Follow Slashdot blog updates by subscribing to our blog RSS feed

 


Forgot your password?
Close
typodupeerror
×
Hardware

Modding The Barton XP To A Barton MP 172

Posted by timothy from the who-needs-a-warranty dept.
Dr. Jackie Lee writes "Don't have the budget for an AMD Opteron? There's always a cheaper alternative with AMD's current Barton processors. In this article, we'll show you how to run these new Barton XP processors in SMP mode."
This discussion has been archived. No new comments can be posted.

Modding The Barton XP To A Barton MP More

Modding The Barton XP To A Barton MP

Comments Filter:

  • Hmf (Score:4, Insightful)

    by BJH ( 11355 ) on Tuesday May 13, 2003 @06:16AM (#5943320)

    Yeah, great - mod your XPs to act like MPs.

    Guess what? You can't guarantee they'll work - people on LKML have refused to help users who have done this, as it simply makes it impossible to determine whether problems are the fault of the kernel or of the CPU itself.

    This is one for the overclockers who couldn't care less about stability, methinks...
    • Most modders/overclockers realise the risks involved and problems which can happen. This is a risk which we take, just like the fact that our warranty will be of no use.

      If you can get an XP to run as an MP and fit two in a system board, and get it to work, then wouldn't it be worth the cost savings to an individual modder?

      Tim

      • Re:Hmf (Score:2, Interesting)

        by BJH ( 11355 )
        My point was, it depends on your definition of "work". Obviously, some avoidable hardware-caused instability is acceptable to you. It's not to some of us.

    • Re:Hmf (Score:5, Insightful)

      by beuges ( 613130 ) on Tuesday May 13, 2003 @06:30AM (#5943363)
      This is one for the overclockers who couldn't care less about stability, methinks...

      Or, maybe, its for those of us who can't afford the hugely expensive 'multiprocessor' chips, but would still like to be able to run a dual-processor system.

      As a programmer, it's always a good idea to have access to a dual-processor system, because there are all sorts of lockups that just wont happen on a single-processor system, but will lock your program up on a multi-processor one. Having a hugely expensive system just to debug an application isn't feasible - if you're able to fool a 'single processor' chip into thinking it's a 'multiprocessor' one, then you've got an essential debugging machine at a fraction of the price

      • Re:Hmf (Score:5, Informative)

        by damiam ( 409504 ) on Tuesday May 13, 2003 @06:41AM (#5943392)
        An Athlon MP costs about $40 more than the equivilent MP. You can buy an older ( 1.5Ghz, fine for testing) SMP-capable proc for next to nothing. Developers don't want a modded version of the top-of-the-line AMD chip just to test their threading.
      • if you can afford the risk, then you definitely can afford a proper MP system.

        as an aside, i ran 2 celerons until the p3's were affordable (and SMP capable). then i bought a pair spec'd at the max speed the main board could handle (933MHz) when ~1.4G were hot (i don't remember exactly, this is well over 2 years ago), and i'm still running the same computer as my primary machine. so i saved myself a lot of money by waiting some time instead of buying the top of the range p3's (which were 800Mhz when i actu

      • Re:Hmf (Score:5, Insightful)

        by cowbutt ( 21077 ) on Tuesday May 13, 2003 @07:27AM (#5943490) Journal
        Er, no, because then you end up wondering whether it's your bug or a hardware problem.

        A friend spent a long time debugging some maths in a game engine he's working on. First of all, he assumed it was his code, then after eliminating that possibility, assumed it was a gcc bug, then eventually found that the problem was caused by the motherboard's voltage settings being incorrectly documented and therefore he'd mistakenly configured it to supply the CPU with the wrong voltage.

        --

      • Re:Hmf (Score:3, Informative)

        by gl4ss ( 559668 )
        you do realise that in a single processor system athlon mp will be still be one single processor.

        you still need 2 of these chips and the motherboard to go with it.

        the point of this mod is that there is NO barton MP's available.
        • You can however get fairly inexpensive Athlon MP chips and Tyan do Socket-A dual processor motherboards for about 120GBP. What does Barton do that's better than the MP chips that came out just over a year ago?

          • Cache (Score:3, Interesting)

            by Andy Dodd ( 701 )
            IIRC, the key improvement of the Barton is that it has more cache (twice as much?) as the previous Athlon core. From what I recall, more cache is a BIG benefit in SMP systems, since you have two CPUs sharing the same memory. i.e. it's a surprise there aren't SMP-ready Bartons on the market because it's a perfect use for the extra cache.

            Either way, it is capable of doing significantly (10-20%) more per clock cycle than the previous Athlon core.

        • Re:Hmf (Score:2, Informative)

          by BJH ( 11355 )
          Not to take issue with you, but Barton-core MPs should be released next week.

          It's in Japanese, but this article [impress.co.jp] has details (and some photos as well).

      • Re:Hmf (Score:2)

        by Lumpy ( 12016 )
        Or, maybe, its for those of us who can't afford the hugely expensive 'multiprocessor' chips, but would still like to be able to run a dual-processor system

        yeah hugely expensive multiprocessor chips...

        Pentium III processors - cheap
        a P-III dual processor motherboard - cheap

        What was your point again? If your excuse isthat you want to use multiprocessor systems they are everywhere for dirt cheap as long as you can live without the bleeding edge. hell the Athalon MP chips that are 2Ghz (oh my god how slow!
        • --Seriously, could you please post a link to where we can find dual-processor motherboards that will run at 600MHz or better, for cheap? Plus the chips to populate them? I'm not being sarcastic, just would like to find an inexpensive way to build a MP machine.
      • cos my dual 1.2ghz PIII cost me £300 to build

        which is less than my gfx card cost!

      • Re:Hmf (Score:2)

        by tupps ( 43964 )
        The cost in going to a dual CPU system isn't the CPU it is the cost of the damn motherboard. Typically the motherboard is close to 4 times the cost of a single CPU board, although it is usually a high end or server board. The CPU's are about 2/3 the cost (each) of the Motherboard.

    • in case of slashdotting (Score:5, Informative)

      by abhisarda ( 638576 ) on Tuesday May 13, 2003 @07:26AM (#5943488) Journal

      here are few mirrors:


      Mirror1 - its got a popup. [curvedspaces.com]
      Mirror 2 - Mirror 2 [lycos.co.uk]
      Mirror 3 -Might get slashdotted. [netfirms.com]

    • Re:Hmf (Score:5, Insightful)

      by theLOUDroom ( 556455 ) on Tuesday May 13, 2003 @07:41AM (#5943525)
      Guess what? You can't guarantee they'll work - people on LKML have refused to help users who have done this, as it simply makes it impossible to determine whether problems are the fault of the kernel or of the CPU itself.

      Bah. You don't know anything for sure, neither to these LKML people. They're just trying to eliminate unknow quantities from their debugging (not like I blame them). For all we know right now (being that the article is slashdotted) there is no difference between the XP and MP except a jumper setting. I honestly wouldn't be suprised if it's true. Why go through all the trouble to fab seperate chips when you can just use external jumpers?

      This wouldn't be the first time someone could upgrade their hardware by connecting two contacts.

      What really matters is determining if there is any actual difference between an XP and an MP. If there is none, then this isn't just for "the overclockers who couldn't care less about stability". There are plenty of others who would love to save $40/cpu.

      Since you don't know either way. You're not really contributing much to the discussion. Yeah, without any testing I wouldn't use a modded chip for critical data, but it's not impossible to verify whether or not the modded chip is stable. "it simply makes it impossible to determine whether problems are the fault of the kernel or of the CPU itself" No, it doesn't. There's a simple solution. Test with both modded XPs and regular MPs.

      What needs to happen is testing. Without that, we don't really know how useful this mod will be.

      • Re:Hmf (Score:5, Insightful)

        by afidel ( 530433 ) on Tuesday May 13, 2003 @07:53AM (#5943555)
        Actually it's bridges on the package that make the difference, AMD laser cuts the SMP bridges on Barton core cpus that are intended to be XP's, these can be either failed MP chips, or much more commonly they are just XP's that were never tested for SMP compliance because the extra testing would take time and additional testing equipment. The problems introduced by running SMP XP's are that the quality of connecting the bridges varies greatly and is never as good as the origional connection that was laser cut, and the fact that you may have a core that AMD already rejected because it failed to work correctly in their SMP tests.
      • Test with both modded XPs and regular MPs.

        First, that assumes that all XPs are alike and will operate the same in the long run, which there's no guarantee. I imagine that a jumper might be the main difference in logic, but the MPs are also supposed to be more heavily tested, and even are binned so that they are lower power use. A chip that won't pass as an MP might be marked as an XP, so you might have a non-obvious fabrication flaw in the unused "MP section" of an XP chip.

        Secondly, how many people are

    • Re:Hmf (Score:2, Interesting)

      by Antiflag1980 ( 673031 )
      I have done this mod to over 250 XP's and have never encountered one that didn't work properly and get exactly the same scores in benchmarks as real MPs. If there is a problem with them running correctly is is not in the processor being labeled as XP it is in the user doing somthing else wrong, like using a single processor and then switching to 2 later on, in windows XP it doesn't always work right, unless you reinstall windows, or something else that they are doing wrong, or AMD's horrible buggy 760mpx ch

  • Urban myth - IBM upgrade (Score:5, Informative)

    by panurge ( 573432 ) on Tuesday May 13, 2003 @06:18AM (#5943327)
    It reminds me of the urban myth of the expensive IBM upgrade that consisted solely of adding a jumper on the clock subsystem.

    Seriously, though, what gives? Is the only real difference between dual and single processing jumper controlled or is the decision made on the basis of testing - in which case, what might go wrong if you mod these things to run as dual processors?
    Even given prevailing levels of cynicism about marketing departments, I would have thought that the potential for chip sales if dual processor boards became more common, would outweigh the loss of margins on the small existing DP business.

    Given the main use for multiprocessor boards, I'd be nervous about a mod that might screw data integrity.

    • Re:Urban myth - IBM upgrade (Score:5, Informative)

      by quigonn ( 80360 ) on Tuesday May 13, 2003 @06:25AM (#5943341) Homepage
      It's not really myth - in fact, when you own one of these big IBM machines (zSeries, pSeries), you can call IBM and buy the unlock code for the additional CPUs that are inside your machine but deactivated/locked. You can also only unlock it for a day or a month.
      • I remember also that Apple had a LCIII, pardon me if I don't remember the exact speed specs, but the motorola chip used was rated for a higher speed then the clock chip was set too. I remember some deal at apple, they didn't really want the LCIII to compeat with some other lightly higher end 68030 mac, so they just popped in a slower crystal.

        There are lots of cases in the computer industry where companies release a high end product, and dummy up the low end product dispite being the exact same thing. I *
        • NVidia GeForce and Quadro cards both use the EXACT same chip. They only differ in the PCI ID. Hence the ability of SoftQuadro to do its magic with driver hacks.

          ATI does it too (9700 vs 9500), although in this case many 9500s failed testing as 9700s and will fry if the dead pipelines are re-enabled. (In this case, it's both about providing a lower-end part and about salvaging "reject" chips - 9700s with bad pixel pipelines had those pipes disabled and then were marked as 9500s.)

        • Reminds me of a story about a friend of a friend who worked at Intel.

          In the days of the Pentium III clawing its way up towards 1 GHz to compete with the early Athlons, this guy had grabbed one of the PIII's they were testing and binning and used it for his own workstation at 1 GHz, many months before the bulk of the reliable yield could be rated that high.

          Thermal issues probably limit overclocking a lot more these days than a few years ago, though.

        • ...the motorola chip used was rated for a higher speed then the clock chip was set...

          That was the Mac IIsi [lowendmac.com]. Ran at 20MHz, but the whole system was rated for 25MHz. Management decided they didn't want it to compete with the IIci. A bit of soldering and a new clock chip, and you have 25% better performance.

          (Is it 'overclocking' when it was designed to run at that speed in the first place?)

          • My mistake, i'm not an expert on vintage macs.... but never the less a good example of a product that was dummied up for the sake of marketing.

            "Buy this IIci, not this lame IIsi. It's got expandability and a 25% faster clock".
        • This is true in other industries as well. Automotive comes to mind. Lots of cars have rev limiters. Beyond that, consider that the 280ZX and the Datsun Maxima had the same motor, but the Maxima was detuned; Ditto for the Camaro IROC Z28 and the Corvette at the time (late eighties.) My 240SX has an artifically low rev limiter and a crappy engine compared to the Japanese market besides so it wouldn't "compete" with the 300ZX (more like blow it out of the water.) Etc etc. I would bet than in every industry fro
    • Are you sure that that is an Urban myth, I'm sure I ve heard it from my father that an NCR tape upgrade was just a jumper setting. Then I believe that an upgrade for Phoenix disc (A couple of MByte the size of a washing machine)was just a switch setting.


      Anybody else identify the expensive jumper setting upgrade.


      slashnik

    • The best take on this issue I heard was a memory upgrade for VAXen that consisted of removing a jumper: the boxes were delivered with the RAM maxed out, but some of it disabled. You could either remove the jumper yourself and void the warranty, or pay for the upgrade and have a Digital engineer to drop by to remove the jumper for you, but keep the warranty.

      Truth/myth? Anyone got any sources?
    • I remember when 20MB MFM hard drives were kicking about on old IBM XT's and their clones. You could fiddle with some of the settings on the controller card and set them up as RLL drives instead. This used to give you a couple of extra megabytes, a big deal in those days.

      I cant remember exactly how we did it now, but I do remember it was different for each controller. You had to load up debug to execute the setup software on the controller board, something like g=c800:5

      Some drives took it OK, others just
    • Back in 1970 or so, UC Berkeley had two CDC 6400s. One had an extra cost instruction used to make process swapping easier, and was used for an experimental timesharing system (which flopped miserably eventually). I had an old program I was fooling around with, FORTRAN, which self modified itself gradually to execute that missing instruction and die, and marked the card deck specifically to only run on machine "A" to avoid the instruction ... it crashed the machine. Turns out the instruction was actually
  • and it's already down. I've been trying to load it since 6:12am.

    Apparently the /. effect is an early riser. Anyone snag the text before she died?

    • Got the text (Score:5, Informative)

      by taff^2 ( 188189 ) on Tuesday May 13, 2003 @06:40AM (#5943386)
      What We Know Thus Far

      More than a year ago, we reported a hack which you can apply to current Athlon XP processors to enable them to operate as Athlon MP processors. The hack was fairly simple and it involved connecting a certain trace on the Athlon XP processor to allow motherboards to recognize them as an Athlon MP. The trace has been cut in the factory so it's just a matter of connecting them via soldering (if you're brave enough), or just putting conductive paint on them.

      The response we received from fellow users were phenomenal and from what we know, many have tried the hack. Although most were successful, some were not that lucky. Here are a few reasons which we can offer at this point in time :-

      # You'll need to ensure both processors are properly modded so that the motherboard can successfully detect and operate the processors in SMP mode. Clean the contacts with acetone or isopropyl alcohol before applying the conductive paint.

      # Certain motherboards would only work with the hack since certain users reported that they were unable to get their SMP setup running even though the processors have been properly modded.

      # Only certain BIOS versions would allow the board to run in SMP mode. For example, we tested the hack on MSI's K7D Master which had no problems working with BIOS version 1.3 and 1.4B3. Newer BIOS versions somehow did not allow us to run the processors in dual mode even though the processors have already been physically modified.

      After our first article, a lot of emails came pouring in and most users were asking if the hack could be achieved with a Thoroughbred core since our initial article was based on the Palomino core. Well, of course it would and we'll be showing some results with a dual Thoroughbred-B core in the later pages. What we've done here is taking another step further with the Barton core. What's interesting with the Barton is that it's now featuring a huge 512KB L2 cache running at full speed. Get a pair of that running in SMP mode, and you'll get a pretty decent workstation-class system going at a really affordable price.

      Now, I'm sure most of you are interested to get your hands on a dual processor system based on two Athlon XP 2800+ (Barton). Find out how you can do that in the next few pages.
      Modding the Barton Core

      The process in modding the Barton core is just about the same as what we've done with the Palomino in previous articles. All you need to do is just to connect the rightmost bridge in the L5 row. That should probably get the processor detected and running in dual mode. However, that's not the end of the story yet as you'll still need to modify a couple of things on the processor to get it running up to speed.

      We know that all the Barton cores are made for 333MHz FSB systems and their multipliers were set to run at its predefined frequency. If you put a Barton processor in older motherboards supporting up to 266MHz FSB, you'll probably get a severely underclocked processor and that's probably what will happen if you have the Barton XP running in an AMD 760MPX motherboard.

      Naturally, the next logical step is to modify the Barton XP's multiplier. Since most server boards based on the AMD 760MPX chipset has no multiplier adjustments for anything more than 12.5x (at least for the MSI K7D Master), we'll have no choice but to hardwire the multipliers ourselves. OK, here are a few rules which we should observe :-

      # The default frequency of the Barton XP 2800+ is only 2.083GHz (12.5 x 166MHz), so we should try to target a multiplier with a lower resulting frequency at 133MHz (266MHz DDR) FSB. It should at least ensure that your processor will boot-up after the modification. Since there aren't any 15.5x multiplier allocated for the Athlon XP/MP core, we have chosen to hardwire the multiplier to 15x on both the chips.

      # Since the Barton cores come with more transistors allocated for the larger 512KB L2 cache, we should see an increase in power consum
    • you may not yet of heard of the twin phenomena of "Europe" and "Asia", I understand these mysterious forces can contribute to network traffic somewhat.

  • Possible DMCA violation? (Score:5, Interesting)

    by Sheetrock ( 152993 ) on Tuesday May 13, 2003 @06:19AM (#5943330) Homepage Journal
    As with the Athlon XP, apparently this processor can be trivially modified to function as an MP chip. I'm curious, however, as to whether this could be perceived as a DMCA violation. If companies refilling ink cartridges can be sued these days...

    But another thing that troubles me somewhat is the idea that, perhaps, people are cheating AMD somewhat by doing this. Maybe the sales of Barton XPs are at a loss, but built with the same die as the MP because AMD doesn't want to blow the cash on retooling. If cheap CPUs are being subsidized by the sales of Barton MPs, aren't people who modify their CPUs performing the equivalent of, say, buying Windows XP Home and stealing Windows XP Server? It seems like a bit of a fuzzy issue, although with the popularity of P2P hardly one that's going to trouble most computer users.

    • Re:Possible DMCA violation? (Score:5, Insightful)

      by fadeaway ( 531137 ) on Tuesday May 13, 2003 @06:22AM (#5943335)
      You trade your warranty for the modification. Sounds fair to me.

      There's no EULA for hardware.. yet..
      • You'd think that would defeat the logic in the lexmark/static control ink-refill case. But, the lawyers are arguing along the lines that the ink chip is an access control device for the printer's firmware (which is software). I'm sure that you could just as easily argue* that the jumper controls access to the processor's multiprocessor microcode and state machines.

        * I still think it's full of crap, but you could still argue
    • I'd doubt that there would be any attempt at prosecution, as the perceived loss isn't as great. At least you've still bought a processor, albeit a less expensive one. If you should happen to toast the processor during the mod, you're prolly gonna buy another one.

      DL an mp3, and the powers that be have already decided you will never, ever buy the album.
      • Actually, they wouldnt necessarily prosecute those individuals who unlock MP, but they would definitely prosecute those who published the information online.

        These MP hacks are circumventing an access control mechanism (in this case, access to MP functionality which you didn't pay for) which is a clear violation of the DMCA.
    • If cheap CPUs are being subsidized by the sales of Barton MPs, aren't people who modify their CPUs performing the equivalent of, say, buying Windows XP Home and stealing Windows XP Server?

      I prefer to see it as the equivalent of buying an XBox (loss leader) and buying no games, but installing GNU/Linux on it. Anyway, last I heard, the MS Windows operating systems were all sold at a huge profit.

      You could also say that it's like buying a lossmaking piece of Microsoft software, and using Wine to run it via Li

  • Personally, I'm holding out for a mod that lets me turn my Barton XP into a Barton GP, because I really want to buy that next armor class.

  • Reasons for SMP (Score:5, Insightful)

    by lingqi ( 577227 ) on Tuesday May 13, 2003 @06:26AM (#5943342) Journal
    not trolling, but a serious question, coming from a ex-SMP-user:

    why SMP nowadays?

    Most OS, including XP, is now reasonably stable to the point where multitasking, including buring a CD, while doing something else is OK. (not to mention that CD burners have buffer under-run protection *anyhow* nowadays) I know this because my laptop, which is a measly P3, can handle all of the multitasking I do, given that I have pumped up the amount of RAM.

    If I was doing stuff for school / research, Mathematica, Matlab, Maple, Spice (ok not 100% sure on spice) are all single processor only. And to be honest, if you are running a 3-day simulation, you really don't want to be playing UT at the same time on the same machine ANYWAY, so that's kind of a moot point. (you should probably not be playing UT period during such times, but that's a time management story that i won't get into)

    If I was doing real work like rendering stuff, I think real work deserves a real SMP system, i.e. with a warrenty.

    I mean, SMP has a pretty hefty price overhead (motherboard, memory if you want ECC, and the extra CPU, heavy duty power supply, another set of heatsinks etc), not to mention that the motherboard / chipset technology is usually a few monthes to two years behind the cutting edge stuff...

    so, what convincing reason do a person have for using SMP right now (especially a ghetto-rigged one)?

    • Re:Reasons for SMP (Score:2, Interesting)

      by Sheetrock ( 152993 )
      It's mostly about the hack value. Everything I'd want to do properly with SMP I wouldn't trust my soldering skills to, such as compiling, rendering, or serving web pages, but I might try something like this if I had spare hardware lying around.

    • Re:Reasons for SMP (Score:2, Insightful)

      by kaamos ( 647337 )
      It really has to do with the bragging rights. I agree, right now I Maple all the way trough my math courses and it is plenty fast on a p3 1.13.

      The only way I could justify this to myself is if I picked up 3dsmax again ON A PERSONAL BASE, ie. not getting paid for it, just for fun, and if the motherboards were cheaper, then it could be fun just for the hell of it.

      On a side note, how is the volume of sound generated by smp systems, you being an ex-SMPer? I was thinking maybe dual vantec tornados 7 turned all

    • Re:Reasons for SMP (Score:5, Insightful)

      by Brento ( 26177 ) <brento@@@brentozar...com> on Tuesday May 13, 2003 @06:43AM (#5943393) Homepage
      why SMP nowadays?

      Sadly, there's still a lot of dog-slow apps. Example: ask Macromedia Dreamweaver MX to test a stored procedure with a hundred parameters, and it'll freeze for 30-90 seconds on all of the P4's I've tried it on, pegging the CPU at 100%. On dual-CPU boxes, that means you can still be productive with the other CPU, and do things like whine on Macromedia user groups while it runs.

      SMP also helps a lot if your box is both a testbed and a design system: mine runs SQL Server, IIS, and I do my design on it, so while I'm working, there's a ton of processes running.

      Don't get me wrong, I still buy the "real" SMP systems with warranties, but just pointing out why I'm doing it, and I'm not rendering.

      • ask Macromedia Dreamweaver MX to test a stored procedure with a hundred parameters, and it'll freeze for 30-90 seconds on all of the P4's I've tried it on, pegging the CPU at 100%.

        Have you tried increasing the nice [upenn.edu] value? Wait a minute...were you playing around with real time schedule policies [systemsix.com]? No wonder your system locked up!

        Seriously, get an OS with a real scheduler, and you won't need SMP. At least not because of stupid problems like this.

    • why SMP nowadays?

      For Linux: not so much of an issue... for Windows: another story.

      To my experience Windows always manages to slow down to a crawl when I'm moving large files around, it just somehow wastes all my processor time by staring at the harddisk.

      With SMP there's another processor to keep the UI responsive. My friend was using a dual 200 MHz PPro for this very reason until quite reasently.

      It might not be a good reason, but it's a reason none the less.
      • ...there goes another kitty...

    • Re:Reasons for SMP (Score:5, Interesting)

      by afidel ( 530433 ) on Tuesday May 13, 2003 @06:52AM (#5943412)
      Mutlimedia, which is the only thing other than games that is pushing the PC at this point will generally get nearly 100% speed improvement from SMP. For instance I run a virtual studio using Reason and have all of my effects and the OS on cpu 0 and the program itself on cpu 1. This allows very low latency and the ability to run more effects without problems. Media encoding (where do you think all those mpeg4 rips come from =) is another task that is frequently done with SMP machines. Other than media though I really can't think of something that a home user would run that would need SMP.
      • Transcoding of media from one format to another (DivX rips for example) is easily partitioned between multiple CPUs. There's a decoding process feeding data to a seperate encoding process. In many cases, between the two there's some scaling and filtering being done, which can also naturally be split to another processor. In some cases the app must be SMP-aware, but in others (AVISynth for example), you'll have these processes naturally partitioned to begin with.
      • I do xvid encodes on an AMD MP box with Linux. I use DVDRip and set all my encodes to run at a nice level of 19. My system is very usable for listening to music, watching movies, webbrowsing, etc. all while I'm encoding up to 3 videos at the same time. And I get between 20 and 60 fps depending on the media and settings.

        Still working out minor timing issues and correct deinterlacing from NTSC sources.
    • If I have everything on one processor, the different calls make the cache nearly useless. With SMP, I can put mathematica onto a processor of its own. With cache allowing for lightening fast ram calls, the speed increase is unbelieveable.
      • With SMP, I can put mathematica onto a processor of its own

        Assuming you can tie a process to one particular CPU. You can do this on Solaris and TruUnix, but neither Linux or *BSD has the processor affinity feature.

        • windows has it (for all versions of windows that support SMP), and Linux does as of 2.5.8-pre3 and can have it sort of hacked on to earlier kernals via This [sourceforge.net] module.
    • why SMP nowadays?
      If I was doing stuff for school / research, Mathematica, Matlab, Maple, Spice (ok not 100% sure on spice) are all single processor only.

      Well, I'm doing programs for research, in biomagnetism field. We have to display up to 1.5 GB of data (in different ways: magnetic map display, channel display, frequency spectrum display) and make different kind of math analysis on them.
      The programs need to be as fast as possible (faster programs mean more subjects analyzed a day), so we write our own
    • Much of braggin' rights, and for the feel.
      I run debian and vmware 4.0, under which there is w98se. I really don't think it could run this well on a single processor, as win98se tends to peg one processor.
      Also make -j4 make c++ coding much tolerable (gcc3.2 seems to be rather slow in compiles, so everything adds, and my coding habit is fix->test, so I recompile lots.)
      As for the feeling. I've never felt desktop like this. I've used 2.8GHz P4 w2k/Linux, but it really didn't cope well with high loads and
    • You are right on many points.

      I have a PIII Xeon, 500MHz, and since upgrading it to dual processors it really does nicely.

      A lot of programs are multithreaded.

      not to mention that the motherboard / chipset technology is usually a few monthes to two years behind the cutting edge stuff..

      That's because for one, AMD takes an extra month or so of testing to qualify a CPU and the parts.

      Because the market is so small for them, only AMD makes DP chipsets for Athlon chips and they don't seem to keep up with VIA
    • Compiling stuff will go roughly twice as fast with two processors (at least for difficult translation units, eg C++ with templates and stuff). Although in that case you could just buy two machines and use distcc / Mosix / doozer / etc etc, so I guess the question still stands. An SMP motherboard with two pricier processors and a bigger PSU is a _bit_ cheaper than two separate boxes I expect.
    • No. 1 reason for an SMP desktop: There's *always* an IRQ available (rule of thumb).
      This results in a more responsive desktop under severe load. The reason for this is exactly that most apps don't use both processors :)

      Personal reason no 1: I am using Gentoo, and compiling is _much_ faster naturally.

      Personal reason no 2: Java development. Runs a shitload of threads, and applications feel much more responsive.

      Apps that actually benefit from SMP: not many. But everything related to Graphics and Video (from

    • Re:Reasons for SMP (Score:3, Informative)

      by rainer_d ( 115765 ) *
      why SMP nowadays?

      A good question. After my dual p2-450 HP Kayak died, and I had the need to buy a new system, I eventually bought another Dual CPU box (DIY Socket 604, 2.4 GHz), mostly because I intend to make heavy use of vmware (workstation & gsx-server).
      I must say that one reason to go SMP was that I wanted a system with little or none of the "BS"-parts that seem to end up on consumer motherboards (RealTec-NICs, max 1.5 GB RAM, 6 USB-slots, Firewire etc - you name it) - and which I have no real us

  • Bad Idea (Score:4, Informative)

    by Anonymous Coward on Tuesday May 13, 2003 @06:30AM (#5943360)
    This is NOT a good idea. The MP cores are subject to much more stringent timing variance restrictions. Stick a couple of XP cores in their place, and you'll be getting random RAM errors (actually RAM->CPU bus errors), as well as slowdown due to cache coherency problems.

    • rated informative, but (Score:2, Informative)

      by lingqi ( 577227 )
      I have *personally* never saw any hard evidence about this. anybody have a link / site / whitepaper / definitive source?

      Now, it is possible that if the SMP portion of the chip have a failure (stuck transistor in a non critical portion, say), they can still sell it as a XP chip, but I don't see why it would require different *timing restrictions*. Believe it or not, it's still just moving data to and from the memory, and having a dedicated path to each CPU even makes this easier.

      Granted there are memory co
      • Having done this with Athy's and Intel chips the ones that are designed to do this can be a pile of problems at times. Very minor differences can cause issues on the intel side liek same s revision but one malaysa and one philipean procs sometimes will not work stabaly with each other. Granted this in intel and that shared bus can be a big issue. Now with athys SMP systems tend to be more picky about there memory to begin with and modded XP's do seem to instigate this effect (8 procs modded 3 had issues)
    • I have a number of AMD SMP systems using non-MP cores and they work just fine, no slowdown no random ram errors no crashes at all.

      You're just making shit up ("cache coherency proclems") to sound like you're informed. Oh wait, you're anonymous coward. So your post is dubious BS by default.

  • Anyone with a guide to PC hardware? (Score:3, Interesting)

    by pchown ( 90777 ) on Tuesday May 13, 2003 @06:37AM (#5943376)
    Does anyone know of a good introduction to modern PC hardware, the different processor cores and so on? I don't really know the major differences between Barton and Palomino. I'd like to know, partly for curiosity, but also because when I build PCs I may otherwise end up with something sub-optimal.

    I'd also be interested in something that explains the structure of a PC system. What is the Northbridge, how is the PCI bus coupled to the processor, that sort of thing.

  • screw that (Score:2, Insightful)

    by joss ( 1346 )
    Now that P4's have hyperthreading I think it makes much more sense to get one of those. That way you dont need a tornado in your box to keep things cool and you get 80% of the benefit of dual processors.

    • Re:screw that (Score:1, Informative)

      by Anonymous Coward
      Hyperthreading does not work if your process is set of tight arithmetic loops like in matrix operations.

      Hence, it is almost useless for scientific applications.

    • Re:screw that (Score:4, Informative)

      by Anonymous Coward on Tuesday May 13, 2003 @07:20AM (#5943470)
      Now that P4's have hyperthreading I think it makes much more sense to get one of those. That way you dont need a tornado in your box to keep things cool and you get 80% of the benefit of dual processors.
      Except that HyperThreading has been benchmarked to improve performance anywhere from 0% to 25% at best, which is hardly as effective as SMP for true multithreaded applications. The benefit of HT is that this gain is essentially "free".

      However, as the only P4 CPUs with HT enabled are the most expensive grades, nobody is choosing HT based on value. Maybe a year from now it'll be a compelling option, but right now it's really a non-starter.

    • Re:screw that (Score:2, Informative)

      by TeknoHog ( 164938 )
      If you're running effectively two processors on one chip, the heat output will increase as well, and you'll really need that tornado to focus all your cooling on one die. I'd rather have multiple CPUs at lower speeds for more efficient and quiet cooling.
  • I've run SMP systems, and I really do prefer them for my workstation. Lots of performance and stablity, and if I do something a little creazy, like compile a 2.5 kernel, it doesn't take forever! Plus, it has that Cyberpunk/Geek factor. The same reason that rackmount phones. Add to that, Dual CPU machines, like Sun SparcStations 10/20s, and Pentium I's and Pro are really cheap and fun to play with.

    O.K., I admit, I'm a Geek...but who here isn't?!?!?!

    BTW, the blooddy site is slashdotted, at 7 am! Wow.

    ttyl

  • hardware fix (Score:1, Funny)

    by Anonymous Coward
    My hardware fix for today has already been met by slashdot. ATI 9800, dual proc barton (3000+'s of course), and all the additional parts to support it all (gotta have a gig of memory, fans and heatsinks galore, cool case mods to make it all the more cool-looking, 500W power source, gotta throw in the audigy 2). If only I had a lot of money and a lot of time on my hands....but instead i'm a broke slashdotter.

  • Death Wish (Score:4, Funny)

    by abhikhurana ( 325468 ) on Tuesday May 13, 2003 @07:18AM (#5943462)
    Boy, some guys sure have a death wish... I mean what else can explain a person posting a story, hosted on his own website, on slashdot?

  • Re: (Score:2, Funny)

    by account_deleted ( 4530225 )
    Comment removed based on user account deletion
  • already /.'ed at 6:30am ??!!

    Dam, you have to get up REALLY early to beat these damn geeks to the good stuff....

  • Chipset ? (Score:1, Interesting)

    by Anonymous Coward
    Ok. Especially when talking about the brandnew 400FSB Barton: Which Chipset to run them on ?

    In genereral, AMD still hasn't solved its chipset problem - worse for the MPs.

    Tons of complaints from users of high end soundcards or videocards (not graphiccards) with SIS and especially VIA based Boards likely due to lousy PCI implementations.

    AMDs own chipsets are terribly outdated and comparatively slow

    nforce2 has no decent Linuxsupport (probably the same for *BSD). A least, as of yet.

    Given the lack of heatsp

  • Questions, external mod possible? (Score:4, Interesting)

    by zakezuke ( 229119 ) on Tuesday May 13, 2003 @08:45AM (#5943810)
    If it's a matter of just bridging a circut trace in order to reconise an xp chip vs an mp chip, isn't there perhaps external to the chip layer solution?

    Either via a motherboard mod.... or CPU socket adapter.

    I ask because the value of the XP chips starts at about $123 per unit (OEM) , where the motherboard starts at roughly $160 per unit. I'd rather modify a motherboard then a pair of chips. Motherboards are bigger, though surface mount, more able to accept wires rather then silver paint.

  • Y'all are forgetting something (Score:3, Insightful)

    by AntiBasic ( 83586 ) on Tuesday May 13, 2003 @02:41PM (#5947420)
    Techies all know that these two CPUs are the same, right? They're wrong.

    It's true that the Athlon MP and XP are built as identical cores at the fab. In fact, you can tweak your L5 bridges on your Athlon XP to enable support for dual processors. Nonetheless, this still doesn't mean they're the same thing. Have you ever wondered why the Athlon MP lags behind the Athlon XP in megahertz? The flagship Athlon MP is only at 2800+ while Athlon XP is at 3200+...

    Athlon MP's are binned Athlon XP's. No two CPU cores fabricated are absolutely identical, and the Athlon MP represent AMD's best product. The goal for the MP line of chips is to have lower temperatures for the same megahertz. This makes their clock ramp-up fall behind the Athlon XP line. Stability is the obvious benefit, however recall also that rackmount servers don't have the same exotic cooling solutions that your desktop may have. Binning is one way AMD ensures a superior product for multi-processor systems. The other way AMD ensures the MP lineup is reliable is that the first Athlon MP 2800+ is going to have a later CPU stepping than the first Athlon XP 3200+.
    • If your claim were really true, they would be using the Mobile Athlon core for the MP processor line, but they're not. The mobile athlon core runs even less temp/power for the same mhz than the MP, but it's cheaper.

      So your argument invites purchasing mobile athlons and using them for MP. In fact, your argument pretty much demands it.

Slashdot Top Deals

"Sometimes insanity is the only alternative" -- button at a Science Fiction convention.

Close