DJ kRYPT's Free MP3s!
    15. Re:Desktop by mczak · · Score: 2, Informative
      Intel Pentium-M 735 $294,21W Comparable Athlon-XP (2600-3000?) $170, ~70W
      That's not really a fair comparison. You can get a Mobile Athlon XP-M 2600+ (mainstream) for $110, and it has a TDP of 45W. Yes, the Pentium-M still uses less power than the Mobile Athlon, but it's nowhere near as dramatic as your numbers suggest - and the price difference is even larger.
    16. Re:Desktop by mr+i+want+to+go+home · · Score: 1
      For my home server I'm using a old 600mHz iMac G3 - low power (approx 6 watts output I think?), quite nimble, and NO FANS!

      As I live in a studio appartment this is ideal. It runs all day and night, and all I ever hear from it is the ide hard drive ticking over every now and then.

      I did look at some VIA based machines, but the Mac was cheaper, and, well....it meant no Microsoft. Love it.

    17. Re:Desktop by BenjyD · · Score: 1

      This whole thing is wildly inaccurate. Rounding errors, ballot stuffers, dynamic IPs, firewalls. If you're using these numbers to do anything important, you're insane.

    18. Re:Desktop by Jeff+DeMaagd · · Score: 1

      It is a bit distorted. I think it means that the idle power consumption by an Athlon is higher than the idle power consumption by Pentium M. Even if the chip is running without HLT, if it isn't using its functional units or hitting to cache, then it takes less power than it would at 100%.

    19. Re:Desktop by forkazoo · · Score: 1

      Adding a fanless power supply, and solid state storage, so that they only way you realise your computer is turned on is the keyboard LED... Priceless.

    20. Re:Desktop by ncc74656 · · Score: 1
      Athlon-XPs don't have easily accessible idle modes. Intel processors do [via HLT opcode]. So when both boxes are idle the PM is taking 6W and the Athlon-XP is basically taking the full power.

      Are you trolling, or are you just unbelievably ignorant? Athlons idle the same way as everything else.

      Though the stability of Intel cpus alone is worth it. Have yet to have my P4 box shutdown because a 10 minute test run of a program [full load] over heats the core...

      Now we know you're trolling...if Athlons ran at full throttle all the time, it wouldn't make any difference whether it was at full load or not. You're contradicting yourself. Go crawl back under your bridge and leave us alone. I keep most of my machines fully loaded with Prime95/mprime, and they run for weeks/months at a time without any problems.

      --
      20 January 2017: the End of an Error.
  19. Overclocking Dothan by mst76 · · Score: 4, Informative

    If you can read French, there's an article on x86-secret where they opened a laptop, installed a big cooler, and overclocked a 2.0 Ghz Dothan to 2.4 Ghz. It remained stable during 2 hours of BurnP6 and stayed under 30 degrees C. The 2.4 Ghz Dothan beat the 3.4 Ghz P4 in all their benchmarks, and is comparable to the Athon 64 3400+.

    1. Re:Overclocking Dothan by AtomicBomb · · Score: 2, Informative

      O/C Dothan is pretty cool even when compare with top of the line P4 or Athlon-64
      x86-secret in English. On the other hand, Althlon-64 3400+ has a core clock speed of 2.2GHz, slightly lower than that of the O/C Dothan (2.4GHz)... Of course, if we talk about processing power/power consumption, Dothan wins by a mile...

      Intel is starting to recover from the CPU design competition. The 3.4GHz clock speed P4 is just unnecessarily too high....

    2. Re:Overclocking Dothan by Kjella · · Score: 1

      O/C Dothan is pretty cool even when compare with top of the line P4 or Athlon-64. (...) Of course, if we talk about processing power/power consumption, Dothan wins by a mile...

      Which begs the question, why isn't it a desktop processor already? Less power = less CPU fan noise, less power drain on system = less PSU noise, less heat in system = less case fan noise.

      Eventually, I think the PIV would be recognized as somewhat of a side-track in processor development. Imagine what Intel could have done if they had gone straight for the Pentium-M. But that's what you can afford to do when you're huge.

      Kjella

      --
      Live today, because you never know what tomorrow brings
    3. Re:Overclocking Dothan by Chep · · Score: 1

      Would you like to place bets on what those 6M transistor are for?
      Now here's hoping Hans de Vries has some spare time... ;-)

    4. Re:Overclocking Dothan by ckaminski · · Score: 1

      Well, their hope was that they could translate some of that hyperthreading into real performance advantage with those uber-deep pipelines they have (what marketdroid thought THAT was a good idea?). When that failed, back to the drawing board...

    5. Re:Overclocking Dothan by TheLink · · Score: 1

      Probably because Intel didn't realize how badly they screwed up till too late. Takes years from plan to tape out.

      Now they have to convince enough suckers to buy Prescotts till they turn things around.

      There's also the server market to worry about. Imagine an SMP Prescott server - 2 X 100+W in a 1-U, or how about quads 8-). Alternative? Does the Dothan/Banias support SMP? Does it work with popular server chipsets? So Dell will probably stick to Northwood for servers.

      Looks like it could be a really good year for AMD, that is if they don't screw up :). They have till mid 2005 whilst Intel does their damage control and emergency stuff.

      I don't see them being able to improve Prescott significantly before then.

      Dothan isn't really much of an improvement compared to Banias. Performance/clock is about the same.

      --
    6. Re:Overclocking Dothan by drinkypoo · · Score: 1

      I'm too lazy to link in the article but right here on the front page of slashdot we just discussed the fact that intel is giving up on Prescott (once they have the new desktop versions of Pentium M out) because it's not scaling any more. Dothan/Banias will apparently become the basis of a new 2-core processor to replace Prescott, the way for which was paved by the HT P4, since people are used to having SMP-capable operating systems now. At least, the people with top-end P4 chips.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    7. Re:Overclocking Dothan by Dun+Malg · · Score: 1
      Which begs the question, why isn't it a desktop processor already?

      It doesn't beg the question, it raises the question. "Beg the question" is an archaic idiom that means "to assume as given the point being argued", e.g. "This painting is trash because it is obviously worthless." The word "beg" in this case means "to improperly take for granted", a usage of the word which we don't see anymore except in this one expression.

      --
      If a job's not worth doing, it's not worth doing right.
    8. Re:Overclocking Dothan by MightyYar · · Score: 1
      Language changes:
      Here is a nice article.

      It used to be strictly improper, but now that use of the phrase has become more and more accepted.

      --
      W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
    9. Re:Overclocking Dothan by Dun+Malg · · Score: 1
      Language changes: Here is a nice article. It used to be strictly improper, but now that use of the phrase has become more and more accepted.

      True, but at this point we still have a sizable number of people using it in the old way, which leads to confusion. Even if the phrase does comes to mean "prompts the question" rather than "assumes the argument", I think it's important to remind people of the original meaning because they may run across it and become confused.
      Besides, I'm an old curmudgeon. This is what we do. :)

      --
      If a job's not worth doing, it's not worth doing right.
  20. Links... by defaultXIX · · Score: 2, Informative

    TrustedReviews

    Digit-Life

    PC Mag

    The a tag is your friend...

  21. Re:FSB @ 100 MHz ? by stephenry · · Score: 4, Interesting

    The large cache is probably more likely used to lower the power consumption of the processor. It's far cheaper (in terms of power) to drive short chip level wires to get commonly used data from a cache than it is to drive long, high capacitance, board level wires to memory.

  22. Linux on Centrino laptops and notebooks by wehe · · Score: 4, Informative

    As soon as the new Centrino generation will be available on laptops and notebooks, there will be Linux information about Dothan machines here.

  23. So The Bottom Line Is... by Afty0r · · Score: 1

    It's about 5% faster at the same clock speeds? Using the same amount of power?

    So nothing spectacular?

    1. Re:So The Bottom Line Is... by klui · · Score: 4, Interesting

      Not sure how to read the comparison table between Banias and Dothan.

      Peak thermal power is 24.5W for Banias and 21W for Dothan. But it looks like average thermal power is 6W for Banias vs. 7.5W. Sleep power, deep sleep power, and deeper sleep power are all higher for Dothan.

      Does this mean if you're not doing a lot of number crunching, your battery life will be less than a Banias system? Maybe the cache has something to do with this... whatever. I would be more impressed if the power requirements were lower across the board. And the lower peak doesn't seem that low. What's the best case thermal advantage going from 130nm to 90nm of an identical circuit?

    2. Re:So The Bottom Line Is... by mczak · · Score: 4, Interesting
      Does this mean if you're not doing a lot of number crunching, your battery life will be less than a Banias system?
      Yes. However, the difference is rather small (keep in mind the cpu is not the only thing drawing power, so that 1.5W difference is really not that much).
      Maybe the cache has something to do with this... whatever. I would be more impressed if the power requirements were lower across the board. And the lower peak doesn't seem that low.
      No, it's not because of the cache (actually, banias/dothan have neat tricks to reduce power draw of the cache). It's a direct result of 90nm vs. 130nm. Peak power draw goes down, because of the lower voltage needed for switching in the transistors. But idle current goes up, because if you reduce structure size, you get more and more leakage current (which, btw, is a huge problem nowadays - leakage current was basically 0 just a short while ago, but it goes up exponentially with smaller structures).
  24. Re:'cut me up' by The_Mr_Flibble · · Score: 1, Funny

    Oh yeah, Over here in good old blighty bmw's are renowned for their drivers inabilty to care about other road users. I think similar to suv's over in the states.

  25. Dothan / Banias Compatibility by NeGz · · Score: 5, Interesting

    I don't suppose anyone knows if it the new Dothan CPUs are compatible with existing Banias notebooks at all? I very much doubt it, but I've not seen it expressly said thus far. It would be nice to be able to upgrade my ~3 month old 1.5ghz Banias to a Dothan, heh.

    Also, if they are not compatible, is Intel planning to take the Banias chips further, or will I be stuck at 1.7ghz max (or is it 2.2ghz?) until I buy a new one?

    Excuse me for being a little ignorant. :)

    1. Re:Dothan / Banias Compatibility by mczak · · Score: 2, Informative

      It is pin-compatible, but intel says you need a new stepping (b-step) of their chipset. I have no idea if these new steppings are already in use, or if it might run even with the old stepping unofficially. Also, the bios might not like it.
      At least a 2.13Ghz P-M (with 533Mhz FSB) is on the roadmap. Not THAT much of an improvement from the now available 2Ghz part though. Still quite impressive nonetheless.

  26. Re:FSB @ 100 MHz ? by haxor.dk · · Score: 1

    Agree - i overlooked that it is xpumped.

  27. Re:Hmmm.... by theguywhosaid · · Score: 3, Informative

    its quad pumped, like DDR but more so. its effectively a 400MHz bus

  28. Re:'cut me up' by Anonymous Coward · · Score: 3, Funny

    hm... "cut me up" seems to indicate that the person has a knife and they're about to carve you like a side of roast beef. Then again, "cut me off" has a rather eunuch connotation to it.

  29. I'm going to miss the good old days of MHz by otter42 · · Score: 4, Interesting
    Yes, I know that MHz were somewhat misleading, but come on, they weren't that bad after all. While it was never true that a Pentium 3GHz was twice as fast as a Pentium 1.5GHz, it was nonetheless faster. And Intel's dominance forced AMD to give similar numbers (rounded up for inflation, of course) that helped us all compare.

    Now I have to wonder which new chips are faster. To (blatantly) steal from an earlier post:

    Will the Pentium 5 X159-XL-SE be more power full than a Pentium 5 X150-Pro-Ultra ? Or less powerfull than a X160-LE ? Does it compare to an Athlon 64 Dual FX-95e 4699+++ ?


    I hate the "consumer electronics" style of naming things, incrementing a model number in order to sell an inferior product. Who here honestly thinks that Intel won't do exactly this when they release a product that bombs? I still remember the to-do when Tom's Hardware published a pre-release review of the Pentium II, showing that it was inferior to a Pentium MMX of the same clock speed.

    What I would REALLY like to se is AMD seize the MHz banner now that Intel has abandonned it. I mean, now AMD doesn't have to give performance "numbers" to convince people to buy it. They could go back to simple MHz ratings, forcing Intel to keep itself honest. After all, we all know that the whole reason we all hated the MHz rating was because AMD had superior performance at inferior speeds and it just wasn't fair. I don't remember too many people complaining when AMD went back to MHz specs with the Athlons. Here's hoping to see it again soon.
    --
    www.eissq.com/BandP.html Ball and Plate System. Amuse your friends. Crush your enemies.
    1. Re:I'm going to miss the good old days of MHz by wwwrun · · Score: 1

      While I quite agree that abandoning MHz as a way of expressing cpu performance is causing confusion (Opterons are already a case in point), what other choice is there? Dothans and future chips from AMD and Intel will be clocking slower than P4s for some time to come, whilst clearly outperforming them. How are the illiterate masses to know what is the 'best' chip for them?

      Using something other than MHz is necessary, but that doesn't mean it isn't going to get silly, as you say. The temptation for Intel or AMD to use daft naming schemes to make their chips seem new and exciting (and better than the competition, natch) will be overwhelming.

      (p.s. It is fairly reasonable for the first of a new generation of cpu to perform worse than the older ones at the same clock speed. Changes such as longer pipelines harm performance at a given speed, but are essential to allow the new chip to clock higher in the future.)

    2. Re:I'm going to miss the good old days of MHz by ckaminski · · Score: 1

      How reasonable is it? The Pentium wasn't slower than the 486, 486 wasn't slower than the 386, and the 386 surely wasn't slower than the 286... The Pentium Pro was an anomaly. It was rushed. Intel was getting slaughtered by Alpha and others, and losing (can't say lost, they didn't really have any to speak of) quite a good bit of marketshare in the workstation market (which Intel wanted to expand into). Hence a processor that in it's slowest version, 150 mhz, was not much faster than a Pentium 133.

    3. Re:I'm going to miss the good old days of MHz by LWATCDR · · Score: 1

      Actually if I remember correctly there where some people that claimed the Pentium was slower than the 486 when it first came out. I could be wrong since it was a long time ago. And of course you could get into the whole 486 DX-50 vs DX2-66 debate. I do rember when the DX2 chips first came out. Oh the crys of foul since they where not really 66 of 50 mhz machines but use a clock doubler to run the core faster than the bus. How Bogus we all thought :)

      --
      See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
    4. Re:I'm going to miss the good old days of MHz by ckaminski · · Score: 1

      How silly, indeed!

    5. Re:I'm going to miss the good old days of MHz by bhtooefr · · Score: 1

      Corrections:

      The 386 WAS slower than a 286 - many people claim that a 286/25 was on par with a 386/40. Also note that the 386 was designed BEFORE the 286, and the 286 was designed because the market wasn't ready for the 386. Also, from what I heard, the Pentium was slower than the 486. The Pentium Pro was by nature slower than the Pentium, but its 32-bit optimizations pushed it past the Pentium - but since everyone ran old 16-bit apps, it seemed slower. Look at where that architecture's gone though - it's made it to 2.0GHz, and 2.4GHz overclocked, from a humble beginning at 133 (not 150) MHz. The Pentium ramped to 233MHz (a 266MHz OC was really pushing it) from 60MHz.

    6. Re:I'm going to miss the good old days of MHz by NutscrapeSucks · · Score: 1

      I can tell you read an article in "PC Magazine" 10 years ago and are misremembering the details.

      The Pentium Pro kicked the regular Pentium's ass. It was even faster than the current Alpha chip on release. It was a great server/workstation chip, and had enough legs that Intel just released a 2.0 Ghz P6 chip today and is going to the core for its mainstream desktop line.

      The only problem was that the PPro was not as fast with 16-bit DOS/Win3.1 code. Not slower. So the mags said that for ordinary users, it wasn't worth the price.

      --
      Whenever I hear the word 'Innovation', I reach for my pistol.
    7. Re:I'm going to miss the good old days of MHz by toddestan · · Score: 1

      Actually if I remember correctly there where some people that claimed the Pentium was slower than the 486 when it first came out.

      If I remember right, the first Pentiums were hugely expensive, so everyone just bought the high end 486's. Also, the first Pentiums were a disaster, it wasn't until the Pentium 75 and socket 5 that Intel got it right. Why get a Pentium 60, when for less money you can get a cooler, faster, 486DX4-100? But stick that 486 against a Pentium of the same speed and the Pentium was a clear winner.

    8. Re:I'm going to miss the good old days of MHz by ckaminski · · Score: 1

      The original P54C took the major hit because they sold it as "workstation class" and it ended up having such a nasty floating point bug that it killed it's early adoption (nevermind that the Pentium was a DOG compared to the new AXP 150 at the time). The consumer market (in the form of AMD, I believe) started with the clock-tripling. Once this was seen as a viable option, you set the stage for the eventual Pentium Pro debacle, where it was Intel's king of the hill for about 2 months before some Pentium MMX CPU's started kicking the crap out of it.

    9. Re:I'm going to miss the good old days of MHz by ckaminski · · Score: 1

      No, not "read", worked with. Intimately. The PPro 200 was the only chip in the (initial) family that was of any great significance. I'm not sure which Alpha was "released" at that time, I'm confused as to which preproduction model I had in hand, but the PPro could not touch the AXP 233. The AXP 150/166 was already approaching 2 years old when the PPro came out. P6-150's were marginally better than the P5-133's. 32 bit software only.

      Yes, the P6 core has definitely had some pretty long legs. And it was a design I was thoroughly upset at Intel for dumping for that bastard P4 architecture. 20 stage pipelines, who'd a thunk it? Itanic on the desktop. almost 3 years, and they've gone a Ghz and a half? :-)

  30. Re:Hmmm.... by tomstdenis · · Score: 1

    Yeah but that's such a scam. My cpu has a 800Mhz FSB. Nice right? Goes perfect with my 400Mhz DDR memory....

    All this quad/double pumping sounds nice but hides a problem that really is the bottleneck. LATENCY!

    That is my cpu can only send control info over the bus at 200Mhz. So while it can send 6GB/sec to my Northbridge if I fetch stuff from random addresses I effectively have a 200Mhz bus....

    I'm sure what will save the Dothan is the huge 2MB cache [provided it has a decently high hit rate].

    Tom

    --
    Someday, I'll have a real sig.
  31. Marketing by Ed+Avis · · Score: 4, Funny
    From the article^H^H^H^H^H^H^Hpress release:
    The new chip opens up advanced applications to notebook computer users like full-screen high-definition video playing and mobile videoconferencing, while keeping the PC relatively light, Intel said.
    Isn't that just what they said for the previous line of processors? If this new chip is needed for videoconferencing, are we to conclude that the previous chip couldn't do it, contrary to what Intel said at the time?

    (My point is, reprinting inane press releases does nobody any good.)

    I'm surprised that the marketing department missed the upcoming opportunity to label dual-core CPUs with 'twice' the clock speed, as is done for bogomips.

    --
    -- Ed Avis ed@membled.com
    1. Re:Marketing by ruud+awakking · · Score: 1

      stty erase ^H Thanks.

    2. Re:Marketing by smcv · · Score: 2, Informative

      Bogomips *are* an objective measure, though. They're how many million times per second Linux can run a particular busy-loop, used for high precision timing (basically the same idea as how you did delays in old BASIC programs, i.e. FOR X%=1 TO 100000:NEXT X%, adjusting the large number down if you had a slow computer, or up if you had something blindingly fast like the 1.8MHz 6502 in a BBC Micro; ah, those were the days :-)

      In other words, they're an objective measure of how fast your CPU can achieve absolutely nothing, hence the name bogomips (= millions of bogus instructions per second).

      Old x86s generally do about a bogomip per MHz, newer ones (Pentium and up) do 2 bogomips per MHz due to different pipelines and such, so yes, they really *can* do twice as much nothing per clock cycle.

      Of course, how fast a CPU can spin round and round a redundant loop has little relation to how much actual work it can do, so the only things bogomips are useful for are high-precision timing and pointless boasting.

      (Different CPU architectures run different busy-loops in that part of the kernel, so in any case bogomips aren't directly comparable between architectures anyway. My G4 manages a little less than one PowerPC-bogomip per MHz.)

    3. Re:Marketing by Ed+Avis · · Score: 1

      I know what bogomips are for, my point is that on a two-CPU system printing the 'total' number of bogomips by adding the two CPUs together is even more bogus ;-P.

      --
      -- Ed Avis ed@membled.com
  32. OS isn't the issue... by blorg · · Score: 1

    ...the problem is that different tasks are impacted differently by "MHz/FSB/Cache/etc" (not to mention the design of the chip). Obviously performance is also impacted by other system components (memory, hard disk, graphics card, etc). It's impossible to come up with a "single common rating #" - rating for what, exactly? There are however various benchmarks dedicated to measuring performance over 'typical' tasks (e.g. business applications, graphics, Quake III framecount, etc.)

  33. Re:Hmm Here come the Germans by gosand · · Score: 1
    For some reason I don't think it is a coincidence that intel basically stole BMWs numbering scheme...

    heh. If they did, the suits from BMW will be on them like Ballmer on a ham sandwich. They don't like *anyone* messing with their branding. You can't even use the three M colors together in anything BMW related unless you have special permission. The Germans don't play around.

    88 ///M3 owner

    --

    My beliefs do not require that you agree with them.

  34. Summary of Article (Tom's Style) by Anonymous Coward · · Score: 1, Insightful

    In this article
    [fullscreen ad]
    click for more...

    we review the
    [fullscreen ad]
    click for more...

    new Pentium M
    [fullscreen ad]
    click for more...

    processor.
    [fullscreen ad]
    click for more...

    (ad nauseum)

  35. Re:FSB @ 100 MHz ? by jeffgeno · · Score: 1

    "Quad-pumped?" Is that anything like a DVDA?

  36. Digital Restrictions Management in Dothan by Adam+J.+Richter · · Score: 4, Interesting
    From The Register:
    The dark side of Dothan also reared its head, briefly. When asked why Intel was introducing a new naming scheme for Centrino, Chandrasekhar replied that the numbers represented more of "goodness measure" and reflected features that were not necessarily "performance enhancing", such as Le Grande. Le Grande is Intel's contribution to TCPA-compliant lock-down computing,and allows large media companies to impair the user's ability to exchange media files, such as their favorite songs. So you can see why Le Grande isn't "performance enhancing", and quite the reverse. [...]

    Here is a link about TCPA as a threat to free software.

    Slowing TCPA adoption is enough of an benefit to me to prefer a TCPA-free processor even if it costs $50 more for the same performance. I just hope I'll have that option for a while, as Intel is not the only company promoting TCPA.

  37. Re:FSB @ 100 MHz ? by javiercero · · Score: 1

    Huh.... no. The cache is made of SRAM, which is very very very very expensive in terms of power consumption and dissipation.

    The power dissipated due to board level capacitance/inductance issues could be almost marginal in comparison to the power needed to feed the 2MB on chip cache. Those millions of transistors are quite a sink since they are on most of the time (I am sure they may use some power saving design and the cache may be implemented using dynamic ram even).

    The cache is there to increase performance due to the overall lower clock speed. Which makes sense because the latest PIV parts have been rather unbalanced machines.

  38. Re:FSB @ 100 MHz ? by stephenry · · Score: 1

    The cache probably may well be made out of Static RAM, however, it almost certainly would not be made out of dynamic RAM as that requires additional power to refresh whereas the SRAM does not. The only form of power consumption in SRAM, except for switching, is the static power dissipation resulting from leakage current. This is usually very small (however it may be quite significant on a 90nm process). It's pretty unlikely that the reason to include a 2MB cache was based solely on the need for increased performance, although it would undoubtably have an impact on it. If it was, all of Intel's top of the line Microprocessors would have them.

  39. Re:FSB @ 100 MHz ? by bhtooefr · · Score: 1

    Only 1/32 of the cache needs to be powered on a Banias, and just from the die photos of Dothan, my guess is that only 1/64 needs to be powered.

  40. Looks OK from here by Russellkhan · · Score: 1

    Never noticed that effect before. I knew Tom's split articles up into too many pages, but I've never seen the ads.

    Try looking at it my way: with Privoxy (Cross-platform, Open Source) or Proxomitron (Windows only, freeware).

    --
    Information doesn't want to be anthropomorphized anymore.
  41. Re:'cut me up' by welsh+git · · Score: 1

    > Is this the same as the expression "to cut (someone) off" that we have in the US?

    If you explained what "to cut (someone) off " meant, then we'd be able to answer!

    To cut someone off would be the phone company or the electricity company or the cable company cutting off someones supply.

    In the Uk , to "cut someone up" means a car driver pulling sharply in front of someone.. I assume from context this is what yours means ? :)

    --
    Sig out of date
  42. better than a hairdryer. by Stephen+Samuel · · Score: 1
    Back in the early '80s some friends of mine started a company called Myrias. The company was originally intending to build a massively parallel supercomputer (their original aim was a 4K processer (68020) cluster). I used to joke that they'd be able to calculate realtime holograms and pop popcorn at the same time.

    Now you can do the same thing with a 4 CPU system.

    --
    Free Software: Like love, it grows best when given away.