https://www.techspot.com/news/77313-do-need-re-review-core-i9-9900k.html

Do we all need to re-review the Core i9-9900K?




Read it if you want to, but the gist is that Intel has lied to everybody again and has gotten caught at it again.

EXTREME THERMAL THROTTLING is the overt sin and the final real production 2019 Core i9-9900K has got itself a really bad case of of overheating and gross throttling when run at Intel "suggestions".

This is the same top of the line chipset that got Macbooks using last years chipset put into freezers which was what was finally required to finally get the advertised performance numbers.

Lots of egg on the faces of the early reviewers that got sucked into this one yet again --- and they are talking on line about that particular situation quite a bit right now too.  

If you can't trust the Intel early evaluation units not to be cherry picked & tweeked to hide some really big known problems, how can you do these early reviews at all?  

Reviewers are using the shill word again, aimed at each other no less.   Honest reviewers are being used as inadvertent shills by Intel ...... and they do not like that at all, not one little bit.

Or, more correctly they do no care to be called out by their peers so publicly.


...... Intel, if I am your shill, you need to be PAYING me you stupid arseholes .......    



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



https://www.techspot.com/review/1744-core-i9-9900k-round-two/

Techspot leads the pack in redoing their "early review" that were done initially using Intel early review units specially prepped for reviewers.  

Today we're revisiting our original Core i9-9900K review and updating it with 95 watt TDP limited results, basically results based on the official Intel specification and not at Intel's "suggested usage".  For better context about this please read our opinion article from earlier this week titled “Do We Need to Re-Review the Core i9-9900K?”.

The short version of this is that motherboard makers are currently getting blamed for running the 9900K out of spec, when in reality we strongly believe it’s Intel who’s cheating on their own spec and pushing all their board partners to run the 9900K at the default clock multiplier table, rather than at the official power spec.

Whatever the case, out of the box the 9900K isn’t running at the Intel spec, it’s essentially being overclocked by Intel and this has caused power and thermal results to go through the roof. So in today’s re-test we'll be showing how the Core i9-9900K performs when adhering to the Intel specification and comparing that data to the current out of the box experience.

It doesn’t really matter where you stand on this, having a resource that shows how these configurations compare under the same test conditions is useful information in our opinion. For the unlimited testing the MSI MEG Z390 Godlike has been used and for the 95-watt limited testing I used the Asus ROG Maximus XI Hero, loaded up the "xtreme" memory profile and opted to use the Intel settings which enforces the 95-watt TDP. So let’s get into the results...

Benchmarks

First up let’s look at the Cinebench R15 multi-threaded scores. Previously we found the 9900K breaking the 2000 pt barrier, however with the TDP limit in place the score is reduced by 14% down to 1763 pts and that places it roughly on par with the Core i7-7820X and crucial, meant it was a few percent slower than the 2700X. Already you might be getting a sense of why Intel was happy for board partners to run out of spec.



htps://static.techspot.com/articles-info/1744/bench/Blender_Short.png" alt="" title="" border="0" />








Let me put a cap on it for you ----- AMD and Intel are really about equal, all things considered when they are clocked the same way (either both clocked normally or both overclocked).   The only thing Intel does better is to CHEAT better on setting up early review units, making misleading "technical suggestions" for the tests and on writing the base benchmarks themselves .....    

You certainly should not be overclocking the Intel processor and not overclocking the AMD processor and calling it a proper technical benchmark review.

https://www.gsmarena.com/kirin_990_to_be_built_by_tsmc_7nm_process-news-34141...



Huawei introduced the Kirin 980 chipset with AI features, and 1.4 Gbps download speeds in September. The platform made its way to the Mate 20 series and is likely to appear in the next-gen Huawei P flagships. Although the SoC is still new, the Chinese company is already working on the follow-up, and according to industry sources, preparations for Kirin 990 has already begun.

Reports from China claim TSMC will build the chipset on the 7nm lithography process. Industry insiders claim the subsidiary manufacturer HiSilicon has invested around CNY200 million in research and development, which translates to over $28 million/€25 million.

The Kirin 980 was the first chipset built on the TSMC 7nm lithography process with the first Cortex-A76 cores. While Kirin 990 will have a similar architecture, including the number of transistors and power consumption rates, it will have something to be advertised as a first - the Balong 5000 modem, certified for 5G speeds. Production should begin in the first quarter of 2019 and that’s when we expect more detailed info to emerge


OK, China runs a slightly different cell tower system, one built by Huawei using their own standards (which are not compatible with USA cell tower system standards, but are not really strongly less capable).

Given that differences exist, the fact that Huawei and company may have ripped off the Qualcomm 5G system technology and yet still have actually beaten the USA carriers to widespread 5G in their own China market is something that is worth noting.

Soon, it will not be a question of ripping off anybody else, Huawei will soon have achieved their own technical innovations and be moving on separately, making their very own technical progress.

....... change, she keeps right on coming, don't she ?

https://semiaccurate.com/2018/11/09/amds-rome-is-indeed-a-monster/

AMD’s Rome is indeed a monster
Some details, some performance claims, and more



So in the end the numbers for Rome are still a bit unclear. The core count doubled, the FP resources have doubled, the internal data paths have doubled, and lots of other detail changes. The memory channels have stayed the same though and so for anything but trivial benchmarks, you are unlikely to get a 2x increase in performance. But you will get well above 70% more performance out of a Rome which puts it significantly ahead of Intel’s Ice Lake architecture which comes more than a year later. Better yet AMD has PCIe4 on Rome, something Intel won’t have until mid-2020.

AMD is on track to deliver Rome in Q2 of 2019. When SemiAccurate said it was a monster, we weren’t kidding, Intel has nothing to answer this with and won’t until 2022. By then AMD will have two more generations out, assuming both sides execute their roadmaps perfectly.

Until then, Intel has no chance in servers and they know it.


It is fitting, that Charlie Demerjian is back to say clearly he was not joking, he wasn't trolling, that his internal sources are accurate and that all the Intel shills who attacked him in blogs really did not earn their money that Intel pays them to be Intel's attack Dobermans.

Charlie is flat out calling for the Fat Lady to come out to sing as old style Intel is "game over" in servers inside 4 months time and Intel has no chance to recover any time in the foreseeable future if they stick to any of the published Intel product roadmaps.   Charlie points out that Intel doesn't even supposedly make any move to even try to compete against what is in the picture above according to their own published timeline until 4 years have passed (and by that time AMD will be on 5nm lithography chiplets and actually working on 3nm stuff).

I have to disagree with you, Charlie, Intel has the mega bucks to simply design and create ARM DyamIQ chiplets of their very own chiplet design ASAP and to have TSMC to run them for Intel at 7nm and to do the exact same thing AMD has done.   They have the ARM design license to do so already (yep, they do).    And they have full access to the ARM Holdings base chiplet design as well .....

Intel has shown the ability to MOVE VERY QUICKLY when the pain gets bad enough, and being dropped into an active volcano will get the big lizard to moving along right promptly and at a pretty durn good rate of speed too .......

Intel has one (1) each of the most modern ASML 7-5-3nm lithography cells already installed in their R&D facility.   Intel could get ASML to come in and set one (1) full production line for them so they could make their own design of chiplets in house at Intel.    One full production line can make a whale of a lot of chiplets when an ASML full production line is run 24 hours a day 7 days a week.  

That same chiplet could do double duty as their initial Intel consumer duty chiplet if they were smart about it.    AMD was smart enough to do that (and they are very busy right now doing just that).


BTW, Intel has announced that all new processors will be based off their new form of chiplet design system, that Cascade Lake will be a chiplet design with 2-4 chiplets.   So Charlie was right, Intel had indeed given up on "Intel 10nm" completely and is seeking out a new method of chipbuilding entirely.

 FYI this:

This time I bet they pay attention to ALL the comments, thought ......

is unreadable on anything but a larger desktop screen.  I never tried to read your posts until today and that's some small red text to see on a tablet or phone.

Eegore wrote on 11/13/18 at 08:09:53:




You custom edited in a 6 font size in your response, which is indeed unreadable.  

The original was a 9 font size, which is readable on just about all full sized monitors.

A little more care when responding is indicated least you mis-apply things accidentally.

Eegore wrote on 11/13/18 at 15:33:27:



Since you are quoting, the correct size will come right along for the ride, unless you changed it on purpose to something smaller.

The original font size.  Small red font sample.

Use the quote function, then simply delete the portions you don't want to respond to

If you quote, just don't play games with what you quote and the system will accurately transfer color and font 1-for-1 with the original post when you quote it.

Playing little games with the quoted material in an effort to criticize something is disingenuous at best.

The Cafe is not the Tall Table.    Asking disingenuous questions, making unsupported "statements of fact",  asking for "help" repeatedly on topics you have no intention of doing are all symptoms of the Troll Disease, something we trust you are not infected with.

At the very mildest, you are leaking over from Tall Table behaviors, badly.

https://www.neowin.net/news/microsoft-details-more-issues-with-the-windows-10...

https://www.howtogeek.com/fyi/windows-10s-october-2018-update-breaks-mapped-n...

https://www.tomshardware.com/news/intel-14nm-cpu-shortage-motherboards-hurt,3...





Synopisis:  The flow of bugs and flaws continue with MS's error ridden Fall Upgrade, some of the new flaws coming out of the woodwork now will affect the all important Enterprise Users and Microsoft has no fix right now for these issues (fix is promised for next year sometimes during a later main system upgrade --- this means it is a very basic function bug that requires a fairly large rewrite of a chunk of network software).   Recommendation being passed around right now is NOT TO INSTALL THE FALL UPGRADE for Enterprise Users.

Anger is still prevalent, just consider the opening paragraph from this fairly opportunistic Linux "replace your Windows" distro.  

https://betanews.com/2018/11/15/deepin-15-8-linux-distribution-available-for-...

As more and more people wake up to the fact that Windows 10 is a giant turd lately, computer users are exploring alternatives, such as Linux-based operating systems. First impressions can be everything, so when searching for a distribution, it is important that beginners aren't scared off by bewildering installers or ugly and confusing interfaces.

https://www.makeuseof.com/tag/linux-finally-good-enough-replace-windows/

Yes, MS has a large image problem and right now everybody and his brother are highlighting every new bug MS puts out as MS does their standard "fumble through it 3-4 times before it jells" software development routine.

Mark Shuttleworth at Ubuntu has been contacted by IBM/Red Hat to see if some sort of a technology consolidation is conceivable/possible between the main Linux groups.   Mark has said previously that cutting down on multiple variants of key systems was worth consideration "for technical consolidation purposes" and as long as he and his team remains firmly in charge of the Ubuntu distro Mark does see a real need to have a unified commercial Linux technology base going forward.   Between the 3 companies IBM/Red Hat and Ubuntu do make up the vast majority of Commercial Linux suppliers.

All the Linux distros are feeling threatened by Google who is working so diligently on Fuchsia OS and by MS who is now saying they are going to totally redo Windows inside the next 2 year time period.  MS will be basing this Windows redo in part on FOSS Linux because of MS's need to rectify some of the nagging errors in the Windows core code base (and their Windows programmers apparent inability to stop making mistake after mistake after mistake).

MS has completely changed their stance on Linux inside the last 2 years, joining the core Linux Federation as a main supporting member and MS is beginning to utilize Linux code freely inside Windows where it fits with customer use demands.  

In the last quarter, MS has released Clear Linux, a pay me $10 MS modified Linux distro that is intended to bridge some of the gaps between traditional MS and Linux worlds.   MS has also released complete chunks of very old MS code into the FOSS world, including all the old versions of MS DOS but stopping short of XP, which would actually be of some use to the REACTOS project.


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


https://www.tomshardware.com/news/intel-14nm-cpu-shortage-motherboards-hurt,3...

OBTW, Intel is now shortfalling 2 million chips a month now -- that is a very LARGE shortfall in chip production which will be reflected in a larger increase in AMD machines shipping this Christmas season, with this shortfall situation continuing and it will be carrying forward into Q1-Q3 2019.


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


https://www.theverge.com/2018/11/16/18098230/microsoft-windows-on-arm-64-bit-...

Microsoft opens the door to better Windows on ARM apps

Native ARM 64-bit Windows app support has arrived at last.

Microsoft is removing one of the big limitations of Windows on ARM this week by allowing developers to create 64-bit ARM (ARM64) apps. Developers will be able to recompile existing win32 or Universal Windows Apps to run natively on Windows 10 on ARM hardware. That means 64-bit app performance should get a lot better, as long as developers take the time to recompile.

Microsoft is now relying on developers to use its tools to improve its Windows on ARM efforts. That’s a situation the software giant has found itself in before, relying on developers to create Universal Windows Apps for Windows 8, Windows 10, and Windows Phone apps for a variety of new touch-based hardware. It’s hard to say whether 64-bit app support will really help move Windows on ARM into the mainstream, but it’s certainly laying the ground work for a bigger push by Microsoft.

WINDOWS ON ARM HAS MADE SLOW AND STEADY PROGRESS
Windows on ARM has been progressing steadily over the past year, but performance and app compatibility are still big issues. Windows 10 includes an emulation layer for x86 apps running on ARM processors, and it’s the way you’ll experience most desktop apps on one of these machines at the moment. Emulation is never ideal, so if developers recompile their apps to run natively on Windows on ARM then we’ll start to see just how well these laptops can compare to traditional Intel-powered devices.

As Intel continues to struggle with its 10nm processors, competition from ARM processors has closed the performance gap significantly. Apple compares its latest iPad Pro gaming performance to an Xbox One S console, and benchmarks show it’s competitive at CPU tasks. ARM is also promising laptop-level performance from its Cortex-A76 chip design in 2019. The chip design company has been claiming that ARM processors next year will compete with Intel’s Kaby Lake range on laptops.

....... uh, guys, next year starts in like 6 weeks from now ????