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.

Comments

First rule of journalism.

[ckatko]

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.

Re:First rule of journalism.

[DamonHD]

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

Re:First rule of journalism.

[rdelsambuco]

SIC!

Re:First rule of journalism.

[asylumx]

SIC!

[sic]

Re:First rule of journalism.

[BigBuckHunter]

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.

The articles linked end with periods. The headline ends with a clickbait, troll, sensationalist shit-up-the-internet question. Ars used to be better than that.

Re:First rule of journalism.

[gweihir]

Well, any tech article that proclaims something "dead" or asks whether it is "dead" usually is just a sign of a brain-dead writer. 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.

Re:First rule of journalism.

[sconeu]

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

Re:First rule of journalism.

[Z00L00K]

Desktop computers are dead, tower computers aren't dead.

But the improvements on CPU technology seems to have slowed to snail pace the last few years. Lack of competition combined with less pressure from the market seems to be the cause - computers seems to have reached a flat spot on the push for improved performance for many applications.

Re:First rule of journalism.

[PopeRatzo]

Well, any tech article that proclaims something "dead" or asks whether it is "dead" usually is just a sign of a brain-dead writer.

Every technology that has ever been invented is in use by someone, somewhere for something. So you're right. A technology is never "dead".

Re:First rule of journalism.

[Gr8Apes]

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.

Re: First rule of journalism.

[Anonymous Coward]

That's like associating editing with Slashdot!

Re:First rule of journalism.

[pla]

"Used to".

Ars hasn't been better than that since... Well, since the days when Slashdot was better than having clickbait make it to the FP. :)

Re:First rule of journalism.

[Anonymous Coward]

SIC!

[sic]

That's sic(k) bro!

Re:First rule of journalism.

[unixisc]

Somehow, this headline looks like the exception that proves the rule

Re:First rule of journalism.

[NoNonAlphaCharsHere]

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

Re:First rule of journalism.

[Anonymous Coward]

Sick man, sick! Haha! How charade you are!

Re:First rule of journalism.

[gweihir]

Indeed. My take is that AMD will now catch Intel and maybe move a tiny bit ahead (10-20%) in the years to follow. Intel will find those 10-20% as well eventually, but that is basically it for the AMD64 architecture. Not that I am complaining, I think the raw computing power is pretty awesome. Software wastes most of it though, and frameworks, interpreted languages and clueless coders are the main reasons.

The only real option, baring some fundamental breakthrough (not even on the horizon, caches, pipelining and branch prediction are fairly old tech, but were just to expensive for regular computers up to now) is massively more and simpler cores, i.e. ARM. But for them, the software side is not there at all, because most coders have no clue how to do multi-threading well.

Re:First rule of journalism.

[kuzb]

...which is why ars is no longer worth reading. Everything is clickbait with them.

Re:First rule of journalism.

[Anonymous Coward]

When the mainstream media consists of fake-newsters like CNN, HuffPo, NYT, ABC, and the like, Ars Technica IS hard-hitting, investigative journalism.

Re:First rule of journalism.

[JanneM]

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.

Re:First rule of journalism.

[gweihir]

I am aware of that. Even going to parallelized software done well will not give us much more, but there may still be some real gains to be had in areas like gaming, simulation and classifiers (often misnamed "learning"). One of the nice things of ARM though is that you can have different cores and mix them and that it generally draws much less power. But yes, for many tasks that have no speed-up or really bad speed-up when parallelized, we are now possibly seeing close-to endgame performance. This is not really a surprise either, every tool is finished at some time and cannot be improved anymore.

Re:First rule of journalism.

[93 Escort Wagon]

But the improvements on CPU technology seems to have slowed to snail pace the last few years. Lack of competition combined with less pressure from the market seems to be the cause - computers seems to have reached a flat spot on the push for improved performance for many applications.

Thing is, even for people who actually use a computer to do work - how many tasks are really CPU-limited anymore? Obviously there are some niches where a faster CPU will improve the efficiency of their workflow... but that's a small percentage of an ever-declining overall percentage.

Re:First rule of journalism.

[Gr8Apes]

There is a class of problems that super computers are awesome at. Somethings can be run almost infinitely parallel, e.g. large matrix operations, search routines, protein folding, etc. Another class of inherently ordered taskings depend completely on the speed of the processing, to a point. A large enough set of ordered taskings can be run in parallel if efficiency is not a concern.

The biggest problem with games is most of the coders, sadly, don't have a clue about running decision trees in a parallel system. It requires an entirely different mindset to code such a system.

Re:First rule of journalism.

[ChrisMaple]

I've just been using some advanced image sharpening software. Waiting an hour between runs to see if I should adjust some input parameter is tedious. I could use a 1000X speedup, and I don't see that happening in my lifetime.

Re: First rule of journalism.

[pchasco]

An overbroad and irrelevant statement. First, "most coders" don't need to understand running decision trees in parallel in order to implement such a thing. Just like every coder doesn't need to understand the intricacies of the x86 CPU or have a PhD in mathematics. For these specialized aspects of a games architecture you hire specialists. Second, most home PCs are running quad to octa core CPUs. It doesn't take a genius to eek out all the performance you can per clock. This is a well understood problem and there is no shortage of solutions. Developers are still constrained by the hardware, plain and simple.

Re:First rule of journalism.

[epine]

For a long time articles on Ars by Jon Stokes pretty much set the standard for enthusiast rehash. Plenty of real journalism could only wish to be as good as much of what Jon wrote back in the day.

Can't say, though, that I'm as impressed with his recent output.

Why I "Need" an AR-15

But I'm even less impressed with this:

The AR-15 has to go: Sorry, Jon Stokes, but your toy isn't more important than people's lives

Re:First rule of journalism.

[brokenpineapple]

Maybe it's just Intel thats dead.

Re:First rule of journalism.

[Anonymous Coward]

Or intels answer the phi

Re:First rule of journalism.

[Anonymous Coward]

but those articles don't claim that "the desktop CPU dead" at all. So the only claim thereupon is with a questionmark, hence false.

Re: First rule of journalism.

[Anonymous Coward]

Ars has had good content, and continues to do so. The problem is that Ars was at the heart of the Gamergate issues, and their forums are filled with angry lunatics that dogpile anyone who goes against the narrative. I declined to renew my subscription because of that bullshit.

Re:First rule of journalism.

[AmiMoJo]

Same thing seems to be true of GPUs. A 5 year old card from the HD7000 range is still pretty competitive. It's only really when you start to crank up to 1440p or 4k that newer cards really start to be a worthwhile upgrade, or if you really need the latest DirectX support / Freesync.

The only thing that might make up upgrade from a 5 year old GPU at this point is that the fans ramp up when scrolling in Chrome at 4k. Performance is fine, it's just that it gets a little bit warmer than newer models.

Re:First rule of journalism.

[TheRaven64]

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.

Re:First rule of journalism.

[TheRaven64]

Even when things are CPU limited, you hit diminishing returns because you care about the reciprocal of the speed. A task that takes an hour and a half is a significant bottleneck. Double the speed and it's still a bottleneck at 45 minutes. Double again, and 22.5 minutes is annoying. Double again, 11.25 minutes is a pain. Double again, 5.63 is a tolerable delay. Double again, 2.82 minutes isn't really much of an improvement. One of the projects that I work on took about an hour and a half to do a clean build when I started working on it. Now it takes me about 5 minutes on my laptop. If I do the build on one of our nice build servers (24 cores, 768GB of RAM, big SSDs), then it takes 1-2 minutes, which isn't a speedup that I care about most of the time.

Re:First rule of journalism.

[zwarte piet]

Well you need that power to keep those 150 tabs with flash ads running.

Re:First rule of journalism.

[Anonymous Coward]

No, the answer could be yes. Otherwise it would not be a question. If the headline was "Has Chris Katko stopped beating his wife?", the answer "No" is not one we would like to think applies.

The law you refer to is dependent on the type of question and the ideological slant of the article author or site editor. E.g. "Is Netcraft Dead?" on slashdot. or "Is Linux a failure?" on theregister by Orlowski.

Re:First rule of journalism.

[Gr8Apes]

Good to know, I'm still running a HD57xx series in my work rig, and intending to move to a 4K monitor soon. I've actually been considering a 960, but you're indicating I may be able to view more options for price/noise/heat comparisons. Noise/Heat are my primary considerations after meeting the core performance requirements of driving a 4K screen.

Re: First rule of journalism.

[Gr8Apes]

An overbroad and irrelevant statement. First, "most coders" don't need to understand running decision trees in parallel in order to implement such a thing. Just like every coder doesn't need to understand the intricacies of the x86 CPU or have a PhD in mathematics. For these specialized aspects of a games architecture you hire specialists.

Except most game shops don't hire those specialists, which is why so many suck in this area.

Second, most home PCs are running quad to octa core CPUs. It doesn't take a genius to eek out all the performance you can per clock. This is a well understood problem and there is no shortage of solutions.

Apparently it takes more of a genius than those coding most games today. I'm running a hexcore with 12 threads, and in any game I've played recently (admittedly a smaller number every year) I've yet to see a game peg the CPU at all. A core or 2? Yes. 6+ cores? Never. Has the game slowed down? Absolutely. I can run a restricted Handbrake job, limited to 6 or 8 threads) in the background without affecting game play at all.

Re:First rule of journalism.

[Anonymous Coward]

There needs to be a fundamental breakthrough in software development technology, which brings even the clueless idiots capable of doing decent multi-threading.

Unfortunately, getting rid of all the clueless idiots thinking they are coders is not possible :(

Re: First rule of journalism.

[pchasco]

Again, the bottleneck is not likely the CPU in this scenario. Memory bandwidth and latency are key factors in seeing the CPU fully utilized. What's a the typical memory latency these days on x86? 12 cycles? Games must make tradeoffs in how many tasks can be run in parallel vs how efficiently they access ram pages. Cache misses kill performance on pipelined CPUs. Work is organized to minimize cache misses. This requires that some tasks are run in serial, essentially creating an upper limit on how parallel the games code can be run.

Re: First rule of journalism.

[Hognoxious]

It doesn't take a genius to eek out all the performance you can per clock.

Hickory dickory dock...

Re:First rule of journalism.

[gweihir]

I agree on both counts. Unfortunately, intense research into this direction (parallelization, not getting rid of clueless idiots) has been running for more than 50 years with basically no good results except for some search and simulation problems that are easy to parallelize. I think most of that research should be stopped, and we should instead research how to get rid of clueless idiots.

Re: First rule of journalism.

[Gr8Apes]

And yet video processing (100% cache misses as each frame is loaded from that super fast spinning disk that puts RAM access to shame) works just fine loading all available cores to near 100% normally using 1 working thread per logical core. I can do the same using multiple other domains. I have done so with a networked game long ago running in Solaris, and have coded multiple clients to work concurrently on a single data set. So sorry, everything you're pointing out is related to the single-threaded synchronous brain-dead approach used by the core of most games, not the physical limitations of the actual system. Writing a robust asynchronous queued message passing system with managed worker pools with synchronous results is not simple, and most games have enough bugs as it is that adding such a mechanism would melt minds during debugging. And no, functional programming isn't a magic bullet for that scenario.

Now do cache misses kill performance? Of course. Do games hit this limit? Show me a well-written one on today's hardware. Note that you have to exclude all GPU access in this context, as we're talking about CPU/memory performance, not memory to GPU and GPU processing, which can be admitted bottlenecks. Do terrible internal architecture and data structures kill performance in games today? It's almost guaranteed. You can write less than 10 lines of code that process data that takes 1000 times longer than making a few minor procedural access changes, on the exact same data structure. The problem with most programmers is that they have no understanding of how memory and data access works, nor the costs involved.

Re:First rule of journalism.

[lsatenstein]

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.

AMD's new ZEN cpu chips are superior to anything Intel is offering. Intel, its time to chase AMD, and while you are at it, chase AMDs pricing too.

Re:First rule of journalism.

[Agripa]

Thing is, even for people who actually use a computer to do work - how many tasks are really CPU-limited anymore? Obviously there are some niches where a faster CPU will improve the efficiency of their workflow... but that's a small percentage of an ever-declining overall percentage.

Transcoding, simulations, and my email client doing searches are CPU limited for me but the biggest time wasters are poorly programmed applications like Firefox.

Re:First rule of journalism.

[david_thornley]

It's easy to get rid of clueless idiots, once you can identify them. The hard part is finding developers who aren't clueless idiots, and so lots of software development is done by clueless idiots because they're available.

Re:First rule of journalism.

[raremediumwelldone]

That's the problem. Your average user does what with their computer? Facebook, YouTube, types something up on Microsoft Office, checks their email, etc. My father got a hell of a deal on a laptop with an AMD A6 APU in it. It flies through everything he gives it to do, because, he uses it for tasks like I listed above. Sure, there will always be high-end high-requirement people wanting 8k 8X antialiasing in their games and 120+ fps for their freesync monitors etc, but... probably 80 - 90% of users couldn't care less. Even today, you could give your average user a Core 2 Duo or Core 2 Quad from 2008 and it would laugh at any tasks they gave it to do.

Re:First rule of journalism.

[Anonymous Coward]

>> SIC!
> [sic]

Sputnik!

Re:First rule of journalism.

[Anonymous Coward]

Apart from some hashes and compressors which are (mis-)designed to be inherently sequential and turing machine emulation, in which situation the longest dependency chain is more than 1% of its total op count and cannot be improved upon?

Come on. 1 second of waiting means one million random memory lookups chained in such a way that each depends on the previous one.

However there are at least two things that make parallelization endgame seem closer than it is:

1. A parallel algorithm might need more than 4 times the number of operations a straightforward sequential algorithm does. Thus, with current cpus parallelization doesn't pay off much yet.

2. Our curren't cpus and languages aren't designed for automatic function-call-level parallelization. No wonder, that doesn't make much sense with 4 cores either. Some could say it's pretty bold move to require all existing software to be rewritten too.

Re:First rule of journalism.

[gweihir]

And that is the problem: Most managers are clueless idiots as well, so the "once you can identify them" part is the hard one.

[sic]?

[Anonymous Coward]

I'm not entirely sure why "[sic]" is written after the line,

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 I had to guess I'd say because of the phrase "in brave new . . . world" without the use of an article, but I'm not sure why they put the annotation so far away, at the very end of the sentence.

Re:[sic]?

[Anonymous Coward]

I'm not entirely sure why "[sic]" is written after the line,

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 I had to guess I'd say because of the phrase "in brave new . . . world" without the use of an article, but I'm not sure why they put the annotation so far away, at the very end of the sentence.

TFA does in fact read:

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.

Please see WIkipedia:

The Latin adverb sic ("thus"; in full: sic erat scriptum, "thus was it written")[1] inserted after a quoted word or passage,

[sic] after a word indicates misspelling, [sic] after a sentence in this case indicates a missing word.

Re:[sic]?

[pla]

Except that still doesn't make it right, because there is no word missing.

Just because Huxley chose to use an article, doesn't mean every subsequent use of that phrase needs to as well.

No.

[Hylandr]

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'.

Re:No.

[DamonHD]

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

Re:No.

[Jawnn]

I'd say it depends more on how you define "get it done". I can perform any of my regular chores on my laptop, but not as quickly nor as easily. My desktop has more grunt per dollar spent on the basic platform, and has three displays and a much friendlier mouse and keyboard. For my workload, looking from one monitor to another is a lot more effective than alt-tab, repeat.

Re:No.

[Anonymous Coward]

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.

WTF are you talking about? Since when do a "proper" keyboard, mouse and monitor have anything to do with performance? My quad-core i7 laptop is well more than plenty for the development work I need to do. What kind of "real work" are you doing that can't be done on a laptop?

Re:No.

[Hylandr]

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.

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!

Re:No.

[fluffernutter]

I can't imagine buying a laptop for a machine that sits in my basement to hold a lot of files for my family to use. Or a machine to back up that machine.

Re:No.

[Anonymous Coward]

I take it you don't type much over 40wpm?

Re:No.

[DamonHD]

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

Re:No.

[Anonymous Coward]

A couple of years ago, I was shopping for a desktop computer, and ultimately bought an AIO. Upgraded it recently - found out that it uses laptop components, like CPU, memory and so on. Aside from gaming, is there a compelling reason for computers not to use laptop components? The AIO has worked excellent for me - particularly since it stays in a dust rich environment, and I don't need to move it anywhere, but there isn't a reason for it to sip more juice.

So I'd say yes, the desktop CPU IS dead. You're not getting much of a performance boost, but you are still consuming a lot more power, so there's not a good reason to keep making those.

Re:No.

[SoftwareArtist]

I think you misread the headline. It asks, "Is the desktop CPU dead," not, "Is the desktop dead." This is nothing to do with keyboards and mice. It's about whether there are CPUs that are specifically designed for desktops, ones that are a lot more powerful than the ones in laptops.

Re:No.

[Hylandr]

Looks like another headline edit.

Going to have to start making screenshots. CNN does the same thing to it's headlines.

Re: No.

[Anonymous Coward]

I type around 70-80 wpm on my MacBook. About the same as a desktop keyboard.

Re: No.

[Anonymous Coward]

then it's likely you'd benefit from a real desktop keyboard, unless you've got really small hands..

Re:No.

[epyT-R]

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..

Re:No.

[Hylandr]

Now THIS looks promising!!

https://hardware.slashdot.org/...

Re: No.

[Anonymous Coward]

My analysis, video work, and development get on just fine with a decent portable. The resolution is high enough that I haven't connected an external display in years except where I need two screens as opposed to mirroring. I use an external mouse just minimize RSI.

I certainly seem to be getting paid well for something that isn't real work.

Re:No.

[GonzoPhysicist]

I guess the flip side is that it's small enough to put in your fridge while doing that rendering.

Re:No.

[epyT-R]

haha.. Kinda defeats the purpose though. A fridge isn't exactly portable.

Re:No.

[david_thornley]

Most things people do with computers really don't need all the memory and CPU power of a high-end desktop. The software is usually limited by human speed for input and can take advantage of human speed for output (you really don't need to give a response in less time than about a tenth of a second). I expect single-user desktops to become a niche market, and they'll tend to be based on servers (which do benefit from additional power).

Re:No.

[Hylandr]

For the vast average business use yes. But when when you start developing locally, gaming, rendering audio / video \ image editing you're going to want a powerful PC.

There's nothing quite like making an edit to an image and have to wait 15 minutes to see the result. It's painful. People that make giant movie posters or touch up old scanned-images for example work with giant file sizes that's well beyond editing a photo or video you took with your phone.

It sounds like you mean well, but lack the exposure to the real heavy lifting PC's do.

Re:No.

[Anonymous Coward]

Heh, even with identical performance I prefer a desktop simply because it tends to be quieter. Maybe 15 years ago it was the other way around but not anymore.

Re:No.

[david_thornley]

Right. Something of a niche market. I'm in that market, in fact, and what I really want from my work computer is faster compile and link times. Most people don't do the things you mentioned, so we're in a market niche.

Re:No.

[Hylandr]

I agree, though that niche is really huge still. :)

[sic]

[Anonymous Coward]

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

Benchmarks look good though

[Anonymous Coward]

While it seems like a small upgrade the benchmarks look promising. The support for Optane storage on the Z270 chipset will hopefully be worth it too, as long as it doesn't cost and arm and a leg to get an Optane storage device.

Re:Benchmarks look good though

[HornWumpus]

Much recent bottom line performance can be attributed to faster RAM and a faster chipset to support it.

CPUs have been RAM pipe constrained for decades.

ARM Processors coming to Desktops?

[randomErr]

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.

Re:ARM Processors coming to Desktops?

[Luthair]

If we're talking Windows running x86 software on ARM, I doubt it. I have hard time believing we're not going to be seeing Netbook 2.0 here. The top ARM processors aren't as powerful as the commonly used x86 processors (which incidentally people claim the base Macbook isn't powerful enough), then add a translation penalty.

The second half of this equation is also, if the manufacturer goes for the cheaper, slower CPU they will also do the same for every other part and end up with a slow piece of junk.

Re:ARM Processors coming to Desktops?

[tepples]

I have hard time believing we're not going to be seeing Netbook 2.0 here.

If it's Netbook 2.0, count me in, because a 10 inch laptop was small enough not to be quite as obvious of a target for thieves compared to a 13-17 inch laptop.

The top ARM processors aren't as powerful as the commonly used x86 processors

But are they more powerful than the 1-core, 2-thread Atom CPUs from the netbook era, especially when skipping the translation layer by running software recompiled by its publisher or (in the case of free software) by the user's copy of GCC or Microsoft C++? One reason Surface RT 1 and 2 failed was that Microsoft deliberately locked out publishers and free software users from recompiling their Win32 desktop applications for ARM ISA, instead requiring all third-party applications to be what are now called UWP applications.

Re:ARM Processors coming to Desktops?

[Anonymous Coward]

People will complain if the streaming services fail to work. DRM is the new killer application, sadly.

Re:ARM Processors coming to Desktops?

[caseih]

I have my doubts. Like I've said before, I have a drawer full of various ARM devices that all turned out to be less useful in real life than they looked on paper. The main problem is that there is just no standard for ARM socs. Each one requires a custom kernel and distribution. They don't have common hardware trees, and most importantly they lack a common, open boot loader. So you're always fighting with some custom uBoot. Would far rather have a normal EFI bios in there and have the ability to boot off of common devices like usb sticks, optical drives, hard drives, etc.

I was trying to have an ARM board act as a router for me, but in the end I got a lot more utility out of a cheap tiny Intel machine for just 3 times the cost. I could use a standard, stock distro on it, and it had loads of RAM and lots of processing power. Way more utility for me.

That said I have a couple of cool ideas for a Raspbery Pi. But actually I think I'd prefer a tiny x86 board that had a lot of GPIO on it.

So no, I don't expect ARM to replace x86 on the desktop, or even the laptop, any time soon. In fact I just bought a fairly cheap x86 (Atom or something) Windows 10 tablet that is surprisingly useful, and can even run standard desktop x86 Windows apps if you really needed that.

Re:ARM Processors coming to Desktops?

[erapert]

Agreed. The thing that is holding ARM back is just how standardized x86 is and how fly-by-night and slipshod ARM's infrastructure is.

Re:ARM Processors coming to Desktops?

[Rockoon]

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.

Re: ARM Processors coming to Desktops?

[xiando]

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.

Re:ARM Processors coming to Desktops?

[0100010001010011]

slipshod ARM's infrastructure is.

That and the quazi-open status of most chips is what kills it for me. I had a SheevaPlug years ago that was great. It ran my house's HVAC and web server and did some light downloading. The uBoot was fairly straight forward and the Kirkwood chipset made its way into Debian. I never had a reason to replace it so I didn't.

Recently I got a CubieBoard since it billed itself as "Open Source Hardware". It was shit. Nothing on it was open. uBoot was a mess. It only ran specific versions of Ubuntu that didn't have anything 'opensourced' from a Chinese manufacturer. It overheated and shut itself down constantly. Then had the same experience with a OLinoXino and Wandboard.

With what I spent on useless ARM paperweight I wish I just got myself an old low powered embedded x86 board. BIOS and UEFI may not be all that, but for the most part it's consistent.

I've held out hope for years for someone to introduce a cheap ARM laptop that had a 12 hour run time.

Re:ARM Processors coming to Desktops?

[farble1670]

I have my doubts. Like I've said before, I have a drawer full of various ARM devices that all turned out to be less useful in real life than they looked on paper. The main problem is that there is just no standard for ARM socs. Each one requires a custom kernel and distribution. They don't have common hardware trees, and most importantly they lack a common, open boot loader.

^^^ This.

Folks that can't understand why it's such an effort for their carrier to update the Android OS on their device, or why they can't just compile AOSP and flash it onto their phones should read this.

Re:ARM Processors coming to Desktops?

[Anonymous Coward]

In the mid to late eighties, ARM was a desktop processor - and running enough circles around Intel at the time to be able to emulate x86 in software.

Re: ARM Processors coming to Desktops?

[Anonymous Coward]

Today we have something called ReactOS which is an Android distribution for x86/x86-64 computers.

ReactOS is not Android, nor is is Linux (but there has been co-operation with Wine): it is a clone of Windows from scratch.

Re: ARM Processors coming to Desktops?

[Anonymous Coward]

ReactOS from Wikipedia. More information on http://ReactOS.org

"ReactOS is a free and open-source operating system for x86/x64 personal computers intended to be binary-compatible with computer programs and device drivers made for Windows Server 2003."

Development began in 1996, as a Windows 95 clone project, and was continued as ReactOS in 1998, with the incremental addition of features of later Windows versions. ReactOS has been noted as a potential open-source drop-in replacement for Windows and for its information on undocumented Windows APIs. As stated on the official website, "The main goal of the ReactOS project is to provide an operating system which is binary compatible with Windows ... such that people accustomed to the familiar user interface of Windows would find using ReactOS straightforward. The ultimate goal of ReactOS is to allow you to remove Windows and install ReactOS without the end user noticing the change.

Re:ARM Processors coming to Desktops?

[Anonymous Coward]

But as more windows software becomes universal and therefore delivered as some form of il to be jit:ed your objection will be less valid for the majority of computer users.

Re:ARM Processors coming to Desktops?

[AmiMoJo]

As performance applications get better at making use of multiple cores it will get easier for ARM to match and eventually out-perform x86.

I've lost the link now but there is a video on YouTube of a guy who underclocked his Intel CPU from 4GHz down to 1GHz. In some games it makes barely any difference, because they care more about having multiple cores than about each core being really fast. ARM, being low power and cheap, is well positioned to take advantage of that and have had 8 core mobile CPUs for years now.

Re:ARM Processors coming to Desktops?

[zwarte piet]

raspi is dead slow for general browsing.

Re:ARM Processors coming to Desktops?

[erapert]

Pi-top costs about $300.
Or you could print your own and put an ODROID XU4 or C2 inside it instead...

[sic]

[Anonymous Coward]

I do not think it means what you think it means.

The future will be some Russian or Chinese chip

[Anonymous Coward]

The future will be some Russian or Chinese chip. Intel is too fat a monopolist to not eventually fall on its own laurels.

The Russkie Elbrus-8S is architecturally interesting and only limited by the 28nm process available to the Russkies. Its also not vaporware but is being delivered to customers RIGHT now. If they can close the process gap with the West they can compete with Intel. But that's a VERY big "if." Right now though at 28nm it at least fulfills national security functions....

Re:The future will be some Russian or Chinese chip

[Archangel Michael]

The future is already here, in the form of ARM CPUs. They are pretty much without a whole lot of the baggage the INTEL processors have, even if it is "emulated" at this point. There were a few good break points that were possible during the long run of the x86 where Intel could have dropped legacy support, but failed to do so.

Re:The future will be some Russian or Chinese chip

[MightyYar]

I know transistor counts are way up, but Elbrus is VLIW, and I don't understand how the limitations of Itanium or the Transmeta chip won't still be a factor. It's not as if the Russians are the first to try VLIW.

Re:The future will be some Russian or Chinese chip

[erapert]

The way things have been is not necessarily the way things will continue to be.

Re:The future will be some Russian or Chinese chip

[Anonymous Coward]

It's not a pure VLIW. It's a dual VLIW for starters. Transmeta suffered from a low transistor count couple with an outdated fabrication process. The Russians have at least not made the transistor count mistake. It's a BIG die and needs a BIG heat sink. I'm not going to pretend that the Elbru-8S can at present compete with a modern i7. i3 maybe. It's at least 5 years behind where it needs to be but at least the architecture is novel and they are working on a 14nm follow on. They need to pump a lot more money in but at least the design school exists.

Re:The future will be some Russian or Chinese chip

[Rockoon]

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.

Re:The future will be some Russian or Chinese chip

[Anonymous Coward]

Already all Russian hardware engineers who could speak English from the Elbrus project have been working for Intel for over a decade, if I'm not mistaken. Meaning that Intel CPUs already have the best / most practical features of Elbrus.

Re:The future will be some Russian or Chinese chip

[Anonymous Coward]

Your reason is "the next chip will be from russia or china because things can change?" are you a fucking retard? people make future predictions based on current and historical information, not random thoughts prefixed by "maybe or possibly."

Current and historical information tells us US or Korea will be the leader in the next generation of processors. current and historical information tells us that china and russia have been trying for the last decade to come up with something that at least matches current shit, and even though they have full access to current tech, they can't even copy everyone else - their latest is 10 years behind, and they don't show signs of improvement. They literally have proven themselves incapable.

How about India? Why not? Lots of smart IT-type people over there. I predict that India is going to be the next CPU manufacturing powerhouse. Why? already answered. lots of smart IT-type people. Zero history of creating anything advanced or revolutionary in technology, nothing to make me think they will, but hey - I just think maybe they will. It's a feeling - ya know.

  you're a fucking idiot.

Inability to analyze data and trends. Random unfounded assumptions w/o basis. I'm guessing you're into religion.

this douche's next day at work (Seinfeld sound):
good morning team. This is Bob, the developer that has not had a single successful project in his lifetime, and has had every effort he has been engaged in consistently fail, and has a proven track record of being bad at his work. He will be the team lead from now on, as I feel he is going to be very successful in rolling out our new product. Why you ask? Because things often change.

you. are. a fucking retarded idiotic stupid moron. slice your fucking veins you fucking doorknob.

Re:The future will be some Russian or Chinese chip

[Miamicanes]

That's because dropping what you call "legacy support" would have erased their main advantage over ARM -- raw balls-to-the-wall no-compromise performance. In car terms, an i7 is like a maxed out Tesla Model S P90D with "ludicrous mode" switched on... by comparison, an ARM is like a Chevy Volt.

Re:The future will be some Russian or Chinese chip

[Archangel Michael]

Take a look at Cavium ... 48 cores x 64bit at 2.5 ghz.

http://www.cavium.com/ThunderX...

Re:The future will be some Russian or Chinese chip

[jon3k]

You mean the CPU being absolutely trounced in the watt/FLOP benchmarks by Intel?

Key benchmark left out

[Anonymous Coward]

One of the biggest benefits of moving from Sandy Bridge to Haswell or Skylake was improvements in frametime which reduces perceived hitching in games. This article would be more convincing if they measured those differences and single core performance as well. It was well worth the upgrade for me upgrading from an overclocked 2600k to a stock i5-6600k with a GTX1070. With VR, frametime is a huge factor and one of the reasons Sandy Bridge setups don't meet minimum requirements, despite how well they perform in traditional benchmarks.

Bottleneck?

[Anonymous Coward]

When is the raw CPU speed a bottleneck anymore? If you are like me, not often. Other than a few edge cases, I have rarely had to sit around for processing power. A few years ago, my real bottleneck was disk I/O or network bandwidth. Now with solid state, I only deal with the network and with 20Mb down 5Mb up and low latency, I'm rarely waiting on anything anymore.

Now, I am sure there are many of you who could use more horsepower at times, but we've seen the old law of diminishing returns and consequently far less innovation that we saw in the Good Old Days(tm).

Re:Bottleneck?

[Hylandr]

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.

Re: Bottleneck?

[Anonymous Coward]

When is raw CPU speed a bottleneck?

In a Digital Audio Workstation (DAW), among other things. Lots of near-realtime DSP going on. Fortunately it parallelizes well, but some of my sessions will bring my 4770k to over 80% total CPU usage on every core.

Re:Bottleneck?

[DamonHD]

Quite a lot of what I paid attention to in banking work was that bottleneck, and it has been an issue since the earliest days of computing (my old prof would roll his eyes and talk about data stalls on the MU5...).

So one job of making stuff run well is to cut bloat and make more of it fit in cache, have fewer branches/misses in inner loops, and reduce data flows generally where possible.

Actually, I'm enjoying the ATMega328P with NO caches and a whole 2k of RAM! B^>

Rgds

Damon

Re:Bottleneck?

[AuMatar]

Games, all the time. I'd love if a turn of Civ6 took less time. I'd pay more for that then I would a network or disk speedup.

Re:Bottleneck?

[Rockoon]

When is the raw CPU speed a bottleneck anymore?

Whenever the user waits for a computation to complete. Lots of filters in photoshop, for example, are waited for by the user that is applying them.

Re:Bottleneck?

[the_Bionic_lemming]

Cyberlink Power Directory slams my i7-5930K to 100% use for a long time when I compile a movie together.

Re: Bottleneck?

[Bryan Ischo]

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.

Re: Bottleneck?

[Anonymous Coward]

But I'm not like you and I'm grateful for that.

Re:Bottleneck?

[Agripa]

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.

That will be a tough row to hoe unless DRAM is replaced with something else. Increasing integration helps with bandwidth and parallelism but helps little with sense amplifier latency. That leaves increasing memory parallelism and caching as the only options and we have been doing that.

Some old high performance computer designs replaced DRAM with SRAM but at considerably increased cost and lower density.

joshtops doesn't know about ellipses?

[Anonymous Coward]

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.

Re:joshtops doesn't know about ellipses?

[dgatwood]

So kind of like a hiccup, but misspelled.

Desktops aren't dead

[ErichTheRed]

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.

Re:Desktops aren't dead

[barc0001]

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).

Re:Desktops aren't dead

[Bigjeff5]

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.

Re:Desktops aren't dead

[Kjella]

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.

No, but neither is he likely to use a proper desktop, Thin clients and virtualization are so much easier to deal with if you consider fixed locations and centralized control to be a feature. Sure there are those with particular workstation or input/output device needs but not the average corporate desktop. If it wasn't for gaming I think the desktop would be relegated to a small, small niche.

Re:Desktops aren't dead

[Joe_Dragon]

say a call center running Thin clients over dual screens is a lot of network bandwidth. In some settings you may need dual / 1 big screen to be able to show lot's of info at the same time.

Re:Desktops aren't dead

[theendlessnow]

AMD Annouces Ryzen

AMD: We're now in second place! Second place!!

Re:Desktops aren't dead

[johannesg]

For me, docking stations and big monitors allow me to use my laptop in a reasonably comfortable work environment.

The article is about development of new high-end CPUs. If it turns out we have reached the end of the line in what's possible, then laptops will be hit equally hard. Mobile CPUs are not that far behind. For the rest of your rant, great job at not understanding the article.

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.

Yes, I as a consumer totally want other companies to have full control over my data and my life in their 'cloud'. Hiccup? "Sorry, according to our terms we have no responsibility or obligation to even try to keep your data safe" - and that's that for a lifetime of photos (memories!), legal documents, etc. Or how about "a random person whose identity we won't disclose has filed an unspecified complaint against your account and now we have closed it without possibility for recourse", aka. Facebook syndrome? And I so love lag when I'm gaming, I really want to pay an online company so I can have even more.

There is most definitely still a usecase for desktops at home. I don't need a tiny box with integrated tiny useless keyboard, integrated tiny useless screen, and so many problems getting rid of its heat that it needs subpar parts just to be able to run.

Re: Desktops aren't dead

[Anonymous Coward]

The placing doesn't matter. What matters is that you're making money.

Re:Desktops aren't dead

[Anonymous Coward]

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).

I also have an i5-2500k that has been overclocked to 4.5ghz since I built the machine a few years ago. There is literally no reason to upgrade my CPU yet. I have upgraded other components such as video card (MSI 970 OC Edition) and have 2 SSD's along side my 2 large HHD's.

Re:Desktops aren't dead

[lsatenstein]

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).

If 55 inch color TVs with 4k support can be had for under $1000, why can't we have a desktop with 32gigs ram, and some SSD slots for $500.00

Re:Desktops aren't dead

[Immerman]

You can. 32GB of DDR3 ram is only ~$160. Mediocre case, power supply and motherboard will get you to maybe $250-300, leaving almost half your budget for a decent CPU. Or an adequate CPU and a small SSD to fill one of the many available slots to get you started.

If you want a high end gaming PC for that price... tough, you'll need to roughly double it for a powerful video card, more storage, and probably a better CPU. But that has nothing to do with TVs because the technologies are almost completely independent of each other. CPUs, RAM, and SSDs all at least use largely the same technology, optimizing similar transistors and manufacturing technologies for similar purposes. TVs use very little of any of that technology, probably all purchased as modules to integrate with the screen itself - which is a technology that, in manufacturing terms at least, is largely unrelated to the other three.

Meanwhile the switch from HD to 4k is largely a question of shrinking a highly reliable 2D array of giant honking transparent transistors to fit 4x as many slightly smaller giant honking transistors in the same space. Once you upgrade your production line and iron out the reliability issues, there's very little price premium in actually producing a 4k versus HD TV of the same size. Yeah, you probably need 4x the internal RAM and CPU power to drive it, but you're still pretty much talking about the equivalent of a lower-end smartphone, and at this point there's minimal price difference between what's needed for a HD and 4k TV, at least compared to the cost of the screen itself. Heck, "Smart" TVs are becoming ubiquitous largely because there's hardly any price premium for moving another several steps up the "low-end computing" ladder and having excess power to burn for offering pseudo-tablet functionality.

So, yeah, it's nice that you can get a 55" 4k TV for $500, but it has essentially nothing to do with high-end PCs, except for the fact that you can now complement it with a relatively cheap 55" 4k monitor.

Re:Desktops aren't dead

[jon3k]

A cash register is likely going to be some sort of PC

Funny enough, I see these more than anything lately.

Missing the point

[Anonymous Coward]

It's the underlying 2D Si CMOS technology that's hitting a wall, not CPUs. Marginal gains will continue to decrease for a while as we scale, then we'll probably see a bit of a "wild west" in CPU technology (though little of it might see mass production) before we finally hit our stride with the next great tech.

Yay

[Anonymous Coward]

Built in DRM crap. I'll pass.

Re:Yay

[Anonymous Coward]

I'm assuming it will still be thepiratebay compatible.

On the bright side the Haswells will go down

[Anonymous Coward]

in history as a truly historic chips setting as they have the benchmark for a solid 4-5 years now. The 4770 was released in 2013 and there still isn't an application out today that can make a 4770 owner feel the need to upgrade. Haswell owners certainly got a ton of bang for their money.

Isn't this always the way?

[DarkOx]

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.

Disasters like Netburst aside, is this not the usual pattern.

1) Invest millions in designing a new architecture. Incorporating everything learned about CPU design in the past, try and open as big advantage over your nearest competitors as possible.
2) profit.
3) Make minor revisions to protect your advantage and create an excuse for the high performance market where your biggest margins are to buy new parts
4) profit some more, and with greater margin
5)... repeat as long as competition / existing design allows.
*a) start work on something new leak vapor specs etc to slow adoptions of anyone elses parts
*b) release next big thing
*c) return to 1)

Firefly-RK3399

[xororand]

This ARM board looks promising:
64-bit
4 GiB of RAM
32GB of eMMC flash
802.11ac WiFi
Ethernet
2x M.2 PCIe
USB 3
$199
https://www.kickstarter.com/pr...

or higher end, and much more expensive AMD A1100 series Opteron:
http://softiron.com/products/o...

Re:Firefly-RK3399

[Anonymous Coward]

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

Re:Firefly-RK3399

[erapert]

What're the IPCs of each?

Re:Firefly-RK3399

[Rockoon]

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.

Re:Firefly-RK3399

[fnj]

That's about the best Arm64 board I've seen, but still comes up short despite being overpriced. No SATA, and no DIMM slots for real RAM expandability.

There's nowhere to go

[mveloso]

For CPUs, there's really not a lot that left to do. Stream video? Load Facebook? I'm pretty sure the older chips do that just fine.

The real action is in GPUs.

Re: There's nowhere to go

[Anonymous Coward]

Feel like getting VST audio plugins running there, 'cause they're CPU-killers.

Re:There's nowhere to go

[iggymanz]

for other than gamers and CAD operators? my old computer does movies just fine....

Re:There's nowhere to go

[ChrisMaple]

Quadruple precision floating point. Accelerated matrix math and convolutions. Hardware Bessel functions.

Thats a Review Now?

[Luthair]

Two trivial benchmarks really?

The big change is in "Intel ME" backdooring, price

[Anonymous Coward]

Unless you like the idea of a remote-controlled computer built into your CPU/mobo architecture that you CANNOT DISABLE OR CONTROL, Haswell or Ivy has 99% of the power and feature set as these new Gen6.5 chips at roughly 1/2-2/3 the price.

Lets compare

[Anonymous Coward]

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

Re:Lets compare

[Anonymous Coward]

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.

Re:Lets compare

[Anonymous Coward]

Then the question is why, in 2017, can't I buy an 8700? with a Passmark score of 30000 for $850 at 150+W? What if I'm not concerned with price or power consumption, just raw computing power? Yes, if this continues, the desktop CPU will be dead. Not of natural causes, mind you, but of foul play.

Re:Lets compare

[Anonymous Coward]

Of course, the Kaby Lake chip is not an "Extreme" product line. Those are coming out later in 2017 and will likely be closer in price to that Sandy Bridge chip. It remains to be seen how much it will actually beat it in performance, though.

Re:Lets compare

[Kjella]

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.

Re:Lets compare

[Malc]

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

Re:Lets compare

[Anonymous Coward]

Nobody interested in a K-part CPU gives a shit about integrated video capabilities.

Re:Lets compare

[dunkelfalke]

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.

Re:Lets compare

[sexconker]

Works just fine on my i7-2600k system because I've got an AMD video card that handles it way better than Intel's Quick Sync or whatever they call it ever did or will in the foreseeable future.

Re: Lets compare

[Anonymous Coward]

You mean like the core I7 6950x? Because that's exactly what you're asking for, and it exists

Re:Lets compare

[Anonymous Coward]

In terms of performance, for most applications, you're right. However that Xeon will probably (depends on usage) use close to twice the power a comparable modern CPU uses, which makes me wonder: in the long run, is it actually cheaper, assuming it will be running at least a couple of years? Guesstimating with some numbers from the top of my had I'm thinking the answer will be no (as in it's not cheaper).

Re:Lets compare

[Anonymous Coward]

This is largely why we still use CPUs released 8 years ago at DC. Our workload is not CPU intensive, only I/O bound but even for that the I/O capacity of those are more than we can push through other bottlenecks in the system.

Power consumption is slightly more, but the cost of total system is less than new barebone chassis, so we think we can run those old servers for few decades to even breakeven.

the server chassis is almost free as in beer, peanuts compared to the rest of the config

Hyperbolic? NEVER!

[Gravis Zero]

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. -_-

Laptops are good for transit users

[tepples]

I can perform any of my regular chores on my laptop, but not as quickly nor as easily.

I guess it depends on whether your job lets you ride a bus or train as opposed to driving or cycling. Transit users can make productive use of commute time, for which a laptop is more efficient than not having a suitable computer at all.

Re:Laptops are good for transit users

[DamonHD]

Well, there's a good point, yes.

So I can get my work done at my desk or kitchen table or on the sofa or in bed, as well as on the train and when hot-desking (since my company doesn't have 'an' office, so we meet up ~1/week). Having a desktop in all of those places would be ... impractical.

And as to the GP point about "getting it done", yes bigger more ergonomic displays would be good sometimes, but impractical in many of the places I work, as before.

And in terms of keeping me waiting: I often find that the work can be tuned or partitioned to keep run time fast enough to avoid being annoying. (First make it run, them make it run right, then make it run fast enough not to get in the way!)

Rgds

Damon

Re:Laptops are good for transit users

[Anonymous Coward]

I too have an Air, 11". Cheeky little thing...
I've been doing some EE Development work on it, I'm mostly at the CAD part now. I start off with some brilliant idea in the morning that upwelled during the night, and a couple of hours later, I'm down on the boat, still working on it. There really isn't the room on the boat for a full blown Desktop, with the added complication of feeding it Alternating Current... in the middle of the Bay. This, by the way, is not a new way of working for me... 25 years ago, I was using a PowerBook 100 and MiniCAD+ 3. Smaller boat, though.

One little trick- I often need to draw up a small part or gizmo. It turns out that the $1 Stylus that I bought for the thoroughly unsatisfactory iPad works just ducky on the MacBook Air Trackpad. And with the distracting Wifi turned off, I get a solid 10 hours on a charge.

Comment removed

[account_deleted]

Comment removed based on user account deletion

The desktop might be dead to you

[Anonymous Coward]

If you are a person who likes the underpowered hardware they use in laptops and tablets, or those 2-in-1 things, because you need portability, then maybe you don't need a desktop.

If you need substantial graphics horsepower, liquid cooling, full size expansion cards, spinning media, or just like to upgrade and expand your system then the desktop is not dead. Which applies to gamers, graphic artists, CAD drafters, etc.

Also and this can't be stated enough - the CPU is not the limiting factor anymore for those that use desktops.

Maybe dying?

[duke_cheetah2003]

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!)

Re:Maybe dying?

[oh_my_080980980]

Are people reading challenged. The article was not talking about the end of desktop computers. It was talking about CPU performance. Whether or not there will be performance increases like those in the past is doubtful without adequate competition. The latest line of Intel processors is proof of that. That's all.

Re:Maybe dying?

[Anonymous Coward]

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...

Re:Maybe dying?

[DamonHD]

Um, I don't think it's about CPU performance in general (server and supercomputer CPUs continue to exist and generally outperform desktops in many metrics) but about the dynamics of one *segment* of the CPU market, one that used to be dominant.

Or am I reading it wrong, too?

Rgds

Damon

Re:Maybe dying?

[duke_cheetah2003]

Are people reading challenged.

You certainly are. The blasted TITLE of this article reads: Is the desktop CPU dead?

Think before you post, eh?

Re:Maybe dying?

[duke_cheetah2003]

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.

Re:Maybe dying?

[AHuxley]

The single GPU can just get to 4K without SLI. The next 4K GPU generation for games will need a bit more CPU power.
Until 4K at the very best setting is one GPU ready and needs a new CPU, the CPU profit taking will fill in the release gap.
Solutions exist for the very best in art, photography, move, broadcast media.
So the games are pushing for 4K but thats a gpu and lcd generation away from been perfect at the max quality settings and top frame rates.

Re:Maybe dying?

[Bryan Ischo]

But it would cost enough in R&D and tooling that Intel would not find enough buyers for the faster chips to make that back and still turn a profit. It's just that hard to increase speeds now that absolutely all of the low hanging fruit have already been picked, and most of the medium-hanging fruit too.

The number of customers who will pay handsomly for faster chips is an ever shrinking pool, and there is not enough money left in that pool to fund the R&D necessary for large increases in processor speed.

Re:Maybe dying?

[InfiniteLoopCounter]

It's tough enough already to read the title of the summary and now you're expecting others to read the article?

Re:Maybe dying?

[Anonymous Coward]

From your post:

"Is the desktop computer dead?"

From the title:

"Is the Desktop CPU Dead"

CPU computer.

Think before *you* post.

Re:Maybe dying?

[Anonymous Coward]

Thanks /. for eating my "!=" characters, they were between CPU and computer.

Re:Maybe dying?

[Anonymous Coward]

Good work proving his point. You could have tried reading the comment he was answering before posting

Re:Maybe dying?

[Anonymous Coward]

Actually the solution is not a faster cpu but rather "The machine" as hp calls it.

Re:Maybe dying?

[duke_cheetah2003]

Your slip is most amusing. Because that was exactly what I was going to point out.. what's the difference? Desktop CPU / Desktop Computer, you really can't have one without the other, can you? So they're rather interchangeable, IMHO.

Amuses me you're willing you argue it this far. There's no difference. To speak of one is to speak of both. Silly.

Re:Maybe dying?

[Agripa]

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.

Upgrade when you can double or quadruple the amount of RAM for the same price.

Re:Hyperbolic? NEVER!

[oh_my_080980980]

Yo moron, the title said "Is the desktop CPU dead?" not desktop computers. The article was talking about the performance of the desktop CPU being at a stand still. Try reading the article sometime.

Re:Hyperbolic? NEVER!

[Anonymous Coward]

42

Intel Afraid of Antitrust

[Anonymous Coward]

Hopefully the prices still drop.

What desktops need most, and what Intel has been focusing on, is better power efficiency. Imagine being able to drop a full speed i7 into a tablet or other mobile device (not just a laptop) that can then be docked and used as a workstation with full human interface usability.

Re:Intel Afraid of Antitrust

[zwarte piet]

Yes, a quad core laptop with a passmark of >10000 and tdp of 1 watt or so would be a great thing. Imagine if we can go without fans and charge once a week.

Re:Intel Afraid of Antitrust

[Anonymous Coward]

There are limits to thermodynamics that prevent you from doing a lot of computations in a very lower power. We're not really close to those limits, but the returns on this will diminish.

Still rocking my Core2Quad and my AMD 8core

[Anonymous Coward]

EFI-less BIOSes, and no fancy shenanigans or motherboard upgrade treadmills.

Why the need to constantly buy processorss?

Re:Still rocking my Core2Quad and my AMD 8core

[zwarte piet]

Cause software gets slower and slower.

Hyperbolic

[OrangeTide]

Seems like a low threshold to describe something as "dead". I stumbled on the stairs the other day, luckily journalists didn't start writing my obituary.

Not dead, just plateaued.

[jxander]

Any of the last several generations of Intel CPUs can run any modern application just fun. Up to and including top-end gaming.

There is no incentive to innovate, so there is no innovation. Desktop CPUs will remain in a holding pattern until something happens to force their hand.

Re:Not dead, just plateaued.

[Agripa]

There is no incentive to innovate, so there is no innovation. Desktop CPUs will remain in a holding pattern until something happens to force their hand.

Intel has innovated but in ways other than increasing single threaded scalar performance like increasing vector performance, lowering cost, and lowering power which is a requirement for increased transistor densities. If there was an easy way to continue increasing single thread scalar performance, then it would have been applied to server CPUs and in some cases it has like diamond composite heat spreaders.

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

[Joe_Dragon]

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.

Re:Hyperbolic? NEVER!

[Anonymous Coward]

I remember many similar stories on Slashdot and other computer-related sites from the past 10 years. I said back then that "no, they are not dead" and I will repeat this now too: they are not dead. I'll probably wait another 10 years so I can say it again: they are not dead.

you'd need to get one anyway

[fbhua]

It seems that the motherboard manufacturers are holding out when it comes to incorporating much-needed I/O features. Consider the following:

- USB 3.1 Gen2 Type A and C
- Thunderbolt 3
- M.2 Gen 3 x4
- U.2

Unless you are willing to shell out top dollars for the top-of-the line Z170 series boards by Gigabyte or Asus, you can't get those features. It seems that they think that only gamers need these features, and Z170 chipset is all about overclocking.

Intel 200 series boards will start appearing in CES 2017. Hopefully lower-priced / non-OC boards would have these features and new system builders will pair them with a kaby lake cpu anyway. What's the alternative?

In a bizarre turn of events, old USB 3.0 has been renamed USB 3.1 gen 1, and the newfangled USB C can be plain old USB 3.0, 3.1 or Thunderbolt. So now you have cables that look the same, but perform very differently.

4 cores, yet again

[Moof123]

Big bucks and still 4 lousy cores. Huge amount of R&D went into 10% overall performance increase compared to Skylake (or really anything semi-recent). I want more damn cores, and drop the useless GPU that is wasting silicon area. 6-8 kickass cores should be the norm these days, but intel wants a massive premium for that.

Yes, I know that most software only uses up to 4 cores today. But I don't care. More cores being common will be a big incentive for software developers to find ways to use that untapped power. Build it and they will come.

Re:4 cores, yet again

[DidgetMaster]

I recently upgraded my 5 year old 3770K (quad core) desktop to a new 6800K (hex core). I am working on a data management system that can utilize as many cores as are available. It not only can do more things at once (e.g. multiple separate queries) with more cores, but it can also break up a single query so that parts of it can run in parallel on separate cores. By doing this, a single large query can often complete faster on 6 cores than it can on 4 cores. Even though each core on the 6800K had a lower clock speed than the 3770K, I was still able to perform queries about 50% faster because of the 50% more cores. I assume that improvements in each core itself over the past 5 years just about made up the difference in lower clock speed.

Re:4 cores, yet again

[Anonymous Coward]

I will counter your claim and say that most software today doesn't use 4 cores but 4 cores allow software to use whole cores since the os can do its thing on another core. There are plenty of applications that actually are properly multithreaded/parallel that can use all cores you throw at them but i wouldn't say they are the majority.
You don't actually have to pay intel that high premium for more than four cores if you accept lower clockspeed.

The majority market want discrete gpu:s so these comes first and then a couple of months later the non-gpu version comes.

Re:4 cores, yet again

[Agripa]

Big bucks and still 4 lousy cores. Huge amount of R&D went into 10% overall performance increase compared to Skylake (or really anything semi-recent). I want more damn cores, and drop the useless GPU that is wasting silicon area. 6-8 kickass cores should be the norm these days, but intel wants a massive premium for that.

The huge amount of R&D, and it was not that huge compared to the gains, went into designing a core which could be used on servers, workstations, and desktop products. If you want a server or workstation CPU, then market segmentation means that speed cost money. How fast do you want to go?

Dropping articles makes you sound like Indian.

[Hognoxious]

There should be a "the" before "brave".

Re:Dropping articles makes you sound like Indian.

[Zontar The Mindless]

The book's title is Brave New World. No article, definite or otherwise.

Re:Dropping articles makes you sound like Indian.

[Anonymous Coward]

Actually... what Huxley was ironically quoting went thus:
"How beauteous mankind is! O brave new world, That has such people in ’t!"
So Brave New World starts with an interjection. A little FTFY:

"The i7-7700K is the first desktop Intel chip in, O brave new, post-"tick-tock", world..." actually has a ring and rhythm to it.

Re:Dropping articles makes you sound like Indian.

[Hognoxious]

One, that makes no sense in the context - there's no relation to the book's subject matter so why allude to it?

Two, if he meant to use it that way why didn't he capitalise and italicise it like you did?

Re:Hyperbolic? NEVER!

[Bigjeff5]

As long as Intel is top dog in the CPU market, the Desktop CPU is indeed dead.

However, only an idiot would think that means the Desktop PC is dead.

This is the perfect opportunity for another party to sweep in and innovate the CPU market. Whether it will be a traditional x86 style CPU manufacturer like AMD, or an ARM or RISC style conentder, who knows. But the longer Intel stagnates, the larger the opportunity will grow.

Intel boads don't have the pci-e lanes for that

[Joe_Dragon]

Intel boads don't have the pci-e lanes for that and even with the Intel 200 series boards you are still pushing a lot over the pci-e 3.0 X4 dmi link. With the other X16 going to video.

With AMD they need to have X16 + usb, pci-e storage, etc on there own lanes. To crush Intel and that is the low end the higher end seemes to be X16 X16 + chip set stuff on there own.

Let's hope AMD's RyZen will cause some progress...

[ffkom]

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

Re: No SATA and no RAM expandability

[xiando]

SATA could be fixed. The CPU doesn't support SATA but it does have limited PCI-express so motherboards could just add controller chip. The RAM limit can't be solved that easily, most ARM chips are made with mobile phones in mind and they are generally limited to 4GB or something like that, the most I've seen is 6GB. Motherboard makers just solder on the max amount supported.

I do love that these small boards have neat things that desktop computers just don't have, like 4k camera support and hardware x264 video encoding. How many laptops are sold with anything beyond a garbage 720p camera? You can probably count them on one hand if they even exist.

But.... Xeon?

[King_TJ]

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.

Re:But.... Xeon?

[Moof123]

Workstation GPU's are garbage. Only speaking from what I see Dell stick into the machines we have in my design center, they are crap. I got a brand new workstation when I joined 3 years ago, and the GPU was listed for about $500, and did have 4 mini displayport outputs, but could not drive 4k screens, and had major issues driving 4 1920x1080 screens. My GTX750TI at home was more powerful and drove 4k no problem, and got about 1/3 as much. The only saving grace for my workstation is that it can take up to 64GB of RAM, something consumer machines at the time could not. In my cases I needed every bit of that.

Much of the workstation market is like that, the same CPU performance can be had for 2-3x the price. Alternatively you can get even more cores for the price of a good used car that are not available at the consumer grade. I've heard of ZERO processors wearing out, and I seriously doubt there is anything actually better about the fabrication of a Xeon compared to a consumer CPU. Wanna bet if the workstation SSD's are actually any more reliable?

Re:But.... Xeon?

[twistedcubic]

and...ECC RAM. You can get motherboards with ECC support for almost any AMD desktop processor, but for some reason Intel wants ECC for Xeons only.

Re:But.... Xeon?

[Agripa]

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.

I have been using "workstation" class processors on my desktops since the K6-3 where "workstation" means large amounts of DRAM, ECC support (I curse Intel's marketing segmentation), and lots of fast local storage.

I think what has changed is that previously consumers needed a desktop class processor for routine tasks and passive entertainment and now a mobile processor and often mobile form factor is sufficient. On the other hand if you want to get serious work done, you still need at least a desktop form factor and often a workstation class processor.

Re:But.... Xeon?

[Anonymous Coward]

xeon for autodesk is laughable at best. I went for single thread frequency as core number is irrelevant in autocad.

https://www.pugetsystems.com/recommended/Recommended-Systems-for-Autodesk-AutoCAD-134/Hardware-Recommendations
Built our engineers system and used gtx980ti as well. Saved $3k AUD over a xeon/tesla combo. Tested with the engineer sitting near by as better on the cheaper higher clocked system.

Autodesk is single threaded. Dumb I know, so is xeon for this then. Xeon for gfx? Lol wrong platform,

Re: It's because AMD quit

[xiando]

I see it as pretty obvious that x86/x86-64 CPUs stalled out because Intel decided to milk the market due to the lack of competition (from AMD). Look at Intels offerings in the Xeon E7 Family for examples of exactly why I say this is obvious. It's not like chips that are far better than what is currently offered to average consumers do not exist, they do - they are just priced outrageously.

If AMD delivers with Ryzen and offers something with a good IPC and lots of cores at half the price of Intel then perhaps they will lower their prices on some of their low-end chips.

My guess is what will really force them to finally innovate a bit will be pressure from ARM. Hardware x264 and x265 video decoding and x264 video encoding has been standard in ARM chips for years. Intel just got x264 decoding. They don't have 265 decoding and they don't have any hardware video encoding. I could go on but my point is that Intel has fallen way behind because they figured they didn't have competition.

More CNN fake news

[Anonymous Coward]

Fake news CNN's arseholetechnica 'expert authors' (not) = CS college dropouts (obese goiter victim Peter Bright) or no qualifications @ all fake it till you make it (Jeremy Reimer). What's their next CNN narrative gonna be? How Russians hacked our powergrid?? Washington post's globalist own (Jeff Bozo, king clown) failed that already.

TechReport Review is Favorable

[branchingfactor]

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."

Re: It's because AMD quit

[Kjella]

Hardware x264 and x265 video decoding and x264 video encoding has been standard in ARM chips for years. Intel just got x264 decoding. They don't have 265 decoding and they don't have any hardware video encoding. I could go on but my point is that Intel has fallen way behind because they figured they didn't have competition.

Not sure what you're smoking, H264 encode and decode support has been there since Sandy Bridge 6 years ago and Kaby Lake does H265 Main10 decoding in UHD resolution as well as 8 bit encoding. Maybe you've used a poor media player?

Intel's Real Problem: Gay Branding!

[Anonymous Coward]

Who in the right Homo sapiens sapiens mind wants a cpu names "Kaby Lake"?

No one!

Hay Intel, maybe name the next sucking air processor "Timmy SUX 9000" in honor of Captain Queer over at Apple Inc.

Ja ja

Re:ZEN ZEN ZEN with more pci-e at the same price o

[Anonymous Coward]

Zen is bullshit until shipping units are in stores and reviewers have actual retail product to bang on. AMD's been feeding us bullshit since bulldozer and there's no reason to buy the hype now.

AMD gives you PCIE lanes and features, sure, but it means fuck-all if the single thread performance of their top-end units barely competes with an i3 - That is the reality of pre-Zen AMD systems.

Single thread performance is still king in the PC space until somebody invents magical fucking fairydust compilers that make arbitrary shit code reasonably scale across many cores with little effort - Because that's the only thing that's going to make having lots of slower cores do any good.

So, if you don't mind, I won't eat the marketing hype and cherrypicked benchmarks.

Zen turns out to be great? Sure. Let AMD sponsor some esports events, do branding deals with streamers, and run tasteless ads with dumb product names. Who gives a fuck. What we need, what the market needs, is a sub 200 dollar part (150-170 really) that will be a no-bullshit replacement for a high-end i5 you'd see in a decent gaming computer. Well.. And we price competitive, feature parity chipsets/motherboards with drivers that aren't awful.

A course the answer is YES

[bongey]

Because according to /. editors DJT will cause the end of the world on January 21,2017.

Ars Technica?

[ruiner5000]

They review CPUs now? Come on.

Good!

[iamacat]

CPU performance is the least critical part in a modern computing scenario. Intel and others should be worrying about GPU, physics engine, AI engine, RAM speed, flash speed, wireless data speed and above all power efficiency. If you have a classic supercomputing problem, try a desktop sized box of intel computing sticks and some big fans. Linear scalability of RAM and storage will give you superior performance to a single CPU even with mediocre performance of each unit.

Re: It's because AMD quit

[Bryan Ischo]

I disagree. We're pretty far along the asymptotic curve at this point. And Intel can only increase processor performance at the rate that people are willing to pay. Maybe they could double processor speeds but the R&D and tooling costs would require that they charge 4x the price of current processors for it. The thing is, so few people are willing to pay significantly for more CPU power (because the number of users for whom CPU speed is a significant issue has declined precipitously as CPU speeds have increased), that Intel would never make back its investment. So it doesn't try.

The thing is that the longer that this is true, the more that gradual investment by AMD will allow them to catch up to the state of the art that Intel has. Which will then drive the prices down considerably. AMD will never make much money, but then when AMD finally regains performance parity with Intel, Intel won't be making much money any more either.

So you can expect top end consumer oriented CPU performance to remain stagnant but prices to drop considerably. I'd still call that a win.

Re:ZEN ZEN ZEN with more pci-e at the same price o

[Anonymous Coward]

FX-8300 currently sells for $120. If they kept the price, but improved performance and efficiency, I'd buy immediately instead of Intel's offerings.

Not quite there yet

[KozmoStevnNaut]

Personally, I'm waiting for the 7.77GHz i7-7777K.

The SKY is falling

[Anonymous Coward]

What a bad headline.

I still remember network PC was suppose to kill desktop PC

Newest Quads are 30% Faster than Sandy Bridge?

[BrendaEM]

I am so glad that I bought a good motherboard. I do Rhino 3D, Photoshop, Premier / music recording / and some after-hours gaming.

My 4-year old stock-clocked processor is only about 30% slower than a new quad. In fact, I just put my that motherboard/processor in a new rackmount case. I am still editing the rebuild video; but case is here: https://www.youtube.com/watch?...

I've built/assembled/configured about 104 computers since 1990. I have never reused a motherboard/processor before. I am hoping that AMD's 8-core pan out, and are affordable. Intel has had things too easy for too long. Intel my not be the whole desktop for long.

Not dead y-

[Mal-2]

Desktop CPUs aren't dead, they just smell funny.

That smell is from the unwashed masses using their phones for 90% of what they used to do on a laptop. (Laptop chips are designed like the desktop chips, just with less power consumption and a lower clock rate.)

When is the last time you could take a top of the line machine that's six years old, and a mid-range but brand new machine, and be hard-pressed to tell them apart? Well, that's where we are now. Intel is having a hard time winning on the "buy an $800 computer to save $25 in power every year!" concept. Laptop noise, size, and battery life continue to improve, but if you're not unhappy with yours, why hassle replacing and reinstalling and all that? A new one won't be that much faster – unless, of course, you bought cheap in the first place.

Not just the CPU

[Anonymous Coward]

Desktop computing is dead. Mobile computing, cloud computing and computing as a service are the future. Those ugly machines occupy a niche nobody is interested in. Laptops will linger for a while but desktops and towers are dead.

Re:Not just the CPU

[Shirley Marquez]

I don't think desktop computing is dead. It's just turning into a niche market. There will continue to be people who need the maximum amount of computing power available, along with multiple displays and input devices, and those people will continue to buy desktops.

Re:Not just the CPU

[Agripa]

I don't think desktop computing is dead. It's just turning into a niche market. There will continue to be people who need the maximum amount of computing power available, along with multiple displays and input devices, and those people will continue to buy desktops.

Especially when the fragile unrepairable laptops have chickelet keyboards and glossy screens.

Obligatory Monty Python reference

[Shirley Marquez]

It's not dead. It's just pining for the fjords.

Re: It's because AMD quit

[tigersha]

Agreed. The chips do exist, but Intel is gouging the market.

I have a 2 year old 14 Core Broadwell Xeon, giving 28 threads. I got an engineering sample for about 350Euro but usually they go for 2500+. It runs rings around anything else I have worked with. Intel could sell those for much less if they want to. Even the Mobo is not much more expensive than a high-end i7 board, it runs on a straight X99 board, albeit without ECC RAM. The chip can do ECC RAM on a server board. It does not even eat that much power. As a virtualisation workstation it totally kicks any i7's butt into orbit, passmark is about 18000. The chip can also run in dual-CPU config, which would give me a 56 thread 36000 Passmark machine, with a 500 Euro motherboard that can go up to 256 GB DDR4 ECC RAM. If Intel brought down the prices of their high-end Xeons instead of forcing chumps like me to buy their somewhat risky engineering samples they would sell zillions of the things.

And that is the previous generation, the newer v3 chips have even more cores per CPU.

Re:Let's hope AMD's RyZen will cause some progress

[Anubis IV]

Right now, Intel just has no financial incentive to innovate.

Sure they do. It's called ARM, and it's already bigger than Intel's offerings. That's why they finally had to bite the bullet and license the tech so that they could remain competitive. Starting this year, they'll be using their industry-leading processes to make what are quickly becoming the industry-leading chips...that they didn't design.

Intel is way ahead of everyone else when it comes to pushing the bounds with lithography, but when it comes to chip design, it seems as if they've pushed this one about as far as it wants to be pushed, whereas ARM is still going strong with year-over-year improvements. It may not be sustainable in the long-run, but it's already enough that Intel has every reason to try their hardest to innovate.

Re: No SATA and no RAM expandability

[Anonymous Coward]

What would a 4K webcam on a laptop be any good for? Besides making illegal sex tapes from non-consenting, unwitting strangers you have sex with.

Otherwise, I wanted to point out H264 encoder is increasingly common in x86 CPU's built-in GPU, soon to be H265 encoders if not already.

The idiom predates Huxley's book.

[Ungrounded Lightning]

... there's no relation to the book [Brave New World] 's subject matter so why allude to it?

"Brave New World" is an idiom (for historical periods that are more utopian than the periods preceding them) that predates Huxley's famous book (which put an ironic and dystopian twist on it).

The sentence uses the pre-Huxley meaning of the idiom and doesn't make a visible reference to the book (though such a reference, and the dystopian newspeak twist, is unavoidable). To be grammatical it requres the article, thus the "[sic]".

Re:The idiom predates Huxley's book.

[Hognoxious]

"Brave New World" is an idiom (for historical periods that are more utopian than the periods preceding them)

If it's just a phrase (and not a title) it wouldn't be capitalized.

"I found myself in a bleak house" isn't the same as "I found myself in a Bleak House".