Donate!
Welcome, Guest. Please Login or Register :: View Members
Pages: 1 ... 9 10 11 12 13 ... 31
Send Topic Print
2020 -- new Intel failures & successes (Read 12299 times)
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #150 - 11/20/18 at 05:07:38
 

https://betanews.com/2018/11/20/microsoft-pulls-buggy-office-updates/

https://betanews.com/2018/11/13/microsoft-windows-10-october-2018-update-back/


Et tu, Office? After pulling Windows 10 update, Microsoft does the same for Office



Microsoft's update procedure for Windows 10 has been a little, er, wobbly of late. The Windows 10 October 2018 Update proved so problematic that it had to pulled, and even the re-released version is far from perfect.

Now it seems the cancer is spreading to Office. Having released a series of updates for Office 2010, 2013 and 2016 as part of this month's Patch Tuesday, Microsoft has now pulled two of them and advised sysadmins to uninstall the updates if they have already been installed.

See also:
* Microsoft pulls the problematic Windows 10 October 2018 Update
* The re-released Windows 10 1809 is still buggy -- and some fixes won't arrive until next year
* Re-released Windows 10 October 2018 Update breaks Apple iCloud

In both instances -- KB4461522 and KB2863821 -- Microsoft says that the problematic updates can lead to application crashes. While this is not as serious a problem as, say, data loss, it does little to quieten the fears that have been voiced about the quality control Microsoft has over its updates.

For the KB4461522 update, Microsoft says:

Notice:

After you install this update, you may experience crashes in Microsoft Access or other applications. To resolve this issue, uninstall the update by following the instructions in the "More information" section.

This update is no longer available.

It's a similar story for KB2863821:

Notice:

After you install this update, you may experience crashes in Microsoft Access or other applications. To resolve this issue, uninstall the update by following the instructions in the "More information" section.

This update is no longer available.

If you have installed either of the updates, the advice is to remove them as soon as you can. You'll then have to sit back and wait for Microsoft to release updated, non-buggy versions -- and it's impossible to say when they will appear.



Pay attention, please.

Microsoft has a problem.   LOOK AT THE PICTURE.  As a company, older employees have been being forced out over the last few years, moving the age of the "in charge" personnel lower and lower and lower.  

Middle managers as a group have been removed and the recently promoted Team Leaders assigned to their Management Control and "internal to the team Quality Control duties".

Microsoft is a fine example of the new nation-wide "Due To Downsizing & Cost Cutting Millennials were put In Charge" syndrome.

These 20-30 somethings think differently than Joe Normal, and in this case they see nothing wrong with "moving on" somebody else's software problems to the next step in the process.  

A millennial has a somewhat narrow view of "his job" and they do their job, not everybody else's.

Microsoft is checking their software releases now, and is putting the brakes on errant software releases and pulling back recently released OS and Office releases right and left at the moment.



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



https://www.zdnet.com/article/linus-torvalds-after-big-linux-performance-hit-...



Linux kernel founder Linus Torvalds: "When performance goes down by 50 percent on some loads, people need to start asking themselves whether it was worth it."


Major slowdowns caused by the new Linux 4.20 kernel have been traced to a mitigation for Spectre variant 2 that Linux founder Linus Torvalds now wants restricted.

As noted by Linux news site Phoronix, the sudden slowdowns have been caused by a newly implemented mitigation called Single Thread Indirect Branch Predictors (STIBP), which is on by default in the Linux 4.20 kernel for Intel systems with up-to-date microcode.

STIBP is one of three possible mitigations Intel added to its firmware updates in response to the Spectre v2 attacks. Others included Indirect Branch Restricted Speculation (IBRS), and Indirect Branch Predictor Barrier (IBPB), which could be enabled by operating-system makers.

STIBP specifically addresses attacks against Intel CPUs that have enabled Hyper Threading, its version of Simultaneous Multithreading (SMT)


Speaking of the Wintel boys screwing up all over again (accidentally ???) so much so that their very latest patches that were contributed to the Linux Kernel intended to combat Meltdown and Spectre are now being pulled by Linus himself, acting directly, effective today.

Both Wintel companies are now "a Foundation Level player" in FOSS Linux and are supposed to be able to be trusted to submit a patch to be included in the next kernel release.  

Linus and company had correctly patch fixed Meltdown and Spectre 6 MONTHS ago (back when Intel was saying it wasn't their processors and MS was saying they had no OS problems to fix) and for the Wintel boys themselves to submit new patches so very very late now, well the patches themselves should have been more thoroughly inspected before being released, AND THE INTEL/MICROSOFT PATCHES SHOULD HAVE BEEN THOROUGHLY VETTED BY THE LINUX SYSTEM MAINTAINERS BEFORE GOING INTO THE LINUX KERNEL with hindsight being all 20-20 as it often is (and as it could be said it is as if some of the black bag boys in the marketing department at Intel may have possibly actually written some of these particular rotten patches, apparently).

These latest Wintel patches actually cut the overall Linux performance by 30 to 50 percent depending on the actual workload, for no apparent gain in protection over the original patch set put out by Linus and his boys months and months ago that successfully mitigated the issues at a "less than 5 to 10% processing delay" processing effectivity cost.

Linus is saying, politely but bluntly, that simply turning off speculative execution completely and stopping multi-threading completely and removing that entire code block from Linux would have less of a slowdown effect than the recent Wintel patches have done when they are implemented "as submitted".  

In any case, all Wintel kernel submissions will now be looked at more carefully in the future because "automatically trusting the guys that built the stuff" really does seem to not be working all that well in the harsh light of reality any more.    One of the Wintel boys seems to have become pretty much incompetent at driver programming and the other one of the Wintel boys is struggling in the marketplace so badly their motivations always seem to be somewhat suspect now-a-days.

A repeated theme is still showing up here, that the latest Intel processors really don't get much better, but the tests and the various background systems are getting tweeked by Intel to make it seem like the newest Intel processor generation is getting a lot better (but really isn't).

Example:  you write a 30 to 50% slow down into the patched Linux drivers, then you can magically fix that problem (the one that you actually created) in the next processor generation's software release and then your Marketing Dept. can claim a 45% improvement in processing power ---- does this sound at all familiar to you?  

Throw on a couple of extra super fast cores that overheat the chipset (overpower existing laptop cooling systems) and inside 10 minutes thermal throttle the whole mess down to way slower than it was before ---- does this sound at all familiar to you?

Jack all your prices up 50% in early October then for a Black Friday sale you can sell a few of them at the original MSRP as "a really good sale price" ---- does this sound at all familiar to you?

And then you can also tweek the hell out of the testing benchmarks and then hire a lying test house for an additional 15-25% worth of "processing improvement" numbers that you know that Joe Normal type people will all initially see touted very loudly in your initial product release claims, but by the time the press embargoes are all over and the real benchmark results are being released by independent testers nobody in the buyers marketplace will ever see or ever remember it was all a bunch of lying BOGUS BS to begin with---- does this sound at all familiar to you?

Those end users will just remember reading the original exaggerated claims and they will now see that you have gotten all of those same exaggerated claims printed up now on the outside of the product cartons.

Linus and Linux won't play along with any more bogus lying that you Wintel MMC boys do any more, you sorry ol' Mickey Mouse Club F---up boys, you clumsy clumsy stupid Wintel idiots  ......
 
Since honesty and competence are key parts of being a Linux Foundation Level Member, Microsoft and Intel should both forfeit the Linux Foundation Member title and forfeit the $500,000 Linux Foundation Level membership fee  and simply go back to being "standard level" members and get treated that way going forward.  If they cannot be trusted to consistently ACT like Foundation Level Members, then they shouldn't hold that title going forward.



Roll Eyes


Oh, by the way, on December 3rd (less than TWO WEEKS after angrily slapping his palm and cutting off all the latest Wintel patches, Linus and the boys have finished fixing the new Meltdown and Spectre variants with NEW LINUX KERNEL PROGRAMMING  ---- at roughly the same ~5%~  processing penalty cost of the original Meltdown and Spectre patches that they did over six months ago to fix the first wave of Wintel security exposures.
Back to top
 
« Last Edit: 12/04/18 at 07:25:40 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #151 - 11/20/18 at 19:00:19
 

https://www.anandtech.com/show/13620/huawei-server-efforts-hi1620-and-arms-bi...





For at least four years now, Arm has been pushing its efforts to be a big part of the modern day server, the modern day data center, and in the cloud as a true enterprise player. Arm cores are found in plenty of places in the server world, with big deployments for its smartphone focused Cortex core family in big chips. However, for those same four years, we have been requesting a high-performance core, to compete in single threaded workloads with x86. That core is Ares, due out in 2019, and while Arm hasn’t officially lifted the lid on the details yet, Huawei has already announced it has hardware with Ares cores at its center.

Huawei Is A BIG Company .....
Normally at AnandTech when we discuss Huawei, it is in the context of smartphones and devices such as the Mate 20, or smartphone chips like the Kirin family. These both fall under Huawei’s ‘Consumer Business Group’, which accounts for just under half of the company’s revenue. One of Huawei’s other groups is its Enterprise wing, which is almost as big, and it creates a lot of custom hardware and silicon using its in-house design team, HiSilicon. HiSilicon’s remit goes all the way from smartphones to modems to SSD controllers to PCIe controllers and also high-performance enterprise compute processors.core

...... And It Makes Server CPUs
Last month, Huawei’s Enterprise Group lifted the lid on its fourth generation data center processor. Part of the TaiShan family, the Hi1620 would follow hardware such as the Hi1616 in being built using Arm IP. The new Hi1620 was announced as the world’s first 7nm processor for the data center, with the Ares cores being what would drive high-performance for its deployments.

While Huawei didn’t have any Hi1620 at the show, it was promoting the fact that it will be a cornerstone in its portfolio, and lifted the lid on a number of key parts of the chip.

The new Hi1620 will feature 24-64 cores per socket, running from 2.4-3.0 GHz. Each of these cores will have a 64KB L1-Data cache and a 64 KB L1-Instruction cache, with 512KB of private L2 cache per core. L3 would run at 1MB/core of shared cache, up to 64MB. On a scale of a consumer Skylake core, that means more L2 cache per core, but less L3. No word on associativity, however. One of the key question marks is on performance: a lot of vendors are hoping for an Arm core with Skylake-levels of raw performance.

Memory is set at 8 channels up to DDR4-3200, and the chip will support a multi-socket configuration up to 4S, with the coherent SMP interface capable of 240 GB/s for each chip-to-chip communication. The 4S layout would be a fully connected design.

IO for the Hi1620 is set at 40 PCIe 4.0 lanes, which is less than the 46 lanes on the Hi1616, but those ones were rated for PCIe 3.0. The Hi1620 will also have CCIX support, as well as dual 100GbE MACs, some USB 3.0, and some SAS connectivity.

The package listed is 60x75 mm BGA, which gives no real indication to the chip inside. But that’s a lot of balls on the back, and the package is larger than the 57.5x57.5 mm design from the last generation. Huawei states that the Hi1620 will be offered in TDP ranges from 100W to 200W, with the varying core count, but chips will be offered that can be fine-tuned for memory bound workloads.

There are still plenty of unanswered questions, such as the interconnect, but we really want to get to grips with the microarchitecture of Ares to see what is under the hood. A number of journalists at the show were predicting that Arm should be having a "full disclosure" event in the first months of 2019 to lift the lid on the design of the core.


OK, Ares is coming out in products very soon now, the A-76 based designs are being pushed forward at least six months early by Huawei and Qualcomm and Apple and they will come out on all levels of platforms.   Qualcomm is outing the 8150 SoC the first week of December, i.e. in 4 weeks so this should key ARM to disclose some more information.

This is Anandtech doing the early reporting, doing some very early pre-release reporting, saying a "full reveal" is going to be done by ARM very soon, around December-January time frame now that several customer products are "in-production" at TSMC for bulk revenue purposes.    Realize that according to all the past generations ARM secrecy and full non-disclosure remains in place until ARM reveals new generation detail information AFTER the first customer product is produced.

Note please:  this first one is being outed as a serious 24 to 64 core multi-multi chip built up RACK SPACE level chipset and then it is going to be outed again as a laptop chipset and then outed again as a tablet/phone chipset using lower, relatively more normal core count numbers.    

Look to see Microsoft Windows 10 support and Google Chromebook support it correctly right out of the gate on this one, since those partners were part of the A-76 development program at ARM.   Fuchsia will land early on this one too, just as soon as Fuchsia itself hits full Beta (which will be in the first half of next year, according to a few specific Huawei leakers).   Huawei, Qualcomm and Apple are the early producers, with Huawei leading in rack space implementation and in mid-range phone implementations.  

Apple and Qualcomm have already led the A-76 wave into primo phone uses ..... but only on their very heavily customized "built on ARM technology" premium plus phone chipsets.

This is not a Qualcomm or Apple fully customized "privatized" design job, it is a bone stock ARM standard design that anybody can license and can build in quantity at TSMC at 7nm (and soon at 5nm off the upgraded production cells).


Roll Eyes


Intel is growing more and more brand new competition all over the place now ........



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



https://www.anandtech.com/show/13614/arm-delivers-on-cortex-a76-promises



The Cortex A76 is a very solid CPU – Deimos & Hercules will follow up next year and the year after that.
Arm had already teased the successor to Enyo (Cortex A76) with the reveal of Deimos and Hercules. Here Arm promised 15-20% performance increases in the next generation. Arm’s strength here lies in actually delivering an overall excellent package of performance within great power envelopes. Also while this part of the PPA metric isn’t something consumer should inherently care about, Arm is able to also keep the CPUs extremely small.

We’ve just recently seen Arm’s new server core in the wild – Ares should be the infrastructure counterpart to Enyo/A76 and part of the recently announced Neoverse family of CPU cores. It’s not hard to imagine 32 or 64 of cores of this caliber on a single chip. Overall, we’re looking forward to more exciting products in the next several months – both in the mobile and infrastructure spaces.


Look at the graph just above, realize that most oriental suppliers really only buy into the larger bi-annual bumps so all the players simply ignore the next bar as "nice, but not doing that one" as doing a bump costs $$$$$.    "Skipping an ARM generation" is becoming the norm as the smaller interim bumps don't yield enough progress to make it worth the time and money to do them.   This is why the chart compares the bumps that it compares, they are the bumps that oriental suppliers are actually doing.

Prediction:   Players will do this A-76 bump, skip the Demos bump and do Hercules as that seems to be the frequency that the oriental suppliers are keeping.    Slamming a much stronger set of systems memory and a bigger AI block on to the old stuff during the time period you are skipping a bump seems to give an appropriate amount of product progress at a far lower licensing cost.
Back to top
 
« Last Edit: 12/01/18 at 00:54:52 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #152 - 11/23/18 at 17:17:17
 

https://www.tomshardware.com/news/microsoft-delays-windows-10-october-2018-up...

Microsoft has its own issues with shoddy misapplied Intel Drivers

The Windows 10 October 2018 Update's rollout has been a comedy of errors. First, it had to be pulled because of a file system flaw and incompatibility with Intel audio drivers. Then, while it was still exclusive to members of the Windows Insider Program, an issue with compressed folders was discovered. The update finally debuted in mid-November...but only for people whose systems don't run certain software or hardware.

The first set of delays arrived for people using Trend Micro's OfficeScan and Worry-Free Business Security software or certain AMD GPUs. Those were followed by another delay for people using the latest version of iCloud, which has problems with Shared Albums on the Windows 10 October 2018 Update. Now, according to Myce, systems with Intel display driver versions 24.20.100.6344 and 24.20.100.6345 have also been excluded.

There is some good news for Microsoft, though, which is that the latest delay appears to be Intel's fault. The company explained on its website:

"Intel inadvertently released versions of its display driver (versions 24.20.100.6344, 24.20.100.6345) to OEMs that accidentally turned on unsupported features in Windows.  

After updating to Windows 10, version 1809, audio playback from a monitor or television connected to a PC via HDMI, USB-C, or a DisplayPort may not function correctly on devices with these drivers." (This is separate from the other Intel issues.)

Microsoft said it's "working with Intel to expire these display drivers, including coordinating with OEMs, and will provide an update on the resolution in an upcoming release." It's not clear exactly what features Intel rushed to support with the drivers, or if the fix will involve releasing those features to Windows 10 users, but right now the issue's just the latest in a string of problems with the Windows 10 October 2018 Update.


Between the two halves of Wintel and all its various sub-vendors throwing both bad and "inappropriate" drivers around willy-nilly and in a completely out of time sequence manner, what real chance do you, the poor PC consumers, actually have for any form of peaceful year in 2019 ???

                                                                                                 Roll Eyes                    


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


https://www.forbes.com/sites/jasonevangelho/2018/11/26/microsoft-breaks-windo...



Microsoft may want to consider renaming the now-infamous Windows 10 October Update version 1809.   First it had to be pulled from the public due to critical file-deleting bugs. Then it was finally reissued in the middle of November.   With all the original "default apps" bugs still broken but MS had decided reissued the software anyway.

Machines with Intel HD Graphics drivers are losing their sound (again). iCloud syncing is still broken. And now Microsoft says Windows Media Player is broken in the latest cumulative update.


Microsoft seems to think that because they SAY something is fixed that magically it becomes fixed ......  except when they admit they didn't fix all of it, or eventually that they can't fix it at all until next year sometimes.

"After installing this update, users may not be able to use the Seek Bar in Windows Media Player when playing specific files," Microsoft writes in its known issues document. "Microsoft is working on a resolution and will provide an update in an upcoming release."

As with most bugs of this nature, it's difficult to determine how widespread the problem is. It's certainly not a crippling one (just install and use VLC instead), but it's baffling to me that a problem like this can be introduced into such a mature piece of software.


Listen to the author's thoughts on having even more backup than just having a full set of non-Mickey applications installed, like VLC.   He also disparages Mickey's excuse that it is all because they have to cover a bunch of different configurations --- THIS IS A BOGUS EXCUSE --- and then he gives an example of somebody else that does cover ALL of the same configurations as Mickey does (and more) but has no constant stream of failures like Mickey does.

That is a dizzying and mind-numbing number of hardware configurations, and I will never envy the position of software developers deploying stuff on that kind of scale. However, perhaps Microsoft should ease back on the "it just works because we said so" mentality for at least the next few months. This bug is not a major issue, but it is yet another bug serving to severely tarnish the reputation of Windows 10 and Microsoft's quality assurance methods.


Windows bugs multiply, making dozens of new bugs to plague somebody else ....

As a gentle reminder, I've been using Ubuntu Linux for nearly six months now.   I haven't lost any files. My audio never stopped working, I can choose new default apps at any time. My laptop has never rebooted unexpected for an update or any other reason. Seriously, I continue beating this drum because, if you don't truly have to use Windows 10, there are superior alternatives out there.   by Peter Bright, Arstechnia.com


Now my own gentle reminder,  Debian Linux goes through 3 tester rings to become Debian Stable, Ubuntu then takes Debian Stable and puts it through another set of test rings for interface mods before releasing it as a Ubuntu short term release.  The best of the best short term Ubuntu releases are then combined to become a Ubuntu Long Term Release.  

Then Linux Mint then takes the best most stable most current Ubuntu Long Term Release and puts it through another set of testing rings for any Mint interface concerns before releasing the very best of the best Ubuntu Long Term Release as the new stable core of Linux Mint.   Then Mint holds on to that "best of the best version" of Ubuntu for the next 3 Mint revisions, making completely sure all improvements passed forward really ARE improvements before letting the next Mint update go out into the wild.

Windows 10 is 100% pure quill "buggy trash" by comparison.   Seriously, I can only remember ONCE IN 10 YEARS when Mint had to take back anything for any reason, and that was for a bad driver that was put out by NVIDA.  

That bad NVIDA driver got pulled inside 3 days after going live .....

Wink


Back to top
 
« Last Edit: 11/29/18 at 04:21:32 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #153 - 11/29/18 at 04:48:26
 

Some long thoughts on progress ......

There are many articles saying that transistors will stop shrinking after 2-3nm, they will hit the wall then even if using the gate all around tricks.   We are currently at 7-5nm with 3nm gate all around coming in a couple of 3-5 years.  

No more transistor size progress will be possible at that time using silicone anything, supposedly.

ALL SUPPLIERS are now using some ARM core IP technology, yes, even including good ol' Intel.   Wintel's attempts to maintain their desktop and server monopoly will erode accordingly.

"Monopoly erosion" from the Intel perspective also includes Microsoft's intentional use of the newest, most powerful ARM processors and those which are using class beating AMD processors in their own Surface line of products.  This change has already begun.  

Microsoft does not intend to let Intel suck them down the toilet bowl swirl whirlpool with them .....

Cores will become more energy efficient and will MULTIPLY like rabbits everywhere.   This puts counter pressure on the old multi-threading tricks that Intel had piloted and on all the forms of predictive execution that Intel has depended upon for up to 30-40 percent of their recent performance boosts.

(I keep saying that Intel has made no real silicon progress in their cores in the last longest time, which is what has prompted Intel to just recently start using ARM core IP in their newest multi-multi-multi core products).

There is no purpose served by multi-threading in a multi-core world with literally dozens of separate cores that can do the much faster single thread workloads on separate physical cores. Remember, single thread workloads on separate cores are easily sandboxed for improved security.

So, Intel has begun to back off on multi-threading as it opens up Pandora's box for like a dozen serious security risks that Wintel is finding simply too hard to fix without killing processing speed and efficiency.  

Intel however is still currently grasping at predictive execution, despite all the security issues that it has.   It is the only move "forward" that they have when using their current x86 14nm production technology.

I find it stupidly amazing that Intel still touts some bi-annual 10-15% "processing improvements" while having to release security patches that kill 20-30% of Intel processing speed on certain sorts of work loads.

In the last few months, major users like AMAZON have welcomed AMD server products because of strong cost/performance improvements.   (AMD uses a modified form of ARM core to do their x86 style code)

However, the newest class of "pure Linux natural" fully stock ARM server core designs are showing up out there now as well.

Just this past few weeks, full production on box stock forms of ARM based server processors bearing 48-148 plus cores have started production currently giving slower performance per core compared to big hot running Intel cores but being so much better in aggregate performance, yes, yielding much better overall processing performance while costing much much less to run, booking 40% less in direct energy consumption and 50% less in site cooling costs.    

Oh yeah, the box stock ARM stuff cost less than half as much as Intel cored products to make and to buy too.   And yes, they cost less than the AMD based server products as well -- but they do not run legacy x86 programming nearly as well.   Folks find that moving over to running on Linux based software is quite easy though, and using Linux 100% makes the new ARM servers more efficient than "mixed programming" anyway.

You are at the cusp of the promised ARM revolution, but it will be relatively short lived as lithography shrinkage is running on up to the end of its road inside the next 5 years.


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


Much much money and research time is now being spent on AI and Quantum Computing ......

Back to top
 
« Last Edit: 12/01/18 at 20:39:58 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #154 - 11/29/18 at 13:24:37
 

Forbes recommends buying the new generation Chromebooks instead of current broken Win 10 based units .....

https://www.forbes.com/sites/kevinmurnane/2018/11/27/the-windows-10-october-u...



Windows 10 is always getting small-scale updates that have to be rescheduled so they don’t interrupt your work. In addition, Microsoft adds large collections of new features to Windows 10 with major updates that roll out in the spring and fall. The October Update, also known as version 1809, is this year's big fall update. It’s been a disaster from day one. Watching Microsoft continue to struggle and to fail to fix the October Update has been like watching a slow-motion train wreck. Unless there’s a Windows 10 program that you absolutely have to have, you don’t have to deal with this mess.

Consider getting a Chromebook.

Before the October Update released, Microsoft warned that trying to install the update if there wasn’t enough room on the drive could cause the install to fail and crash the system. The company didn’t reveal how much space was needed, didn’t explain how to recover from the problem, and didn’t address the question of why the update didn’t check available storage space before it tried to install.

Users who manually installed the October Update in the days before the official rollout began reported that the update was deleting their personal files stored in Windows 10’s default directories like Documents, Photos and Music. Microsoft halted the planned rollout. About a week later the company said the problem was fixed and the rollout would resume as soon as the fix was tested with Windows Insiders. Meanwhile it became known that Windows Insiders had reported the file deletion problem many times through the spring and summer. Microsoft either ignored the warnings or didn’t see them. The October Update rollout didn’t recommence until mid-November.

An early November update broke the ability to set some programs but not others as default applications. For example, How-To Geek reported that Adobe Photoshop could no longer be set as the default program for opening an image file. The problem has not yet been fixed.
On November 13, Microsoft reported that an update broke the seek bar in the Windows Media Player. It remains broken.

On November 16, Microsoft reported a number of problems with the October Update including VPN users losing internet connectivity in some circumstances, iCloud Shared Albums failing to sync or update, and mapped drives failing to reconnect after logging into a Windows device. None of these problems have been fixed.

Had enough? Don’t want to continue biting your fingernails hoping your system isn’t borked after Windows 10 finishes updating? Tired of waiting around for the latest update to finish or to eventually be repaired?

Give Chromebooks some serious consideration.

The days when chromebooks were nothing more than hardware platforms for the Chrome browser are long gone. Today’s high-end chromebooks are fully functional machines for both work and play. They run Chrome OS, Android and Linux apps seamlessly, side-by-side on the same desktop. They’re the most flexible devices you can buy. They’re also extraordinarily secure, boot lightning fast compared to PCs and Macs, and you never have to deal with updates.

The update problems plaguing Windows 10 are not an issue for chromebooks. You never have to wait for an upgrade because Google constantly updates Chrome OS in the background. Moreover, if you power a chromebook down when you’re finished using it, the Chrome OS’s Verified Boot system guarantees that you’ll always have the current version of the operating system running on your machine when you power it back up.

Verified Boot also insures that any malware that managed to get through the Chrome OS’s state-of-the-art security and modify the operating system will be eliminated when the system boots up. Powering up a chromebook every time you want to use it isn’t nearly the annoyance it is with other systems. Windows 10 and MacOS machines usually take somewhere between 30 and 90 seconds to boot. As I reported in another article, my Pixelbook takes about 8 seconds to arrive at the password screen and is ready to go in another 4 to 5 seconds with several tabs open in the Chrome browser.




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






Intel is dumping their ongoing plans for Cannon Lake mobile chipsets and possibly the next generation of Intel mobile processors which was Whiskey Lake or some such strange sounding moniker.   All of these were supposed to be 10nm Intel domestic produced parts off of the now out of date (non-working) Intel 10nm lines.  

Intel is now planning on executing some sort of long term processor reorganization and doing some more production line rebuilds to support their new plan which is yet to be officially announced.

This news comes from a raft of Intel suppliers who have had cancellations of this and that rain in upon their heads this past week.

Why, do you ask?   Because ARM and AMD had "overcome by events" these planned Intel generations with current announced and being produced 7nm AMD products and with even more "easily foretold to be soon released products" from several different ARM suppliers making the long delayed Cannon Lake and Whiskey Lake product lines "preemptively non-competitive" out in the future should they ever actually be built.

Right now (using Germany's biggest tech store Mindshare.de as a market predictor) the new AMD 7nm processors are selling in at a 2.5x ratio to Intel's old traditional 14nm processors.    This is said after tracking this trend for two full rolling quarters of 2018.   Rolling quarters is a good way of tracking the increase or decrease as the sharp increase in Intel pricing really really showed up well in this metric.  

Intel is shedding market share like a kitty shedding fur in spring time right now  .....

So, Intel would rather empty out their bulging warehouses at this stage, rather than build any more "to be obsoleted" stuff.   They are getting their premium prices at the moment, so this is maximizing short term profits, something Intel is pretty good at doing.   By not building what they will not need Intel saves some manufacturing capacity for those things that they can actually sell when produced off their 14nm process lines.  

Intel is also renovating some more of these nearly obsolete 14nm manufacturing production lines, taking them out of service temporarily for lithography processing based upgrades.

We await the coming Intel announcements of new Intel 7nm products produced made up out of chiplets that are built in volume using turn key production lines from ASML and "built on ARM technology" designs from ARM Holding.   Intel will choose to build the center chiplet themselves on their own 14nm lines but the myrad smaller workhorse cores will have to be built at TSMC at 7nm in the short term in order for Intel to be competitive to the already announced AMD product lines.

To buffer things some (and speed up the new Intel introductions) look for TSMC to produce some large quantity of wafers of chiplets in bulk for Intel to use during calendar 2019,
(or at least until the new Intel lines are set up, tuned and running well ......)


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


Interesting Rumors    (having 3 different sources pushing about the same rumor gets me interested -- where there is smoke there is at least a little fire of some sort going on).

Rumor #1    Microsoft is coming out with a new Windows lite to compete with Chromebooks.   Yeah, we hear this every other year and eventually a fizzle happens because Windows 10 is such a huge screwed up pig it can never become light and fast like Chrome OS.

Rumor #2    Microsoft is taking Open Source Chromium Browser to make the core of their new Edge replacement browser.

Rumor #3    Microsoft is branching off a build of Chromium OS off similar to what Amazon has done.   This one could become Rumor #1 and Rumor #2 very easily.   All 3 rumors added up sound .......  mebbe possible in some mixed fashion ???

Please remember, this is all legal and fairly up & up as Microsoft is both a Chromium Partner and a Linux Foundation Level Member.

The rumors have an unappealing twist to them, MS and Intel have both show signs of having hit the point of "giving up" on their old standard proprietary stuff as it has become very unwieldy to keep updated and it isn't really competitive or secure any more.

Bad news here for the new Fuchsia OS as Google will wind up giving Fuchsia away to MS and Intel at the same time they release it to the FOSS people.   Google has not released the license terms for Fuchsia OS yet so some little wrinkles may happen to keep the Wintel boys actively sucking up on Google's work.


Roll Eyes


Confirmations:

https://www.windowscentral.com/microsoft-building-chromium-powered-web-browse...

https://www.theverge.com/2018/12/4/18125238/microsoft-chrome-browser-windows-...

https://www.neverware.com/#introtext-3      This is a Chromium rip off by a commercial company, MS could do similar to this or just buy these guys complete.

https://www.howtogeek.com/397210/the-future-of-windows-what-are-polaris-and-w...

https://www.laptopmag.com/articles/microsoft-killing-edge-google-chrome


Back to top
 
« Last Edit: 12/09/18 at 17:17:44 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #155 - 12/04/18 at 06:51:14
 

How loyal are Intel fans really ???

About 50% loyal and that 50% number is falling rapidly .......


How loyal are MS fans ???

There aren't any 100% loyal MS fans right now (ones that still have a functional brain, anyway)    
Mickey needs to do some housecleaning and floor scrubbing ASAP as EACH ERROR THEY MAKE FROM THIS POINT FORWARD BECOMES ........
    (bad, really really bad for their image)

It is very bad when Mary Jo herself says Mickey's development system is so rotten they actually intentionally sent out known bad MS software to the public consumers SIMPLY TO COLLECT FAILURE RATE STATISTICS ON IT so they can prioritize what to fix first for their business customers.

this pile of Win 10 shite is sooooo big now they need to use sampling statistics to get a handle on it

Microsoft now wants to adapt and use Linux code and Chromium code and other good FOSS software code that still works very well to create a new MS OS system which has a development name of Windows Core / Polaris.   This sounds neat to MS, but MS really needs to add something of real value to this mix from the very beginning or else folks will just use the various FOSS software packages directly and forgo even having MS around .......  especially if MS keeps on being totally buggy crap the way it is right now.

Look to see license terms for all the FOSS softwares to shift to keep MS from trying to own their hard work.

The new IBM/Red Hat/Ubuntu/Linux Foundation agreements in principle that you keep hearing about are intended to preempt MS's efforts to control the standards for Linux right now and going forward into the future.  

Letting MS take any form of leadership on Chromium or Linux or Android is a very very bad idea.   Historically, MS has operated by a join, modify, document and then license/limit/extinguish methodology that they have used for over 50 years now to steal FOSS code, going all the way back to the Unix DEC/VAX era.  

If MS can take over your tech for free (or steal your ideas) they will surely do so.    And later on after a decade has passed they will actually turn around and sue you claiming THEY invented it in the first place (based upon having written up and published and maintained standards for what you had done in the first place).   MS employs a bevy of lawyers to do just this activity and they make 10's of Billions each year off of the Android community doing it ......
Back to top
 
« Last Edit: 12/23/18 at 03:05:19 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #156 - 12/04/18 at 23:50:23
 

https://www.howtogeek.com/397210/the-future-of-windows-what-are-polaris-and-w...



Microsoft wants all Windows devices to be based on the same operating system, and it’s building Windows Core OS to be that operating system. A Microsoft job posting on LinkedIn says Windows Core OS (WCOS) will be “the OS shared across all new devices.” That same Microsoft job listing also refers to Windows Core OS as a “new operating system” and says the OneCore team at Microsoft is involved.

Windows Central‘s Zac Bowden spoke to sources at Microsoft about Windows Core OS back in 2017. Here’s how they described it:

Windows Core OS (WCOS for short) is a new, modern version of Windows and is a monumental step forward in making Windows a truly universal OS. In short, WCOS is a common denominator for Windows that works cross-platform, on any device type or architecture, that can be enhanced with modular extensions that gives devices features and experiences where necessary.

Windows Core OS is the next step in making Windows 10 fully modular. It’s a single base operating system for all devices. Rather than Windows 10, Windows 10 Mobile, and the Xbox operating system being based on OneCore but still different, they’ll all be running Windows Core OS.

According to ZDNet’s Mary Jo Foley, Windows Core OS won’t natively run Win32 applications—in other words, traditional desktop software. It runs Universal Windows Platform (UWP) apps. However, Windows Core OS is modular. Support for classic desktop applications could be added as a module, and probably will be. But support for traditional software isn’t a core part of the system.

As Foley also points out, Windows Core OS was previously known as AndromedaOS inside Microsoft. That’s because it may launch on the rumored “Andromeda” mobile device Microsoft is working on, which may include dual screens. This device was rumored for launch in 2018, but Foley writes that Microsoft has gone back to the drawing board. It may never even be released, so don’t hold your breath
.

Roll Eyes
Back to top
 
 

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #157 - 12/06/18 at 20:26:49
 

https://www.pcworld.com/article/3325329/components/qualcomms-snapdragon-8cx-f...

https://gizmodo.com/the-snapdragon-8cx-is-qualcomms-first-purpose-built-chi-1...




But next year Qualcomm plans to launch its first processor designed specifically for Windows PCs. The Qualcomm Snapdragon 8cx isn’t a modified smartphone chip. The company says the 7 watt, octa-core processor will offer performance that’s on par with a 15 watt Intel U-series laptop processor, but using just half the energy at 7 watts.

We’ll probably have to wait a little while to truly put that claim to the test — Qualcomm says it’s sampling the chip now, but it won’t show up in consumer devices until the third quarter of 2019.


Huawei and Mediatek and Rockchip will also be in line to put forward laptop capable chipsets for both Win 10 lite and Chromebooks, fully intending to slice off a piece of Intel's overpriced fat juicy market share .......      

And don't forget to count Apple, with their A12x and A13x super chips that will run Apple/Mac OS.

Back to top
 
 

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #158 - 12/07/18 at 02:08:40
 

https://www.theverge.com/circuitbreaker/2018/12/6/18129678/qualcomm-8cx-mozil...



Qualcomm’s new, purpose-built Snapdragon 8cx is the company’s chance to finally power a Windows laptop worth buying. Battery life and integrated cellular connectivity (read: your laptop always has an internet connection) have always been the company’s strengths, but its previous Snapdragon 850 didn’t have always have enough raw horsepower to muscle through a web browser full of web apps.

We won’t know if that’ll truly change until we’ve spent some quality time with real Snapdragon 8cx-powered computers — of which none have been announced so far — but I did just get my hands on Qualcomm’s Snapdragon 8cx reference design laptop, and my first performance impression was promising.

When I fired up a beta of a new Firefox browser that’s been overhauled to natively run on ARM-based computers, I was able to easily open up a dozen browser tabs and scroll through them without any major issues. That’s while a PowerPoint presentation and a Windows Store instance hovered in the background, and while mirroring to an external monitor. It wasn’t a buttery smooth experience, but it wasn’t a laggy one either, and it felt like I had headroom to spare.

IT FELT LIKE I HAD HEADROOM TO SPARE
And I’m not just talking about a dozen instances of Google. I’m talking about websites with a decent amount of page load and some ongoing demand on a computer’s memory and processor, including The Verge, CNET, a pair of YouTube HD videos, a Google Sheets spreadsheet, and a fully loaded instance of Tweetdeck with all my social media columns. It did take a second for the spreadsheet to load properly, but that could have been connectivity — I was surfing on a fairly limited 20 Mbps Verizon LTE connection at the time.

Obviously, this isn’t nearly enough testing. I didn’t run a single quantitative benchmark. Also, the friendly on-site Qualcomm rep wouldn’t let me access the Windows Task Manager to see just how much my browser load was stressing the new chip.

And there’s not a lot to say about the reference design itself. It looks and feels like a pretty generic, mid-range, thin but a little plasticky Windows convertible, with a fairly dim screen and a camera bump around back. None of these things will necessarily be true of final devices from proper PC manufacturers.

We’ll hopefully have the chance to test more in mid-2019, when Qualcomm expects the first 8cx devices will come to market. For now, the early Firefox performance is a good sign.




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



https://www.theverge.com/2018/12/6/18128648/microsoft-edge-chrome-chromium-br...

Microsoft is rebuilding its Edge browser on Chrome and bringing it to the Mac



The Verge understands Microsoft has been considering this move for at least a year, and a lot of the push has been from consumers and businesses who wanted the company to improve web compatibility. Edge has been improving on this front, but even small compatibility issues have caused headaches for users along the way.

A move to Chromium will immediately solve these nagging web compatibility issues, and it aligns Edge with Chrome and other browsers that also use Blink.

Microsoft has also heard loud and clear from businesses that want the company to support a modern Edge browser across all versions of Windows. Many businesses have machines running Windows 7 and Windows 10, in a mixed environment. As a result, Microsoft is bringing Edge to Windows 7 and Windows 8, decoupling it from being exclusive to Windows 10. Edge will become a downloadable executable across all supported versions of Windows, and it means Microsoft will be able to update it far more frequently than before. It’s not clear if this will be monthly, but it will certainly not be tied to every major Windows 10 update anymore.

Another big part of overhauling Edge involves developers. A lot of web developers use a Mac to develop and test sites, but Edge doesn’t exist there, and it’s currently difficult to test Microsoft’s web rendering engine on a Mac without dual booting Windows. Microsoft is now bringing Edge to the Mac. We understand it’s not a move designed to grab more market share specifically; it’s more about making it easier for developers to test Edge. Microsoft hasn’t committed to a specific date for Edge on the Mac, but we expect to see it later next year.

CHROME ON WINDOWS WILL GET BETTER WITH MICROSOFT’S HELP
All of this work means that, ultimately, the browser engine that powers Chrome will get better on Windows. Microsoft is committing to contribute web platform enhancements that will improve both Edge and Chrome on Windows, including things like touch performance, accessibility features, and support for ARM-based versions of Windows. Microsoft has been working closely with Google engineers to help support a native version of Chrome on Windows for ARM, and this will now be available soon as a result of that work.

Microsoft is only just starting to disclose this platform shift to other companies involved in the Chromium project, and the company isn’t ready to start distributing daily builds of Edge running with Chromium just yet. Those beta builds will start early next year, before Microsoft makes the necessary changes in Windows 10 to shift Edge toward Chromium. We expect to see Windows 10 move to this Chromium-based version of Chrome sometime in 2019.

Microsoft now wants to collaborate with Apple, Google, and everyone else who also commits changes to Chromium. “If you’re part of the open-source community developing browsers, we invite you to collaborate with us as we build the future of Microsoft Edge and contribute to the Chromium project,” says Belfiore. “We are excited about the opportunity to be an even-more-active part of this community and bring the best of Microsoft forward to continue to make the web better for everyone.”



Pay close attention, Google ---- letting MS take ANY FORM of a leadership role in FOSS Chromium Browser or FOSS Chromium OS means they will try to execute a "fork it plan" to privatize their version of the FOSS IP and then legally maneuver to take the whole thing over inside 3-4 years.   Please remember what they are still doing with Android, charging people a yearly fee for an "agreement" simply so they will not to be sued by MS's many many lawyers ......

Listen very carefully to the Mozilla folks, they went through this "let's be buddies" charade with MS back when they were still called Netscape Browser ...... and they have yet to recover from the evil that MS did to them back then ......  

Then listen to Apple, as they remember well hosting Bill Gates for a one day visit and taking the time to explain "Macintosh Windows" and that new fangled Apple Mouse stuff to him.

Back to top
 
« Last Edit: 12/11/18 at 19:58:59 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #159 - 12/07/18 at 18:43:13
 

https://hothardware.com/news/microsoft-windows-10-fail-cumulative-update-surf...

Microsoft Windows 10 Fails Continue As Cumulative Update Cripples Surface Book 2 With BSOD
(if you are lucky .....  and a complete bricking of the machine if you are unlucky)

Microsoft has just encountered a string of bad luck with Windows 10 updates in recent months. It had to pull its October 2018 Update after it went rogue by deleting users' files. Most recently, a Windows Update borked the Windows Media Player in Windows 10.

Now, we're having that a rather routine release of cumulative updates resulted in trouble for owners of Microsoft's very own Surface Book 2. You wouldn't expect for a cumulative update (KB4467682) for Windows 10 version 1803 (April 2018 Update) to have harmful effects on a box stock Microsoft Surface machine, but in the case of some Surface Book 2 owners, the update released last week has sparked a number of Blue Screen of Death (BSODs) on systems.



Surface Book 2 owners have been complaining not only on Reddit, but also on Microsoft's own answers forum. Some lucky owners have been able to restore their Surface Book 2 machines to normalcy by uninstalling KB4467682, however, some unfortunate souls have found that going that route only presents them with a bricked machine upon rebooting.

"Well, for some reason removing the update bricked my surface," wrote Luis10e10 on Reddit. "Now I can't even get into the OS. Just have blue screens. I will have to reinstall windows."

"Also having this problem," added seinberg. "WTF Microsoft? Seriously - your flagship laptop device, and you're sending out updates that cause widespread BSOD. I can't even uninstall the update without it failing."

The cause of the problem hasn't been pinned down at the moment, but some people are pointing to potential problems with Intel display drivers, the onboard Bluetooth controller, or Windows Hello. With that being said, Microsoft has officially addressed the issue in an update on the support page for KB4467682:

MS writes After installing this optional update some users may get a blue or black screen with error code, “System thread exception not handled.” A resolution for this issue will be available in the December 2018 security update release.

The company says that is now actively blocking Surface Book 2 owners from receiving the update.

Whatever the cause for the issues, Microsoft has a serious problem on its hands. It's one thing for Windows 10 installs to fail and cause issues on PC hardware that is anything but homogenous given seemingly infinite configuration options, but for it to happen on a flagship product produced by Microsoft is a bit harder to swallow.



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


https://www.techradar.com/news/microsoft-halts-yet-another-windows-10-update-...



Microsoft’s Windows 10 update woes are continuing, with the company having to halt the recent Windows 10 November 2018 cumulative update (KB4467682) after reports it was causing PCs to crash and display the infamous ‘Blue Screen of Death’.

At first, the errors appeared to only affect the Surface Book 2, Microsoft’s own hardware, which lead to the company halting the update for those devices.

However, Microsoft has now revealed that the problem is more widespread, and it has now pulled the update from all devices.

In a new support post, Microsoft explains that “After installing this optional update some users have reported getting a blue or black screen with error code, “System thread exception not handled.” As a precaution, we have removed this optional update from Windows Update and Microsoft Update Catalog to protect customers.”

Pulling an update due to it causing PCs to crash is embarrassing enough at the best of times, but Microsoft has had a brutal few months recently where it has had to halt the rollout of a number of updates, including the major Windows 10 October 2018 Update.

The update that has caused the new problems is an optional cumulative update that packages a number of ‘fixes’ into one download. Hopefully this means that it won’t have affected quite as many people as other major updates. Still, it’s not a good look for Microsoft.

According to the company, “fixes and improvements will be available in the December 2018 security update release and will include a resolution for this issue,” so if you are affected, uninstall the update and wait for a new one. That’s if the next update doesn’t get pulled as well. Going by Microsoft's recent form, we can't rule that out.


Please note that this mess occurred AFTER MS began doing full QC review / inspections prior to releasing any new updates.  

MS's very best reinforced QC inspection efforts could not detect and stop this cumulative addition to the corporate shite pile --- THIS WAS AN ERROR THAT BLUE SCREEN KILLED ONE OF THEIR VERY OWN PREMIUM SURFACE BRANDED MACHINES, yes that one, the one that had just now gotten its Good Housekeeping status put back to "Recommended" for the very first time this calendar year after being "Not Recommended" for over two years due to multiple quality issues leading to multiple multiple customer complaints ......


   ..... Hey, we had two QC guys assigned to this project, where did they get off to?  

Oh my God, is that somebody's fingers with a wedding band sticking out of that huge reeking pile of shite  ?????  


Auuugghh !!! .......   somebody get a shovel and a stretcher, quick !!!


Back to top
 
« Last Edit: 12/13/18 at 04:46:18 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #160 - 12/08/18 at 22:41:05
 

https://arstechnica.com/gadgets/2018/10/report-intel-is-cancelling-its-10nm-p...


Earlier last month, it was reported that Intel is cancelling its troublesome 10nm manufacturing process. In an unusual response, the company has tweeted an official denial of the claims.

Development of Intel's 10nm process has been difficult. Intel was very ambitious with its 10nm process—planning to increase the transistor density by something like 2.7 times—and wanted to use a number of exotic technologies to get there. It turned out that the company had bitten off more than it could chew: yields were very low, which is to say that most of the chips being manufactured were defective.

In a bid to recover, Intel is now in reality striving for a less ambitious 11-12nm scaling (though still more than double the transistor density of its 14nm process). It has one oddball processor on the market: the Cannon Lake core i3-8121U which is built half on its new 11-12nm process with the GPU half being vended from AMD.  

Unusually for this kind of processor, the built-in fully integrated Intel GPU has been completely disabled. That's because the Intel GPU is not working; we've heard it claimed that the Intel GPU uses different designs for their logic than the Intel CPUs, and these Intel GPU designs are proving particularly troublesome.   Paying for a built in Intel GPU and then paying again for a second full GPU from AMD and putting all three major chipsets on the same daughter board means Intel is paying out over 3x more $$$ simply to be able to say they got their 10nm product into production ...... finally ......

The company's most recent estimate is that 10nm will go into volume production in the second half of 2019. The report from SemiAccurate cites internal sources saying that this isn't going to happen: while there may be a few 10nm chips, for the most part Intel is going to skip
10nm and go directly to its budding 7nm process.


Note please that SemiAccuate got it right in all the particulars they had listed so far, but Intel simply did not want to admit to such a total 10nm defeat so Intel has muddied up their troubled waters as much as they possibly could.    

Paying 3x more for the double GPU equipped "10nm" processors combined with the extra cost for a special run motherboard to hold the doubled up monstrosity seems to be an extravagant price to pay for a little bit of corporate image protection, but remember Intel's overall stock price isn't based upon reality, it is based on that carefully maintained corporate image.



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



https://www.extremetech.com/computing/278095-intel-may-have-10nm-hardware-in-...

According to SemiAccurate, Intel has now made significant changes to its 10nm to push the tech out the door, relaxing its own Intel design rules and changing the nature of the implementation. It’s not clear how significant these changes will be. SemiAccurate earlier argued that the new rules would leave Intel’s 10nm more equivalent to a 12nm process node, but its most recent update argues that “the new downgraded ’10nm’ process from Intel will not take as big a hit from the removal of this tech as SemiAccurate said earlier, but it will still take a hit.”

SemiAccurate should be taken with a grain of salt, but the idea that Intel would tweak 10nm to get it out the door more effectively isn’t surprising. Nor is the idea that the company might have had to back off its initially-aggressive plans for 10nm in order to make up for lost time, but the resulting hidden 12nm feature size puts Intel about equal to the best efforts from Global Foundries (which in turn were abandoned by Global Foundries recently in an all out push for getting some new direct burn 7nm EUV lines).

Business-wise, 2018 hasn’t been a good year for Intel; the company has been rocked by security problems from Spectre and Meltdown and pinched by manufacturing issues related to the ramp of its long-delayed 10nm process. Earlier this year, the company told customers to expect 10nm shipments to be pushed back all the way to Q4 2019. It’s repeated that guidance several times since, but the timeline may have improved in recent days.

Last week, Intel acting CEO Bob Swan published a letter in which he wrote “We’re making progress with 10nm. Yields are improving and we continue to expect volume production in 2019.” This omitted the specific reference to “Holidays” 2019 that Intel had previously given, but it wasn’t clear if the company was changing its guidance or Intel was simply choosing not to emphasize that it would be the better part of a year before it could ship 10nm parts in volume, this in a letter that was supposed to be stock holder positive and forward-looking.


Back to top
 
« Last Edit: 12/11/18 at 08:15:21 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #161 - 12/09/18 at 15:43:16
 

https://www.anandtech.com/show/13683/intel-euvenabled-7nm-process-tech-is-on-...

Intel: EUV-Enabled Intel 7nm Process Tech is reported to be on track for 2020 - 2021
by Anton Shilov on December 6, 2018 10:00 AM EST

Originally planned to enter mass production in the second half of 2016, Intel’s 10 nm process technology is still barely used by the company today. Currently the process is used to produce just a handful of CPUs, ahead of an expected ramp to high-volume manufacturing (HVM) only later in 2019. Without a doubt, Intel suffered delays on its 10 nm process by several years, significantly impacting the company's product lineup and its business.

Now, as it turns out, Intel’s 10 nm may be a VERY short-lived node.

Why?  Intel’s 7 nm tech is on-track for introduction in accordance with its original schedule.  Intel’s 7 nm production tech will use extreme ultraviolet lithography (EUVL) with laser wavelength of 13.5 nm for select layers, reducing use of multipatterning for certain metal layers and therefore simplifying production and shortening cycle times. As it appears, the 7 nm fabrication process had been in development separately from the 10 nm tech and by a different team. As a result, its development is well underway and is projected to enter HVM in accordance with Intel’s unannounced roadmap, the company says.

Murthy Renduchintala, chief engineering officer and president of technology, systems architecture and client group at Intel is quoted to have said at the Nasdaq's 39th Investor Conference:

“7 nm for us is a separate team and largely a separate effort. We are quite pleased with our progress on 7 nm. In fact, very pleased with our progress on 7 nm. I think that we have taken a lot of lessons out of the 10 nm experience as we defined that and defined a different optimization point between transistor density, power and performance and schedule predictability.  So, we are very, very focused on getting 7 nm out according to our original internal plans.”






The Intel exec reaffirmed the company plans to start HVM production of client CPUs using its 10 nm process technology in 2019, with datacenter products following on a bit later. That said, the Intel exec is very clearly not admitting Intel will be skipping any of its already announced 10 nm products, but instead simply implies that its 7 nm products may hit the market earlier than we might expect up until today (i.e., four years after the 10 nm).

“One thing I will say is that as you look at 7 nm, for us this is really now a point in time where we will get EUV back into the manufacturing matrix, and therefore, I think, that will give us a degree of back to the traditional Moore’s Law cadence that we were really talking about,”

“With 7 nm we are going back to more like a 2X scaling factor and then we will really moving forward with that goal.”

Intel has never disclosed characteristics of its 7 nm fabrication tech, but a major reduction of multi-patterning usage as well as a more traditional 2X scaling goal vs. 10 nm indicates a more extensive usage of EUVL.



According to ASML, one EUV layer requires one EUV step-and-scan system for every ~45,000 wafer starts per month. Therefore, if Intel plans to use EUVL extensively for 10 to 20 layers, it will require approximately 20 to 40 EUVL scanners for a fab with a 100,000 wafer starts per month capacity. Considering that Intel is not the only company with plans to use EUVL in the 2020s, getting the number of EUVL scanners it might need for HVM at multiple fabs may be a challenge.


Where can Intel put all these many state of the art 7nm machines, an existing place with all the needed infrastructure already sitting there, unused?     Why at Fab 42, the Fab plant they had built originally for 10nm which never worked right and was never fully completed for real production.

Understand that Intel will not doing much for 10nm for most of next 2019 year, and will require part of another year's worth of lag time (2020) before any sort of fully tuned 7nm Intel production can begin.  All this simply adds credence to the rumor Intel SIMPLY HAS TO BUY 100's of thousands of full sized wafers per month of their own 7nm designs (to be built by TSMC on TSMC equipment) simply to tide Intel over until Intel can get their own production equipment up and running.

AND WE REPEAT -- understand that Intel will not doing much for 10nm for most of next year (2019) nor for the majority of yet another year's (2020) worth of lag time before any sort of full volume 7nm Intel in-house production can begin.   Production of Intel chipsets at TSMC will have to take place (whether Intel ever admits to it or not).

TSMC, Qualcomm, Huawei, Apple, Mediatek and Samsung (all producing 7nm designs right now) will not be standing around idle during this same 2 year period of time.    One can presume these folks will be rolling over to 5nm and below at about the same time as Intel is finally cranking up their own 7nm plant, finally.



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



Intel announces their updated Product Roadmaps through 2022


2018      will have a 14nm Coffee Lake refresh.

2019      will have a 10nm Sunny Cove introduction on the bottom end of the Intel line up (with secure all single threaded performance on many many cores)

2020       Willow Cove to come in to fix whatever problems still exist in Sunny Cove, with many many cores with secure single thread actuation which will move all the way up to the top of the product listings.

2021      Golden Cove, a 7nm changeover period   (this is three years late compared to the ARM boys and will be 2 full generations behind at that point in time)


What Intel is laying out right now puts them 1-2 generations behind ARM and ARM chipset suppliers all the way out into the future.    But what it does do is to AVOID buying hundreds of thousands of wafers monthly from TSMC ---- I think Intel finally got their quoted prices from TSMC and it gave them a small stroke when they read it.



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



https://www.theverge.com/2018/12/12/18137401/intel-foveros-3d-chip-stacking-1...

Intel also released their proposed chip layouts .......





The first frame is now, the second frame is 2019, the third frame is a special construction that is CPU on top of CPU on top of CPU on top of CPU ......   you get the picture.

The last stacked CPU one has grown an instant moniker from the PC press people.

THE TOWERING INFERNO

Intel cores already suffer from overheating and thermal throttling, so what do you think this beast will do, what with the four deep stacked CPU on top of CPU on top of CPU on top of CPU ????   (you are asking for all the processor waste heat to move on up THROUGH the chain of CPUs to get to the heat sink)


   Intel is really grasping at straws now .......

Back to top
 
« Last Edit: 12/15/18 at 04:12:27 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #162 - 12/11/18 at 19:42:23
 




Progress in laptops due to the Intel AMD processor wars .......

This is an under 3 pound 17" display "LG Gram" brand laptop with a touch screen, lighted keyboard and a 13 hour battery life.  

It is quite expensive at the moment being the first one out of the gate and all, but the desirable portions of the feature set will roll down to less expensive units within a year or two.

HINT:  if you think this is good, wait until you see what you can get next year this same time.


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


BAD NEWS

Intel has started spontaneously rolling back to 22nm whenever they lack 14nm capacity to fill current orders.    And, as long as the machine builder is willing to take the chips at a discount to fill their original 14nm orders, Intel sees themselves as having ZERO responsibility to tell anybody about the substitution.  

Intel feels that their customer (the machine builder in question) obviously knows about the substitution as a different socket is needed to use the cheaper older chipset.

And if that machine builder does not happen to change his part number or notify his customers or properly review his advertising throughout all his distribution channels, who is at fault for that?  

Not Intel, according to Intel, anyway.         Roll Eyes

Machine builders do feel that they are being strong armed into taking the 22nm chipsets, as if they don't take them they will miss the Christmas selling season.  

The machine builders also feel the default "do nothing" pathway is a total "keeping mum" be completely silent secrecy --- which means the final consumers will see A advertised, will decide to buy A and will actually get C inside the product, with C performing at a slightly reduced capacity and using more energy.   Since the real improvement for 14nm was always very minimal compared to 22nm the customers may never really notice they got screwed.

GUYS, THE REAL ANSWER IS YOU ......  YOU MUST INTENTIONALLY BUY AND USE AMD PROCESSOR PRODUCTS !!!!
Know you are getting what you think you are getting (and that you will always get what you are PAYING for).

Intel isn't acting trustworthy going forward on telling you what is what with processors being acetone rag wiped,  re-labled and sold for something else all over the place now-a-days ......

https://www.techpowerup.com/247773/intel-to-move-select-chipset-fabrication-b...

https://www.anandtech.com/show/13714/intel-adds-b365-chipset-to-lineup-the-re...

https://www.tomshardware.com/news/intel-b365-chipset-specs,38235.html

And yes, these articles do mention the use of the acetone rag and the relabeling of 5 year old warehoused chipsets to fill in for production shortfalls ......


Back to top
 
« Last Edit: 12/19/18 at 04:22:14 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #163 - 12/17/18 at 05:54:10
 

Microsoft sends out a special update announcement and offers to "Completely refresh your Windows 10" installation as a separate special update ......

(just type "Refresh Windows 10" into your browser search bar)

Why ?????   The recent years of endless messed up nightly update stuff has left many machines limping along and Mickey wants the chance to READ YOUR HARDWARE afresh with their newest hardware sensing programs and then to apply only the newest (most fixed) generation of MS drivers and newest most fixed Win 10 software to what you actually have for old hardware.

And why did I have to discover this wonderful offer?   Mffy wife's machine had recently stopped recognizing its roller ball mouse and anything else that was attached to the laptop's USB ports.   Yup, the machine quit finding its own original OEM USB ports.

So I tried out MS's kind offer as I found it to be suspiciously TIMELY and by inference perhaps a necessary step in healing the items Mickey had done to the wife's mysteriously broken Win 10 machine ......

....... and yes, after 2 nights of secret Santa updates, the wife's machine suddenly had USB ports and all the hardware attached to them again.  The main printer worked again and the little label printer worked again.   Best yet, her favorite roller ball mouse regained full function as well.  The USB based multiport device even started working again and that Amazon Basics multiport (originally advertised to use the stock MS drivers) had mostly broken over a year ago .......

Life is good once again and perhaps Mickey actually picked up a few of their many times repeated foul balls by doing this.    

However, I will contrast Linux Mint Mate OS that for YEARS AND YEARS has hit no nasty foul balls and had nothing to go running around scurrying after while the crowd of spectators enthusiastically hissed, hooted and boo'd at them.

This might offer you a chance to get your machines tended to (in some sort of generic fashion) so consider perhaps doing it if your machine is currently nonfunctionally sick.    DO NOT DO IT IF YOU HAVE NO SERIOUS ILLNESSES AT THIS TIME AS IT WIPES OUT ALL MISC. SOFTWARES (including the misc. little shite like your old versions of MS Office, just to name a few).   Dinking around with updates like this often involves losing old software, but MS sez get used to it as they remove all unauthorized software stuff routinely now-a-days.


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


Oh by the way, TSMC is shipping the first trial lots of 5nm SoCs to three different suppliers this month.   This is good news as that means that 5nm is gonna be rolling on into 2019 on schedule ......

When this flies later on in first half 2019 when full production hits then Intel will be 3-4 full generations behind the pace as they still will be using 14nm with 22nm as their production short fall fill in ......
Back to top
 
« Last Edit: 12/19/18 at 07:43:21 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & sucesses
Reply #164 - 12/20/18 at 12:42:10
 

https://www.techspot.com/news/77963-tsmc-gets-approval-build-3-nanometer-fact...

Read it, it simply says that 3nm production is gonna happen inside 3 years .....

(translate:  7nm is a main node, 5nm will tend to get skipped, 3nm is a main node)

Back to top
 
« Last Edit: 12/20/18 at 21:22:08 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Pages: 1 ... 9 10 11 12 13 ... 31
Send Topic Print


« Home

 
« Home
SuzukiSavage.com
04/24/24 at 23:07:17



General CategoryThe Cafe › 2020 -- new Intel failures & successes


SuzukiSavage.com » Powered by YaBB 2.2!
YaBB © 2000-2007. All Rights Reserved.