Slashdot Mirror

← Back to Stories (view on slashdot.org)

26 Desktop Processors Compared

Posted by kdawson on from the two-bakers-dozens dept.

theraindog writes "The number of different CPU models available from AMD and Intel is daunting to say the least. The Tech Report's latest CPU review makes some sense of the landscape, exploring the performance and power consumption characteristics of more than two dozen desktop processors between the $999 Core i7-975 and more affordable sub-$100 chips. The article also highlights the value proposition offered by each CPU on its own and as a part of the total cost of a system. The resulting scatter plots nicely illustrate which CPUs deliver the best performance per dollar."

31 of 192 comments (clear)

  1. Seems pretty clear: by fuzzyfuzzyfungus · · Score: 3, Informative

    AMD has pretty embarrassing performance on the high end, which makes your choice there downright trivial; but is an excellent value in those niches where they have an entry.

    1. Re:Seems pretty clear: by mdm-adph · · Score: 3, Insightful

      Not embarrassing, just not as fast. However, a Ferrari is vastly more powerful for the money, but a Corvette is still the much better deal -- do I really need to explain this on here?

      --
      It is by my will alone my thoughts acquire motion; it is by the juice of the coffee bean that the thoughts acquire speed
    2. Re:Seems pretty clear: by caerwyn · · Score: 4, Informative

      No, but the $300 i7 920 surely is...

      --
      The ringing of the division bell has begun... -PF
    3. Re:Seems pretty clear: by mdm-adph · · Score: 3, Insightful

      Well now, they are faster. They're just not $800 faster. Nowhere near it.

      Not to mention that the corresponding hardware you have to buy with a Core i7 (the expensive motherboard, the expensive triple channel ram) makes it even more not worth it.

      --
      It is by my will alone my thoughts acquire motion; it is by the juice of the coffee bean that the thoughts acquire speed
    4. Re:Seems pretty clear: by Anonymous Coward · · Score: 3, Insightful

      Plot isn't that clear to me at all.

      Price/performance is slope on that graph, and AMD looks pretty good at first look, occupying a band mostly on the top-left.

      They ought to have plotted performance on the independent axis vs. price/performance on the y axis. And put that sucker on a log-log plot. Then we'll see who's got the best price/performance for various performance levels.

    5. Re:Seems pretty clear: by mister_playboy · · Score: 3, Informative

      The Corvette ZR1 has more horsepower and is less expensive than any current Ferrari.

      I love Ferrari... but the Corvette needs no apologies at all!

      --
      Do what thou wilt shall be the whole of the Law ::: Love is the law, love under will
    6. Re:Seems pretty clear: by characterZer0 · · Score: 4, Funny

      It does if you want to turn.

      Or if you want to not look like a redneck.

      --
      Go green: turn off your refrigerator.
    7. Re:Seems pretty clear: by Anonymous Coward · · Score: 4, Insightful

      Not if you factor in the motherboard price difference - about $100 for x58 over AM2+. The i7 is brilliant for heavy, multi-threaded workloads but I question how much of that performance is really worth it for an average consumer or even Slashdot reader.

    8. Re:Seems pretty clear: by athakur999 · · Score: 5, Insightful

      The Phenom II X4 955 beats the i7 920 in 3 out of the 4 games they tried. The only one it lost was Cryis Warhead and it was a narrow loss (48 vs. 46 FPS).

      These price difference of these two chips is about $35 on Newegg. I think for gamers, getting the X4 955 and putting that extra $35 towards the video card will net better results. This isn't counting the additional cost of DDR2 vs. DDR3 memory which has minimal effect on performance right now but still has a big price difference.

      --
      "People that quote themselves in their signatures bother me" - athakur999
    9. Re:Seems pretty clear: by Znork · · Score: 3, Insightful

      Personally I don't find it that trivial; never before has the best choice been so extremely dependent on what you need the performance for.

      Say you've got some eminently parallel task, like ray tracing. With the huge price/performance difference between low end and high end you might beat the high end in performance by buying two cheap systems rather than one expensive. Look at the i7-940; it's just barely twice as fast as the cheapest CPU, yet it costs six times as much. That price would easily accommodate a cheap motherboard and memory and you'd still be shelling out less.

      On the other hand, say you have very few parallel tasks, then you may still be as well off with a cheap CPU. Without several tasks you're not going so see much difference between a dual core CPU with good per-core performance and a quad core. Not an entirely rare situation when talking about desktop systems.

      Or you might be in the sweet spot of latency sensitive parallel tasks, possibly applicable to some games, in which case a more expensive quad core CPU might definitely be what you're looking for.

      And add to that the need for actual application specific benchmarks to determine actual performance as opposed to generic benchmarks... well, I wouldn't call the choice trivial.

    10. Re:Seems pretty clear: by Rockoon · · Score: 5, Interesting

      I'd also question its price/longevity scale as well, since the sort of folks who are going to drop $700+ on a basic (no video card, no monitor, no hd's) high performance system are probably the same sort of folks who are going to scrap that box as soon as Larrabee hits.

      My AMDx2 3800+ is starting to show its age, but there is no way I am going to buy a high end part with Larrabee so close.

      --
      "His name was James Damore."
    11. Re:Seems pretty clear: by hairyfeet · · Score: 5, Insightful

      Not to mention that for most folks that "ludicrous speed" is simply overkill. I've been building my customers a mix of Pentium Duals and AMD 7550s and all they can talk about is how blazing fast they are. Because for most of the stuff folks are doing today, even games, the CPU is rarely the bottleneck and passed "good enough" a few years back.

      So at least for my customers and myself (I liked how the last AMD 7550 ran so quiet so I built one for myself) it is more about "bang for the buck" and when I can build a nice AMD 7550 with 4Gb of RAM, a 780 board along with a 300Gb Sata 2 and a nice black case for $281 shipped it is just nuts to blow all that extra cash in this economy to get a bigger epeen. For most of us even the bottom of the line dual cores are sitting idle a good 80% of the time, so why blow the extra cash? I just checked process Explorer and with 9 tabs plus Comodo Internet Security on XP X64 I've still got 3.2Gb of RAM free and am only using an average of 1.4% CPU, plus I got another 1Gb of RAM on the $50 HD4650 for watching videos.

      Considering my first machine was a VIC20 where I had to peek and poke everything all this extra power is nice, but I doubt I'll be saying "you know what? I need more power!" for quite awhile yet. So for a box that cost a hair over $500 counting the OS and GPU I'm a happy little camper. While having ludicrous speed might be good if I was doing major CAD or graphics work, for the things that myself and my customers do with their PCs there just really isn't a point. Hell I gave my nearly 5 year old 3.6GHz P4 to my oldest and he is blasting zombies in Left4Dead even as we speak, so even that CPU is "good enough" for what he wants to do with it. The "bang for the buck" you get nowadays even on the low end is frankly unreal!

      --
      ACs don't waste your time replying, your posts are never seen by me.
    12. Re:Seems pretty clear: by Latinhypercube · · Score: 3, Funny

      Wait until your kid is 3d rendering in Maya and you will be screaming for 8 core Xeons.

    13. Re:Seems pretty clear: by poetmatt · · Score: 3, Interesting

      That's just the motherboard, let alone the processor or ram cost differences. Not to mention they are about to change the socket again rather soon, if I recall correctly (aka 1 year).

      AMD changes sockets a lot more gradually, letting people actually, you know, upgrade.

    14. Re:Seems pretty clear: by geekoid · · Score: 3, Interesting

      WHen I can I buy the highest end I can and use it for 5 years.
      I might upgrade a video card. After 5 years, it gets deprecated into the house hold pool.
      I have 5 year old computers running games like team fortress 2 and wow just fine. Not the highest end graphic performance, but fine enough.

      When I build my next box, It will have 16Gigs ram, minimum. Hopefully 32. This will probably stretch it's use out to 8 years.

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    15. Re:Seems pretty clear: by DAldredge · · Score: 3, Informative

      The consumer i7 chips do not support ECC RAM while the AMD ones do. That could be important to some.

    16. Re:Seems pretty clear: by LordVader717 · · Score: 3, Insightful

      That's one way to go, but the most economically sound way to do it has always been (slightly) more frequent upgrade cycles and lower range hardware.

    17. Re:Seems pretty clear: by ceoyoyo · · Score: 4, Insightful

      In science we call that honesty. You can make a tiny little difference look REALLY BIG by cropping your graph so that it only shows a very tiny range. By showing the origin tiny differences look, well, tiny.

    18. Re:Seems pretty clear: by steveha · · Score: 4, Insightful

      My issue with the graph is someone needs to take a class on "how to make a graph". 0% performance and $0 cpu.... why? Was there a $0 cpu? Did any of the cpus get a 0%?

      Probably because the person who made that graph for The Tech Report wanted all the proportions to be honest.

      Did you ever read the book How to Lie with Statistics? Or the book How to Lie with Charts? Or a nice, short blog post called Graphs That Lie?

      When you chop out some of the "wasted space" in a graph, you distort the graph. Unless people are careful and check where your axes begin, and then mentally visualize where the axes go, they'll get a misleading idea of the data from the graph.

      Suppose the bottom part of the graph was sliced off, at the 90% line, to make you happier. Imagine what it would look like. The AMD X2 6400+, sitting at the 100% line, would have very little white space under it; and the Intel i7-920, sitting a bit below the 200% line, would now appear to be ten times faster than the AMD X2 6400+. The numbers would be the same, but the visual impact would be that the Intel chip totally blows away the AMD chips.

      The graph is good the way it is.

      I'll meet you halfway, though: it wouldn't have hurt for them to have put in a second chart, zooming in on just the most crowded areas.

      steveha

      --
      lf(1): it's like ls(1) but sorts filenames by extension, tersely
    19. Re:Seems pretty clear: by Falconhell · · Score: 3, Interesting

      The most economically sound upgrade path for me has been sell after 6 months, getting 95% of purchase price back-my people like buying my 6 month old systems, and I like buildiing new ones!

  2. Correlation is positive by JoshuaZ · · Score: 3, Insightful

    The first take away I get is that there is an actual, substantial positive correlation between cost and performance. This is a good thing. If I were in a cynical mood I would have guessed that the correlation would have been small or non-existent. The other thing to note is that there are some CPUs that by this metric are clearly just not very worth it where their are cheaper ones that perform better. So, more expensive generally means better, but not always. So CPUs are sort of like wine?

    1. Re:Correlation is positive by TinBromide · · Score: 3, Funny

      yup, just like cars (Once upon a time, you could spend 36k USD for a pt cruiser), houses, internet subscriptions, insurance, meat, items on ebay, and just about anything that has a varied price range and is subject to advertising, brand loyalty, or outright cons.

      I've always tended to get right behind the cutting edge, processors usually take a dive in price and speed, Like when i bought my processor, i paid 300 for a quad core intel last summer, the next step up was 395, then 600, then 1000, but the 1000 wasn't 3.3x faster, the 600 wasn't 2x faster, etc. However, the $150 intel chip was 40% slower, so the price/performance sort of follows a log function where the trick is finding the right place before the price starts to skyrocket in relation to performance. I'm sure that if you could find a bunch of old pentium 1 processors on the super cheap, in bulk, you might get the same price/performance as a new mid-level intel chip, but why would you want to run a beowulf cluster to get the same throughput? (Ignoring the price of all of the other components, since the article is price/cost of the cpu alone, that was the metric i went by as well).

      I wonder what the price/performance ratio of a pentium 1 that cost $0.01 is.

      Well, I suppose that running a beowulf cluster is it's own reward.

      --
      Is it sad that I am more likely to recognize you and your posts by your sig than your name or UID?
    2. Re:Correlation is positive by Anonymous Coward · · Score: 3, Insightful

      Aside from the possitive correlation, the other main point I would take away is the apparent 'knee' in the chart at about the $300 mark.

      Seems that is the point of diminising returns.

  3. A better solution. by LWATCDR · · Score: 5, Interesting

    What is the cheapest CPU that can playback 780P flash well?
    That is probably a good CPU for 99% of the population. Flash is a resource hog and is likely to be most intensive thing that most people use.
    The next step up would be to list several games and see what is the cheapest CPU that can play them at say 60FPS at good settings with a $99 video card.

    If your a video editor, hardcore PC gamer, transcode a lot of video, or run CAD get the fastest CPU you can afford.
    So hard core types should buy I7s and pretty much everybody else should buy AMDs once you take into account ram and motherboard prices.

    Also if you are planing on running virtual machines AMD are often a better choice. Intel doesn't support virtualization on a lot of their CPUs while I think AMD does on their AM2 and up CPUs.

    --
    See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
    1. Re:A better solution. by wjh31 · · Score: 3, Insightful

      if all you want is 720P flash, pretty much anything on the market will do. My wifes laptop with a C2D T2080 @ 1.73GHz manages it fine. when comparing it against other desktop processors of similar performance on here: (http://www.cpubenchmark.net/cpu_lookup.php?cpu=Intel+T2080+%40+1.73GHz), and then looking it up on ebuyer.com, suggests that a £40 proccesor is fine for 720p flash.

  4. Re:Only Intel compatible .... by Microlith · · Score: 3, Insightful

    They'll cover ARM when someone sells a motherboard with a socket I can stick a 2+ GHz quad-core ARM in and get performance equal to an equivalent AMD or Intel chip.

    As it stands, ARM is irrelevant outside the embedded/pda/non-x86 netbook scope.

  5. What I'm seeing... by jd · · Score: 4, Interesting

    The performance scales sub-linearly with the price, and ends up almost flat at the extreme end. This means you need to examine the cost of SMP vs. a more powerful CPU. Two X2 6400+ chips in an SMP should give you about the same performance at the same cost as one i7-920, after you add in the extra for the upgraded chipsets and mobo.

    More powerful low-end chips become more and more effective when SMPed versus their higher-end rivals. The other benefit of going SMP is that you have fewer cores sharing the same cache, therefore increase the number of distinct tasks you can perform in parallel effectively without cache-flooding.

    Of course, you can't SMP forever - the largest SMP array you can make before the system slows down by more than the CPU increases performance is 16-way. Even before then, you lose linear scaling fairly early on. So you end up balancing the different CPUs against the different methods of arranging them to get the best performance for your money.

    --
    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. /.'ed by dr_wheel · · Score: 5, Funny

    Apparently, the tech report should have benchmarked their web server before putting this article up.

  7. Article doesn't really talk about features by vux984 · · Score: 5, Informative

    Its too bad the article doesn't talk about things like Execute Disable, Virtualization support, etc. For a power user audience like /. these are important considerations.

    For me not being able to install Xen, or Windows 7 XP mode, etc are complete deal killers. I want CPUs with those features, especially when shopping "value CPUs".

    Getting something like an E8190 is a mistake that will bite a /. power user in the ass eventually even if it is a few bucks cheaper than an E8200 and delivers the same performance, at the same wattage, etc...

  8. Re:Ancient CPUs (upgrade comparison) by dr_wheel · · Score: 4, Informative

    Tom's cpu chart is a great tool, but they don't generally compare older chips to newer ones. They also change the testing credentials from time to time, so there's no real way to directly compare old vs. new.

    Anandtech has a new cpu benchmark site that compares everything from a single-core atom up to the top-of-the-line core i7. They've also recently added two pentium 4 chips to the mix so you really can directly compare them to the newer stuff. Check it out:

    http://www.anandtech.com/bench/

  9. Intel Atom 330 by Eil · · Score: 3, Informative

    I've been something of an AMD fanboy ever since the Athlon came out, but I just bought an Intel Atom 330 for a lightweight file server, and I have to say I'm thoroughly impressed. 64-bit, dual-core, virtualization extensions, and low-power to boot for around $80 which includes the motherboard. Simply unbeatable.

    Also wanted to mention that these guys have easy-to-read benchmark charts of a wide variety of CPUs. Certainly more than the 26 in TFA. Benchmarks don't tell the whole story of course, but it's a good start for quick-and-dirty comparison.