Slashdot Mirror


Intel 800 MHz FSB Processor Family Review

David writes "Techware Labs recently had the opportunity to spend some time with Intel's new 800 MHz front-side bus (FSB) processor family. The review includes a overview of the features in this processor family, Intel's new Springdale and Canterwood chipsets, and an analysis of processor scaling within this family. The article focuses on how the relationship between CPU and video card affect various aspects of performance."

137 comments

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

    Who the hell are Lilu and Niles?

  2. Re:So fucking what? by wilfire · · Score: 1

    oh well give it a few mounths and it will be cheap enough for us poor students to afford.

    --
    Anti gravity, but don't positives and negatives attract, humm a flaw me thinks.
  3. Ha! by Anonymous Coward · · Score: 0

    AMD is only running at, what, 400?
    You've got a long way to go

    1. Re:Ha! by Schugy · · Score: 3, Informative

      AMD has Opteron and Athlon 64 with Ultra Low Latancy controllers. Previews say that performance often is not a matter of maximum bandwith (3,2GB/s Athlon 64 and 6,4GB/s Opteron) and the Athlon 64 performs just like a P4 xxxxC with FSB 800 in memory dependent benchmarks like q3a. Have fun :-)

  4. Enough! You insensitive clod! by Anonymous Coward · · Score: 5, Funny

    I already feel bad enough having bought 5 486's at $2000 each. Now, my ebay auction to sell one for $10 got 0 bids. I mean, a 99.5% decerase in value is bad enough - you don't have to rub it in by telling me about the latest in computing!

  5. 100Mhz-133Mhz-800 Mhz FSB by Anonymous Coward · · Score: 0

    I have used Mother boards of 100 Mhz and 133 Mhz,suddenly seeing 800Mhz blown me away,anyway nice to see that its 4 x 200 Mhz in the article.

  6. Spending some time with .... by watzinaneihm · · Score: 5, Informative

    Spending time is nice but This is the page I think they should have linked to .
    Basically in the review they compare different chips (2.4Ghz, 2.8Ghz) etc. against each other all with 800Mhz FSB

    --
    .ACMD setaloiv siht gnidaeR
    1. Re:Spending some time with .... by crisco · · Score: 2, Insightful
      Heaven forbid they benchmark the CPUs against earlier P4s at the same clockspeeds. Or compare different memory technologies / mboard chipsets.

      Pages and pages of pretty graphs and charts all to tell us that yes, higher clock speeds mean higher performance.

      --

      Bleh!

    2. Re:Spending some time with .... by Carnivorous+Carrot · · Score: 1

      This is all fine and good, but what I want to know is is the Alienware 3.0 GHz with 800 MHz FSB and Radeon 9800 Pro for $2,100 a good deal?

      Of course, it's $2800 with the 22" CRT (is that a good CRT?) and the blue mouse and blue keyboard.

      --
      "Has [being a kidnapped teenage girl, raped repeatedly for months] changed you?" - Katie Couric to Elizabeth Smart
    3. Re:Spending some time with .... by Anonymous Coward · · Score: 0

      um, no.

      pricewatch has the cpu around $420, so unless you plan on spending $1600 on the motherboard/gfx, you're better off buying it piece by piece, by like $800 or so.

    4. Re:Spending some time with .... by TopShelf · · Score: 1

      About a month ago I got a 3.0GHz from IBuyPower with a Radeon 9500 Pro (looks way good enough for me!), 1GB RAM and no monitor (no need to replace the one I have) for around $1,500.

      Having come from a 300MHz IBM Aptiva, I'm pretty freakin' giddy these days!

      --
      Stop by my site where I write about ERP systems & more
  7. Bah by Anonymous Coward · · Score: 5, Funny

    In 1980 I had a 1.023 MHz Apple ][+ and I could type ~70 WPM. Intel is pushing 3+ GHz chips and I can still only type ~70 WPM.

    1. Re:Bah by KingDaveRa · · Score: 5, Funny

      Ah yes, but could you play an MP3, encode a movie, search for the cure for cancer andtype at ~70WPM?

    2. Re:Bah by verbatim_verbose · · Score: 5, Funny

      Well clearly you're upgrading the wrong parts. Go overclock your keyboard, silly.

    3. Re:Bah by Some+Dumbass... · · Score: 4, Funny

      In 1980 I had a 1.023 MHz Apple ][+ and I could type ~70 WPM. Intel is pushing 3+ GHz chips and I can still only type ~70 WPM.

      Clearly the solution to this problem is a benchmark-specific optimiation to your typing. Try typing just the word "I" and see how many wpm you can get...

    4. Re:Bah by Elie+De+Brauwer · · Score: 1

      You forgot to search for extraterrestrial intelligence, calculate mersenne primes and try RC5-72 keys.

    5. Re:Bah by PCeye · · Score: 1

      The computer is not the problem. You need to upgrade your fingers.

    6. Re:Bah by Carnivorous+Carrot · · Score: 1

      > In 1980 I had a 1.023 MHz Apple ][+ and I could
      > type ~70 WPM. Intel is pushing 3+ GHz chips and
      > I can still only type ~70 WPM ...if you're lucky. With all the scripting going on inside applications these days, if you could do 120 then, you might not be able to do 120 now.

      --
      "Has [being a kidnapped teenage girl, raped repeatedly for months] changed you?" - Katie Couric to Elizabeth Smart
  8. Tom's hardware had it first by nounderscores · · Score: 4, Informative

    Read all about it here

    ___________________________________
    The spiders are coming

  9. Targeting this: by GoldMace · · Score: 0, Troll

    LOL @ Intel Ad saying Targeting This: Pentium 4, Xeon, Itanium when it's clear we should be targeting this new processor instead.

    1. Re:Targeting this: by frs_rbl · · Score: 1

      ...which happens to be a P4 3GHz

      --
      This is not my opinion. Actually, it's not even an opinion. And I'm nowhere to be seen near it
    2. Re:Targeting this: by rusty0101 · · Score: 1

      And I thought they ment painting a target on those processors so we finally make some use of them, even if it is only as targets at the rifle range.

      -Rusty

      --
      You never know...
  10. Re:Lilu by Anonymous Coward · · Score: 0

    She was named after Leeloo, but I liked how Lilu looked better.

    Tank-u!! She was so cute!

  11. too much power != good by Anonymous Coward · · Score: 3, Interesting

    The more powerful the chips intel pushes the less effcient the coder becomes, i remember when i used to tweak my programs so they would run optimally on a slower machines, now a days its like you need 192mb and 500mhz for word processing. People need to get back to the old school days when a 486/66mhz and 4mb RAM was minumum. I can understand how games evolve and more power is needed, but it's not just games that have this high requirement these days.

    1. Re:too much power != good by Anonymous Coward · · Score: 2, Insightful

      The more powerful the chips intel pushes the less effcient the coder becomes, i remember when i used to tweak my programs so they would run optimally on a slower machines

      Yeah, I said that too when the PII came out. Sure there is always going to be bloat in code, especially in large projects. But you are more than welcome to go to ebay and get an 8088 or an Apple II and enjoy a machine that fits your computing needs (floppy drive or tape drive your pick).

      Me, I would like to have a computer fast enough to do things like audio/video editing, real time ogg encoding, or whatever. I surely would not mind buying a computer today thats 4x faster than these new P4s for about $1000. I'd find a need for it or enjoy the lack of bloat feeling, who cares?

      Although I have had 0 formal training in programming, one thing I've read and have incorporated into my coding is early optimization == bad. 1st write good code, then find out where the bottlenecks are (if any) and then optimize those bottlenecks. There are even great profiling tools out there to help you do these things.

    2. Re:too much power != good by Angry+White+Guy · · Score: 5, Funny

      Although I wholehartedly agree with your comment about the proliferation of inefficient coding styles, the additional power allows us to write higher abstraction languages, almost to the point of natural-language programming. More structure introduced into programming can only mean a shift to programming in english, and that can only mean less poor spelling on slashdot.

      I see it as win-win

      --
      You think that I'm crazy, you should see this guy!
    3. Re:too much power != good by Anonymous Coward · · Score: 0

      Why not write a book! Choose COBOL!

    4. Re:too much power != good by Anonymous Coward · · Score: 0

      People need to get back to the old school days when a 486/66mhz and 4mb RAM was minumum.

      People said the same thing when a 486 was cutting edge. "Awe... &$&# all this power. It's turning our coders into lazy bums."

    5. Re:too much power != good by Tancred · · Score: 2, Insightful

      Some people need all the processing they can get. It's allowing some of us to do things we couldn't before. If you don't need it, don't buy it. And you can always stop buying bloatware and write your own software or optimize your favorite OS program.

    6. Re:too much power != good by guinsu · · Score: 1

      So how many mp3 players and video editors did you have for your 486? How many gigabit ethernet cards did your 486 support?

    7. Re:too much power != good by Anonymous Coward · · Score: 0

      it does seem like the whole process is evolving to where at some point, it will be similar to startrek.

      "COMPUTER! Please display a schematic of Alpha2, at time index 17.38, and supply callouts cross referenced with the database Riker4271"

      Programming is basically problem solving. As programming becomes ever more abstracted, more people can participate in complex problem solving.

    8. Re:too much power != good by chez69 · · Score: 1

      Gosh I'm sick of wankers like you that claim that we don't need faster processors. There are folks that do more with their computer then bitch on slashdot.

      --
      PHP is the solution of choice for relaying mysql errors to web users.
    9. Re:too much power != good by drsmithy · · Score: 2, Interesting
      People need to get back to the old school days when a 486/66mhz and 4mb RAM was minumum.

      Why ?

    10. Re:too much power != good by vldmr_krn · · Score: 1

      The more powerful the chips intel pushes the less effcient the coder becomes.

      This is a good thing. When less skill and effort is needed to code, the more programs get released, because designers take the time they would have spent on code optimization, and spend it on other things.

    11. Re:too much power != good by Anonymous Coward · · Score: 0

      Since most programmers can't express an idea in natural languages (i.e. can't really speak english), we can infer that all that extra processing power will still be wasted by running buggy programs at higher inneficiency levels.

      If you think I'm way off mark, I'll just point you towards java.

  12. For those to lazy to read :) by Anonymous Coward · · Score: 1, Informative

    Conclusion:

    The Intel 800 MHz FSB Family of processors truly lives up to its name. After looking at the results which tests are CPU limited and which are video card limited, the data concludes that there is a certain balance between the dependency of the video card and the processor. As the graphics get more intense, the performance becomes more dependent on the video card. The Intel 3.0GHz 800MHz FSB is definitely cutting edge and the CPU really shines during programs that require the most calculations and complexity. For example, the Unreal Tournament Botmatch is CPU dependent because the program requires many calculations and factors of each bot or player. On the other hand, there are other programs where the CPU is not needed as much, such as programs that require complex lighting and transform calculations by the video card or other video card specific features. The power of the CPU allows it to shine in conjuncture with older APIs or engines, such as DirectX8 or the Quake3 Engine, which are not as dependent on the video card. These programs allow the CPU to have more of an influence on the performance, creating greater improvements. With brand new programs or software based on DirectX9, the graphics are so intensive that a stronger video card is needed, with the CPU playing a lesser role (at least the high end CPUs, this may be different with lower end CPUs). This causes a less noticeable effect on improvement with greater CPU speeds, but that will only improve over time when the development of video cards are increased. In the future, the software based on DirectX9 won't be so dependent on the video cards due to new technology and the CPU will be able to shine in these regions as well. This comes to show that the potential of the Intel 3.0GHz 800MHz FSB is still uncharted and performance with the CPU should dramatically increase when coupled with future video cards. After seeing the capabilities and performance during strenuous tests, the Intel's family of 800MHz FSB processors have earned our Editor's Choice Award.

    1. Re:For those to lazy to read :) by Anonymous Coward · · Score: 0

      I'm too lazy to read all that, can't you like break it down to a simple sentence or something??

      Please?

  13. It'd be nice... by Mister+Black · · Score: 4, Interesting

    It'd be nice if they normalized all their charts with some current non 800FSB proc+board so I can see how much of an improvement there actually is.

    --

    You are standing in an open field west of a white house, with a boarded front door. There is a small mailbox here.
  14. Obligatory Pentium Jokes by Anonymous Coward · · Score: 2, Funny

    Q. How many Pentium designers does it take to screw in a light bulb?
    A. 1.99999289345, but that's close enough for non-technical people.

    Q. The Pentium conforms to IEEE standards for floating point math. If you fly in an airplane designed using a Pentium, what's the correct pronounciation of IEEE?
    A. Aiieeeeeeeeeeeeeeeeeee!

    Q. What's another name for the Intel Inside sticker they put on PCs?
    A. The warning label.

  15. On Performance... by paranoidsim · · Score: 1

    Well, with the recent slashdot article [http://slashdot.org/article.pl?sid=03/05/20/14222 43&mode=nested&tid=137] explaining how useless CAS Latencies are in concerts to memory bandwidth, increasing the FSB and CPU to Memory Latency is not a good thing, its a great thing. Intel has my money!

    1. Re:On Performance... by Pulzar · · Score: 4, Informative

      That article showed that lower latency doesn't mean higher bandwidth (and this is only true if your original latency is low enough, mind you!), but it didn't consider overall performance. Latency has indeed an impact on the performance -- look at Tom's Hardware article on performance improvements when Intel's PAT is enabled. All PAT does is lower latency by 2 cycles.

      --
      Never underestimate the bandwidth of a 747 filled with CD-ROMs.
  16. you should wait by nounderscores · · Score: 1

    and then sell them as nostalgia jewelry.

    Just so long as you didn't buy them for $2k in this century, you should be fine.

    ___________________________________
    The Spiders are coming

  17. dualies by bloosqr · · Score: 4, Interesting

    Anyone know why the dual xeon motherboards aren't flipping to 800mhz? I would think that a faster bus would make more of a difference on dual processor boxes. Currently the memory bus is at 2*266 = 533 and I think it will move to 667 either late this year or early next year. On the other hand, i've noticed iwill is now selling a dual xeon motherboard for $300 and the chips are not much more expensive than their "normal" equivalents. (Note I am talking about the dual processor xeon chips not the "made of pure gold" 4 processor xeon mp chips)

    -bloo

    1. Re:dualies by barureddy · · Score: 1
      I'm not sure why the xeons don't have 800mhz fbs yet. They have just went up to 533fbs from 400fbs. Some of your facts are incorrect. Your math is wrong in caculating fbs. It is 133fbs quad pumped to 533 or (133*4=533). Intel decided to skip 667fbs and just go for the 800fbs. Though SIS did make a chipset that supported the 667fbs.

      If your still here can you clarify:
      "Note I am talking about the dual processor xeon chips not the "made of pure gold" 4 processor xeon mp chips"
    2. Re:dualies by FeeDBaCK · · Score: 1

      Intel usually takes a bit more time to release new technologies into their Xeon line. This allows them to iron out any major bugs in the system before shipping them off to large corporations, which rely on this hardware to be very stable in comparison to Joe Gamer.

      --
      wolf31o2 Developer, Gentoo Linux Games Team
    3. Re:dualies by doormat · · Score: 2, Informative

      They will be in early 2004 it seems...

      http://www.theinquirer.net/?article=9823

      --
      The Doormat

      If you're not outraged, then you're not paying attention.
    4. Re:dualies by bloosqr · · Score: 3, Insightful

      Ahh the 266*2 was wishful thinking :). You are right about its 133*4, apologies about that.

      My understanding is that xeon mp line is for their
      4-way based motherboards. The main advantage is they have a meg of cache on them. But the normal processorshave 512k the same as the new p4's I believe.

      The xeon mp motherboards are $2k and the processors are about $2k each (pricewatch 1.6ghz/1meg cache i.e made of gold :)

      In any case the normal xeon dual systems are actually not that much more than buying a 875pe
      motherboard and processor. Btw here is the road map I found on the inquirer. Apparantly the xeon mp's are going up to 2.8ghz/2 megs of cache and the normal xeons are going up to 3.06/1 meg of cache and selling for $700.

      Here's the weird part, while it looks like intel skipped 667 fbs for the PIV line, the xeon line will "ramp up" to 667 early next year.

      In anycase I'm probably going to build a "normal" xeon/iwill running at ~2.66 which comes out to really not much more than a normal PIV/865/875 series. The selling of 800mhz memory/bus speeds on the PIV line while keeping the xeon line at 533/667 makes no sense to me. I was going to wait until a new set of mbs/chipset came out for the xeons but it doesn't look it will happen.

      -bloo

    5. Re:dualies by bloosqr · · Score: 1

      Ahh thanks much! Thats what I was looking for..

      -bloo

    6. Re:dualies by akuma(x86) · · Score: 2, Insightful

      Multiprocessor bus speeds and CPU frequencies always lag behind uniprocessor systems. It takes much longer to validate multiprocessor boards when compared to a uniprocessor system. This is because the number of things that can go wrong goes up exponentially with the number of CPUs on the board. The typical customers of multiprocessor systems value this sort of reliability even more than performance.

    7. Re:dualies by n3r0.m4dski11z · · Score: 0

      dude you should seriously consider a dual amd system. im running a dual mp 1600+ with 512 ram and i paced it with my freinds xeon p4 at 2ghz and mine was about the same. they have alot more cache sure but for like about 2000$ more i would gladly spare it.

      i am planning on getting opterons when they fall to about 250-300 per proc and hey, for 600 more i have a quad system, if anyone puts out a mobo that is...(read none yet). the rams cheaper, the procs are cheaper and the mobos are cheaper. and on top of it all -- 64bit.
      i used to cry heat! heat! but properly ventilated mine runs at 35*c up to 47 at full tilt. (volcano 7s) i used to be an intel man all the way but performance is better than the slight cooling issue.

      --
      -
    8. Re:dualies by bloosqr · · Score: 1

      I think at the ~1.4/~1.8 ghz speeds the AMDs were a good deal (especially at the time the only dual p's were pIII's that were reasonably priced ) so the AMD chips were clearly the better buy. I am not so sure this is quite as true any more. The dual motherboards for xeons (on atx sized mb) are going for $300. The 2.4 ghz chip is going for $240 and the 2.66 chip for $319 and you get free hyperthreading and you (if you write GPL/open source code) get free intel compilers.. Okay so the 2600+ mp chips are going for $216 and the 2400+ for $166. Tiger has the mpx boards going at $190. You're right actually you may be able to build decent dual mpx athlon systems at $1000 a box! The intel ones come out at ~300 or so more.. but you do get a faster fsb and "hyperthreading"!

      -bloo

    9. Re:dualies by Anonymous Coward · · Score: 0

      The Xeon price seems low for some reasons:
      - Xeon doesn't have 200x4 FSB
      - you're looking at the cheapest Xeons, with half the cache of a real one
      - the new P4s use hyperthreading

      Personally, I don't give a flying fuck about HT, I'm still waiting for a regular 3+GHz/800 P4 so I can build a REAL dualie. Looks like I'll have to be saving up for a pair of Xeons though...

  18. Re:WHERE ARE THE WMD'S? by Anonymous Coward · · Score: 0, Troll

    Hey buddy, people in the Congo want to know when they get their US intervention. What's that? No, they don't have oil...hey! Where are you going?

  19. You know it's time to upgrade... by splerdu · · Score: 5, Funny

    When the new processor's FSB is higher than your rig's CPU clock =(

    1. Re:You know it's time to upgrade... by Anonymous Coward · · Score: 0

      Uhh, that was three years ago for me.

    2. Re:You know it's time to upgrade... by Anonymous Coward · · Score: 0

      hah! How about when the FSB is TWICE your current cpu? Or you just built a machine for someone that had more RAM than you have free HD space?

  20. Buy the 2.4 by wpmegee · · Score: 5, Informative

    The 2.4c will be finding a home in my box soon because of it's amazing overclocking.

    At this forum (click on Intel cpus) almost everyone has successfully overclocked theirs over 3Ghz on air, with most hitting 3.2 or 3.4 (and don't forget a 1 Ghz fsb).

    A popular motherboard to go with it is Abit's IC-7 with the i875 chipset. The processor and motherboard are just $180 and $145 respectively over at Newegg, so don't waste your money on 3.0s.

    1. Re:Buy the 2.4 by barureddy · · Score: 1

      You want to keep your memory timings and clock speed in synchronization (1:1 ratio) to attain max performance (I don't think PAT is enabled at any other setting than at 1:1). So unless you have ddr500, you are taking some preformance hit by changing to another ratio (ex. 5:4) to keep your system stable. So the recommended overclockers are 2.6 and 2.8 for their higher multipliers.

    2. Re:Buy the 2.4 by Anonymous Coward · · Score: 0

      True but money talks. That $50 you save can go towards another 256MB of HQ memory, an extra 30GB 72000 drive, a Video capture card, or even a decent case or power supply. I'd rather have one of those then a 5-10% faster machine wouldn't you?

      For those with the money get a 2.6, but for everyone else that extra $50 buys a nice addon.

  21. No Athlon XP benchmarks? by Alereon · · Score: 3, Insightful

    What good is benchmarking the new P4-C processors without comparing them to Athlon XPs, or even older P4s? Really, you can just multiply the performance of a P4-C 3.0Ghz by 0.8 to get a guesstimate of the performance differences within the family; what really matters is how they perform in comparison with the competition.

    1. Re:No Athlon XP benchmarks? by fitten · · Score: 1

      Benchmarks with the older processors were done a while back. It's not that hard to dig up the older benchmark articles and see what was posted back then. The only real variations would be video cards but you can always use numbers from benchmarks that don't use the video card to compare (RAR, MP3 encoding, SPEC, etc.)

    2. Re:No Athlon XP benchmarks? by Alereon · · Score: 1

      Chipsets and memory architecture can also be expected to have a singificant performance impact. I'll bet that a P4-HT 3.06Ghz is going to be significantly faster on an i875P chipset with dual-channel DDR266 than on it's original platform.

    3. Re:No Athlon XP benchmarks? by Anonymous Coward · · Score: 0

      There are benchmarks comparing the Athlon XP 3200+ vs the Intel P4-C 3 Ghz. Here is an article that does that, showing the P4-C beating the Athlon XP 3200+ in 6 out of 10 benchmarks. The problem with benchmarks, though, is they can unfairly take advantage of particular features of the hardware to show a result that is unrepresentative of what you would see if you were using it. For example, the benchmarks that Tom's Hardware use strongly favor processors with 'Hyperthreading' and SSE2 instructions. The P4 has those features while the Athlon XP does not, benchmarks which favor those features will indicate a better result with the P4-C. This would be fine if all, or even most, software supported hyperthreading and SSE2 but the reverse is true so the benchmarks which exclusively favor these features are therefore unrepresentative of the software in use.

    4. Re:No Athlon XP benchmarks? by Alereon · · Score: 1

      This is why it's important to use benchmarks that represent real world tasks, such as games or actual application benchmarks. If it shows the performance that a user could be expected to see in real life, then it's probably a fair benchmark.

    5. Re:No Athlon XP benchmarks? by ProtonMotiveForce · · Score: 1

      There are plenty of benchmarks out there - the 2.8 generaly beats, ties, or spanks the "3200+" depending on what benchmark.

      That's some misleading rating AMD has going, there.

    6. Re:No Athlon XP benchmarks? by Alereon · · Score: 1

      And there's plenty more where an XP 2800+ ties, beats, or dominates the P4 2.8Ghz. Athlon XPs generally come out ahead or equal in gaming, completely dominate in general usage and office applications, COMPLETELY own in hard integer and floating point math, and lose quite badly in hard 3D rendering (to distinguish from gaming) and media encoding. Basically, if it doesn't use SSE2, the Athlon XP wins.

    7. Re:No Athlon XP benchmarks? by ProtonMotiveForce · · Score: 1

      No, there aren't. The 2.8C beats the crap out of the 2800+. Try looking at the benchmarks, next time.

      AMD is way behind, period.

    8. Re:No Athlon XP benchmarks? by Alereon · · Score: 1
  22. heat by martin · · Score: 1


    wow look at the heat sink and fan...

    http://techwarelabs.com/reviews/processors/intel _8 00mhz_fsb/pictures/setup1.jpg

    I bet ya need brick in the case to stop the thing taking off.. :-)

  23. 8-bit? by TheRaven64 · · Score: 1

    The article states that the 800MHz bus gives 6.4GB/s bandwidth. A quick calculation shows that it is therefore only 8 data bits wide. Is this normal for a modern microprocessor? Or is it just Intel perpetuating the MHz myth into a new area?

    --
    I am TheRaven on Soylent News
    1. Re:8-bit? by Pulzar · · Score: 1

      The bandwidth is 6.4GB/s, not 6.4Gb/s. I.e. it's 6.4 giga-bytes, so the widths of the bus is 8 bytes (or 64 bits).

      --
      Never underestimate the bandwidth of a 747 filled with CD-ROMs.
  24. Fixed link and more information by Alereon · · Score: 1

    http://techwarelabs.com/reviews/processors/intel_8 00mhz_fsb/pictures/setup1.jpg The P4-C 3.0Ghz can be expected to consume about 100Watts at full load, which is far more than most processors. Only the P4-HT 3.06Ghz and the original P4 2Ghz came close, also coming in at about 100W. AMD's hottest running processor, the Athlon XP 3200+, consumes 76.8W.

    1. Re:Fixed link and more information by Billly+Gates · · Score: 1
      The powerpc processors that apple uses only use I think about 20 watts. This is getting redicoulous. I thought the itanium was bad.

      Just goes to show that risc is better not to mention that both Intel and AMD overclock their processors for marketing reasons. I prefer no fans at all and have a quiet pc. Also look at nvidia? 20 db fans!! Come on? Its useless for games because of the noise.

      Give me a 10% decrease in performance anyday for a more reliable and quiter solution. If the cooling fan fails your cpu is toast.

    2. Re:Fixed link and more information by compwiz3688 · · Score: 1

      If the cooling fan fails, only AMD is toast, Intel's P4 can slow itself down (P3's crash but not burn) when overheated. Haven't you seen THG videos? Also, the current motherboards will power down when AMD CPUs go over a certain temperature, although you'd loose any precious unsaved data.

      Then again, it's been a while since they made the video. I wonder if the current Northwood will crash and burn. Anybody dare to try? :)

      I do have to agree though, another fan in the case is another breeding place for the dust bunnies.

    3. Re:Fixed link and more information by Alereon · · Score: 2, Informative

      No modern CPU will continue to operate through a catastrophic cooling failure. The P4 will crash when it throttles below 25% (errata), and will completely shutdown if the heatsink is removed. The P3 will burn up without a heatsink, unless the motherboard shuts it down (I don't know of any motherboards that do this). An Athlon XP will be shut down by the motherboard before damage occurs (ideally). Regardless, what possible set of circumstances could occur that would cause your heatsink to come off your processor without destroying it or the motherboard in the process?

      THG's infamous video was clearly faked, or at least misleadingly edited. The P3 was shown running at 38C after it crashed, which is lower than it would run WITH A PROPERLY OPERATING COOLER. The P4 was also shown clocking back up to full speed immediately after the heatsink was merely SET on top of the processor, this is completely impossible. The processor would have shutdown due to the temperature it was running at, and regardless it would have taken at least a few seconds for the processor to be cooled enough by the heatsink before it clocked back up, and it wouldn't be cooled at all if the heatsink was merely SITTING ON TOP, unmounted!

    4. Re:Fixed link and more information by drsmithy · · Score: 1
      The P3 will burn up without a heatsink, unless the motherboard shuts it down (I don't know of any motherboards that do this).

      All intel CPUs have had the ability to automatically shut themselves down in the case of excessive heat since at least the 486.

      The P4 is the first one that did this _gracefully_ (ie: gradually slow down then stop instead of just stopping), but AFAIK they all do it.

      A P3 will almost certainly crash if the HSF is removed, but it probably won't be destroyed without excessive uncooled running.

      An Athlon XP will be shut down by the motherboard before damage occurs (ideally).

      Most motherboard sensors can't react quickly enough to effectively throttle an Athlon.

      Regardless, what possible set of circumstances could occur that would cause your heatsink to come off your processor without destroying it or the motherboard in the process?

      One of the plastic lugs giving way from the weight and pressure of a HSF in a tower case.

      The HSF being knocked loose during shipping and the owner turning the machine on without realising.

      [...] and it wouldn't be cooled at all if the heatsink was merely SITTING ON TOP, unmounted!

      The THG video may or may not have been faked. However, there's no reason a HSF wouldn't work to some extent without being clipped on.

    5. Re:Fixed link and more information by ProtonMotiveForce · · Score: 1

      Umm, how exactly does a processor manufacturer "overclock" their own processor, since by definition it should run at the speed they say it should?

      That makes as little sense as the rest of your meandering inanities do.

    6. Re:Fixed link and more information by Alereon · · Score: 1

      All intel CPUs have had the ability to automatically shut themselves down in the case of excessive heat since at least the 486.

      The first Intel desktop processor to feature an on-chip thermal diode was the Pentium II. The P4 was the first processor to throttle or shutdown itself based on CPU temperature. Earlier processors USUALLY didn't fry from overheating, simply because they didn't produce that much heat and it was thus difficult for them to get hot enough to die, but it DID happen. If you remove the heatsink from a Coppermine P3 while it's running at maximum load, it will FAIL, not shutdown. I think it's quite likely that it would be permanently damaged in the process, even if it doesn't burn up Tbird-style.

      Most motherboard sensors can't react quickly enough to effectively throttle an Athlon.

      They don't bother throttling, they just cut power if the CPU hits 70C (or another user configured temperature). They can, in fact, react quickly enough to prevent a CPU in which the heatsink is removed during maximum load operation from frying. There was a video released by AMD in answer to the THG video which showed this in action.

      The HSF being knocked loose during shipping and the owner turning the machine on without realising.

      In such a case, the motherboard is guaranteed to be destroyed already, and probably also the cards the heatsink hit on the way down. Realistically, the heatsink just doesn't come off without catastrophic impact or force to the system, something that just plain doesn't occur very often.

      However, there's no reason a HSF wouldn't work to some extent without being clipped on.

      A heatsink that is not affixed to the processor cannot cool the processor, because it is not in effective contact with it.

    7. Re:Fixed link and more information by drsmithy · · Score: 1
      The first Intel desktop processor to feature an on-chip thermal diode was the Pentium II. The P4 was the first processor to throttle or shutdown itself based on CPU temperature.

      Section 3-8, page 26 of this disagrees.

      The processor protects itself from catastrophic overheating by use of an internal thermal sensor. [...]

      That's for the Pentium Pro. I'm alsmost certain the same functionality existed all the way back to the 486, although I'll admin my recollection is a bit rusty, so I may be wrong.

      Last-ditch termal protection was certainly present in Intel CPUs before the P4.

      There was a video released by AMD in answer to the THG video which showed this in action.

      I somehow suspect such a video would be displaying ideal conditions, both in terms of the proximity of the thermistor to the CPU and the reaction speed of the motherboard's monitoring software.

      The point is, this is something that should be built into the CPU, not tacked on afterwards by hardware integrator's whose primary interest is in making it as cheap as possible.

      In such a case, the motherboard is guaranteed to be destroyed already, and probably also the cards the heatsink hit on the way down.

      Fact is, I've seen machines shipped where the HSF has come off the CPU during transit. Put it back on and the machine powered up fine.

      Realistically, the heatsink just doesn't come off without catastrophic impact or force to the system, something that just plain doesn't occur very often.

      Not necessarily. The ridiculously high-tension clips on some HSFs can eventually snap the plastic lugs that hold them onto the CPU - this is much more prevalent with the models that only hold on via a single lug each side instead of all three. Again, I've *seen* this sort of spontaneous failure (well, more accurately, heard the snap from inside the case).

      A heatsink that is not affixed to the processor cannot cool the processor, because it is not in effective contact with it.

      Of course it's in contact with it. This tricky thing called gravity pulls it down onto the CPU. Now, this contact is far from thermally ideal and probably wouldn't be good enough for extended use, but it is enough to transfer some heat away from the CPU. If there's some thermal compound there it's even better.

    8. Re:Fixed link and more information by Alereon · · Score: 1

      That's for the Pentium Pro. I'm alsmost certain the same functionality existed all the way back to the 486, although I'll admin my recollection is a bit rusty, so I may be wrong.

      The Pentium Pro was not a desktop processor. It was a server processor, analogous to Xeon. And, again, while the P2/P3 supported temperature monitoring, they DID NOT have any kind of internal throttling or shutdown. When they got too hot, they crashed because stuff stopped working at that temperature. Do it enough, or let them run really hot for awhile, and they died. Pentium and 486 had no thermal monitoring capabilities whatsoever.

      I somehow suspect such a video would be displaying ideal conditions, both in terms of the proximity of the thermistor to the CPU and the reaction speed of the motherboard's monitoring software.

      All AMD processors since the original Palomino core have had an onboard thermal diode. It's up to the motherboard to shut them down at a given temperature, but the temperature being worked off of IS the core temperature, on boards that support such overheat protection.

      Again, I've *seen* this sort of spontaneous failure (well, more accurately, heard the snap from inside the case).

      Such failures are rather unlikely. If you don't drop the system, the heatsink isn't going to come off. If it comes off in transit due to rough handling, you'd probably know it by the large object rattling around in the case. Regardless, overheat protection IS featured on current AMD processors and motherboards, so it really is a moot point.

      Now, this contact is far from thermally ideal and probably wouldn't be good enough for extended use, but it is enough to transfer some heat away from the CPU. If there's some thermal compound there it's even better.

      Spec mounting pressure for a CPU is usually around 30lbs of force. Gravity is exerting what, 0.5lbs on your average heatsink? There is most assuredly not enough contact to make a heatsink just sitting on top of a CPU noticeably better than no heatsink at all.

  25. Not the best article ... by vorwerk · · Score: 4, Informative

    I've been planning to upgrade my computer at the end of this month, and have been keeping a pretty close eye on the 865/875 motherboard and chip performance reviews. This article didn't really enlighten me as much as the following Tom's Hardware reviews:

    here

    and

    here

  26. If anyone says this is for gamers only... by Anonymous Coward · · Score: 0

    They are full of SH!T. I would love 10 GHz CPU right about now. Why? I game yeah. But I also encode DivX which is highly processor intensive. I do some graphics rendering. And most of all I am a coder. Yeah... and when I do full compiles on my projects I hate sitting around. Even if I'm just recompiling my kernel, the faster I can get that the happier I am.

    Right now I want an Athlon XP 2500+ just because its soo affordable (under $100 US), but this is a good thing--this faster FSB.

  27. Actually 200Mhz QDR [also, bits vs. bytes] by Alereon · · Score: 5, Informative

    The FSB on a P4-C is actually clocked at 200Mhz, but data is transferred four times per clock cycle, boosting the effective bandwidth to equal that of an 800Mhz FSB. Latencies are, however, still equal to that of a 200Mhz FSB.

    I believe the problem with your calculation is that you calculated that the bus is 8 bytes wide. 8 bytes is 64 bits, the standard bus width on modern systems.

    1. Re:Actually 200Mhz QDR [also, bits vs. bytes] by TheRaven64 · · Score: 1
      I believe the problem with your calculation is that you calculated that the bus is 8 bytes wide. 8 bytes is 64 bits, the standard bus width on modern systems.

      Note to self: Wake up before trying to do maths...

      --
      I am TheRaven on Soylent News
  28. This matters? by Anonymous Coward · · Score: 0

    This was new news about 4 months ago.

  29. take heart by nounderscores · · Score: 1

    the FSB is only 200mhz, quad pumped out to 800mhz (that means it shifts 4 times the data per clock cycle)

    Of course if your rig is running at less than 200mhz, who cares? so long as it plays the games you like.

    ___________________________________
    The Spiders are coming

  30. um what?! by NedTheNerd · · Score: 1, Insightful
    im probably not the first one to mention this but this review sucks. They didnt test anything but other 800 buss processors nor any AMD processors. I used to think the point of a review was to see how something performed against everyone else not against itself . . . Not to mention they didn't even overclock the codamn things. Now I know some people out there are bitching oh but who wants to see it overclocked, overclocking makes things unstable (if you dont run mprime ;). Now I realize that point but then you think about it again and realize the only thing they test is GAMES!

    In all my rantings I hardly mentioned the FSB issue when did buss speed become more important than processor speed? but maybe thats just intels way of taking the light off of the idea that they are having a little trouble pushing thier chips faster, and thats to be expected when you cant make it faster make it better. too bad all FSB gets you is 5 FPS in your precious GAMES (yea I play em too!).

    oh well I guess I am just too demanding . . .

    1. Re:um what?! by kramer2718 · · Score: 2, Insightful

      Bus Speed IS more important than processor speed. The bus is what keeps the CPU supplied with data from the memory. If you have a very slow bus it doesn't matter how fast your CPU is, it will have to wait on memory accesses.

      Smart caching can keep values in the cache that will be accessed frequently and smart compiling can execute the code in an efficient sequence (so that a lot of memory accesses can be done at once), but even still the gap between bus/memory performance and CPU performance 200/400/800 MHz vs 1/2/3 GHz is so great that this still slows down execution quite a bit.

  31. Just... by Pinguu · · Score: 1

    thought of something quite funny of the top of my head :)
    They should replace the 'Intel Inside' logo with 'Warning! May contain Intel!'. :)

    --
    --
  32. where are the comparisons with non-HT P4s? by Anonymous Coward · · Score: 0

    this review really missed the mark and left unanswered what we really wanted to know, which is how does the 2.4C, 2.6C, 2.8C compare against the plain 2.4, 2.6, 2.8?

  33. Re:too much power - efficiency by Anonymous Coward · · Score: 1, Interesting

    That's a popular but ridiculous idea. The less time the programmer needs to spend tweaking and writing incredibly painful assembly language, the more time he has to meet the actual user's needs for new software and new features.

    And by the way, you're calling a 486/66 "old school" made me laugh! To me, old school is a 1 mhz 6502, or maybe a 2 mhz z80.

  34. Not gonna read it by scumdamn · · Score: 1

    I really tried to get interested, I figured this article would give me some good weekend reading, and I clicked the link.
    But when I get to the second page and I still haven't gotten anything out of it but a little history, and the content on that page is a glorified paragraph, I'm not sticking around for more. If you want me to read something, keep it succinct, don't put a paragraph per page, and dammit, don't make me look at 15 ads just to get some benchmarks.

  35. The real question is... by jd · · Score: 1, Troll
    Can it divide?


    Seriously, this looks like a decent processor, with a reasonable chipset to go with it. I'm a bit concerned with Intel opting (again!) for overclocking hardware, rather than improving it. Especially when they keep insisting that overclocking is hazardous to the hardware, and make every effort to stop other people doing the very same thing.


    The 800 MHz FSB is just an overclocked 200 MHz FSB. Given that AMD just has to build a real 400 MHz, overclocked only once, or a real 800 MHz FSB to flatten Intel's offering, this is a chancy way to go. It was a reasonable gamble with the 486DX range, because there wasn't much competition on the chipsets. Anyone remember the 486DX-50? That was a raw 50MHz chip, no overclocking, that outperformed the 486DX-66 with ease. The supporting hardware was too expensive, and Intel never bothered working on making it cheaper. It was much more cost-effective to just keep overclocking the DX-33 further and further.


    Today, it's a different story. AMD isn't dependent on Intel, the way Cyrix and the other Intel-clones were in the 486 days. This is a tougher fight, and it's one Intel is losing, every time someone produces a faster processor or chipset. (Or even just a more reliable one.)


    Having said that, this looks a great system, and one that may well contribute to Linux being in more of the top 500 supercomputers.

    --
    It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
    1. Re:The real question is... by bigbadwlf · · Score: 1

      This is a tougher fight, and it's one Intel is losing

      It looks to me more like AMD is losing.
      Either way it's a bad thing - competition is good.
      Then again, maybe it's a good sign that we each see it differently.... it would seem to indicate that it's too close to call.
      No matter which chip you prefer, as long as they're racing neck-in-neck it's better for all of us.

    2. Re:The real question is... by jd · · Score: 1
      I agree. I guess I'd prefer them to be racing neck-and-neck with better hardware, rather than putting existing hardware into overdrive, but - hey - it's better than no hardware at all.


      (I guess I watch too much F1, where "neck-and-neck" means 240 mph, 1 inch apart, engines on the verge of meltdown, a thousandth of a second seperation, and the drivers complain that there's no serious competition.)

      --
      It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
    3. Re:The real question is... by Ozric · · Score: 1

      I have a working 486DX-50 in my closet. My first system... and I can't part with it. Its an EISA bus too. Was a smoker in its day.

    4. Re:The real question is... by drsmithy · · Score: 2, Insightful
      I'm a bit concerned with Intel opting (again!) for overclocking hardware, rather than improving it.

      This word, "overclocking", I do no believe it means what you think it means.

      That was a raw 50MHz chip, no overclocking, that outperformed the 486DX-66 with ease.

      Hate to break the news to you, but pretty much every CPU in use today uses exactly the same asymmetrical bus/CPU that the 486 DX2s did, that you are calling "overclocking".

      Oh, and a DX/50 would only outperform a DX2/66 in tasks that were bus-bound. If the bus wasn't a limiting factor, the higher clocked CPU would be faster.

      The supporting hardware was too expensive, and Intel never bothered working on making it cheaper.

      Yes, they did - it's just that by that time (ca. second generation Pentium CPU - P90, P100, P100 and P133) the 486 was obselete.

      This is a tougher fight, and it's one Intel is losing, every time someone produces a faster processor or chipset. (Or even just a more reliable one.)

      I've yet to see any manufacturer produce a more reliable chipset than Intel. Even the ones that are faster only tend to be so because they lobbed into the middle of an Intel release cycle and aren't outperformed until the next model (in general).

    5. Re:The real question is... by jd · · Score: 1
      Overclocking is to take hardware designed and specified to operate at a performance P, using a clock S, and to increase that clock speed S beyond the formal design specificaton of that hardware, in order to get an increase in P.


      If the hardware is rated at 200 MHz, but is being wired to an 800 MHz clock, you are overclocking the hardware. It is running 4 times the speed for which it was designed, and for which the components are rated.


      Overclocking in the "field" is limited to ramping the clock for entire components or entire motherboards. But it is perfectly legitamate to refer to running any sub-component at greater speeds than rated as overclocking, because that is what it is.


      The 486-50 would outperform ANY 486-66 for ANY task, because it was designed to run at the faster speed. The 66 was not. The 66 was merely a 33, with a local clock-doubling.

      --
      It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
    6. Re:The real question is... by drsmithy · · Score: 1
      Overclocking is to take hardware designed and specified to operate at a performance P, using a clock S, and to increase that clock speed S beyond the formal design specificaton of that hardware, in order to get an increase in P.

      So what makes you think the P4 wasn't designed and specified to ramp to a quad-pumped 200Mhz bus ?

      If the hardware is rated at 200 MHz, but is being wired to an 800 MHz clock, you are overclocking the hardware.

      So what makes you think the hardware is only rated for 200Mhz ?

      The 486-50 would outperform ANY 486-66 for ANY task, because it was designed to run at the faster speed.

      Untrue. See, for example, this. The DX50 was certainly faster than a DX2/66 at some things, but depending on the task, the higher clockspeed of the DX2/66 gave it an advantage in tasks where the bus speed was not a limiting factor.

      The DX2/66 was a 66Mhz chip connected to a 33Mhz bus. The CPU itself was rated for 66Mhz, the bus it was plugged into was not.

      Or are you going to try and claim that every intel (and AMD, for that matter) CPU since the 486DX/50 is overclocked because they all have internal clock speeds higher than their bus clock speeds ?

  36. Comparison ? by Elie+De+Brauwer · · Score: 2, Insightful

    What is really missing in the article is the comparision betwoon other cpu's running at 533 and/or 400 MHz. How can one interprete the benchmark results if there is no comparison to another product ? It's like saying that something is 600.1 gigaquats without defining a gigaquat.

    1. Re:Comparison ? by heff · · Score: 1

      I agree, there was even a lack of comparison between the non HT p4's. At least we know what the difference between the 2.08 and 3.0 p4 is.. whatever that's worth.

      --

      --

      |-_-| . o O ( bEef!)

    2. Re:Comparison ? by WaXHeLL · · Score: 1

      I thought this article was solely to analyze the scaling of processors of this family, and less of how well the processors perform in comparison to other processors out there.

      --
      The troll with karma.
  37. Not worth the effort to read. by Anonymous Coward · · Score: 0

    This review is pretty weak. The tech info is not much more than you could get reading Intel press releases.

    They don't compare the results to any known quantity, but that's fine, since they don't really get any useful results. The tests are mostly irrelevant to actually testing the processors. The conclusion is "we don't really know how fast it is". And then they give it an award of some kind.

    The only really interesting thing about it is that it hits a new low in obnoxious presentation.

    Not only do they not have a reader-friendly (or "printer-friendly" as it's often called) version, they also fail to provide an index page or even an indication of how many pages it has been pointlessly split into. They seem to have taken great care to hit the page division exactly wrong, with each page just a little too long to fit in a browser window.

    The little menu-bar animation adds the finishing touch to the overall aesthetic of rancid cheese.

    Conclusion: After seeing the capabilities and performance during strenuous tests, this Techware Labs review has earned my Coward's Choice Award.

  38. Intel Won't Upgrade Xeon FSB by Hiro2k · · Score: 2, Insightful
    Anandtech

    Read the article at anadtech. It's the roadmap for Intel. And discusion of all the processor currently in the market. They Discuss why the Xeon isn't getting the nice FSB upgrade even though they need it the most.

    From a performance standpoint, things make even less sense once we start looking at Intel's plans for the Xeon MP line for 4-way configurations. Intel's Xeon MPs will continue to use a 400MHz FSB through the first half of 2004, even as the processors reach speeds as high as 2.60GHz. Intel's reasoning behind this is mostly to allow for a single upgrade path without forcing enterprise customers to upgrade their motherboards as frequently as we have to on the desktop side, but it's clear that the performance of the processors will be limited by FSB bandwidth. Remember that Intel's MP solutions rely on a shared bus protocol, meaning that in a 4-way Xeon MP server all four processors must share the same 400MHz FSB. This essentially quadruples the FSB bandwidth requirements of the server, and the Xeon MP happens to be the processor that will have the least amount of FSB bandwidth out of all of Intel's enterprise CPUs.
  39. the mysteries of the universe REVEALED! by null-sRc · · Score: 1

    from the article as a pro listing in the conclusion: "Revealed that the limiting factor is more of the video card and less on the CPU" WHAT A SURPRISE!!! all this time i was playing DOOM3 on my dual xeon 2.8 server wondering why i was only get 1fps with my ATI XPERT@Work 8meg!!! Thank's for revealing the mystery guys ;) i guess they don't play games much..

    --
    -judging another only defines yourself
  40. As any TRUE geek could tell you... by Grog6 · · Score: 1

    .. Obviously the 8048 CPU in your keyboard can't keep up with the newer CPU's. They're only running about 4MHz or so. :)

    --
    Truth isn't Truth - Guliani
  41. Not the best article ...Aopen. by Anonymous Coward · · Score: 0

    " I've been planning to upgrade my computer at the end of this month, and have been keeping a pretty close eye on the 865/875 motherboard and chip performance reviews. This article didn't really enlighten me as much as the following Tom's Hardware reviews:"

    Well I'm in the same boat.

    I'm looking at an Aopen for $159.00 retail. In this soft economy it looks like a reasonable solution. Anyone know how well it works with Linux?

    1. Re:Not the best article ...Aopen. by Anonymous Coward · · Score: 0

      Do not buy anything from Aopen. Tech supports is terrible, all of the instructions are written in "engrish", and my ddr 400 corsair memory will not work with the board. It would be nice if I could get some firmware updates from aopen, but that is not going to happen. I RMA'd the board and am getting an ASUS for cheaper.

    2. Re:Not the best article ...Aopen. by Anonymous Coward · · Score: 0

      Well then I'm looking for a middle of the road, Athelon board that will take up to an XP. I will not accept an nForce board due to both the problems people have had, as well as the fact that you need proprietary binary drivers to get maximum use out of it. Integrated is OK as long as it can be disabled. $150-160 is the sweet spot. I expect to keep for about three years. Reasonably future proof with a modest amout of backwards compatiability. (accept both SATA and regular ATA. have PS/2 and keyboard port). Serial and parallel isn't necessary.

  42. Self Refletion. by Eric_Cartman_South_P · · Score: 1
    The article focuses on how the relationship between CPU and video card affect various aspects of performance.

    I just read that line in the article, got very excited, and then it hit me like a ton of bricks just how much of a fucking geek I am. :)

    That said, I love Intel improvements. Why? I wait three months, and buy an AMD ship that gives between 90% and 110% performance and costs a thrid less.

  43. Back to matching pairs of RAM modules? by timeOday · · Score: 1

    That is a pain. Can't they put two controllers and double the bus to each stick? Or something?

  44. How is this benchmark useful? by Theovon · · Score: 1

    If, for whatever reason, you were limiting yourself to only 800mhz FSB Pentium 4 processors, I can see how this article MIGHT be of some value. But what if you want to compare against Athlon? What if you want to compare 800mhz FSB against 533mhz FSB? I mean, I would like to know how a 3ghz with 533 compares to a 2.4ghz with 800. If you really have some shopping to do, this article is useless.

  45. Clueless by Wesley+Felter · · Score: 1

    DDR and QDR techniques are not overclocking; a 4x200MHz bus produces virtually the same bandwidth as an 800MHz bus would. (Notice how these systems have over 5GB/s of memory bandwidth.)

    AMD is already ahead of Intel with on-chip dual channel memory controllers and HyperTransport.

  46. Re:WHERE ARE THE WMD'S? by Anonymous Coward · · Score: 0

    Or the 20,000 killed after we didn't back them in '91, or the 10,000 civilians killed this time around or the UNKNOWN numbers of soliders blown to bits. What about them? No one cares, we went to war and won YAY! Wave your flags YAY!

  47. Re:Enough! You insensitive clod! by Rob+Simpson · · Score: 1

    I doubt it. At that price, I'd probably buy one if it were in good condition - it is difficult to run a lot of old software on modern computers. Plus, in this 31C heat, any computer that doesn't double as a blast furnace is appealing.

  48. Boycott Intel by Anonymous Coward · · Score: 0

    it seems like most people around here don't know yet another reason to boycott Intel .... http://www.inminds.co.uk/boycott-intel.html

    1. Re:Boycott Intel by Anonymous Coward · · Score: 0

      Oh great, more religous zealotry.. they're taking over Wal-Mart, now they're trying to take out Intel

  49. Re:Obligatory Pentium Jokes [From 1945] by ProtonMotiveForce · · Score: 1

    Someone's lost touch with the date.

    How about that wacky Gomer Pyle? And whoa, what about that Ollie North, eh, eh, hehe :)

  50. 6.4 GB or 5.97 GB? by Anonymous Coward · · Score: 0


    Max bandwidth =
    = (64 * 800.0e6) / (8 * 1024 * 1024 * 1024)
    = 5.97 GB/s


    Here 64 is number of data bits (8 bytes) on the bus.

    On the other hand, if you consider 1 GB = 1.0e9,

    Max bandwidth =
    = (64 * 800.0e6) / (8 * 1000 * 1000 * 1000)
    = 6.4 GB/s


    The second value is incorrect, (kinda like hard disk manufacturers definition of GBytes)

  51. Use the power that is there by tjstork · · Score: 1


    It takes SQL Server a Quad Xeon Machine to stuff slightly less than 10,000 inserts a second into a table. Yet, writing raw records into a file can happen hundreds of times faster than that.

    --
    This is my sig.
  52. Re:Enough! You insensitive clod! by HoboMaster · · Score: 1

    Man, I'd buy a 486 for $10... don't sell it on eBay though, because to find it, someone would have to specifically be looking for a 486, and the chances of that are slim... Just find a computer geek, almost any one of them will give you at least $10 for it, probably more if youask.

    --
    Remember kids, tin foil doesn't work, so use LeadHat.