Slash Boxes

SoylentNews is people

posted by takyon on Wednesday September 02 2015, @11:59PM   Printer-friendly [Skip to comment(s)]
from the month-long-news dept.

Intel's first two 14nm Skylake desktop chips, the Core i7-6700K and Core i5-6600K, were launched at Gamescom in Germany and reviewed back on Aug. 5. When the processors are set at 3 GHz, Anandtech benchmarking showed 2-3% instructions per clock (IPC) improvement over Broadwell, and over 5% IPC improvement from Haswell to Skylake. Skylake's overclocking potential seems better than both Haswell and Broadwell. One explanation is that the fully integrated voltage regulator (FIVR), which had caused heating issues in Haswell and Broadwell chips, has been removed. FIVR is expected to be reintroduced in 2017 in the generation of chips following "Kaby Lake". Skylake processors will support both DDR3L and DDR4 memory, but the performance benefits of DDR4-2133 over DDR3-1866 are minor and inconsistent.

Anandtech testing found that discrete gaming performance decreased slightly (1.3%) versus Haswell when both CPUs were limited to 3 GHz. The issue might be cleared up with firmware updates and further benchmarking, but the finding sets the tone for Skylake: Not much new for desktop users, but potentially important improvements for mobile users. Skylake on the desktop does look better when compared to older CPUs, with Skylake beating Sandy Bridge by 25-37%, and Skylake should consistently reach higher clock rates than Haswell and Broadwell.

i7-6700K/i5-6600K Review at Tom's Hardware

Following a leak of embargoed Skylake "full" lineup details, Intel pushed out its fact sheet early. Anandtech has its coverage and analysis of the full launch. You may also be interested in media improvements.

Skylake "Iris Pro Graphics 580" (GT4e) will see the introduction of a fourth "slice" of execution units (EUs), bringing the high-end from Broadwell's 48 EUs to a total of 72 EUs. However, none of the chips announced today have a fourth slice. Some of the U-series (Ultrabook) chips come with "Iris Graphics 550", with three slices and 48 EUs (GT3e). Even if a GT4e part was available, a 50% increase in EUs would not mean a 50% increase in graphics performance.

Skylake GT3 and GT4 chips will come with 64 and 128 MB of eDRAM respectively (hence GT3e and GT4e). Integrated graphics and "specific workloads" are improved by the presence of eDRAM.

Skylake chips operate across a wide range of power envelopes, from 4.5 Watts on Core M (Skylake-Y) to 91 Watts on Skylake-K. Battery life on mobile platforms will be improved, particularly when watching video due to better hardware decoding.

German IT media house Heise reports weird benchmark results [in German] for the new "Skylake" generation of Intel Core CPUs. The tested CPU exhibited more than a factor-two speedup over the previous Broadwell generation for single-thread tasks in a specific benchmark, but the performance did not scale with more cores; it even dropped below single core performance.

With somewhat stagnating clock rates in the past few years, the path to optimization has recently been to improve the Instructions-Per-Clock rate, through widening of the execution path, and, more importantly, through improved branch predition. But with major improvements in the Sandy Bridge and Broadwell generation branch predictors, Intel had reached a point of diminishing return in this area.

Do they have an unexpected ace up their sleeves, or what explains this mystery?

Original Submission #1Original Submission #2

This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2) by Snotnose on Thursday September 03 2015, @12:10AM

    by Snotnose (1623) on Thursday September 03 2015, @12:10AM (#231492)

    Back in the 80's I went to a trade show and was talking to a grizzled old vet who must have been in his 40s (I was in my 20s). We were wondering if CMOS or NMOS was going to win, CMOS was the dark horse as it was so vulnerable to static electricity. One of the few things we agreed on was the process wasn't going to shrink much more than another 2-3 generations.

    14 nm. That's what, 3 atoms or something? It's farking amazing how far it's come.

    The 3 symptoms of laziness: 1) think of something tomorrow 2)
  • (Score: 2, Informative) by Anonymous Coward on Thursday September 03 2015, @12:12AM

    by Anonymous Coward on Thursday September 03 2015, @12:12AM (#231493)

    Binary blob required for Skylake now.

    Have fun...

  • (Score: 2) by opinionated_science on Thursday September 03 2015, @12:56AM

    by opinionated_science (4031) on Thursday September 03 2015, @12:56AM (#231503)

    any word on the FLOPS rating for the chips? LINPACK would be nice, but theoretical peak DP/SP would be useful, especially if the graphics units are usable.

    • (Score: 2) by takyon on Thursday September 03 2015, @01:29AM

      by takyon (881) <> on Thursday September 03 2015, @01:29AM (#231514) Journal

      Here's word on Skylake graphics having about 1,152 GFLOPS. It might not reflect real world, so take it with a grain of salt: []

      Gen 9 Graphics: Gamers Rejoice
      While processing capabilities are increased, graphical prowess in the new architecture has increased more significantly. The new graphics architecture, called the Gen 9, is more performance-oriented, and has embedded L4 cache of upto 128MB. It will feature support for DirectX 12, OpenCL 2.0 and OpenGL 4.4. According to Intel, the top-of-the-line Intel HD530 embedded GPU is able cross 1,100 GFLOPS (Floating Point Operations Per Second) of graphic processing output. This, in essence, is a very high level of gaming output, as it falls about 200 GFLOPS short of XBOX One's graphic processing capability.

      Compare to 768 GFLOPS for Broadwell, 640 GFLOPS for Haswell, 256 GFLOPS for Ivy Bridge, 130 GFLOPS for Sandy Bridge, 43 GFLOPS for Iron Lake (never heard of it).

      What's more, Skylake HD 530 is a "GT2" part with 24 execution units, not GT3 with 48 execution units or GT4 with 72 EUs. I'm not sure why it's described as "top-of-the-line" above... maybe because the article was published before GT3 came out and GT4 is nowhere to be found. I can tell you that performance does not scale linearly with increasing number of slices... so GT3 will not reach 2,300 GFLOPS and GT4 will not reach 3,450 GFLOPS.

      [SIG] 10/28/2017: Soylent Upgrade v14 []
      • (Score: 3, Informative) by meisterister on Thursday September 03 2015, @02:25AM

        by meisterister (949) on Thursday September 03 2015, @02:25AM (#231529) Journal

        Ironlake was the GPU integrated into the actual 2nd generation i-series chips (ie. Westmere).

        Westmere is kind of the forgotten generation because it was sandwiched between Nehalem and Sandy Bridge.

        (May or may not have been) Posted from my K6-2, Athlon XP, or Pentium I/II/III.
  • (Score: 0) by Anonymous Coward on Thursday September 03 2015, @02:51AM

    by Anonymous Coward on Thursday September 03 2015, @02:51AM (#231534)

    If I'm in the market for a 15" laptop running on Core m5 or Core m7, is it worth waiting for Skylake? Would the difference in price likely to be above the price/performance regression line?

    • (Score: 2) by takyon on Thursday September 03 2015, @03:38AM

      by takyon (881) <> on Thursday September 03 2015, @03:38AM (#231545) Journal

      I would wait for Skylake over Broadwell since systems will be out in the coming weeks and months. Or take advantage of Broadwell inventory clearance by keeping an eye on a deal site [].

      Skylake's improvements on mobile should be real enough. Here is some of what they are advertising over Broadwell in terms of performance: [] []

      30% better battery life? 10-30% better graphics performance? I didn't put too much of this info in the summary because it is still really uncertain. There were a flurry of Skylake laptop product announcements yesterday so we should see benchmarks for mobile soon.

      Intel has supposedly gotten manufacturers on board to add more wireless capabilities - meaning wireless charging and maybe something like WiGig for peripherals. That might factor into your decision.

      If you're unsure about going with Intel and are willing to wait a year, AMD's Zen is due for 2016. They are promising 40% improved instructions per clock by fixing the bad threading scheme they had before. I haven't seen any Intel Skylake vs. AMD Carrizo comparisons (on integrated graphics), but AMD will not want to lose the integrated graphics race to Intel. Intel should be releasing "Kaby Lake" next year, and nobody knows what that will look like. It could end up like "Devil's Canyon" - essentially a rebrand with 100-200 MHz higher clock rates.

      [SIG] 10/28/2017: Soylent Upgrade v14 []
      • (Score: 0) by Anonymous Coward on Thursday September 03 2015, @03:50AM

        by Anonymous Coward on Thursday September 03 2015, @03:50AM (#231548)

        Thanks - 30 pct better battery life alone is prolly worth the wait. IF they can deliver that.

        • (Score: 2) by mtrycz on Thursday September 03 2015, @08:37AM

          by mtrycz (60) on Thursday September 03 2015, @08:37AM (#231630)

          I'd argue that *most* of the battery is drained by the led screen, and it's drawing continously. So 30% on a component that is already drawing a small amount and stays idle most of the time, probably shouldn't be a big deal.

          It's a great feat in and of itself, but not that important in a practical/consumer point of view. I personally always went with previous generation clearances (getting actually better performance for the price), but as GP suggested, wait for the benchmarks reviews and comparisons.

          In capitalist America, ads view YOU!
  • (Score: 0) by Anonymous Coward on Thursday September 03 2015, @03:19AM

    by Anonymous Coward on Thursday September 03 2015, @03:19AM (#231541)

    Dare I ask why a "fully integrated voltage regulator" is a good idea?

    Voltage converters use relatively bulky inductors and capacitors. While you can make those components smaller by raising the frequency; you lose some efficiency during switching.

    I suspect they want to make sure the chip gets a calibrated Power Supply output resistance as was required by the Pentium D processor (at about 4milliohms).

    • (Score: 3, Informative) by takyon on Thursday September 03 2015, @03:43AM

      by takyon (881) <> on Thursday September 03 2015, @03:43AM (#231546) Journal []

      Prior to Haswell, voltage regulation was performed by the motherboard and the right voltages were then put into the processor. This was deemed inefficient for power consumption, and for the Haswell/Broadwell processors Intel decided to create a fully integrated voltage regulator (FIVR) in order to reduce motherboard cost and reduce power consumption. This had an unintended side-effect – while it was more efficient (good for mobile platforms), it also acted as a source of heat generation inside the CPU with high frequencies. As a result, overclocking was limited by temperatures and the quality of the FIVR led to a large variation in results. For Skylake on the desktop, the voltage regulation is moved back into the hands of the motherboard manufacturers. This should allow for cooler processors depending on how the silicon works, but it will result in slightly more expensive motherboards.

      The TL;DR of this is while the FIVR was more efficient by all accounts, and apparently a lot smaller too, it moved an obvious source of heat onto the chip. A good idea in theory, but not in practice. Hence, Skylake will have better overclocking capability than Haswell and Broadwell. But Intel is not done with the FIVR concept: it will reportedly reintroduce it in 2017, with the 10 nanometer shrink following Kaby Lake.

      [SIG] 10/28/2017: Soylent Upgrade v14 []