Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Intel Hardware

Intel Quad-Core Price and Performance Showdown 115

ThinSkin writes "The folks over at ExtremeTech have had enough time on their hands to benchmark Intel's entire quad-core lineup to determine which has the best performance for the dollar. While prices range from $183 to $1399, the real bargain is with Intel's latest Core i7 architecture which outpaced many other more expensive processors. For comparison's sake, Intel's fastest dual-core CPU was thrown into the mix and was, at times, not even competitive, which suggests that we're beginning to see more and more multi-threaded applications take advantage of four cores."
This discussion has been archived. No new comments can be posted.

Intel Quad-Core Price and Performance Showdown

Comments Filter:
  • Whoopee (Score:1, Redundant)

    by TheLink ( 130905 )
    One of those multipage reviews again.

    Anyone have a summary?
    • Re:Whoopee (Score:5, Informative)

      by Z00L00K ( 682162 ) on Tuesday December 16, 2008 @09:42AM (#26132397) Homepage Journal

      Summary in short is that the Core i7 series is the way to go unless you just run office apps in which case the dual-core processors are sufficient.

      The Q-series seems to be expensive and slow compared to the Core i7. And unless they can make a considerable price reduction on them it's no idea to select a Q-series processor.

      • Re:Whoopee (Score:5, Insightful)

        by Retric ( 704075 ) on Tuesday December 16, 2008 @10:43AM (#26133065)
        It's also wrong, I have a Core 2 Q6600 the slowest chip in their roundup and I never tax all the cores. Games don't need that much CPU and I can encode video while playing games so I don't care how long it takes.

        Unless you are spend all day running benchmarks it's silly for most people to spend all that much on a CPU.
        • I use to have 7 computers. I did a lot of work for World Community Grid and was averaging around 20 results a day. I purchased a q6600 motherboard and soon found that it could do more than 10 times as many results as my slowest computer(ironically a celeron at the same speed of 2.4GHz) I now have 4 quad computers and averaging 40+ results a day and have experienced a significant reduction in my electricity bill.
        • I was playing Battle for Middle Earth (an RTS) and was trying to find out what component was affecting performance. I had a desktop with a mediocre processor and good graphics card and a laptop with a good processor but mediocre graphics card.

          The high res, high quality graphics staggered the laptop, but showed fine, in small quantities, on the desktop. But when I played on medium graphics settings on both computers, when actual battles came and there were large amounts of units on the screen, it was the l

          • by Retric ( 704075 )
            At some point in time processing power made a difference when gaming, but saying a single 1.6GHz P4 had troubles says little about a quad core cpu at 2.4Ghz with 8Mbytes of cache. For a tiny fraction of people buying a 20,000$ workstation is a great idea, but for most things like playing FPS at normal resolutions it's just not that taxing for modern CPU's.
      • Re:Whoopee (Score:4, Interesting)

        by Anonymous Coward on Tuesday December 16, 2008 @11:36AM (#26133681)

        Nobody in their right mind should get an i7. A 30% performance-per-clock increase over the Core 2 series is not worth doubling the cost of the CPU and motherboard. DDR3 is also more expensive than DDR2. On top of that, Intel are getting into the power-sucking height of the Pentium 4s again; the Core i7s have a TDP of 130 watts. For any desktop use - including highest-end game machines - anything another other than a Core 2 Duo is just a waste of money.

        • Yeah, you can't ignore the total platform cost. I can get 8GB (eight!) of DDR2 for under $100. I'll take that, a budget Intel mobo (P43?), and a Q6600 over an i7 system right now.

          And if gaming is your thing, the graphics card is what matters.

        • I just got 20,676 in 3Dmark Vantage. I can run crysis max settings 1920x1200 with 4x AA and i get 40fps. before I couldnt even have everything max at 1920 with no AA, same GPUs but a e6700 dual core. so it made a huge difference for me and was worth it.

          Why does everyone on slashdot act like high end gaming doesnt exist or is unimportant? I thought this site was for nerds...

      • by Idbar ( 1034346 )
        My summary is: I don't have a clue because the memory configuration is ambiguous and incomplete for the i7.
    • Re:Whoopee (Score:5, Informative)

      by Hatta ( 162192 ) on Tuesday December 16, 2008 @11:39AM (#26133723) Journal

      While we're bitching about the format, why the hell are they connecting the points on the line graph? Their X axis is meaningless, the order of those chips is arbitrary, so the slope of the line connecting the points is absolutely meaningless.

      This is what bar charts are for.

      • Re:Whoopee (Score:4, Informative)

        by Mike1024 ( 184871 ) on Tuesday December 16, 2008 @05:53PM (#26138891)

        While we're bitching about the format, why the hell are they connecting the points on the line graph?

        Or, given that we're comparing price and performance, a scatter plot.

        I decided to replot some of the graphs properly. Here are the results [uwcs.co.uk].

        • Thank you! I don't know anyone that compares price/performance using a double line chart.

          Well, I guess that's a lie. Whoever made up the idiotic charts for this article is probably a first.

  • If going with the i7-920 is it better to go with 1066MHz ram or 1300Mhz? I plan to overclock the chip since lots of people have had great success with air cooling.
    • Always depends how much you have to spend on it. Buy a MB that can support higher memory clock in case you want to upgrade your mems in the future.
    • Re: (Score:2, Insightful)

      by Tiger4 ( 840741 )

      I just gotta say it. Overclocking is sooo lame. For a 5% improvement in performance you risk losing 50% of service life and reliability. And then end up buying another cpu, which totally blows the performance/price comparison all to hell. You might as well just spend the money increment more and buy the next one that actually has the reliable performance the overclocking gets you. Some people just can't be comfortable unless they are living on the edge, even artificially.

      But of course, you do get techi

      • Wrong. If done properly (especially the cooling), OC'ing does not reduce the useful life of a processor. It is common knowledge that processors come factory clocked lower than its limit.

        The problem is the idiots who just put the open case in front of an AC vent, shoot the voltage settings in the BIOS all the way up, and pray for the best.
        • Re:RAM Question (Score:4, Insightful)

          by Tiger4 ( 840741 ) on Tuesday December 16, 2008 @11:07AM (#26133331)

          "If done properly"

          But that is the whole problem. I understand the fun of fiddling around with a system and experimenting with something new. If you want to pay for that kind of fun, go for it. We all know you can push something up to the ragged edge, and over, and have a good time doing it.

          But really, it should just work, reliably every day, and without having to watch a temperature gauge or worry if the water pump is going to give out or chase dust bunnies out of the heat sink. And by the way, how much did you save when you bought the cheap processor AND the cooling equipment to keep it alive?

          • by ifrag ( 984323 )

            From your comments, I'm guessing you haven't tried to up-clock anything even relatively modern. The term overclock itself is sorta silly since the advertised frequency is based on some paranoid assumptions.

            If we're talking about Intel chips here, they have always had relatively good protection against burning themselves up. In the past I know AMD has had some chips where you actually were taking a risk, but this discussion is about Intel. These newer chips are basically idiot proof, you probably could no

            • Suicide Question (Score:3, Insightful)

              by Ostracus ( 1354233 )

              How do you "turn off" electromigration? [wikipedia.org]

              • Re: (Score:1, Insightful)

                by Anonymous Coward

                Electromigration-aware design [wikipedia.org].

                Intel has already been experimenting with fabrication & design techniques to improve electromigration performance [intel.com].

              • by slittle ( 4150 )

                I first heard about the electron migration problem as a reason for not overclocking back in the 386 days.

                And yet my Celeron 300A has been running stable (first under OS/2, then XP) at 450Mhz since I bought it, 10 years ago.

                Record uptime was ~650 days under XP, before a disk failure got it. And that disk was essentially "DOA" (visible bad sectors) but rather than RMA it I decided to see how long it lasted (obviously it's not a very important box).

                PowerMax and NTFS/chkdsk recovered the initial damage and mar

          • This and your previous post indicate that you haven't updated your opinion about overclocking in quite some years. This is a time where Intel have no high-end competition and routinely sell mainstream chips that are easily as capable as ultra-high-end models, but aren't sold that way simply for lack of demand.

            You are incorrect on every point: Overclocking isn't even remotely dangerous, nor is it difficult. Chip failures never happen; the worst case scenario is that you must lower your overclock or reset
            • Re:RAM Question (Score:5, Informative)

              by Detritus ( 11846 ) on Tuesday December 16, 2008 @04:38PM (#26137945) Homepage

              Like I tell people at work, if it doesn't have to produce the correct results, we can make it run as fast as you wish. Just because your system seems stable, doesn't mean that some obscure part of the chip isn't failing in a subtle manner. Intel has insanely expensive test jigs to ensure that their parts meet published specs at their marked speed. You have what?

              For games, who cares. For real work, it's absolutely unacceptable.

              • Re: (Score:1, Flamebait)

                by Spatial ( 1235392 )

                Intel has insanely expensive test jigs to ensure that their parts meet published specs at their marked speed. You have what?

                I have Intel's linpack test. You know, the one Intel use for torture testing their CPUs. I think it might suffice for my home use.

                For real work, it's absolutely unacceptable.

                Nobody was suggesting anything of the sort. Overclocking is the domain of home users, this is so banally obvious it doesn't even need to be stated. What kind of retard overclocks their workstation? Nobody.

          • Re: (Score:3, Informative)

            I understand the fun of fiddling around with a system and experimenting with something new.

            no you don't, as demonstrated by the rest of your post.

            not all processors/memory are made equal. some handle being overclocked better than others (certain models of high performance memory are designed by the manufacturer specifically to be overclocked). CPUs rated at different clock speeds within the same family are usually manufactured the same way, and often from the same wafer. manufacturers separate processors i

          • There's absolutely no reason NOT to overclock Core2s. The small investment required to avoid cut-rate bare-bones parts gives you faster clock at stock/lower voltage with lower temperatures and much higher reliability.

            For goodness sake, you're using a CPU that has to perform reliably in machines cobbled together by *OEM* suppliers who have to slash their costs to the bone and use the cheapest possible components in order to not go immediately out of business. There's enough headroom in the designed ratings

            • 100% stable at stock voltage

              Queue some petulant twat whinging about how it might, maybe, undetectably, subliminally, possibly, almost be unstable in 0.0000001% of circumstances (maybe) and therefore you should never ever use it oh my god you're overclocking, dear GOD!!!

            • There's absolutely no reason NOT to overclock Core2s.

              What if I want a quiet-to-silent system that can run with minimal cooling and use as little power as possible to do its job?

              • by deroby ( 568773 )

                Then UNDER-clocking might be your thing =)

                Ha! You thought this was binary : to overclock or not to overclock, well sonny, I've got news for you, there's a new guy in town !

                My friend's been having this 1GHz PIII running on 750MHz for over 6 months now, and it's AB-SO-LU-TE-LY 100% stable !

                (ps: genuinely true story : the BIOS battery-connection is broken somehow and the thing boots by default on 750MHz. Although I explained how to set the right values in the BIOS, he simply got tired of it and claims "it's ju

      • I have a Q9450 at home (2.66 x4 stock) which I have OC'ed to 3.2GHz. I have actually been able to do this while LOWERING the stock voltages of the chip. It runs completely stable on prime stress testing. I have just a mid-range copper and fan heat sink which keeps the CPU at a lower temperature than it was at the stock speeds and voltages on the heatsink assembly that came with it.

        By my math (which you should not trust, I suck at math), that is around a 20% increase, and may even prolong the life of t
        • I would rather have a quiet and cool chip that I could slightly underclock. I'd love a silent media-PC to slap Linux on.

          • Any of the 45nm Core 2 Duo series would be great for that. You could underclock (and undervolt) them significantly while still retaining good performance.
          • Re: (Score:2, Interesting)

            by lytles ( 24756 )

            i've got:
                amd 64 x2 at 25W (BE-2350)
                690G motherboard with onboard graphics (ASUS M2A-VM HDMI)
                $165 total in dec 2007

            and it runs fanless fine (tho i do have the fan hooked up idling and thermally controlled most of the time). doesn't look like they're still selling the BE-2350, and not sure if there's a current equivalent or if you can accomplish the same thing by underclocking.

            • The Athlon X2 4850e and similar chips have replaced the BE series. I think they're better performance, and I'm hoping they are "real-world" 45W.

              You should be able to bring those down close to 15W as well.

      • Why is this idiot getting modded up? Widespread ignorance is fun! Listen up mods. Nothing this guy said is true; literally nothing. What the guy described is a fantasy that NEVER happens, he is simply an imbecile with outdated preconceptions. People don't tend to read about things they don't like, and he seems to be following this trend quite nicely with an opinion on overclocking straight from 1994.

        I'm sure it was perfectly valid back then, but in no way does it apply to the modern Intel chips we are
  • Print Link (Score:4, Informative)

    by HaloZero ( 610207 ) <protodekaNO@SPAMgmail.com> on Tuesday December 16, 2008 @09:26AM (#26132191) Homepage
    Full print article should anyone not want to deal with the multipage click-through: http://www.extremetech.com/print_article2/0,1217,a%253D235027,00.asp [extremetech.com]
  • by phr1 ( 211689 ) on Tuesday December 16, 2008 @09:39AM (#26132345)
    The Nehalem/i7 uses DDR3 which is a lot more expensive per GB compared to DDR2 and not available in as high capacity. It has more bandwidth but its latency (which matters more) is about the same. The usual desktop mobo is limited to 2 dram modules per channel. DDR2 boards usually have 2 channels (4 sockets max) while DDR3 boards have 3 channels (6 sockets). But 4GB DDR2 modules are around $100 (link [newegg.com]) while DDR3 currently maxes out at 2GB. So you can populate a Phenom or Core 2 mobo with 16gb of ram for $400 but you can't put that much on a normal consumer i7 board for any amount of money. 2GB DDR2 parts are a lot cheaper still, you can put on 8gb (4x 2gb) at around $15/gb, $120 total. Right now a 2gb DDR3 part is $50-ish, 3x as expensive (link [newegg.com]). It helps that you can put 6 of them on a board (12gb total, $300) but you have to take the cost difference per GB into account with 2GB parts, and comparing with 4GB DDR2 parts there is $/GB parity but lower total capacity (4x4gb vs 6x2gb). And of course when 4gb ddr3 does come out, it will bring a welcome increase to 24gb total capacity, but it will be WAY expensive for quite a while (the 4gb ddr2 modules that are $100 now were $500+ for most of this year).

    I just don't understand why there aren't more consumer boards with a lot more sockets, using FB-DIMM or registered DDR. You have to go to server boards for that ($$$).

    • by fuzzyfuzzyfungus ( 1223518 ) on Tuesday December 16, 2008 @09:51AM (#26132521) Journal
      I don't think that there is much demand for consumer boards with that many slots. 4 sockets, the consumer usual, will do 4GB for peanuts, which is overkill for anybody running a 32 bit OS, and covers an overwhelming majority of consumer use cases. With 2GB modules, also fairly cheap, a 4 socket board will hit 8GB, which is enormous for consumer purposes.

      Beyond that, if you want FB-DIMM or registered DDR, to support some enormous number of sockets, you'll need to go with a memory controller that supports that. With AMD or newer intel, that means buying a server processor(since the memory controller is on die), which will be expensive. Older intel still meant using a server chipset, also not cheap. And silicon aside, more sockets=higher BOM costs, more difficult signal routing, possibly more board layers, and so on.

      There is a market of "people who want server/workstation boards at consumer prices"; but I suspect that the "consumers who need more than 8GB of RAM" market is pretty small. And, in any case, more sockets costs more to manufacture.
      • and I think we know better than "no one will ever want more than 640K" or 4GB as the case may be. The latest trendy accessory is ultra expensive ($25/GB) Intel X25-E flash drives and a lot of the motivation for buying them is inadequate ram capacity in the host computer (since the flash disk costs more per GB than RAM which is 100x faster, though volatile).

        Yeah a lot of people are still running 32 bit OS's, but almost all desktop hardware now being shipped is 64 bit-- we're in something like the tail end

        • Who said anything about "ever"? We all know that the price of RAM goes down, and the capacity goes up, and we all know this, so what's the problem? 8GB is a lot of RAM for a desktop today, and in the future the price of those 4GB DIMMs will decrease. Or buy them today for a total of 16GB, and a lot less cost than a flash disk (and are people really using these as RAM or as "fast" non-volatile storage? Because even with modern max write cycles that seems like a silly thing to do)

          By the way, part of the r

      • will do 4GB for peanuts, which is overkill for anybody running a 32 bit OS, and covers an overwhelming majority of consumer use cases.

        I think having 2 Gigs is awesome enough. It's four thousand times as much as my first computer.

        To give you a sense of perspective: a three-mile bike ride down to the university and the same three miles back, that's roughly 10 km/day. Scale that number up by the same size factor and we get 40000 km. Instead of just going down to the local university, I could be circumnavigating the earth. ... And get off my lawn! ;)

        Seriously though, what use can one put > 2 GB to? Right now, I spend 1 GB on disk cache

      • I have no problem filling up 8 gb of memory. I run several virtual machines on different sides of my compiz cube in Ubuntu.

        But I'm a fucking weirdo. I doubt very much that anyone else is using that much ram that often. :)

    • Re: (Score:3, Insightful)

      I just don't understand why there aren't more consumer boards with a lot more sockets

      Desktop user never needs more then 3GB, because this is what Vista/XP support. Mainboards, unfortunately, are created with XP/Vista (32-bit) in mind.
      • by rudeboy1 ( 516023 ) on Tuesday December 16, 2008 @11:12AM (#26133403)

        Actually, I had to look that up recently. It's not 3GB, it's 4GB. Here comes the science:

        "Microsoft Windows XP Professional, designed as a 32-bit OS, supports an address range of up to 4 GB for virtual memory addresses and up to 4 GB for physical memory addresses. Because the physical memory addresses are sub-divided to manage both the computer's PCI memory address range (also known as MMIO) and RAM, the amount of available RAM is always less than 4 GB.

        The PCI memory addresses starting down from 4 GB are used for things like the BIOS, IO cards, networking, PCI hubs, bus bridges, PCI-Express, and video/graphics cards. The BIOS takes up about 512 KB starting from the very top address. Then each of the other items mentioned are allocated address ranges below the BIOS range. The largest block of addresses is allocated for today's high performance graphics cards which need addresses for at least the amount of memory on the graphics card. The net result is that a high performance x86-based computer may allocate 512 MB to more than 1 GB for the PCI memory address range before any RAM (physical user memory) addresses are allocated.

        RAM starts from address 0. The BIOS allocates RAM from 0 up to the bottom of the PCI memory addresses mentioned above, typically limiting available RAM to between 3 GB and 3.4 GB."

        I actually learned something last week, thought I'd pass it on...
        *Cue the "The More You Know" logo*

        • Actually, I had to look that up recently. It's not 3GB, it's 4GB. Here comes the science:

          RAM starts from address 0. The BIOS allocates RAM from 0 up to the bottom of the PCI memory addresses mentioned above, typically limiting available RAM to between 3 GB and 3.4 GB."

          I actually learned something last week, thought I'd pass it on...
          *Cue the "The More You Know" logo*

          Looks like you ended up arguing his point. 32-bit Vista/XP are limited to a little over 3GB. How much over 3GB is determined by your hardware.

          • Looks like you ended up arguing his point. 32-bit Vista/XP are limited to a little over 3GB. How much over 3GB is determined by your hardware.

            Unfortunately sometimes the hardware screws you even if the OS doesn't. 32-bit Linux can support and use up to 64 GiB of RAM, but I just put 4 GiB in my Thinkpad and the OS only reports a little over 3, even with HIGHMEM64G set in the kernel config.

            Turns out to be a limitation of the Intel chipset. I guess Intel figured 3 GiB is enough for anybody's laptop.

            • by Kagura ( 843695 )
              32-bit OS can only support 4gb of RAM total, meaning installed RAM + video card RAM maxes out at 4gb. That's why you're not getting the full 4gb of motherboard RAM.
              • by swillden ( 191260 ) <shawn-ds@willden.org> on Tuesday December 16, 2008 @03:48PM (#26137283) Journal

                32-bit OS can only support 4gb of RAM total, meaning installed RAM + video card RAM maxes out at 4gb. That's why you're not getting the full 4gb of motherboard RAM.

                Wrong, twice.

                First, video card RAM is not mapped into the kernel's address space and is irrelevant.

                Second, 32-bit OSes can use Intel's Physical Address Extension (PAE) [wikipedia.org] mode to support up to 64 GiB of RAM. Linux does this. 32-bit Linux actually has three different memory management modes:

                • HIGHMEM off: This is the traditional "split address" mode, which assigns part of the 4 GiB address space to the kernel (for kernel use and for mapping devices) and part of it to user programs. Linux uses a 3/1 split by default, with 3 GiB available to user processes and 1 GiB available to the kernel.
                • HIGHMEM on: This sticks with 32-bit pointers but avoids permanently mapping chunks of the address space. The kernel gets preferential use of the lower 896 MiB of address space, but can also allocate from ZONE_HIGHMEM as needed, and user space memory can also be allocated from anywhere. Bottom line: with HIGHMEM on, the Linux kernel will use all of 4 GiB, but no more.
                • HIGHMEM64G on: This enables PAE mode, which allows the kernel to use page tables to map different parts of up to 64 GiB total RAM into per-process 32-bit (4 GiB) address spaces. No single process can use more than 4 GiB in this mode, but each process can use up to 4 GiB, using all of the available RAM. This comes at a small performance hit in terms of CPU cycles.

                Most desktop Linux distros today (inluding Ubuntu) ship kernels with HIGHMEM on, but not HIGHMEM64G, avoiding the PAE performance cost, but enabling the use of up to 4 GiB of RAM -- assuming the hardware supports it, which my laptop does not.

    • I just don't understand why there aren't more consumer boards with a lot more sockets, using FB-DIMM or registered DDR. You have to go to server boards for that ($$$).

      It's not just the matter of having the sockets, the chipset must also support the additional RAM. Most consumer chipsets either support 4GB or 8GB only. Some like the atom support 2GB only iirc. Even some serverboards max out at 8GB (esp those supporting the Xeon 3000 processor)

      • My hundred-euro Gigabyte board supports 16GB of DDR2 out of the box. 16GB should easily be within reach of any consumer.
    • Re: (Score:1, Troll)

      by Mad Merlin ( 837387 )

      I just don't understand why there aren't more consumer boards with a lot more sockets, using FB-DIMM or registered DDR. You have to go to server boards for that ($$$).

      One word: Windows.

      Nobody uses 64-bit Windows, and (sadly) Windows users make up the majority of the computer buying population, so board makers don't see any reason to provide for more than 4G of RAM. At the current DDR2 prices, I'd love to drop 16G or 32G in a new system, but it's simply not an option with consumer boards, as you've noticed.

    • 3 2 2 2 2 4 3 3 6 4 2 100 3 2 2 16 400 7 2 2 8 4 2 15 120 2 3 50 3 6 12 300 2 4 2 4 4 6 2 4 3 24 4 2 100 500

      You made a mistake: that 3 is really only 1.5.

    • by Kjella ( 173770 )

      Like you say, you can get 16GB of RAM on mainstream motherboards. I can edit my digicam pics and HDV home vids all day long on 4GB, so what application would benefit? I'm mostly:

      a) bandwidth-bound to Internet
      b) GPU bound
      c) CPU bound
      d) storage space bound

      Being memory bound is like way, way down on my list. Maybe if you have a ton of layers and effects in print-quality posters or HD movies you'll notice, but I can't manage to find a reasonable 16GB working set if I tried. Even the little Java hogs I run usual

      • by Detritus ( 11846 )
        I sometimes run data analysis applications that could easily use 32GB if it was available. The programs are small but the data sets are huge. There is a large class of applications that could run faster or handle bigger problems if there was more RAM.
    • Do you know where I might be able to find cheap RDRAM? Back in '02, this totally seemed like a good idea...

      The computer next to me has 64MB(I think) of EDO-RAM

      • by Detritus ( 11846 )
        I looked into expanding the RAM on my old Pentium IV system (Intel D850GB) and it was much cheaper to replace the old motherboard with a new motherboard that supported DDR2 RAM. RDRAM is still available, but it's damn expensive. The annoying thing is that I couldn't find anything at a reasonable price that supported ECC RAM like the old motherboard.
  • Last I checked, DDR3, which i7 requires, still had quite the premium over DDR2. Also, the motherboards, last I checked, started from about $300, while Core 2 motherboards can be obtained of course much cheaper.

    And the benchmarks are a bit silly. How many people spend most of their time encoding video and running photoshop plugins? Gaming, OK, but other than that more realistic benchmarks for the average user would be more appreciated. Not just benchmarks that are designed to showcase multi-core processors.

    • Re: (Score:2, Insightful)

      by iainl ( 136759 )

      What benchmarks would you like to see? Because off the top of my head, media encoding, photoshop stuff and games are exactly the sort of things that home users would be doing to require that kind of CPU oomph.

      Sure, there are plenty of other things people spend lots of time doing, but most of those (at least as far as home apps go) would run just fine on practically any reasonable new processor.

      • Re: (Score:3, Interesting)

        by RT Alec ( 608475 )
        I'd like to see how the systems (CPU/RAM/MB etc) perform with a 64 bit OS with a simulated workload. How about:
        • FreeBSD 7 (AMD64)
        • MySQL 5.1.30
        • Apache 2.2 (worker MPM)
        • PHP 5.2.8 (or HTML::Mason)

        What is the responsiveness of the system under load? Openssl speed? bonnie++?

        • by iainl ( 136759 )

          All of which is a perfectly reasonable thing for people to run on a quad-core box; indeed practically tailor made for it.

          But that's a webserver, not a desktop app or game like the article is talking about.

  • For Canadian prices, (Score:1, Informative)

    by Anonymous Coward
    For Canadian prices, add $40-140 (wtf, man)
  • by DarthVain ( 724186 ) on Tuesday December 16, 2008 @09:59AM (#26132615)

    Its a bit stupid to do a price performance without any consideration to anything else. Someone else mentioned DDR2 VS DDR3 and price and availability, also if anyone does one whit of looking you will see that you can have a nice dual-core or quad-core motherboard for like 150-200$. I looked a few weeks ago for the 7's and the cheapest was like 350-375$, most seemed to be over 400$. Which to me, is simply nuts.

    Anyway like a good slashdotter I didn't RTFA, so it may be that the 7's are the bees knees. However I would caution anyone from basing their decision on a benchmark and a price tag, as there is more involved that that. Anyway my two cents...

    • Outfitting the Qxxxx's with 2 GByte ram and the i7's with 3 doesn't the TFA remotely balanced.

    • by Ecuador ( 740021 )

      I looked a few weeks ago for the 7's and the cheapest was like 350-375$, most seemed to be over 400$. Which to me, is simply nuts.

      This statement means a lot to people who lived in an era of absolute Intel dominance (i.e. before 2000) and had to pay whatever Intel asked, if they wanted a decent processor. I hope for the sake of all of us that AMD does not go away...

  • by DrDitto ( 962751 ) on Tuesday December 16, 2008 @10:22AM (#26132851)
    The analysis is flawed in my opinion since it doesn't consider overall system cost. I just built a machine yesterday for use as my desktop. The Core i7 mainboards and DDR3 memory really push up the price right now. I considered the following configurations:
    • Quad-core Q6600 65nm 2.4GHz ($183), Gigabyte LGA775 mainboard ($105), 4GB DDR2 memory ($41): Total $336
    • Dual-core E8400 45nm 3GHz ($165), Gigabyte LGA775 mainboard ($105), 4GB DDR2 memory ($41): Total: $311
    • Quad-core Nehalem Core i7 920 ($299), Gigabyte LGA1336 mainboard ($245), 4GB DDR3 memory ($125): $659

      At 2x the price, Core i7 was not a consideration for me at this time.

      The choice between the E8400 and the Q6600 was a tough one. I could have gone either way. Quad-core is great for threaded applications like media encoding. But the E8400 outperforms the Q6600 for the majority of what I do (including Photoshop CS3). I am not convinced that threading will be widespread enough during my 3-year upgrade cycle. A common argument on the forums is that the Q6600 can be overclocked to 3GHz such that single-threaded is the same as the E8400. While I do not overclock, the E8400 supposedly can easily get to 4GHz on air.
    • Re: (Score:2, Informative)

      You do NOT want to use 4GB ram for an i7. It has to be in multiples of 3GB since the i7 uses a triple channel ddr3. Some mobos have 4 slots (eg Intel DX58SO), however populating that last slot will sacrifice triple channel with single channel performance.
      • however populating that last slot will sacrifice triple channel with single channel performance.

        No, it won't. For some time now, Intel's RAM controllers have had some way of interleaving that buys you most of the performance of the 'correct' X-way setup, even with mismatched bank sizes.

        That said, Intel have lost the plot if they think that encouraging this sort of thing is a /good idea/.
    • by msu320 ( 1084789 )
      The choice between the E8400 and the Q6600 was a tough one. I could have gone either way. Quad-core is great for threaded applications like media encoding. But the E8400 outperforms the Q6600 for the majority of what I do (including Photoshop CS3). I am not convinced that threading will be widespread enough during my 3-year upgrade cycle. A common argument on the forums is that the Q6600 can be overclocked to 3GHz such that single-threaded is the same as the E8400. While I do not overclock, the E8400 suppos
      • by DrDitto ( 962751 )
        I have no doubt that CS3 is multithreaded. However numerous benchmarking sites have shown the E8400 outperforming the Q6600 in Photoshop CS3. Obviously this will depend on what aspects of CS3 were exercised during the benchmark. All multithreaded codes have serial portions, and I suspect that a substantial portion of CS3 is still serialized.

        CUDA filters will work the same regardless if you are dual- or quad-core (well, there is some CPU overhead for transferring data to/from the GPU, so maybe this wo
    • The choice between the E8400 and the Q6600 was a tough one. I could have gone either way. Quad-core is great for threaded applications like media encoding. But the E8400 outperforms the Q6600 for the majority of what I do (including Photoshop CS3).

      Totally off-topic, but how hot does your E8400 run? In a cool environment, my non-overclocked CPU ran at 49C idle and 63C active with the stock cooler. I bought an Arctic Cooling Freezer 7 Pro (my first and only foray into non-stock cooling) and saw that drop to 40C idle to 49C active.

    • I wish I had never made a choice at all. I had high hopes when I got a quad core system, but the current state has disgusted me.

      It turns out I still can't run matlab, watch a flash app, and move the mouse all at the same time in Vista on my Core2Quad HP box. Maybe if I'm running 3 instances of matlab maybe I'll see an improvement the quad core over say my dual core laptop, but based on everyday uses the upgrade in systems was an infuriating waste of money.

      Sometimes the quad core system can't even handle u

  • http://www.extremetech.com/image_popup/0,,iid=223387&aID=235027&sID=27866,00.asp [extremetech.com]

    This would mean that my Geforce2 256 is better at games than my ATI 4870, which does not seem quite right.
  • Um, I'm sorry but isn't Xeon still an Intel brand? There are quite a few offerings in the Xeon line that are quad core. In fact, I'm building a Socket 771 machine now with dual Xeon procs, and was interested to see how the Xeon quad 2.5GHz was going to stack up, (what I can afford) but no. Fail.

    • You're right, and since I'm only looking in the server market, there doesn't seem to be any hosts with i7's out there. Once you get beyond 1 or 2 servers, the price/performance is a great determining factor for cluster machines. The consumer and server cpu's usually do have a direct mapping. not sure what this is for the i7 940.. or if the price for that would be similar.
    • I believe that they were only looking at the consumer line-up of quads (Kentsfield, Yorksfield, Core i7), and not the server lineup (Xeons).
    • Save your dosh. The xeon will have nehalem in Q1.
    • As far as I know they're the same chips with different names. For example, the Xeon 3110 is the exact same as the Core 2 Duo E8400.
  • If you're concerned about money but still want a screamer, I'd recommend the Xeon 3550 (the exact same as the Q9450, except reputed to run at lower temps, taken from better batches, etc). When I purchased one, it was the same price as the Q9450.

    Pair this with decent DDR2 RAM (not DDR3, it's still expensive and not worth the gains) and you'll save a bunch o' money. In my case, the system overclocks reliably to 3.6GHz.

    So if you want something better than the cheapest option, but don't want to spend bazill

After all is said and done, a hell of a lot more is said than done.

Working...