Slashdot Mirror


First Look At Intel Tejas & Socket 775

Anonymous Indian writes "The snoops at Anandtech have unearthed some details and photos of Intel's rumored Tejas 90nm CPU which draws 150 watts of power, a 50% jump compared to Prescott. It's also got an interesting locking mechanism instead of the traditional metal clip from hell for most processors." There's not much info beyond the photos, but it's still interesting to see what lies ahead for Intel.

137 of 224 comments (clear)

  1. And this is interesting why? by Pike · · Score: 1, Insightful

    The both the slashdot sum-up and the linked article fail to explain why, exactly, I should be excited about this.

    1. Re:And this is interesting why? by alexatrit · · Score: 1, Insightful

      My thoughts exactly. The power consumption is the only thing going for it, and it's understated as it is.

      AnandTech: Here are pictures of the thing. We don't have a clue what it does or how it works, but we presume that the extra power draw is part of Bush's Mars Mission.

      --

      Nothing but the finest in meaningless drivel
    2. Re:And this is interesting why? by Ralph+Wiggam · · Score: 3, Insightful

      The dominant CPU maker is releasing a new CPU and a new socket. That's news for nerds. It may not be the most exciting news for nerds ever, but it's still news. The 150 Watt consumption is somewhat interesting.

      -B

    3. Re:And this is interesting why? by dslbrian · · Score: 4, Funny

      The 150 Watt consumption is somewhat interesting.

      The real question is how freaking big and loud is the cooler that has to deal with 150W???

    4. Re:And this is interesting why? by hxnwix · · Score: 1

      Well then, please cease diluting the fine discourse of those who do care about next generation technology.

    5. Re:And this is interesting why? by squiggleslash · · Score: 4, Funny
      Basicly people are excited because it's a hotter CPU than anything for desktops that's come out before it (which is explained in the write up.) As you probably know, a primary concern of many is that their PC just doesn't pump out enough heat. This is why people overclock their CPUs: while you may be able to get a couple more FPS out of Quake by overclocking 5-10%, that's like nothing compared to the extraordinary heat generated.

      This is also at the heart of Apple's G5 strategy. When they announced the PowerMac G5 had 9 fans, many people were enormously excited because it appeared to mean Apple had found a way of making their machines pump out more heat than any PCs around today. As it turned out, Apple was fooling everyone! Their fans are very slow and they use careful venting to make it appear the machine needs more fans than anything else.

      That's why you should be excited by this...

      --
      You are not alone. This is not normal. None of this is normal.
    6. Re:And this is interesting why? by toasted_calamari · · Score: 1

      Not loud at all, you have to use one of these

    7. Re:And this is interesting why? by bhtooefr · · Score: 2, Insightful

      Intel desktop and server processors. Have you looked at the Pentium M lately? By the way, I didn't think the Athlons ran exactly cool (cooler than any Intel desktop CPU nowadays, though).

    8. Re:And this is interesting why? by topham · · Score: 1


      While trollish, I'll take my G5 with 9 fans over my PC with 3, it's a fraction as loud under normal usage.

      Actually, I can barely hear it at all.

    9. Re:And this is interesting why? by inode_buddha · · Score: 1

      Well, I have a dual Pentium Pro box, each cpu draws about 70 watts IIRC. The machine has *no* fans, just the original aluminum heatsinks. No heat problems yet, but all the same I'd like to find one of these to fit PPro. Just in case, because they have 1 MB cache on them and it wasn't easy to find this machine. In case anyone is wondering, Linux *ROCKS* on it. Anyone know where to find coolers like that for Pro?

      --
      C|N>K
    10. Re:And this is interesting why? by timeOday · · Score: 1

      I think it's a time for a little steam turbine on top of the CPU to pump electricity back to the motherboard's power lead - "Regenerative Computing" :)

  2. That latching thing... by wampus · · Score: 2, Funny

    Looks like a German dental tool.

    1. Re:That latching thing... by chunkwhite86 · · Score: 4, Informative

      Looks like a German dental tool.

      Actually, the Germans are known for their skill with machined parts and their engineering prowess.

      This looks more like a dental tool from .

      Now before you mod me a troll for bringing up Soviet Russia again, let me teach you something. In Soviet Russia, a manufacturing facility's productivity was measured not by the number of units sold, or by customer satisfaction. It was measured by the quantity of raw materials used. The problem with this, is that quality immediately goes down the toilet, and raw material consumption goes through the roof. A soviet era farm tractor, was so unreliable - but contained so much steel - that when Jonev Vladstov (That's John Doe in Russian) bought a tractor, it was worth MORE if he melted it down and sold the steel than it was as a tractor! That's called 'negative value-add' in the economic world, and that's why those old 'In Soviet Russia...' jokes use role reversal as their humor mechanism - because Soviet Russia really was backwards.

      Intel Tejas. There. Now this post is not off topic.

      --
      I'd rather be a conservative nutjob than a liberal with no nuts and no job.
    2. Re:That latching thing... by Fiveeight · · Score: 1

      Is it safe?

    3. Re:That latching thing... by SomeGuyFromCA · · Score: 1

      it puts the conductive paste on the cpu and puts it in the socket...

      --
      if the answer isn't violence, neither is your silence / freedom of expression doesn't make it alright
    4. Re:That latching thing... by JPriest · · Score: 1

      From 130nm to 90nm is a 30% reduction, usually requiring less power. I am suprised that this thing uses 50% more power.

      --
      Saying Java is nice because it works on all OS's is like saying that anal sex is nice because it works on all genders.
  3. So much power by Anonymous Coward · · Score: 5, Interesting

    So much power use from the first 90nm process cpu? There can be only one reason for this... the rumored twin core intel CPU. Let's see how hyperthreading run's on THIS baby!

    1. Re:So much power by Anonymous Coward · · Score: 1

      you mean the second? G5...

    2. Re:So much power by Jesrad · · Score: 1

      I don't see why Intel wouldn't try to benefit from both technologies and offer a processor with HyperThreading AND dual-Core.

      --
      Maybe we deserve this world ?
    3. Re:So much power by Hoser+McMoose · · Score: 1

      The G5 at 90nm may have been announced by Apple, but it's not yet shipping from what I can tell. Given traditional Apple schedules, it will actually start shipping in one to two months time, or roughly the same time frame as Intel starts shipping their first 90nm processor (Prescott, not Tejas, that's a future chip not scheduled for 9+ months).

    4. Re:So much power by BitchKapoor · · Score: 2, Insightful
      To say that SMT is "IBM's version of HyperThreading" is backwards-speak. Simultaneous MultiThreading (SMT) is actually a generic term for any technique which multiplexes threads on a single CPU core by allowing any thread to use any available resource during any clock cycle. HyperThreading (nee Jackson Technology) is Intel's name for their implementation of SMT -- see http://www.intel.com/technology/hyperthread/.

      From my description of SMT, you might conclude that it is able to make the fullest use of CPU core resources of any multithreading approach. Then why are some benchmarks slower with HyperThreading enabled? I have three ideas. First of all, since OSes tend to treat a single CPU with hyperthreading like a dual-CPU system, they may schedule lower-priority processes on one of the threads, resulting in less total CPU time available for the main benchmark process. Secondly, there is some overhead to multiplexing all of these resources amongst two threads, so perhaps Intel had to "tone down" superscalar operation or memory queues to meet clock timing when in HyperThreading mode (I know this is vague). Thirdly, perhaps in ostensibly non-HyperThreaded mode the second thread is actually used for speculative execution, and hence slightly boosts the performance of a benchmark.

    5. Re:So much power by javiercero · · Score: 2, Interesting

      Actually you too are wrong: Hyperthreading allows multihple threads to be active per functional unit. I.e. you could have 2 threads active in a single functional unit, you do not need 2 functional units to have 2 threads active. Hyperthreading does not imply parallel execution as you implied before, it basically means that as soon as one thread stalls in a functional unit there must be at least another thread that can be brought right away to keep that functional unit from stalling.

    6. Re:So much power by nek · · Score: 1

      Well, the new XServe G5 will be using 90nm process:

      "IBM uses a 90-nanometer process to produce the PowerPC G5"

      Taken from the XServe G5 technical specification: Xserve G5 Tech Overview

    7. Re:So much power by matfud · · Score: 1

      Curernt Hyperthreading has limitations. Namley the two "Threads" share the instruction and data cache. The caches have no idea of the SMT happening in the rest of the core. How is this relevent. Well, if your two threads (intel SMT only supports 2 at the moment) have very different memory usage profiles they can cause the caches to thrash as they try to cache what is effectivly random memory accesses. This can really hurt performance. If however the two threads have similar, or complementary, memory access patterns then SMT can work well.

      SMT is about reducing the wasted execution opertunities that current CPU's (with thier multiple execution units) have. It can improve performance (by up to about 30% in benchmarks) but it can also hinder performance. More refined versions may reduce the instances in which poor performance can result.

      matfud

    8. Re:So much power by BitchKapoor · · Score: 1

      Ah, that's it, cache sharing. There you go, that nicely fleshes out problem #2.

    9. Re:So much power by be-fan · · Score: 1

      Mod parent as stupid.

      Hyperthreading and dual core are two different things. Hyperthreading places no extra execution units within the same CPU core. It simply splits up the existing execution resources, and adds a few extra registers to keep track of two instruction pointers, etc. It works quite well, for what it does, which is to make better use of a CPU's existing parallel resources. It is not a replacement for SMP.

      --
      A deep unwavering belief is a sure sign you're missing something...
    10. Re:So much power by Hoser+McMoose · · Score: 1

      Err, not if you're me apparently. I just went through the Apple Online store and brought up the configuration of a bog-standard dual-processor G5. Estimated shipping time, 6-8 weeks. Tried again for the single-processor system with the same result, 6-8 weeks.

  4. Re:150 Watts? by Anonymous Coward · · Score: 1, Funny

    Using a car battery for your laptop?

  5. Different locking mechanism? by Phosphor3k · · Score: 1

    Looks like the same arm I've seen on every other socket motherboard.

    1. Re:Different locking mechanism? by CaptBubba · · Score: 2, Interesting
      It looks to me that the CPU itself doesn't have any pins on it, just little gold contacts. The motherboard has all the pins on it, and the locking mechanism is really just something to hold the CPU down against those pins to make contact.

      That's quite different from the standard today where the arm locks the pins of the cpu into the socket of the motherboard. I could be wrong, but that's what I think I see in the pictures.

  6. 150 watts of? by dave1g · · Score: 2, Interesting

    Is that dissipated heat?

    If so, ouch! that system is gonna need a huge heat sink or water cooling will be needed.

    I would loves to see standardized water cooling parts for computers.

    That way all you would need to do is buy the hosess and your favorite pump.

    1. Re:150 watts of? by Jesrad · · Score: 1

      Bah, that's nothing. My 2.4 GHz microwave oven can dissipate up to 1000 Watts !

      --
      Maybe we deserve this world ?
    2. Re:150 watts of? by Waffle+Iron · · Score: 4, Funny
      Bah, that's nothing. My 2.4 GHz microwave oven can dissipate up to 1000 Watts !

      But microwaves use expensive and smelly burrito heat sinks that only have a service life of about 5 minutes each.

    3. Re:150 watts of? by temojen · · Score: 1
      Is that dissipated heat?

      Unless it emits light, moves rocks uphill, accellerates your car, or charges one whopper of a battery, yes, that's dissipated heat.

      It's called conservation of energy. Here is a NASA page explaining it. It's not rocket science, but it is important to understand for building rockets, bicycling, etc. Along with Conservation of Mass and Conservation of Momentum it forms the trilateral commission that keeps us all from building those "free energy" devices I keep hearing about.

    4. Re:150 watts of? by cabp5 · · Score: 1

      Water cooling is a great idea...until the "enthusiasts" start tinkering and shorting out our their equipment with the spillage

    5. Re:150 watts of? by dave1g · · Score: 1

      I figured that the act of computation requires energy, and the heat given off is just a side effect of our inefficient means of performing computation....

      Isnt there work being done than giving off heat???

    6. Re:150 watts of? by temojen · · Score: 1
      Isnt there work being done than giving off heat???

      Changing the state of a bit is work. Changing it back is un-work. This results in zero net work. At least that's how my professor explained it.

    7. Re:150 watts of? by tomstdenis · · Score: 1

      I'm not an EE but this explanation seemed decent to me.

      A 2,1 gate [say NAND] takes two inputs of X1 and X2 current [and V1/V2 voltage] and emits a single output that is less than the sum x1+x2, v1+v2. That energy has to go somewhere.

      Also transistors work by pushing current "uphill" [e.g. resist current until switched on] the act of forcing electrons "against the grain" [e.g. against a negative base] generates heat much like a stove element generates heat.

      Tom

      --
      Someday, I'll have a real sig.
    8. Re:150 watts of? by Mesaeus · · Score: 1

      Yes, but burritos are biodegradable. Well, MOST of them anyway (yuck)

    9. Re:150 watts of? by jrockway · · Score: 1

      > That energy has to go somewhere.

      Umm, maybe to ground?

      --
      My other car is first.
    10. Re:150 watts of? by tomstdenis · · Score: 1

      Maybe to activate the switch? Transfer Resistor.... transistor...

      hmm.. Could catch on.

      --
      Someday, I'll have a real sig.
  7. tradition by awing0 · · Score: 5, Interesting

    Regarding those metal clips from hell, I've always wondered why Intel and AMD never followed the examples of Alpha, Sun, SGI, etc. machines. They usually have bolt on heatsinks that either bolt the CPU and heatsink together or sandwich the CPU between the mainboard and the heatsink.

    It takes care of the flat head screwdriver ruined mainboard, and there are no clips to break off the socket itself. I'm glad they are finally changing the way x86 sockets work.

    --
    Cthulhu Saves.
    1. Re:tradition by chunkwhite86 · · Score: 1

      Regarding those metal clips from hell, I've always wondered why Intel and AMD never followed the examples of Alpha, Sun, SGI, etc. machines. They usually have bolt on heatsinks that either bolt the CPU and heatsink together or sandwich the CPU between the mainboard and the heatsink. It takes care of the flat head screwdriver ruined mainboard, and there are no clips to break off the socket itself. I'm glad they are finally changing the way x86 sockets work.

      Because some l33t h4x0r would over torque the nuts and damage the cpu, then try to return it to the store. On Alphas, it only about two ft/lbs of torque that you're supposed to tighten them

      And yes, many Alphas DO have an annoying clip mechanism to hold the heatsink on. It is used in addition to the bolts, so that the full weight of the heatsink is not pulling the cpu out of the socket.

      --
      I'd rather be a conservative nutjob than a liberal with no nuts and no job.
    2. Re:tradition by Hoser+McMoose · · Score: 1

      Err, have you ever installed a P4 or an Athlon64/Opteron processor? They DO use locking mechanisms to sandwich the processor between the motherboard and the heatsink!

    3. Re:tradition by Pidder · · Score: 1

      Newer motherboards have 4 holes around the socket where you can bolt heavier heatsinks (e.g the thermalright slk series).

    4. Re:tradition by iMMo · · Score: 3, Informative

      The interconnect used for Sun's USIII CPU is not something that anyone at home would be able to deal with. There's a socket, the CPU, then a mylar pad, then a bolster plate and then the heatsink. The heatsink needs to be torqued down with a precise amount of pressure, or else the CPU can't make electrical connection through the socket to the board.

      You need some relatively major tooling just to get it connected in just the right way, with no carpet fibers or hair in between the CPU/Socket or Socket/Baseboard. These interconnects are not for the average user -- so -- can you imagine having to return your PC motherboard to the factory to have your CPU swapped/upgraded?

    5. Re:tradition by the_2nd_coming · · Score: 1

      they use the LUGs not the holes...at least standard they do, you an buy more expensive heatsinks that will bolt on,.

      --



      I am the Alpha and the Omega-3
    6. Re:tradition by Sigl · · Score: 1

      Along the same line of thinking, why can't there be additional heatsink on the back? Does heat magically radiate only up from the processor? Right now the air on the back side is insulated by the board. Why not run a small heatpipe through the board and radiate some in the cool air underneath? This would especially be useful for processors that sit vertically.

    7. Re:tradition by ColaMan · · Score: 1

      Why not run a small heatpipe through the board and radiate some in the cool air underneath?

      Its a problem when you've got 500 connections to a CPU going through a 5 layer circuit board to all parts of your motherboard - there's no physical (or electrical!) way to clear some space for a heat pipe under the cpu. yet. :-)

      --

      You are in a twisty maze of processor lines, all alike.
      There is a lot of hype here.
    8. Re:tradition by Zugok · · Score: 1

      i was looking for some heavy duty heatsinks and I found some and then looked at the installation of some of them on my motherboard. One of ths sites say that all the P4 motherboards have the option of a bolt down heatsink because of the way the standard heatsink/fan bracket is designed. Only a few of the Althon motherboards do however.

      --
      "I just can't sit while people are saying nonsense in a meeting without saying it's nonsense" J Watson, Sci Am 288:(4)51
  8. Interesting by chunkwhite86 · · Score: 4, Interesting

    It's interesting to see how what was once consideres "high-end" eventually makes it's way into the consumer desktop stuff.

    For example, this chip appears to use a "pinless" package design. Instead of little pins that fit into the socket, it has little ever-so-slightly raised 'nubs'. These 'nubs' simply sit on top of contact points in the socket.

    This pinless design was being used by Compaq for the Alpha CPU as early as 2000, so this isn't a new packaging technology. The only problem Compaq had with it, was keeping all the little 'nubs' firmly in contact with the corresponding points on the socket. They used some sort of plastic clip design at first, which ended up with a high failure rate - not for the CPU, but for the plastic retaining mechanism itself. That's probably why intel is using that big beefy metal retaining clip.

    I guess intel learned something from all those Compaq Alpha engineers they bought a few years ago.

    --
    I'd rather be a conservative nutjob than a liberal with no nuts and no job.
    1. Re:Interesting by ImpTech · · Score: 1

      Unless I'm way off, this is what's called a Ball-Grid Array, and yeah, the concept's been around forever. My understanding is that its really hard to mass manufacture these things so that everything lines up and has good connection. And afaik, they're of dubious benefit versus the pin socket we all know and love. Maybe someone who knows more can explain what benefit this new socket has for Intel?

    2. Re:Interesting by temojen · · Score: 1

      No bent pins.

    3. Re:Interesting by WuphonsReach · · Score: 1

      Lower vertical profile? (might be useful for things like laptop motherboards)

      --
      Wolde you bothe eate your cake, and have your cake?
    4. Re:Interesting by CaptKilljoy · · Score: 2, Informative

      Unless I'm way off, this is what's called a Ball-Grid Array, ...

      You're way off, I think. Ball Grid Array refers to a IC form factor that has a grid of contacts on the bottom of the chip carrier. Each of these contacts is pre-filled with a small ball of solder. BGA devices aren't meant to be socketed.

    5. Re:Interesting by chunkwhite86 · · Score: 1

      Bending pins on a CPU can only result from carelessness.

      Right, because we all know no one has ever bent a pin. Because there are no careless computer users.

      lack of pins sounds like a good benefit, particularly for mom-n-pop computer stores who accept returned cpu's from l33t h4xors who try and DIY.

      --
      I'd rather be a conservative nutjob than a liberal with no nuts and no job.
    6. Re:Interesting by gordlea · · Score: 1

      I'd like to see the battery life for a laptop using a processor that draws 150watts...

      --

      Choose yer poison: Prophets or Profits

    7. Re:Interesting by Nogami_Saeko · · Score: 1

      I just installed a new CPU in this new machine - As I was carefully inserting the CPU into the socket, I felt a bit of resistance and had a closer look - 6 pins in two adjacent interior rows were bent marginally.

      Given the location of the bent pins, it's very likely it came that way (it was a retail-boxed CPU) - in any event, it's not always carelessness...

      A few tense moments with a fine exacto-knife blade and some exceeedingly gentle pressure straightened them out and allowed 'em to fit right in so I can type to you now!

      N.

      --
      "Nothing strengthens authority so much as silence." - Charles de Gaulle
    8. Re:Interesting by tiger99 · · Score: 1
      Quite correct, and they are almost impossible to inspect to see that the joints are of good quality, and even more impossible to replace if they fail. IMHO an utterly diabolical invention, sadly (mis)used on a lot of motherboard chipsets and other things.

      The safety-critical industries dread the fact that many new CPUs (microcontrollers etc) and other vital things are now only available in BGA.

      It is actually possible to get a device removed from the board, re-balled with solder, and replaced, but after all that thermal stress it would be sheer folly to trust either the PCB or the chip again.

      Personally, I HATE the things.

      They can be, and are, sometimes socketed for development purposes, but the configuration breaks every rule of reliable connector design. Sensible people get them fitted to a little PCB, with pins attached, so you can plug them in, and throw away the chip and PCB if it breaks, but you can't always do that on high-frequency things.

    9. Re:Interesting by tiger99 · · Score: 1
      It is hard to tell from the photos, but if they are really not pins, I anticipate very serious reliability problems indeed. The design of connectors is not something where you can take liberties, and discard all previously discovered solutions to problems. The existing ZIF sockets are designed to positively clamp against the sides of the pin when you actuate the locking lever, the fixed contact as it engages also slides slightly along the pin with a wiping action. All of that is vital, and it is very necessary to get a gas-tight fit at the interface, or corrosion, or cantamination, makes it fail very quickly indeed.

      I have seen a number of attempts over the years to make face to face contact, and all have been extremely unreliable. One design of "butting connector" had to have both sets of contacts, which were gold-plated, and one half sprung, cleaned very carefully with iso-propyl alcohol immediately before making the connection, and even then there were numerous problems. The bayonet lampholder is a face to face connector, it is tolerable (just) because the hard brass spring-loaded pins tend to dig in to the soft solder pads on the base of the lamp.

      End contact on a large array of tiny pins is a disaster waiting to happen. I suspect an optical illusion, and that they are conventional pins, if not, Intel are shooting themselves in the foot, as they have done on several occasions with packages which broke the rules of good mechanical engineering. I seem to remember certain "slot" rather than "socket" packages, where the whole thing was unstable and would break the processor, or the motherboard, if you were not extremely careful when transporting the PC.

      There again, the mechanical design of the PC is itself a disaster. I guess that IBM must have let Bill have influence there as well as on the software.... Single-part edge connectors are banned in many applications because they are inherently unreliable, in any case it is cheaper to use a 2-part connector because only the vital part of the pins is gold-plated and you don't have to put the PCBs through the gold plating process. Tolerance build-ups between back of case, where the PCB bracket mounts, and the motherboard are still a nightmare.

      As for the truly vile arrangement of PS2, serial, USB etc ports on a modern ATX board, well that is a disaster waiting to happen. I wonder how many people have trashed their motherboard by tripping over a cable?

      The industry is based on producing junk, which is just as well for Bill because the hardware does not embarass his software too much!

  9. Re:150 Watts? by mumblestheclown · · Score: 5, Interesting
    I'm sure they're thinking that this is the best way for them to get ahead in the market to deliver the cpu power that tomorrow's demanding consumer requires. they don't make any money by "wasting" electricity so I'm sure they are doing what in their best interest is a good mix of engineering and catering to consumer needs. so tomorrow's cpu uses 150w. two light bulbs. jet engines use more fuel than reciprocating ones. if they made a bad technological bet, the marketplace will punish them. life moves on.

    the "lack of metal clip", however, is so exciting that I am sure that we are to see perpetual world peace from its announcement any minute now.

  10. Take a closer look... by orthancstone · · Score: 1

    In the pictures, it locks down on the left side and the mechanism covers the CPU. Not many sockets I've seen in the past few years use that, so you must be seeing some special motherboards...

  11. That much power? by SexyKellyOsbourne · · Score: 1, Informative

    The thing I always liked about Intel chips was their low power absorption and their low heat. Though they're a bit pricy in comparison, AMD chips were power-hungry and thus produced heat as if they had uranium cores.

    Intel chips were great for Mini-ITX cube PCs if you didn't want them to burn, as they ran cool enough to easily run with heat pipe technology. They were even better for laptops, since you didn't have them draining the battery like crazy . On the regular PC front, they would famously run cool overclocked to extremes, like from 1.6-2.4 or from 2.2-3.0 on cheap stock cooling alone.

    Now, it seems like they've lost that advantage.

    1. Re:That much power? by Saven+Marek · · Score: 2, Interesting

      > Now, it seems like they've lost that advantage.

      I think that depends what the tradeoff is. Why would the cpu have such a high power output from a 90nm process? Either it's a huge cache inside, or there is more than one core. That sounds exciting. It may be very hot running but it has the advantage of space.

      Just which applications this finds use in, I cannot tell.

      a macgirl web cam. live 24/7

    2. Re:That much power? by Dun+Malg · · Score: 2
      The thing I always liked about Intel chips was their low power absorption and their low heat. Though they're a bit pricy in comparison, AMD chips were power-hungry and thus produced heat as if they had uranium cores. Intel chips were great for Mini-ITX cube PCs if you didn't want them to burn, as they ran cool enough to easily run with heat pipe technology. They were even better for laptops, since you didn't have them draining the battery like crazy . On the regular PC front, they would famously run cool overclocked to extremes, like from 1.6-2.4 or from 2.2-3.0 on cheap stock cooling alone. Now, it seems like they've lost that advantage.

      Hello, McFly! It may be hard for the Intel fanboiss\ to accept, but the Pentium IV has held the title of "hottest/most power hungry chip" for about a year. They "lost that advantage" (?) quite some time ago when they decided the mHz number was more important than efficiency. Besides, if you want actually cool chips, get a VIA/Cyrix.

      --
      If a job's not worth doing, it's not worth doing right.
    3. Re:That much power? by Viceice · · Score: 3, Interesting

      RTFA! Oh wait, this is SlahsDot.

      Realize that this processor is a Prototype, fabbed on a process that doesn't reflect Intels true capabilities. So criticism as to it's heat dissipation is at best pre-mature and at worst, downright off topic.

      --
      Sometimes I wish I was a plumber, then I'd know how to deal with other people's shit.
    4. Re:That much power? by Hoser+McMoose · · Score: 3, Informative

      Err, have you bothered to check the datasheets? The P4 is hardly a cool running chip by any stretch. It may have a wider power consumption range than the AthlonXP, but when the chips were running flat out they sure didn't run any cooler.

      The only reason why P4's used to run cooler than Athlons was because people would stick a 60mm x 60mm heatsink on their Athlon and an 80mm x 80mm heatsink on their P4. Both of these chips consume a lot of power, and both drain laptop batteries like crazy if you use the highest powered parts (Intel actually produces some P4 "mobile" chips with a TDP of 70W!, while AMD's brand new "mobile" Athlon64 chips consume over 60W at full throttle).

      Now, the Pentium M... well that's another story altogether.

    5. Re:That much power? by the_2nd_coming · · Score: 1

      The thing I always liked about Intel chips was their low power absorption and their low heat.

      as compared to what? a steel mill?

      --



      I am the Alpha and the Omega-3
    6. Re:That much power? by bhtooefr · · Score: 1

      If you actually want cool chips, get a VIA C3. If you want somewhat cool chips that perform, get a Pentium M and a PowerLeap adaptor.

  12. AMD Ad by Unregistered · · Score: 1

    Did anyone else get an Athlon 64 ad when they opened the article?

    1. Re:AMD Ad by herrvinny · · Score: 1

      I got one. Bet Intel wishes they were sponsoring that site....

  13. Tejas? by jotaeleemeese · · Score: 3, Informative

    He, that is the Spanish (as written on Spain) for Texas.

    The Spaniards write Mexico as Mejico.

    Just a tidbit for your amusement.

    --
    IANAL but write like a drunk one.
    1. Re:Tejas? by patanish · · Score: 3, Informative

      Thats sanskrit for luminance

    2. Re:Tejas? by Jeff+DeMaagd · · Score: 1

      I thought Intel was using the names of rivers for their code names but it looks like it might not be true for Tejas.

    3. Re:Tejas? by vegetasaiyajin · · Score: 1

      He, that is the Spanish (as written on Spain) for Texas.
      Tejas is the spanish for "tiles", the things used to cover a roof.
      The Spaniards write Mexico as Mejico.
      In spanish I always see Mexico written as Mexico. It is pronounced Mejico, though.
      I always see the U.S. Texas state written as Texas. Sometimes people pronounce it with a "J" (especially in Mexico) as is Tejas and sometimes with an "X" as in Texas.

      --

      My heart is pure, but make no mistake, it's pure evil
  14. Re:150 Watts? by Waffle+Iron · · Score: 4, Informative
    so tomorrow's cpu uses 150w. two light bulbs.

    That's about $10/month; similar to the cost of adding a movie channel or two to your cable subscription. Something to think about for those that will use this CPU to volunteer for distributed number crunching projects.

  15. why it is so .. dirty? by Janek+Kozicki · · Score: 2, Interesting

    that's my first thought when I saw the picture. Why the CPU is looking so... um.. dirty (looks like corrosion)?
    I'm not an overclocker specialist, or whatever, and really I don't want to be a troll here ;) I'm just curious.

    perhaps answer to this question will be modded informative (as this question itself is not ;)

    --
    #
    #\ @ ? Colonize Mars
    #
    1. Re:why it is so .. dirty? by Roguelazer · · Score: 1

      Probably gunk from removing the heatsink, no? I don't think they were in a big hurry to clear off the name of the inside informant...

    2. Re:why it is so .. dirty? by hxnwix · · Score: 2, Informative

      thermal compound such as artic silver. ye olde white paste breaks down, dries out and leaves an insulating crust behind at high temps

    3. Re:why it is so .. dirty? by Dun+Malg · · Score: 3, Informative
      Why the CPU is looking so... um.. dirty (looks like corrosion)?

      Probably just thermal conduction compound residue. You know, that white zinc-based greasy crap that goes between the heatsink and the chip die.

      --
      If a job's not worth doing, it's not worth doing right.
  16. Re:150 Watts? by aldoman · · Score: 2, Insightful

    The only trouble is that power usage inside computers is skyrocketing lately.

    150 Watts CPU, a 100 Watts graphics card and then you have all the fans to cool it, and you'll be lucky if you are under 300W and that doesn't even factor in storage, optical drives, motherboard, audio, communication devices.

    I mean you will need a hell of a power supply unit to provide that sort of power reliably.

  17. But... by NeoGeo64 · · Score: 2, Interesting

    Is it going to be 64-bit? If not, a new chip is really a waste of money.

    By 2006, most everything will be 64-bit, and Intel needs to realize this and stop making 64-bit chips just for the server market.

    1. Re:But... by The+One+KEA · · Score: 1

      Maybe new systems, corporations, scientists and enthusiasts will all be using 64-bit, but I think the vast installed base of 32-bit hardware and software will guarantee that the platform will not disappear anytime soon.

      --
      SCREW THE ADS! http://adblock.mozdev.org/ Proud user of teh Fox of Fire - Registered Linux User #289618
    2. Re:But... by benzapp · · Score: 1

      I think the poster is referring to all new computers sold in 2006 will be 64-bit, and that is correct.

      Many offices were using 286 and 8086 machines well into the early to mid 90's. That doesn't mean those machines were actually sold in 1994.

      --
      I don't read or respond to AC posts
  18. Hyperthreading? No... by UPAAntilles · · Score: 1

    Well, maybe, there have been problems with heat at the 90nm mark, though there are rumors that Tejas might use dual cores However, the hyperthreading thing would be a bad idea. It would mean that you have 4 logical processors, instead of 2. And with 4 threads running in parallel, there's a good chance that a lot of the time, 2 of those 4 will be identical. All hyperthreading would generate is too much heat.

  19. How much power? by NeoGeo64 · · Score: 5, Funny

    1.21 jiggawatts?! Where am I going to find that much power!? It's not possible!!

    1. Re:How much power? by Anonymous Coward · · Score: 1, Funny

      1.21 jiggawatts?! Where am I going to find that much power!?

      News flash: Intel to introduce new line of power supplies for use with high wattage radio transmitters, university particle accelerators, electric subway systems, and Tajos core CPUs.

    2. Re:How much power? by synaptik · · Score: 2, Funny


      >1.21 jiggawatts?! Where am I going to find that much power!?

      News flash: Intel to introduce new line of power supplies for use with high wattage radio transmitters, university particle accelerators, electric subway systems, and Tajos core CPUs.

      But not flux capacitors? Damn... er, I mean, great scott!
      --
      HSJ$$*&#^!#+++ATH0
      NO CARRIER
  20. Re:150 Watts? by October_30th · · Score: 1

    Well, as you said, distributed projects are about volunteering. I wouldn't say $10/month for a project like Folding-at-Home is a waste.

    --
    The owls are not what they seem
  21. How will they differentiate business ($$$) cpus by j_dot_bomb · · Score: 1

    If this is multi core, which I would love. How are they going to find a market segment to charge high prices for ? I want dual. But with either intel or amd its expensive. In AMD's case the board. Intel now at least has hyperthreading which should fix some sluggishness I feel. So what will differentiate a xenon if there are duals in a core ? An extra processor will mean alot less to a workstation or some servers if there is already a dual in a single core.

    1. Re:How will they differentiate business ($$$) cpus by j_dot_bomb · · Score: 1

      Actually I have ECC REG for my single CPU. But not dual channel. A GIG no less. AMD dual boards are min $190 alone. Doing dual CPU without soldering hacks is expensive. Id prefer not to even use the nondual marked AMD's because of stability concerns.

  22. Re:150 Watts? by JustKidding · · Score: 2, Interesting

    ehm, it would actually drain quite quickly. A regular car battery supplies about 14.4 volts when it's full (even though it's called a 12 volt battery), and can supply about 40 Ah, when drained slowly (less when drained quickly). 150 / 14.4 ~= 10 Amps, so it would last only 4 hours, and that's only for the CPU.

  23. Re:Sorry to burst your bubble! by Anonymous Coward · · Score: 1, Interesting
  24. 150 W? by diegocgteleline.es · · Score: 4, Funny

    150 W? What does that Tejas thing have, a Electric Chair Inside ?

    1. Re:150 W? by MonkeyBoy · · Score: 1

      I think they're going to team up with McDonalds and use all that heat to put The McGriddle Inside.

      --

      Moof!

  25. Re:150 Watts? by Anonymous Coward · · Score: 4, Insightful

    You only seem to equate power wastage in terms of your personal $s.

    Try and think a little bit bigger. Think in terms of global warming, energy depletion, entropy etc.

    Using less power (whatever the application) can only be a good thing.

  26. wtf? by autopr0n · · Score: 2, Insightful

    Mod parent as troll.

    Why should it be modded as a 'troll'? just because it betrayed some lack of knowledge about CPU design? I think your post should be modded as troll.

    --
    autopr0n is like, down and stuff.
  27. Dual core hyperthreading...interesting... by UPAAntilles · · Score: 2, Interesting

    I can see them trying, but I don't really see it. Hyperthreading as it is generates a lot of extra heat that isn't exactly proportional to the performance gain (I've noticed this on my own 2.8C). I know there are some dual Xeon systems that have hyperthreading enabled, so they must have partially addressed the problem the inquirer put as...

    "An instruction thread might well favour one kind of execution unit, but who's to say the other threads, the ones that are running in parallel with the 'troublesome' one, won't be favouring the same execution units? Assume you've got two cores per die, and that's two threads running simultaneously. Chuck in Intel's much-touted HyperThreading technology and you'll have four threads being processed in parallel, two per core, all being 'hopped'. Chances are that at least two of them will be using the same execution units, so swapping them round isn't going to change anything."

    But in a dual core system, heat is a much bigger problem, so it would make more sense to up the clock speed over enabling hyperthreading whereas the heat on a dual Xeon would be easier to eliminate because there are 2 processors with their own cooling solutions. There's also the issue that stardard operating systems don't support 4-way systems for residential computing. There's also the fact that Intel COO Paul Otellini said "We'll go from putting HyperThreading in our products to putting dual-core capability in our mainstream client processors over time." That implies that hyperthreading is used as a stepping stone for the consumer, as some programmers optimize their code to be run on hyperthreaded systems, it will also boost performance on future dual core systems. Anyone else have any ideas?

    1. Re:Dual core hyperthreading...interesting... by Dan+Ost · · Score: 1

      There's also the issue that stardard operating systems don't support 4-way systems for residential computing.

      Residential computing? Please explain what you mean by this.

      We've got several machines with dual CPUs with hyperthreading enabled on each
      and Linux (red hat 7.3 with stock kernel) seems to have no trouble handling "4"
      CPUs.

      --

      *sigh* back to work...
    2. Re:Dual core hyperthreading...interesting... by UPAAntilles · · Score: 1

      Computing at home-the average user that buys the preconfiged DELL, HP, Gateway, Apple, and Sony boxes

    3. Re:Dual core hyperthreading...interesting... by bhtooefr · · Score: 1

      Windows XP Home supports (AFAIK) one CPU (HT won't work), XP Pro supports two CPUs (two actual CPUs, or one HT CPU), Server 2003 supports a shitload of CPUs (can be as low as four, however).

    4. Re:Dual core hyperthreading...interesting... by mackkie · · Score: 1

      actually, windows XP home supports 1 physical CPU AND 1 HT CPU but not 2 physical CPU. Windows XP Pro supports 2 physical CPU and will support HT on BOTH for a total of 4 HT. but will not support more than 2 physical CPU. Windows server 2003 will support many more.

  28. Scratch that, after research, it could work... by UPAAntilles · · Score: 1

    I take back this last post, my other post in this thread is better.

  29. Re:It has no pins by Anonymous Coward · · Score: 5, Informative

    And it also may approach the heating capacity of an AMD chip

    No. It's much higher. The 'super-hot Athlon' preconception is way way out of date. It was true comparing Athlon "Thunderbirds" with PIIIs, but hasn't been true since the P4 - the 2GHz Willamette and 3.2GHz Northwoods both dissipate more power than any Athlon.

    The first Pentium was a stripped-together combination of 2 486-like CPUs

    No. You're trying to describe superscalar-ness but not succeeding. The Pentium was the first superscalar x86, having two integer pipelines. All modern x86 CPUs are superscalar (except the VIA C3). Superscalar-ness is not CMT/multicore. CMT appears as multiple CPUs to software.

  30. Opteron runs both 32-bit and 64-bit stuff by Anonymous Coward · · Score: 1, Informative
    What do you mean?

    I've got Mandrake 9.2rc-1 AMD64 on my dual opteron system. It's running legacy 32-bit code (Seti) perfectly fine alongside the newly compiled 64-bit stuff.

  31. Re:It has no pins by Jeff+DeMaagd · · Score: 3, Informative

    The first Pentium was a stripped-together combination of 2 486-like CPUs, with shared parts. To describe this as "two discreet Prescott cores on die... sharing data caches and maybe other units..." brings together this similarity.

    The closest is the Pentium Pro and it had separate cache and core dies, NOT two separate cores, but that was converted later to cartridge (PII) and then later all on one die (later PIII). I doubt that Intel produced an original Pentium anything like this, or in a dual core manner.

  32. I don't believe so... by UPAAntilles · · Score: 1

    What indications are there that everything's going to be 64-bit? It's been 32-bit for a while now, and if Intel doesn't go 64-bit the rest of the world might wait until they do. Intel controls the majority of the market, and they have enough power keep trends suppressed for a while (or try to start new trends, and keep them much more alive than they need to be, *cough* RDRAM *cough*). There's rumors that there will be 64-bit instructions that one could somehow add-on to the prescott, but I don't know how it would work, as Intel is keeping everything tight-lipped.

    Heck, if they wanted, they could create their own 64-bit instructions and pull the rug out from under the AMD64's feet. But I don't see that happening, as everyone would not like that (not just /.ers, but companies like Microsoft which is making a 64-bit version of XP for AMD64).

  33. Going by this, this is the opposite of Apple/IBM by CrackedButter · · Score: 1

    They use less Watts and activily want to reduce this as well plus the cost of owning a mac when comparing electricity consumption would work out cheaper as well.

  34. Turnabout is fair play by sundling · · Score: 1

    I've gone to AMD related articles and gotten Intel adds so many times I got used to it. I'm glad to see the opposite happening. I just hope in Q2 they start advertising so that AMD64 really takes off and 64 bit software will become very mainstream. The reason I say Q2 is that this is when the socket shakedown should be over and the windows alternative will be available.

    Let's face it. Those of us on slashdot are the ones most likely to be utilizing the new 64 bitness. :)

    Paul

  35. Intel already has a 64 bit solution by charnov · · Score: 1

    Intels 64 bit solution is the Itanium (which, along with HP, they invested around $10 Billion). Intel is in a bind in that Microsoft will not tolerate another 64 bit instruction set and Intel would be shooting Itanium in the foot to use AMD64 (which they are a licensee of).

    Maybe they will support AMD64 or maybe they wont, but they will not create ANOTHER 64 bit ISA and we will not be seeing an Itanium for the average user (it is way too expensive to produce and a royal bitch to program for).

    --
    [RIAA] says its concern is artists. That's true, in just the sense that a cattle rancher is concerned about its cattle.
  36. Re:150 Watts? by hxnwix · · Score: 1

    Intel's grasp of what is in their best interest never seemed more dubious. Well, perhaps their extravagant network sector buyout frenzy and the subsequent fire-sale liquidation of all those recently purchased properties might have seemed a worse error than forging ahead with the mhz marketecture on an electrically inefficient 90nm process. Intel reminds me of HP... They both push flimsy PR initiatives, forgetting that they are sitting on huge assets. There is a place for 150, 300 and perhaps 600 watt 32bit cpus... in an eroding corner of the market, with more efficient competitors assuming leadership. This is where Intel's current course terminates - likely with the release of a relatively low clock x86-64 part (which they could have had now). Instead, two to five years of steady marketshare loss and blind pushing of an inferior, wasteful product will intervene.

  37. Re:150 Watts? by mshultz · · Score: 1

    Nope, your currency conversion's backwards. Canadian dollars are worth less than American ones... should be $US 4.93, according to the Universal Currency Converter. That said, I don't know what the going rate for electricity here in this part (Texas) of the US is, so perhaps this conversion won't be so helpful.

  38. Re:150 Watts? by Billly+Gates · · Score: 1

    Is not the G4 40 watts?

    Yikes

  39. 150 Watt? by Eisenstein · · Score: 1

    How many energy-company-stocks do the Intel techies own?

  40. Worse than that by devphil · · Score: 1


    were the late 80's and early 90's, when the ruble was so worthless that people would melt down the various denominations of coins, and get more money back for the metal ingot.

    And at 150 watts, you could probably do the melting on top of one of these new CPUs.

    --
    You cannot apply a technological solution to a sociological problem. (Edwards' Law)
  41. Tejas information by Saville · · Score: 5, Informative

    2/27/2003:
    http://www.extremetech.com/article2/0, 3973,900185, 00.asp
    10/11/2003:
    http://www.xbitlabs.com/news/ cpu/display/200310110 84615.html
    misc:
    http://endian.net/details.asp?t ag=Tejas

    So it looks like it will come in in 2005 instead of the original 2H 2004. It'll have 24k L1 instead of 8k or 16k like current and prescott have. When it is made at 65nm insteadof 90nm it'll have 2megs L2 instead of 1meg.

    It should start eventually run as high as 5Ghz. Maybe that is on the 65nm process years from now? Bus speed should be 1066Mhz (266*4) or 1200.

    It should have some new instructions in order to make life harder for AMD.

    Fortunately for AMD Prescott was already supposed to be shipping at 3.8Ghz, but Intel is a bit behind on their road map too :)

  42. Are these CPU's teflon coated... by Lawrence_Bird · · Score: 1

    so that consumers can also cook on them with easy cleanup?

  43. Looks good, but... by efextra · · Score: 1

    does it run linux?

  44. 150W will not be a problem by Pidder · · Score: 1

    I don't think there will be a problem air cooling this CPU. The only thing that will change is that the default heatsink will have to be upgraded and thus increasing the price by $40 or so. My CPU probably using 80W or so at full load and I can easily cool it with a SLK-900U and a 1000 rpm 92 mm fan. The fan is definitely inaudible. The heatsink weighs in at 1 lb of copper and $50, however. I can see myself cooling this new CPU with the same heatsink and a 4-5000 rpm fan. That's alot of noise though...maybe water will be the default cooling in the future?

  45. Re:150 Watts? by TheSunborn · · Score: 1

    Acording to Motorola, the 1Ghz G4 require 10 Watts. The faster ones require a bit more.

  46. Prescott has 64-bit instructions by supervillain · · Score: 1

    Prescott has 64-bit instructions they are just turned off. At some point Intel will enable them for new chips. The original Pentium 4 had hyper threading in its core but it was turned off until the 3.06ghz chip was made. If Tejas is a generation after Prescott it most likely has 64-bit instructions on the core although they may or may not be enabled.

  47. Re:150 Watts? by jetmarc · · Score: 1

    > so tomorrow's cpu uses 150w. two light bulbs.

    Hey! If we combine this fact with yesterdays news we might just have solved that whole heatsink problem: just let the CPU produce light instead of heat, and then cut rectangular lightsink holes into the case to let the light dissipate..

  48. Maybe not Tejas shown? by Saville · · Score: 1

    http://www.aceshardware.com/forum?read=105063294
    http://www.x86-secret.com/

    Apparently anandtech is showing a prescott?

  49. Links by dave1g · · Score: 1

    Sorry but I am responding to both your posts, dont hate me. I am complaining both about your non link urls and Slahdot's inability to convert them to links...

    If you are going to post urls, you might as well put the link tags in there...

    BUT! why doesn't slashdot auto convert URLs to links?? Come on, forum software, even the really crappy ones, have had that feature for ages.

  50. Links by dave1g · · Score: 1

    Why doesn't slashdot auto convert URLs to links?? Come on, forum software, even the really crappy ones, have had that feature for ages.

  51. Blocked Out? by wideBlueSkies · · Score: 1

    Anyone know what they blocked out in the socket picture?

    What could possibly be there that they wouldn't want us to see? Or is is that they're not allowed to show us?

    --
    Huh?
    1. Re:Blocked Out? by City_Idiot · · Score: 1

      Its says they have because what ever is under there could show who the contact is

    2. Re:Blocked Out? by swankypimp · · Score: 2, Informative

      Intel distributed ten samples of this CPU, probably under condition that you do not talk to the press. (The first rule about Tejas is that there is no Tejas.) This is likely enforced by lawyers and promises of future access to demo units. I'd guess that each CPU is marked with a number or some other unique identifier: if the ID were photographed, Intel would know who leaked the pictures of its top secret processor and could take action.

      --

      --All your stolen base are belong to Rickey Henderson
  52. Re:150 Watts? by AchmedHabib · · Score: 1

    How about, "Time to invest in Zalman"
    The P4 heatsink/fan that came with it was really noisy. But I guess that is only an issue with homebuilt PCs since my IBM P4 at work does not make any noise. I opened it to look at it and it had a no fan on the CPU but a cooling duct connecting it to a big fan at the back of the case.

  53. Re:150 Watts? by AchmedHabib · · Score: 1

    True, but it does put seti@home into another perspective.

  54. Re:150 Watts? by Dan+Ost · · Score: 1

    How many watts does the G5 consume?

    --

    *sigh* back to work...
  55. Re:This is actually a troll, even so by chunkwhite86 · · Score: 1

    It's patently obvious that, in the context of the Soviet Economy, Ivan Ivanovitch would not have been better off melting down his tractor because (a) he didn't have access to a smelter and (b) it wasn't his tractor, it belonged to the collective and (c) there was no internal private market for steel and (d) he would have been shot for economic sabotage.

    My tractor example was a hypothetical. I was not implying that soviet farmers spent their days melting tractors for money.

    I was illustrating an example of the inefficiency of the soviet system - and some history of the "In Soviet Russia..." jokes.

    You might also consider that in many ways Soviet Russia was not backward. ... he Soviet Union produced some fairly decent cameras, some remarkably advanced military aircraft and helicopters (some of which I have been in and survived), some useful surface to air missiles, some working nuclear subs. ... It just forgot the living standards and human rights of its population, failed to deal with mundane issues like delivering food and clothes efficiently...

    Right. Because building cameras, military aircraft, and nuclear subs takes presidence over essential elements of human life, namely food and clothes. And that isn't backwards. Sure....

    --
    I'd rather be a conservative nutjob than a liberal with no nuts and no job.
  56. Re:150 Watts? by be-fan · · Score: 1

    57 watts for the current 2GHz G5. The new 90um G5s will use 71 watts at 3.2GHz.

    --
    A deep unwavering belief is a sure sign you're missing something...
  57. Re:Welcome to 1985... by Sean+Johnson · · Score: 1

    No thanks! I'll wait until the "Mr. Fusion" attachment becomes cheaper.

    --
    >>>>>> Chewie, take the professor in the back and plug him into the hyperdrive.
  58. Re:This is actually a troll, even so by IPFreely · · Score: 1
    Right. Because building cameras, military aircraft, and nuclear subs takes presidence over essential elements of human life, namely food and clothes. And that isn't backwards. Sure....

    Does that apply to the US too?
    (Sorry, couldn't help myself. *waves* Hi.)

    --
    There is nothing so silly as other peoples traditions, and nothing so sacred as our own.
  59. Re:150 Watts? by Billly+Gates · · Score: 1

    No wonder Apple laptops last so long while the windows ones need a recharge.

  60. socket? by Zugok · · Score: 1

    Wouldn't this make the CPU the socket, if the pins are on the motherboard?

    --
    "I just can't sit while people are saying nonsense in a meeting without saying it's nonsense" J Watson, Sci Am 288:(4)51