Slashdot Mirror
Intel's Skylake Architecture Reviewed
Vigile writes: The Intel Skylake architecture has been on our radar for quite a long time as Intel's next big step in CPU design. We know at least a handful of details: DDR4 memory support, 14nm process technology, modest IPC gains and impressive GPU improvements. But the details have remained a mystery on how the "tock" of Skylake on the 14nm process technology will differ from Broadwell and Haswell. That changes today with the official release of the "K" SKUs of Skylake — the unlocked, enthusiast class parts for DIY PC builders. PC Perspective has a full review of the Core i7-6700K with benchmarks as well as discrete GPU and gaming testing that shows Skylake is an impressive part. IPC gains on Skylake over Haswell are modest but noticeable, and IGP performance is as much as 50% higher than Devil's Canyon. Based on that discrete GPU testing, all those users still on Nehalem and Sandy Bridge might finally have a reason to upgrade to Skylake.
Other reviews available at Anandtech, Hot Hardware, [H]ard|OCP, and TechSpot.
The performance increase is going to be negligible until the "new instructions" on the skylake are utilized more in daily software use. Buy today, pay a premium for basically no bump.
Imagine a beowulf cluster of these!
This caught everyone off guard. It is interesting that they released the unlocked K series targeted at enthusiasts first. As a Haswell refresh owner I see no reason to upgrade. I applaud Intel for finally ditching the included fan and heat sink on the K model; any overclocker will immediately install an after market cooling solution.
That's why the Raspberry Pi 2 is a much better solution. Since its SoC is made in China, you know there's no NSA backdoor in it!
Nah, the spying stuff is on the Z170 chipset, hidden away in the Intel Management Engine. Too bad Intel killed off third party chipsets awhile back.
I read the internet for the articles.
I just wish that Intel would make a version that's 8 cores instead with lots of cache rather than waste the space and power on the integrated graphics. If you are gaming with it then you would have a dedicated card since all IGPs pretty much suck, and if you aren't gaming on it there isn't much point to improving it since even the most basic IGP can run video and 2d applications just fine.
Yeah, I thought it was really suspicious when Microsoft heavily promoted the new version of their operating system. Then when hardware manufacturers kept on including wifi and bluetooth in their hardware, without the need for an external card, I knew the only possible explanation was a massive snooping campaign by the NSA.
Slashdot: providing anti-social weirdos a soapbox, since 1997.
Still a deal breaker for me.
My next CPU is sooooo going to be an AMD.
What kind of spying stuff?
The old one really limited sata / pci-e storage / network / usb on the the older chipset.
30% Percent Faster, perhaps, with the wind behind it, and if I don't overclock my rig.
Cinebench 931 VS 694 Multicore.
No, to be fair--it's only 25.4564983888292% Faster.
https://www.youtube.com/c/BrendaEM
Yes, we all prefer a PLA front door.
If you're on Sandy Bridge or newer, don't bother unless you really need the new chipset features.
Benchmarks of course show a small gain, but in the real world I suspect you could do a blind test of Sandy Bridge next to Skylake and you couldn't tell the difference.
Anyone who needs the performance difference shouldn't be on either chip, if you do serious image/video editing, you should be on Xeon anyway with 8+ cores if you make a living doing such work. The cost of such a system is trivial compared to the cost of the employee doing such things.
I have several systems in my office, ranging from a single Q6600 machine and two Core i7-920 machines all the way up to a Haswell Refresh i7-4790k. The difference in general Windows performance between all those machines is minor. Games play, more or less, the same in anything Sandy Bridge or newer, and we don't do anything so intensive to require more power.
Come on AMD, get back in the game so Intel has some real competition. Since Core2Duo came out, you haven't been coming to the party.
It's 2015. I want ECC support on all chipsets. Don't make me buy a Xeon just to get ECC.
https://01.org/linuxgraphics/i...
"No reverse engineering, decompilation, or disassembly of this software is permitted."
Reading AnandTech's review, they make a bold statement at the end:
"Sandy Bridge, Your Time Is Up."
That is an interesting thought, but is it really?
If you need USB 3, if you want some of the other newer chipset features, perhaps. But for performance?
In benchmarks, Skylake appears to be about 25% faster than Sandy Bridge. Sure, if you're doing video encoding all day or other CPU intensive applications, it is... (and if you ARE doing that stuff, why aren't you on Xeon?)
But for most desktop computer uses, you likely won't see any difference between the two. What is worse, is that most of the above gains came from Haswell, not Skylake.
http://www.anandtech.com/show/...
Look at the "Gains over Sandy Bridge" chart on that page. Look at the red lines, then the purple lines. The red lines are the Haswell gain over Sandy Bridge, then the purple lines are the Skylake gains over Sandy Bridge.
Nah, the spying stuff is on the Z170 chipset, hidden away in the Intel Management Engine. Too bad Intel killed off third party chipsets awhile back.
It's hard to pull off a hardware hack that would be visible to many hundreds of engineers.
Go check the hypervisors. It's much easier to inject a back door there.
Surprisingly, rarely discussed and completely undocumented, but E5 16xx 1P Xeons are very unlocked. And nope, not talking about ES versions either.
For example, my 4.5 Ghz (triple radiator) 8 core 1680 v2 that originally lived in a mac pro. There is a 14-core v3 SKU I would love to get my hands on for a new build...which might end up coming from a mac yet again.
Sure, because there's nothing here that could be explained by market trends.
1. Microsoft's monopoly is cracking badly, though perhaps not in the way most of us imagined. If you look at StatCounter's platform stats it's now 55% desktop, 39% mobile, 6% tablets and the desktop has been losing 10%/year the last few years. And people expect apps for their platform, if you're only on Windows or even Mac/Linux too you're now a dinosaur unless it absolutely requires a traditional desktop.
2. The OS is going to become a commodity, they saw what happened with Android once it hit critical mass. Chromebooks are an early warning. Also that XP and Win7 work "too well" so users aren't interested in upgrading, even though it's an expense maybe twice a decade. That MS Office - their stranglehold on the business market - is now on mobile and tablets is clear proof Microsoft knows this.
3. So their old strongholds are breaking down, where do they want to go next? They want to be the middleman between the app developers and the consumers, like Apple's App Store pioneered and Google Play mimic. To do that you need Win10 everywhere. You must get the snowball rolling that to make money you must be on the MS Shop, the same way you could install apps from other sources on Android but the vast majority don't. If you're not on Google Play, you "don't exist".
As for Intel:
1. Mobile, tablets, convertibles, laptops all need wireless connectivity and it's basically just expected features today like network and sound is on desktops, they used to be add-in cards once but was integrated long ago. And fewer and fewer want the hassle of running cables as WiFi speeds go to hundreds of megabits. It's also a simple way for Intel to steal market share by vertical integration, squeezing out third party chips.
2. And here's the kicker people don't seem to understand, Intel doesn't really make desktop chips anymore. Their mainstream chips are laptop spin-offs which get a higher TDP and a few other modifications, the same way their high end chips are Xeon spin-offs. That is also why they sell grossly overpriced desktop chips with better IGP, even though you can do much cheaper with a dGPU. They're just laptop spin-offs that happen to sell well enough to make a desktop version of.
3. So what's the combined effect? Well, you get the laptop features for "free", whether you want them or not. Same way Intel puts an IGP in every chip killing off much of the second hand GPU market, before you had machines that needed any old graphics card and now you don't. Less resale/reuse value means gaming cards in net cost more. It's an indirect way of using their dominance in the CPU business to expand without running into antitrust problems, at least so far.
Or maybe I'm just a NSA disinformation agent out to discredit the revealing of our secret master plan. But you have to admit the cover story is pretty credible, yes?
Live today, because you never know what tomorrow brings
If there's one thing I learned when I came back from AMD-land and started building Intel systems again, it's that you end up buying Xeons anyway. (Not only do they support ECC RAM (whereas the Core i7 doesn't) but for some weird reason they cost less. Those Haswell E3 12xx-V3s are fuckin' sweet.) What's actually sad (relatively, it's not that big of a deal) about your link is the have-to-wait-for-Cannonlake part of it.
What I wanna know is: why is i7 called the "mainstream" one, when the Xeons are better in almost every way (including price) (though not including overclocking, which mainstream people don't do)? If my grandmother or niece were buying an i7 (or even an i5!), I would tell her to get a Xeon instead.
(Or maybe the question is: how come Xeons cost less? If they cost more then everything would make sense. But they're cheaper. WTF.)
2. And here's the kicker people don't seem to understand, Intel doesn't really make desktop chips anymore. Their mainstream chips are laptop spin-offs which get a higher TDP and a few other modifications, the same way their high end chips are Xeon spin-offs. That is also why they sell grossly overpriced desktop chips with better IGP, even though you can do much cheaper with a dGPU. They're just laptop spin-offs that happen to sell well enough to make a desktop version of.
This actually isn't true. The chips that go into all the thin laptops people like these days are use a different mask set than the chips that go in desktops (They are physically different implementations, even though the microarchitecture is the same, but by that logic a Xeon E7 is the same chip as well. They put a GPU in because it easily benefits from the area. Even if you could make the CPU twice as big, the added performance would be rather minimal and would make their big server parts twice as expensive and effectively lower the price they can charge for their server parts. You also ignore the fact that most people don't buy desktops for gaming and won't use a high-end discrete GPU, Intel's not going to give up that market to Nvidia/AMD when they don't need to.
*whoosh*
This actually isn't true
Yes, it actually is true. DT and M segments use the same dies, U and Y are different.
but by that logic a Xeon E7 is the same chip as well
Nope, because that's actually a different die. ...
It's like claiming i7-5xxx and Xeon E3v4 are the same chip, or S2011-3 i7 and LCC E5v3, or i7-4xxx and Xeon E3v3, or S2011 i7 and E5v2, or i7-3xxx and Xeon E3v2, or
640K ought to be enough for anybody.
I'm still gaming on a Lynnfield. So, yeah, just about due for an upgrade.