Tuesday, June 30, 2015

EXCLUSIVE: When Intel will mass deliver 10 nm Cannon Fake technology ?

Intel was supposed to convert its fab in Kiryat Gat, Israel, to manufacture 10nm chips, but delayed the upgrade until December. Just recently, SemiWiki reported an even further delay, until 2016. Why not, if there is no real X86 competition from AMD.In shortly, lets tick once again.
99% BAD HARDWARE WEEK:  22nm Si FinFET announced in 2011, became mainstream in 14nm 2014. Thus, when we can expect 10 nm FinFET technology as mainstream ??

Provided 10 nm announcement to be in 2017, that will give us 7 nm mainstream in 2020 !! 

Google: The cleanest energy is only the energy you don’t use

Google: This time, we’re doing something we’ve never done before: we’ll be building on the grounds of the Widows Creek coal power plant in Jackson County, which has been scheduled for shutdown. Data centers need a lot of infrastructure to run 24/7, and there’s a lot of potential in redeveloping large industrial sites like former coal power plants. Decades of investment shouldn’t go to waste just because a site has closed; we can repurpose existing electric and other infrastructure to make sure our data centers are reliably serving our users around the world.

99% BAD HARDWARE WEEK: We too. Here is some 1000 tons of heavy pollutants it has delivered to our envirnoment up to now ! Plus 100 000 tons od sulphur dioxide and 10M tons od carbon dioxide each year !!

Saturday, June 27, 2015

TICK; TICK;TIC : Intel inherently unable to switch to 10 nm node

Intel will likely release a Core i series processor lineup in 2016 based on the Kaby Lake-series. This would be the 3rd generation processors that Intel fabs at 14nm.

99% BAD HARDWARE WEEK: Well, 10nm is far from REAL 10nm, namely it is only node name. And Intel is notoriously unable to reach even that. Low performance, low price. Only future Kaby worth mention is H -line with 256 MB L4 cache. Intel's bulk Si has hit Lgate problem. FinFET could solve it but introducing its own scaling problems.

The cost per gate for 28nm bulk HKMG CMOS will be much lower than FinFETs even in the fourth quarter of 2017. A similar pattern will occur for 20nm HKMG in 2018 or 2019 when depreciation costs decline.
The data indicates that FinFETs can be used for high-performance or ultra-dense designs but are not cost effective in mainstream semiconductors. There's no end in sight for this situation. Scaling to 10nm and 7nm nodes will entail additional wafer processing challenges for which the industry is not well prepared. Thus, watch your money or performance ! You can't get both at the same time. NEVER MORE !
So, Intel in the future needs one or more TICKs than used to. To make some decent profit returns on big new process investement . In situations when PC market where Intel's X86 dominates, shrinks year after year.

Friday, June 26, 2015


 This is a collection of static SSH keys (host and authentication) that have made their way into software and hardware products. This was inspired by the Little Black Box project, but focused primarily on SSH (as opposed SSL) keys.
Here is Wikibook on Open SSH

The Internet of what ?

Obama Administration Argues For Backdoors In Personal Electronics
99% BAD HARDWARE WEEK: Save the child ! Your Cisco appliances are already safe. An attacker could exploit this vulnerability by obtaining the SSH private key and using it to connect to any WSAv, ESAv, or SMAv. An exploit could allow the attacker to access the system with the privileges of the root user.
Major websites such as MSN.com and Hulu.com have been tracking people's online activities using powerful new methods that are almost impossible for computer users to detect, new research shows. Install https://disconnect.me/ , regain your online privacy and peace of mind. It's a security browser extension developed by ex-Googler that blocks Google, Facebook, other third party sites from tracking online activity and collecting sensitive personal data.

Wednesday, June 24, 2015

What The B-3 Bomber Should Be

99% BAD HARDWARE WEEK: Remember F35's lack of inter communications? Now, you know why. Waiting for B3 arrival. Say in 2018. Or later.

Monday, June 22, 2015

Intel’s 28-core Xeon ‘Skylake’ CPUs to support 6TB of DRAM, LGA-3467 ?

The “Purley” platform will be Intel’s biggest server platform advancement in many years, when the world’s largest chipmaker rolls it out in 2017. Intel’s “Purley” will be a highly-configurable platform designed for enterprise, cloud, HPC [high-performance computing], storage and network applications. Intel “Purley”-based supercomputers are expected to finally hit ExaFLOPS performance late this decade.

Intel Xeon “Skylake”: 28 cores, 6-channel DDR4 memory, Omni-Path fabric
Intel will release three different versions of Xeon processors for its “Purley” platform targeting different applications two years from now – “Skylake-EP”, “Skylake-EX” and “Skylake-F” – according to a report from CPU World. The new chips will feature up to 28 cores based on the “Skylake” micro-architecture with AVX512 instructions and Hyper-Transport technology, up to six DDR4 memory channels (up to two 2400MHz DIMMs per channel are supported, i.e., up to 768GB of DDR4 memory per socket without SMB), up to 48 PCI Express 3.0 lanes as well as two or three UPI channels per socket.

Intel Xeon “Skylake-EX” processors will be designed for high-performance and mission critical machines with two, four or even eight sockets, which means that they will feature up to three UPI links. The processors will introduce new RAS [reliability, availability, serviceability] features such as Instruction Retry (pipeline error protection for integers), Advanced Error Detection and Correction as well as Adaptive Dual Device Data Correction, to make next-gen high-end servers even more robust. According to previously released unofficial information, Intel’s forthcoming expandable processors will support four times higher memory capacity (compared to today’s chips) thanks to “Apache Pass” scalable memory buffer (SMB), which means up to 6144GB (over 6TB) per socket, or up to 24.576 GB of DDR4 RAM per 4S machine. The “Skylake-EX” chips will have TDP of up to 165W.

Xeon “Skylake” processors will use the new socket P0 and will feature flip-chip land-grid array packaging (FC-LGA) with up to 3467 contacts. The final amount of pins to be used is unclear today, but according to unofficial information, it will exceed 3000 balls.


Friday, June 19, 2015

Revealed: The Secret Gear Connecting Google’s Online Empire

 Three-and-a-half years ago, a strange computing device appeared at an office building in the tiny farmland town of Shelby, Iowa. Google runs a data center not far from Shelby, and apparently, someone had sent the switch (pic above) to the wrong place.

The current network design, which powers all of Google’s data centers, has a maximum capacity of 1.13 petabits per second. That’s more than 100 times as much as the first data-center network Google developed 10 years ago. The network is a hierarchical design with three tiers of switches, but they all use the same commodity chips. And it’s not controlled by standard protocols but by software that treats all the switches as one.

A decade ago, Google built its networks like everyone else did. It bought enormous “cluster switches” from companies like Cisco. Inside each data center, these served as the backbone of the network. “They were essentially mainframes,” PortLand held promise for supporting a plug-and-play large-scale, data center network. Emerging mega data centers hosting100,000+ compute nodes.

Google was at the forefront of the move to 10-gigabit Ethernet networks, and the technology is now starting to spread across the industry. Facebook just recently moved to 10-gigabit Ethernet inside its new data center in western North Carolina.
Well, stop guessing. Google uses 40 Gbps Ethernet.

99% BAD HARDWARE WEEK: Pluto switch, Jupiter network, Google seems has some Galaxy networking ambitions ?? Idea os software defined network seems simple, but traditionally, computer networks have not worked this way, forcing technicians to make individual changes to individual devices if they wanted to revamp or expand their networks. Until now, Google has kept quiet about its work. And to a certain extent, it continues to do so. Vahdat will not discuss the particulars of Google’s latest technologies, and the company is not open sourcing any of its work, something Microsoft has done, according to Russinovich. But more than anyone else, the company shows where the world of networking is moving.

Wednesday, June 17, 2015

Hackers hacked

I know how deep we are in our enemies’ networks without them having any idea that we’re there.
99% BAD HARDWARE WEEK: I’m worried that our networks are penetrated just as deeply
. Nice excuse to use Snowden for misuse :)

Tuesday, June 16, 2015

HP Destroys a Dream Computer to Save It

Memristors were barely in sight. A prototype of the new computer could be out next year, Mr. Fink said, based on more conventional DRAM memory. Instead of a special-purpose computer operating system, he said, the Machine will initially have a version of the popular Linux system. 
The major reason for the change is that HP has no idea when, exactly, it will be able to produce memristors in commercial quantities.

Sunday, June 14, 2015

China and Russia cracked British code ?

Code is public method encrypted !!, so british agents in those countries have to be withdrawn.
99% BAD HARDWARE WEEK: What to withdraw ordinary users of public encryption methods.

Friday, June 12, 2015

Falling airplane after missing files

 The troop and cargo carrier crashed outside Seville, Spain, on May 9 after three of its huge turboprop engines froze just minutes into a routine, pre-delivery test flight, killing four of its six crew members.
Investigations are at an early stage but the key scenario being examined is that the data -- known as "torque calibration parameters" -- was accidentally wiped on three engines as the engine software was being installed at Airbus facilities.

Under the A400M's design, the first warning pilots would receive of the engine data problem would be when the plane was 400 feet (120 meters) in the air, according to a safety document seen by Reuters. On the ground, there is no cockpit alert. 
"Nobody imagined a problem like this could happen to three engines," a person familiar with the 12-year-old project said.

99% BAD HARDWARE WEEK: Nobody imagined ? Well, check recent Dreamline's software bug.

Thursday, June 11, 2015

Win 10 will not help Intel

Market watchers say the PC market's decline is continuing, and Intel's spending has been 6-10% ahead of its model for more than a year.
Intel would not see the financial benefits of layoffs for over a year, analyst Jon Peddie told EE Times. He speculated that layoffs of 5,000 positions could translate into $1 billion in cost savings.
As a result of the trip, Deutsche Bank now estimates PC sales will fall 9 percent in 2015 compared to 2014. That’s steeper than the 6% annual drop currently predicted by market watchers at International Data Corp.

Wednesday, June 10, 2015

Kaspersky uncovers Duqu 2.0 after malware attacks its systems

Cybersecurity firm Kaspersky Lab finds three hotels that hosted Iran talks were targeted by a virus believed used by Israeli spies. Kaspersky believes that the hackers were trying to gather intelligence about nation-state attacks that the company is investigating.
99% BAD HARDWARE WEEK:  Certificates authorities be aware, you are the primary target. Beside, "For a security company, one of the most difficult things is to admit falling victim to a malware attack," Kaspersky researchers wrote in their report.

Thursday, June 04, 2015

IMPV8 : Sky Lake and Cannon Lake power supply standard

99% BAD HARDWARE WEEK: Please note Skylake VCCIA input:  1,15V , 78,42A.
And VCCGT power input: 1,06V , 75A

You can see too that dual main voltage input is essential for Sky Lake UNIFIED Desktop and Notebook platforms power supply. Enjoy.

Tuesday, June 02, 2015

Semiconductors industry losing steam ! And Intel 10nm node.

Now only marketing leaks more than the semiconductors industry with each next node.
While Intel never mentioned the part by name in the earnings call they did talk about the 10nm process. Krzanich indicated that the company still hasn’t established the timing for the process. This makes a 2016 debut very unlikely. If the new process proves to be as problematic as 14nm then we’re quite possibly looking at another Haswell like situation with Intel refreshing Skylake parts in 2016. It’s too early to tell for sure but it’s evident that the whole semiconductor manufacturing industry is slowing down, not just Intel.
99% BAD HARDWARE WEEK: No 10nm in 2016 !

This page is powered by Blogger. Isn't yours?