Intel, churning their lesser chipsets nomenclatures repeatedly, renaming a given item with a more powerful chipset name then claiming "xx% in efficiency improvements" as the "new" Intel part number draws less current than the "old" part did .....
        some slightly evil, smelly brown poot progress there


Intel has done this for the last 2  years, all while stuck at 14nm lithography, it is a way of taking the credit for some generic Microsoft programming improvements by doing some slight of hand and then claiming their "new" processors actually did the improvements.

Have some dual core processors benefitted from going from 2 to 4-6 cores?   Yes, some have indeed gotten 10-15% better results on some the rewritten multi-threaded tests, but there has been no real single core Intel improvements in the last 2 years.

Will ARM chipsets eventually lap past Intel's Core i chipsets?    Yes, on two fronts they have done so already --- ARM chipsets have already surpassed Intel's best Atoms and lesser Core Ms.  And then again up in the very top end of the server world the Centriq 2400 has surpassed the upper level Xenon server chipsets using a 48 cluster (2 cores per cluster) ARM design that costs far less than Intel's Xenon servers do while outperforming them clearly for both throughput and energy efficiency.

News Flash 12/19/17   Intel announces new 45 watt Six Core laptop chipsets  More of the same, or perhaps a bigger desktop catalog item simply relabeled after being sorted for defects?

According to the pictures, what we’re looking at is a 45 watt, 6-core chip with a base clock speed of 2.4 GHz.

Meanwhile, CPU Monkey has a listing for the Core i7-8700HQ chip, which seems to have similar specs. It looks like a significant upgrade over the Core i7-7700HQ, with more CPU cores, more cache, and a slightly higher graphics clock speed, among other things. The new chips are also said to support DDR4-2666 memory, up from the DDR4-2400 RAM supported by their equivalent 7th-gen chips.   yep, desktop chips post sorted

Having spanned and beaten all of Intel already (Intel being all stuck at 14nm lithography still) it isn't hard to think that some specifically designed brand new ARM Holdings canned "pre-approved" certified DynamIQ cluster based designs at 7-8nm may indeed lap any specified level of Intel Core i product once ARM (or a "built on ARM technology" licensee) decides to go design a cluster set to do just that.  

This task gets easier each time ARM lithography shrinks yet another full level below 14nm and Intel stays stuck exactly where they have been for the last 4+ years.    

Apple and Samsung just announced that 7nm is indeed going to commence this year with the first wave of SoCs coming from both TSMC and from Samsung.   Global Foundries will stop at 8nm for yet another half year's worth of mini-progress before sliding down to 7nm.





In a sense, Microsoft has just made the eventual death of Intel Core i products more possible by working with ARM's vendors to make the Qualcomm Centriq 2400 and the Snapdragon 835 and the Snapdragon 845 all quite real and very appealing while running Windows 10 on them.

Windows 10 has gotten slightly lighter on its feet as it has moved towards ARM, but it is still relatively bulky and slow compared to Chrome OS and Apple OS products.   But the ARM chipsets have gotten a LOT better lately, so they can handle the task now.

HOWEVER, less of an Intel processor is needed to run Win 10 because Win 10 just got slightly better.    So, Intel will rename their Intel product lines, shift the old chipsets "upstream" a notch namewise, then claim some brown vapor efficiency improvements because the new Core i5 takes less power to "do the same Win 10 job"  as the old Core i5 did.



And the highly competitive Hockey Stick boys all say  Hmmmmmm ........ and now our ARM SoCs can run this stuff?   I already own the license rights to make my up my own "built on ARM technology" versions of stock ARM designs ..... I can go do that right now.

Xiaomi and Huawei have just realized there is a huge rare juicy Intel steak sitting out on the free buffet table and have just made the first moves to go cut their first big piece of meat off of that juicy steak .....

ut the fact you weren't told that they were actually building stuff using the lower grade C in the product while touting grade A chipsets in the initial press advertising releases really really sucks)

Almost looks deceptive..

It is year's end 2017 and the start of 2018 -- time for some PREDICTIONS to be made and tested against reality when it becomes known.

Intel is being affected by something they cannot abide ---- real competitors.    Let's count up the things that have happened and put some predicted 2018 market share declines against Intel PC/Laptop.

ARM 48 core Centriq chipsets in Server Space ..... Intel Xenons all have to expand core counts and die size drastically, net result is Intel's unit cost goes through the roof again in 2018 while trying to compete against the real alternatives.   Intel winds up losing market share due to price/performance .....

20% of new installation sales in SERVER SPACE moves away from Intel


AMD Ryzen simply continues to advance at 10nm, then rolls down to a new lower 8nm lithography node this year and 7nm next year and gets even faster and cheaper to manufacture so Intel loses about the same percentage of new product sales dollars as they did last year for the next 2 years ......

20-25% of new unit sales in PC desktop  (includes both new AMD and Apple/Mac sales)


AMD also uncorks their brand new 8nm laptop chipsets and because of much lower cost and better built in graphics Intel loses 25% of the existing laptop sales market share (again part of this change coming from Apple's using more AMD) causing Intel to lose ......

~ net 25% ~ of new unit sales in mid to high end Laptop


Qualcomm Snapdragon 845 low end laptops running Win 10 ARM stresses Intel in the lower end of things, snapping up lower price sales from bottom and mid-range.    But because of relatively high prices charged for the new Qualcomm based units, the effect is only limited to 10-15% of low end market share and does not affect Chromebooks so much as Wintel units.


Intel strikes back by paying Google to "Chrome support and fully qualify" several still physically larger, stronger,  "better but not cheaper" Chromebook level Intel chipsets, which causes Chromebooks as a class to actually still better and to sell as well as ever in the upcoming year.   Problem is that this pulls still more units out of  Wintel's oldstyle laptop market share as a whole, rounding the laptop market loss numbers up to a roughly 35% decline in old style "traditional" Wintel based laptops.



Intel still fights fiercely over the Chromebook market zone, but is unable to stop several newly designed ARM DyamIQ  competitor chipsets from Rockchip and Mediatek from entering into the Chromebook marketplace and starting to take some share from the lower end of things.


Intel China however continues to gain in Far Eastern share by transferring all remaining 22nm lines over to China, and the Chinese Intel subsidiary then makes hay by flooding the burgeoning Far Eastern markets with older "more slightly obsolete" Intel 22nm designs at new "cheaper than dirt" prices.  

Under Chinese management and using cheap Chinese raw materials and labor, Intel finally becomes more cost competitive.

Resulting new growth from the Far East is so large that Intel actually grows in total while losing ground in America and Europe and South Korea and Japan.   Intel China becomes dominant in Intel International and begins calling the shots in the corporation.

PC/Laptop in USA becomes a relatively minor role for Intel, with over half of Intel's new profits coming from supplying raw components such as radios, location devices and memory to Automotive vendors and to the various other players.  

Automotive and IoT still grows year on year at rates greater than Intel's net losses in PC and Laptop.

Chipzilla continues into the next year based mostly out of China .......  rolling more and more away from the PC/Laptop arena.

https://wccftech.com/intel-developing-new-x86-uarch-succeed-core-generation/

The Intel ‘Core’ legacy will come to an end with Tiger Lake in 2019 – To be succeeded by a lean and mean approach to x86

The reason for doing so is simple, by no longer guaranteeing backward compatibility, Intel could save precious die space by removing the hardware for legacy SIMDs and other legacy features. The result would be a much leaner and more efficient x86 architecture that can deliver a bigger bang for lower resources on-die. According to the source, the architecture will be used to replace the current Core series on the Desktop and Enterprise market. It is possible that Intel shifts the 100% backward compatible x86 towards the server side where legacy operations might actually be required on a continuous basis.

If this news turns out to be true then Tiger Lake would be the last iteration of the current x86 architecture from Intel Corporation. This is expected to arrive sometime in 2019 so we can expect the brand new x86 implementation to land by 2020 approximately. One of the reasons we believe that this rumor is actually true is because Intel has already taken steps in this direction with the advent of Skylake, which can be thought of as the first significant change since the Sandy Bridge era. The focus on mobile and energy efficiency is clearly obvious and the removal of gimmicks such as the Fully Integrated Voltage Regulator (FIVR) (remember that?) shows a much more mature outlook on on-package economics.

Intel is planning on changing up x86 for something else, something that AMD does not license at this point in time ???

Then Intel is planning on legally killing off a chunk of their competition, again, in other words.   Trying to create an advantage jest for themselves, naturally.

Intel also sees Microsoft actively jumping ship for ARM/Qualcomm right now, so they are also perhaps attempting to force MS back into the exclusive Wintel marriage fold again (more vague brown vapor promises, as if MS doesn't know Intel's modus operandi by now).

Intel is also VERY TIRED of losing market share to ARM RISC influenced processors because CISC x86 is so very very bulky and slow compared to the RISC influenced design alternatives.   AMD is teetering on going completely over to the ARM side and you can see RISC influences in the current wave of Ryzen chipsets.

Intel also sees Google playing around with Fuschia OS, a brand new superfast code base that would run best on stripped down RISC chipsets.

If Intel were actually designing a generation of Intel processors to try to run Fuschia at full speed then this would all hang together pretty well (just leaving MS out in the cold this time around).

SO, Intel is again acting self-delusional, if there is any REAL ADVANTAGE to what they are planning on doing, FOSS and Linux will have programmed up a variant for it that will allow both old and new hardware to be repurposed.



Or they can go with ARM DynamIQ chipsets and FOSS Fuschia, and jest let new Intel go off and play all by its lonesome.


===================================================


Intel's plans are proprietary brown murky vapor that won't jell for two more years at least.

On the other hand, Google's Fuchsia "FOSS only" future pathway is getting clearer all the time now ......

http://www.androidpolice.com/2017/12/30/pixelbook-used-test-googles-fuchsia-os/



For over a year, Google has been developing an operating system named 'Fuchsia,' designed to run across a wide array of devices. The company hasn't said anything publicly about it, but it is entirely open-source, so development on the project has been transparent. Simply put, we can see what Google is working on, but we don't know what it will actually be used for.

According to Fuchsia's documentation, the Acer Switch Alpha 12 and Intel NUC are officially supported 'target' devices. This means that Fuchsia has been verified to work on those devices, and they are most likely the top devices to be used for testing. A page about installing Fuchsia on the Google Pixelbook was recently added, which explains the process of placing the laptop in developer mode and booting from a USB drive which would be common during early development.

Does this prove the Fuchsia will someday replace Chrome OS? Definitely not. However, it does mean that Google wants to make sure the OS works on high-end laptops, like the Pixelbook.

Will Fuchsia end up as a stable operating system that runs across a wide array of computers/tablets/phones?   Only time will tell ......    
right now it is beginning to look like it


Me, I think Google has been "arranged to be disrupted" by its enemies once before ostensibly by a newly landed sue-you Sam who had bought them a chunk of some old somewhat partially patented software tech that could be interpreted to impinge on Android in some vague ancient "pre-Dalvik" fashions while good 'ol Sue You was rummaging around in the boxes at some dead parent company's estate sale.  

Google doesn't care to repeat that "Hey, let's go sue Google" experience with the next wave of old company garage sale sell offs.   (and there are quite a few of these garage sales that are coming up in the next 5 years)

Try it again and Google will simply say, "Sorry, here's all that IP you are now claiming to be able to restrict."  Then they will let the world go deal with "sue-you Sam" according to their own local laws while they continue to roll out carefully documented totally FOSS from the get go Fuschia OS.


===================================================


We have manufacturers now spontaneously "announcing" that their hardware is now part of the Fuschia "target" development trials and will be supported from the very beginning by actual Fuschia development and testing.

Hey, for the landed cost of a test unit that is a VERY cheap price to get on to the newest bus in town before it even rolls off the assembly line.

So far we got about a third of the mgf. crew participating voluntarily, willingly and eagerly.   These are mostly the odder, lower cost devices with ARM chipsets and of course a few of the main players with the current generic Intel chromebook chipsets.

The trick is that the lower power ARM processors will run Fuchsia like a scalded ape ..... you don't need a mega processor for Fuschia nor do you need the killing big memory either.    More (faster) memory simply means more tabs able to be open all the time, somewhat faster tabbing too.

Folks seem to want a real alternative to Intel and Microsoft, as the Wintel pair has fouled up their nest fairly badly in the past 2 years ......

HOWEVER, this is all future speak as Wintel still rules in computing right now by a very large margin.   And Windows users are STILL acting "fairly numb" to the repeated screwing overs they keep on getting from Wintel.



Largest effect:   Google will become like Apple, able to tune OS and Hardware to optimize good competitive stuff -- which being Google instead of Apple the good stuff that was built becomes available to everybody effective the very next year.

ALSO NOTE:   two three new hardware makers are now saying they have supplied "target units" for Fuschia development use since yesterday.

https://liliputing.com/2018/01/zhaoxin-wants-to-take-on-intel-and-amd-in-desktop
-space-with-kx-5000-cpu.html



Zhaoxin wants to take on Intel and AMD in desktop space with KX-5000 CPU

Once upon a time Intel and AMD weren’t the only companies producing x86 chips for desktops. VIA Technologies was a minor player in this space up until a few years ago, but the company has largely been focused on embedded chips for the past few years.

Now a company called Zhaoxin, which is co-owned by Shanghai and VIA is announcing plans for a new line of processors that could eventually compete with AMD and Intel’s latest offerings.


It’s not clear if VIA is actively participating in development of the new chips, but VIA has a license to produce x86 processors, which means that Zhaoxin does too.

Zhaoxin has produced a handful of x86 processors over the past few years, but according to reports from eefocus and semi.org.cn, the new KX-5000 is the first to fully support modern technologies including DDR4 memory, PCIe 3.0, and USB 3.1 Type-C.

The 28nm chips will be available in quad-core and octa-core versions clock speeds up to 2.4 GHz. They also feature integrated graphics and support 4K video playback.

Zhaoxin plans to move to a 16nm processor and support speeds up to 3 GHz for its upcoming KX-6000 chips, and add support for DDR5 memory and PCIe 4.0 with the eventual KX-7000.

At this point Zhaoxin’s chips aren’t likely to overtake Intel’s (or even AMD’s) in the global market. But China has been focused on developing home-grown alternatives to chips designed by Western companies for a number of years, so I wouldn’t be surprised to see Chinese PC makers like Lenovo opt for these chips for computers that will be sold in their home market.

In fact, Lenovo is already said to be planning to launch an M6200 computer powered by the KX-5000 sometime this year.

The Chinese government needs updated computers for military and gov uses.   Intel has become completely abhorrent to the Chinese at this point in time due to the Management Engine debacle.   So the Chinese are rolling their own PCs now ......


===================================================

https://techreport.com/news/33018/via-joint-venture-reveals-kx-5000-x86-socs-for
-chinese-pcs



Chinese site EEFocus reports that Zhaoxin Semiconductor, Via's joint venture with the Chinese government, has announced the KX-5000 line of x86-64-compatible chips, which it claims are the first Chinese processors with full integration of platform controller chips and dual-channel DDR4 memory controllers.

The x86-64 KX-5000 CPUs will come in four- and eight-core versions manufactured by TSMC on a 28-nm process. The chips use an SoC design with integrated graphics, video decoding, SATA and USB controllers, PCIe 3.0 interfaces, and a dual-channel DDR4 memory controller. Zhaoxin claims the processors will be able to run at speeds above 2 GHz. While EEFocus didn't share any deeper details of the KX-5000's "Wudaokou" microarchitecture or thermal design power specs for these SoCs, Via's past x86 CPU efforts have been designed with low cost and power efficiency in mind, so we'd expect these parts to compete with Intel's lower-power Atom cores.

The KX-5000 is one part of a larger roadmap at Zhaoxin. The venture reportedly plans a 16-nm follow-up called KX-6000. The company already sells four- and eight-core ZX-C CPUs that can run Microsoft Windows or China-developed operating systems. These chips don't use the SoC design that's the hallmark of the KX-5000, according to Golem.de, so the transition from separate CPU and platform controllers to a fully-integrated design might mark a major advance for Zhaoxin's chips. Zhaoxin's existing processors are sold in the Chinese market only, and we suspect that the same policy will apply to the KX-5000 for the immediate future.

Please note this new Beijing Communist Government owned Zhaoxin Semiconductor effort is making an integrated SoC right off the bat, something Intel hasn't really yet pulled off successfully.

If I were Intel I would expect to see another 10-25% of my international market share taking a hike over the next year, year and a half if things go really really well for Intel and their fixes for their Management Engine go smoothly and well.    Otherwise, it will hurt even worse ......

After all, this is a CHINESE DOMESTIC PC and Laptop chipset competitor, someone who will automatically get a good bit of preference by the Chinese Communist government, naturally, since the Beijing government owns over half of Zhaoxin Semiconductor company and also has interests in TSMC foundry who will be building it first off as well.    

By fiat, the Beijing government could simply say Intel is functionally out of business in China if their current Intel irritation level remains as strong as it is now.

And this all assumes Intel can actually stop stepping on its own limp willy all the time like it has been doing repeatedly lately .....        

That's some analysis right there..
I'll enjoy your following it as it unfolds.

   
Lookie, MS's lips are moving again .......


Intel has confirmed reports that a security vulnerability affecting its processors.



In a statement, Intel says the company doesn’t believe the security exploit could be used to modify, delete, or corrupt data… but stealing your passwords and other sensitive information is probably bad enough, which explains why a massive effort is underway to address the issue through operating system updates.

Intel says it’s working with other technology companies to resolve the issue… and notes that it doesn’t just affect Intel processors.

The chip maker says it had planned to disclose the issue next week, after more of the necessary software and firmware updates were available. But Intel released its brief statement today “because of the current inaccurate media reports.”

So what’s inaccurate?

First, the Intel suggests that there’s no “bug” or “flaw” in its chips, but rather than an exploit has been discovered that can collect sensitive data from “computing devices that are operating as designed.”

That one sounds open to interpretation to me. The chips and software work… but they’re vulnerable to a previously undisclosed method of attack. Is that a bug? Apparently Intel doesn’t think so.

Intel also says “contrary to some reports, any performance impacts are workload-dependent, and, for the average computer user, should to be significant and will be mitigated over time.”

That’s actually in keeping with what we noted earlier today: some activities such as gaming seem to be largely unaffected. But as discovered by Phoronix, some tasks, like PostgreSQL and Redis seem to take serious performance hits on a Linux system with the updated version of the Linux Kernel designed to address the security vulnerability.

So you may or may not notice a big performance hit depending on what it is you’re using your computer for.

Finally, Intel also says that it’s not the only company whose products are affected and that the company is “working closely with many other technology companies, including AMD, ARM Holdings, and several operating system vendors” on the issue.

Despite the implication that AMD chips are affected, an AMD software engineer has explicitly that the company’s chips are not subject to the same types of attacks.

It’s probably not surprising that Intel’s stock price ended the day about 3.4 percent lower than yesterday, while AMD's stock price was up more than 5 percent at the end of the day.

Even today, Intel is trying to deny the issue even exists, much less that they actually own the problems it created ......


The top dogs are slow peddling that information while they sell their stocks....