Slashdot Mirror


Intel Core I7-7700K Kaby Lake Review By Ars Technica: Is the Desktop CPU Dead? (arstechnica.co.uk)

Reader joshtops writes: Ars Technica has reviewed the much-anticipated Intel Core i7-7700K Kaby Lake, the recently launched desktop processor from the giant chipmaker. And it's anything but a good sign for enthusiasts who were hoping to see significant improvements in performance. From the review, "The Intel Core i7-7700K is what happens when a chip company stops trying. The i7-7700K is the first desktop Intel chip in brave new post-"tick-tock" world -- which means that instead of major improvements to architecture, process, and instructions per clock (IPC), we get slightly higher clock speeds and a way to decode DRM-laden 4K streaming video. [...] If you're still rocking an older Ivy Bridge or Haswell processor and weren't convinced to upgrade to Skylake, there's little reason to upgrade to Kaby Lake. Even Sandy Bridge users may want to consider other upgrades first, such as a new SSD or graphics card. The first Sandy Bridge parts were released six years ago, in January 2011. [...] As it stands, what we have with Kaby Lake desktop is effectively Sandy Bridge polished to within an inch of its life, a once-groundbreaking CPU architecture hacked, and tweaked, and mangled into ever smaller manufacturing processes and power envelopes. Where the next major leap in desktop computing power comes from is still up for debate -- but if Kaby Lake is any indication, it won't be coming from Intel. While Ars Technica has complained about the minimal upgrades, AnandTech looks at the positive side: The Core i7-7700K sits at the top of the stack, and performs like it. A number of enthusiasts complained when they launched the Skylake Core i7-6700K with a 4.0/4.2 GHz rating, as this was below the 4.0/4.4 GHz rating of the older Core i7-4790K. At this level, 200-400 MHz has been roughly the difference of a generational IPC upgrade, so users ended up with similar performing chips and the difference was more in the overclocking. However, given the Core i7-7700K comes out of the box with a 4.2/4.5 GHz arrangement, and support for Speed Shift v2, it handily mops the floor with the Devil's Canyon part, resigning it to history.

37 of 240 comments (clear)

  1. First rule of journalism. by ckatko · · Score: 4, Insightful

    If the article ends with a question mark, the answer is "No". Because if they had evidence to say it, they would have just put a period.

    1. Re:First rule of journalism. by DamonHD · · Score: 4, Informative

      I quote:

      Betteridge's law of headlines is one name for an adage that states: "Any headline that ends in a question mark can be answered by the word no." It is named after Ian Betteridge, a British technology journalist, although the principle is much older.

      https://en.wikipedia.org/wiki/...

      Rgds

      Damon

      --
      http://m.earth.org.uk/
    2. Re:First rule of journalism. by sconeu · · Score: 2

      I refuse to believe that the Desktop computer is dead until Netcraft confirms it!

      --
      General Relativity: Space-time tells matter where to go; Matter tells space-time what shape to be.
    3. Re:First rule of journalism. by Gr8Apes · · Score: 5, Insightful

      Also, anybody that expects any real speed-ups from Intel in the next 2-3 years has no clue how long it takes to fundamentally improve a CPU.

      It's been since 2010's release of the 980x that we've only moved up the charts maybe 50% on a per core basis. Note that a 980x is unlocked and can be increased significantly over its stock clocking. A 4790K (the fastest single core performer) can only be OC'd a little bit, so the actual performance differences may actually be significantly less than 50%. And that's just sad given that it's now 7 years later.

      As a final insult, to actually double the performance from 7 years ago, you'll be spending nearly $1500+ for a 10 core 6950, and that's before exercising the considerable headroom of a 980x over that of the 6950.

      --
      The cesspool just got a check and balance.
    4. Re:First rule of journalism. by NoNonAlphaCharsHere · · Score: 2

      Any sufficiently advanced (Intel) technology is indistinguishable from the 4040.

    5. Re:First rule of journalism. by JanneM · · Score: 2

      The only real option, baring some fundamental breakthrough [...] is massively more and simpler cores

      The problem with that approach is that most problems are not infinitely paralleliseable, and some important problems fundamentally do not parallelise at all. You rapidly hit diminishing returns for more cores, and that's before you consider that you need to go beyond a shared-memory architecture beyond a dozen cores or so.

      The newest generation of supercomputers already have big problems finding jobs that actually use all the hardware, and for the next generation people have more or less thrown their hands in the air already and say that except for a few very specialized workloads, the machines will be shared systems, not used for single jobs at a time.

      --
      Trust the Computer. The Computer is your friend.
    6. Re:First rule of journalism. by TheRaven64 · · Score: 2

      Ars hasn't been worth reading since Hannibal left. His were the only articles where I'd read something in a field that I knew about and not only fail to spot any glaring errors, but also learn something new. None of the other Ars authors seems to have even a vague clue about what they're writing about.

      --
      I am TheRaven on Soylent News
  2. No. by Hylandr · · Score: 3, Insightful

    A story comes out like this at least twice a year. The harsh / glorious reality hasn't changed. If you want to get real work done it's going to be on a desktop. Even laptops get docked with a proper keyboard, mouse and at least 1 extra monitor when it's time for heavy lifting.

    Then again one has to wonder at the headline. Tech update 'NEW Cpu!' Combined with the leading question, 'Is the desktop dead'. Will the new Slashdot owners please stop treating these message boards like the alphabet channels and focus on the geek culture? Sure it's yours but can you at least pretend it's not been subjugated by the mainstream entertainment industry?

    Also, any headline that asks a question can be answered with 'No'.

    --
    ~ People that think they are better than anyone else for any reason are the cause of all the strife in the world.
    1. Re:No. by DamonHD · · Score: 4, Interesting

      The harsh / glorious reality hasn't changed. If you want to get real work done it's going to be on a desktop.

      Depends what you mean by "real". Yes, I got paid megabuck(s) in banking to optimise quant algos across cores, CPUs and servers in (eg) the Credit dept at Lehman's, but I find my nominally underpowered MacBook Air (the saleswoman was slightly reluctant to sell it to me when I said I was a dev) to generally be damn good for what I need, including some decent data driven models and analysis, wrapped in not-even-optimised C++ unit tests, and running within a Java-based IDE!

      So, horses for courses.

      Also, I am the happy owner of an RPi that does all the work a Sun server farm used to do for me:

      http://www.earth.org.uk/note-o...

      and I target my primary code to 8-bit MCUs similar to a Z80A form 30Y ago in power, running some nice slim highly-optimised distributed coding.

      Cut your suit to fit your cloth.

      Rgds

      Damon

      --
      http://m.earth.org.uk/
    2. Re:No. by DamonHD · · Score: 2

      and I target my primary code to 8-bit MCUs similar to a Z80A form 30Y ago in power, running some nice slim highly-optimised distributed coding.

      You forgot to add 'up and down hills for 100 miles in the snow on a bicycle backwards while on your way to school.' :) It sounds like what you're working on is mostly text. While a Tablet, phone or laptop can certainly host a terminal window, typing speed is still much faster with a proper keyboard. imho.

      I am actually from Yorkshire and resemble that remark! We fought over our holes in ground...

      But again, my MBAir keyboard is one of the better ones I've used, and I do a lot of typing (including code and words for a living). Laptop ergonomics are not great, but in any case to come back to the original point of the fine article, that hardly has a very strong connection with the CPU type. Or am I misunderstanding you?

      I do live my terminal windows and vi though!

      Cut your suit to fit your cloth.

      Very interesting quote.Following that comparison I wonder how much our smart phones clothe us today?

      If the amount and quality of clothing were expressed in computing power then the first astronauts to land on the moon did so wearing loincloths. now there's a mental visual!

      Very very scanty string thongs.

      The first (Cray X?) supercompter replacements I looked after for an oil major ~20Y ago were ~30MHz CPU and ~256MB of memory (IBM and Sun *nix servers). My not-hugely-smartphone beats those parameters by at least an order of magnitude.

      Rgds

      Damon

      --
      http://m.earth.org.uk/
    3. Re:No. by epyT-R · · Score: 4, Interesting

      So many people without clues these days. Around here, I expect more understanding, even if the reader's needs don't fit the niche.

      If your primary interest is in performance, especially when overclocking, a laptop chassis isn't going to have the thermal dissipation.. Hell never mind that, just try a 5 hour video encode on most laptops.. I wish you luck. They'll hit max temp and throttle big time. I've seen some with warped boards from excessive heat damage..

  3. [sic] by Anonymous Coward · · Score: 2, Informative

    [sic] does not mean what you think it means.

  4. ARM Processors coming to Desktops? by randomErr · · Score: 4, Interesting

    It seem like the major development is switching to portable devices. Will ARM or the new RISC become the new standard in desktops? The Raspberry Pi's are good enough for most people's needs.

    --
    You say things that offend me and I can deal with it. Can you?
    1. Re:ARM Processors coming to Desktops? by Rockoon · · Score: 2

      Plain old RISC was a fad. On general purpose hardware a CISC instruction set with a RISC execution pipeline is hands down better w.r.t. performance, and thats what (for example) both Intel and AMD are already doing.

      As far as ARM penetrating the desktop space, I think its a certainty that the x86 line will eventually fall due to licensing. Intel is losing the FAB edge (they are now arguably just keeping up) and if all these other FAB's cant produce x86, they will still produce something. Maybe ARM takes over the desktop space, or maybe some other yet-to-be-invented architecture, but its going to happen.

      Intel is still the biggest FAB company with the most revenue, but they no longer have a majority share of either production or revenue, and these two metrics will only get worse and worse for Intel. When as is inevitable a non-Intel chip greatly outperforms anything Intel can fabricate, that may finally be the year of the Linux desktop, and then Intel will either go the way AMD did and split/sell off its FAB's, or the way Motorola did and abandon the business for something else.

      --
      "His name was James Damore."
    2. Re: ARM Processors coming to Desktops? by xiando · · Score: 2

      Yes, they are absolutely coming to .. servers and laptops and eventually desktops. Remember, all we need is the right major crisis and the nations will accept the GNU World Order (Many think David Rockefeller said "New World Order" but GNU is actually pronounced new).

      Today we have something called ReactOS which is an Android distribution for x86/x86-64 computers. I have an older laptop that I put Fedora and ReactOS on in dual-boot and this let me do something interesting: Benchmark Android on said laptop using the same benchmarking software you'd use on any Android device. Guess what, that AMD E1-6010 CPU is weaker than my current cellphone.

      Many people will naturally protest that running win32 software on ARM will be painfully slow. While this is true it's also irrelevant for most people. You don't need win32 to browse websites or post on SpyBook.

  5. joshtops doesn't know about ellipses? by Anonymous Coward · · Score: 3, Informative

    I think that this "joshtops" character may not know about ellipses, and is wrongly using "[sic]" where any sensible person would use "..." to indicate that some text was removed from the quoted material. Even then, given how much text is omitted, any reasonable person would probably just use several separate quotations instead of trying to cram it all into one big and mangled quotation.

    1. Re:joshtops doesn't know about ellipses? by dgatwood · · Score: 2

      So kind of like a hiccup, but misspelled.

      --

      Check out my sci-fi/humor trilogy at PatriotsBooks.

  6. Desktops aren't dead by ErichTheRed · · Score: 5, Insightful

    For me, docking stations and big monitors allow me to use my laptop in a reasonably comfortable work environment. But, there are still use cases for desktop PCs, especially those that aren't shoved into the back of an all-in-one monitor. You're not going to let a call center employee in a regulated, locked down environment pull out his iPad or laptop to work, for example. A cash register is likely going to be some sort of PC, same thing with a kiosk or ATM. And at the high end, workstations are meant for "real" work - though most have the Xeon processors in them. It's an interesting time; desktops and thin clients are sort of merging and tablet use is demanding more of CPU manufacturers' attention. And this makes sense - mobile stuff has the constant pressure to be squeezed into smaller spaces, produce less heat, provide more on-chipset functionality and run cooler at the same time. I'm still surprised when I see a Surface Pro or other convertible tablet and remember that there's a full-fat Intel processor crammed inside that tiny case without melting through the bottom!

    I just think the desktop market is maturing and there's less and less that Intel processors and chipsets don't natively provide. PC processors are already insanely fast and powerful for what typical users throw at them. Desktops aren't dead, they're just a niche market these days, but one that is still there. The pundits want to claim that no one wants a powerful client device and just wants all their stuff streamed from the cloud onto a tablet or phone they don't control. I think that's true in the consumer space, but businesses still have use cases for desktops.

    1. Re:Desktops aren't dead by barc0001 · · Score: 4, Insightful

      That's not really what the "question" in the article was implying though. I completely agree that desktops are going to be a thing for ages to come yet (and I have 2), but the question was lazily trying to point out that performance increases on the desktop are seemingly coming to a halt for newer chips. This isn't really a surprise for me, as I've got a 5 year old i5 2500K in my home machine that is keeping pace with even the newer games just fine as long as I spend a couple hundred bucks every 2-3 years on a new video card. Same at the office. We went to assess our 3 year upgrade cycle for workstations and realized we'd only get a 20-25% boost in peak processing power by spending our full per-person budget on new machines and instead decided to keep what we have, switch all boot OS drives to SSD, max out the RAM and get 32" monitors and we STILL have money left over.

      I'm not sure if AMD's got anything in the pipeline that can shake things up, but if they do, this is their chance (again).

    2. Re:Desktops aren't dead by Bigjeff5 · · Score: 3, Interesting

      I'm not sure if AMD's got anything in the pipeline that can shake things up, but if they do, this is their chance (again).

      Some of the official stuff released about Ryzen look pretty spectacular. It's still not clear whether it will be able to beat Intel in total performance, but it's looking damn close, which is really encouraging to me. Furthermore, they are actually introducing new technologies in the chip, rather than slightly polishing old ones.

      I have my doubts that AMD will fully match Intel this cycle, let alone beat them, but it gives me hope for the future. It's pretty clear right now who is resting on their laurels and who is driving to be the future of CPUs.

      --
      Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
  7. Lets compare by Anonymous Coward · · Score: 5, Informative

    Top Kaby Lake Intel Core i7-7700K @ 4.2GHz has a Passmark score of 12800 for $350 at 95W released Q4 2016
    Top Sandy Bridge Intel Core i7-3970X @ 3.5GHz has Passmark score of 12651 for $770 at 150W released Q1 2012

    So yes, it looks like 4 years got us 1/3 less power and 1/2 price for same performance of the top Extreme Sandy Bridge

    http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i7-3970X+%40+3.50GHz&id=1799
    http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i7-7700K+%40+4.20GHz&id=2874

    1. Re:Lets compare by Anonymous Coward · · Score: 2, Informative

      Yep, my own 7 year old PC gets 11,800 on passmark. I recently got from ebay an old and cheap Xeon W3680, which is 7 years old. But overclocked to 4.3 GHz it's still able to mix it with the latest high end desktop CPUs. It's remarkable to see how little difference 7 years has made. In 2010, a 7 year old PC would be obsolete, not still up there with the latest kit.

    2. Re:Lets compare by Kjella · · Score: 2

      Of course the "top extreme" of anything will always be ridiculously poor value for money. A i7-2600k @ 3.4GHz has a Passmark score of 8488 for $317 at 95W released Q1 2011, correcting for inflation it's pretty much same price, same power, +50% performance increase in six years or about 7% annually. That's ten years to double performance, the next generation will have eight times the performance in 30 years. Granted if it was anything other than computers it wouldn't be that bad, but if you compare the 2010s to any other decade since they discovered the transistor in the 1950s it's like hitting a brick wall. The real question is whether it's just because AMD quit or if Intel really has run out of tricks.

      --
      Live today, because you never know what tomorrow brings
    3. Re:Lets compare by Malc · · Score: 2

      How's the hardware accelerated iQSV HEVC encoding at 4K on that Sandy Bridge system? Oh yeah: it's not possible.

    4. Re:Lets compare by dunkelfalke · · Score: 2

      Top Sandy Bridge is a Xeon E5-2690 which has a passmark score of 20699 at 135W TDP.
      I personally have bought a E5-2665 a few months ago, for only 70 Euros. Not bad for a CPU with a passmark score of 12084. Finding a motherboard was a bit difficult, though. Then again used ECC RAM is far cheaper than desktop RAM.
      Top Ivy Bridge by the way would be also a Xeon E5-2690, but this time V2. Passmark says 16546. Costs some serious money, though whereas the still very fast V1 can be bought for about 300 Euros.

      --
      "It's such a fine line between stupid and clever" -- David St. Hubbins, Spinal Tap
  8. Hyperbolic? NEVER! by Gravis+Zero · · Score: 4, Insightful

    How many times do we need people to declare the "desktop is dead!" or some other equally preposterous hyperbolic statement? Does someone feel like /. doesn't have enough hyperbole because I will just die if there is someone like that. -_-

    --
    Anons need not reply. Questions end with a question mark.
  9. Re:Bottleneck? by Hylandr · · Score: 2

    I would like to see the CPU RAM bottleneck get as much attention as the CPU itself has. Unless we get that addressed I really don't see faster chips doing much good.

    --
    ~ People that think they are better than anyone else for any reason are the cause of all the strife in the world.
  10. Comment removed by account_deleted · · Score: 3, Insightful

    Comment removed based on user account deletion

  11. Maybe dying? by duke_cheetah2003 · · Score: 4, Interesting

    I am definitely a bit underwhelmed by the release of the new CPUs from Intel. They're not really all that much better than Sandy Bridge i7s, which is what I have (2 of them.)

    Is the desktop computer dead? Na. But it may be dying. The improvements we've come to expect over the years has definitely slowed down quite a bit compared to previous jumps in performance.

    Have we reached some kind of 'peak' in designing faster and faster CPU's? I definitely think a kick to the pocket book of Intel is this underwhelming release. If Intel and/or other manufacturers cannot convince users to upgrade their computers it could definitely be trouble for the desktop computer. I certainly don't feel like I need to upgrade, my i7-2600 based PC seems to run anything/everything I throw at it, quite well. Lackluster performance in new generation of computers isn't very wise, because you're going to need a bigger jump to convince people to upgrade. It's of course not helping that older Core series (and Core2's for that matter) are STILL running todays browsers, operating systems and various software quite well. Should be noted, AMD Turion X2s are also about on par with Core2's. Still running todays stuff pretty handily. That hurts the manufacturers a lot, used to be you had to upgrade, now its more like, "might be nice to upgrade, but not really necessary." The more times they release something new and it's lackluster, the more it hurts, cuz people will be in the mindset, like me, "That's not a big improvement, I'll wait for the next big thing." I certainly feel no compelling reason to jump to this new CPU. 600mhz of performance, for the price of basically replacing my entire PC? Na, pass.

    One could get the impression the desktop is a dying breed of computer, I suppose. Certainly seems like things are headed in a different direction (mobile computing, tablets, etc) for mainstream consumers. But I definitely feel like the industry can and will cater to whichever group of people will earn them the most profit. That seems to be mobile computing right now. And it seems like the news reflects this. Seeing much bigger jumps in performance in the mobile CPU offerings (Qualcomm's Snapdragon CPU are darn impressive!)

    1. Re:Maybe dying? by duke_cheetah2003 · · Score: 2

      it's not important if software still runs on your computer, the problem is that after 5-6 years, I don't see a need to upgrade, because the new chip is hardly faster. I would expect the calculations to be 2x as fast. So I would love to buy a new pc, yes everything works, but if I have to wait less long for something I would love to upgrade... The only thing happening here, is I pay 1000$ and I probably won't notice any difference...

      This was the entire gist of my post. If new offerings are only a marginal improvement over what I have now, and I'm likely not to notice much of a change in performance, why should I upgrade? And this is a self-perpetuating problem. The more times they release lackluster improvements, the more times we opt not to upgrade, they lose more profit and decide against developing better cuz better isn't selling.

  12. ZEN ZEN ZEN with more pci-e at the same price or by Joe_Dragon · · Score: 2

    ZEN ZEN ZEN with more pci-e at the same price or less.

    Intel may need to go back to there old tricks again to lock out AMD.

  13. Re:Firefly-RK3399 by Rockoon · · Score: 2

    ARM doesn't have anywhere near the IPC of intel chips.

    But thats on purpose. They are targeting different markets. Intel tried to shoulder in on ARM's power efficiency market and hoped that its greater performance would make up for not being better at power efficiency, but all they ended up with was an under-performing x86 that nobody really wanted. They have since backed off on that push and have instead re-focused on keeping ARM from making a big dent in the server space.

    --
    "His name was James Damore."
  14. Re:The future will be some Russian or Chinese chip by Rockoon · · Score: 2

    They will still be a factor, but all strategies are an evolving landscape as transistor counts increase. A pure RISC was a better performing strategy than CISC at one time, and now a hybrid of both is the best performing.... as transistor counts go up things like translation become a non-issue.

    Over time the best way to get the most work into the pipeline per cycle changes. Right now Intel CPU's can pull in at most 4 instructions per cycle into the pipeline (unless that has changed with the latest update) and then only if the instructions have already been decoded and cached. The ingenuity of this accomplishment is of course great, but when complex solutions are used to solve otherwise simple problems, its clear that things are near a threshhold where a new take may win out.

    --
    "His name was James Damore."
  15. Let's hope AMD's RyZen will cause some progress... by ffkom · · Score: 4, Insightful

    ... at either company. Right now, Intel just has no financial incentive to innovate. Maybe that is going to change in 2017.

  16. But.... Xeon? by King_TJ · · Score: 2

    The impression I'm getting in recent years is that we're transitioning towards a computing world where individual consumers primarily want portables, or alternately, "all in one" or super small form-factor desktops which just use mobile motherboards and CPUs anyway.

    The high-end "power users" who tell you they still need a desktop machine for the work they do are best served by a "workstation" class system, vs. a regular desktop PC. The primary differentiation between a "desktop" and a "workstation"? Seems to be the inclusion of a Xeon class processor, originally intended to go into servers. Secondarily, workstations tend to offer the highly costly video cards optimized for use with CAD/CAM and other graphics design packages.

  17. TechReport Review is Favorable by branchingfactor · · Score: 2

    http://techreport.com/review/3... Conclusion: "If time is money for your work, and your work can take advantages of lots of threads, the i7-6950X is the fastest high-end desktop CPU we've ever tested, full stop. If you don't need all of its cores and threads, however, the Core i7-7700K arguably delivers the best gaming performance on the market for about a fifth of the price. Intel's Extreme Edition CPUs have never been good values, but the i7-6950X takes the definition of "halo product" to eye-watering new heights. If the return-on-investment calculations work out for you, though, the i7-6950X is an amazing chip."

  18. Re: Bottleneck? by Bryan+Ischo · · Score: 2

    Professional software development typically has compilation steps that can use all processors at 100% for minutes at a time. I can easily use 100% of all CPUs for 20 minutes straight when compiling 5 million lines of source code scattered across 25,000 files. Which I do several times per day typically.

    Of course, not all of compilation is embarassingly parallel; there is usually a link step at the end which cannot be multithreaded (at least not by current tools) and which just sits there adding another 5 minutes using only a single core the whole time.