AMD bulldozer has achieved insane 8.58Ghz using liquid nitrogen, hitting a world record. But 5.6Ghz is the best we’ve reached with Core i7. The work is done by professional overclocker Hicookie who overclocked his Core i7 3930K to 5.6GHz.

His setup was actually amazing, he used aforementioned Dippin’ Dots refrigerant. Hicookie used an Gigabyte X79 motherboard for making it possible. Gigabyte X79 is first such board to reach a multiplier of 57. Check out the video after a break:

don’t forget, we warned you about the poor video quality

We write latest and greatest in Tech Guides, Apple, iPhone, Tablets, Android,  Open Source, Latest in Tech, subscribe to us@geeknizeron Twitter OR on Facebook Fanpage, Google+:

“The idea is that at a fixed computing load, the amount of battery you need will fall by a factor of two every year and a half” which will shape the future of mobile computing in all forms.

Koomey’s law is more consumer centric, which says that Energy efficiency and thus, battery life is becoming more of a concern than the raw power of new devices thanks to the continuing rise of smartphones, tablets and notebooks. Till very recently, both Moore’s law and Koomey’s law had been valid, but Moore’s law had been invalidated according to several scientists.

Typically, performance improvement can be achieved by either overclocking the chipset ( by delivering more power to the transistors) thus generating more heat, or by simply adding more transistors to meet the new performance requirements. Both the problems of performance and energy efficiency have been answered thus far with a single solution: reducing the size of transistors. This allows for more of them to be placed on a chipset and reduces each transistor’s energy and heat mitigation requirements. e.g. a 45nm chip is more power hungry than a 22nm chip, and so on; Size difference is obvious.

Energy efficiency of transistors is a better concept and benchmark to measure improvements over years.

“This is a fundamental characteristic of information technology that uses electrons for switching… It’s not just a function of the components on a chip”. It stands to reason then, that future technologies will likely be based on this same premise, and thus conform to the theory.

Physicist Richard Feynman tells us Koomey’s theory would hold true for decades to come. In 1985, Richard analyzed the electrical requirements of computers and estimated that the technology could theoretically improve by a factor of one hundred billion (100,000,000,000) beforeit reaches the limit. So far,  we’ve only seen an improvement of around a factor of forty thousand (40,000). Whatever may happen, both Moore and Koomey believe that a technology that efficiently delivered boundless processing power would be magical indeed. Arthur C. Clarke’s third law “Any sufficiently advanced technology is indistinguishable from magic”.

We write latest and greatest in Tech Guides, Apple, iPhone, Tablets, Android,  Open Source, Latest in Tech, subscribe to us@geeknizer on Twitter OR on Facebook Fanpage:

As Intel calims:

With these upgrades, customers will enjoy:

• Better performance for photo, video, and music tasks.
• Better responsiveness for everyday PC activities.
• Better responsiveness for PC productivity at work or at home.

Wait a minute, how is that even possible?

Truth be told, the hardware was already capable of doing more but looks like it was not unlocked out of the box. Intel had disabled HyperThreading technology, more L3 cache for applications on some of the Core i3 processors, and Intel Pentiums, which will be unlocked via this software upgrade. Intel is selling these as $50 upgrade cards.

Pentium 2.8GHz G6951 is sold with 3MB of L3 cache and has HyperThreading disabled. By purchasing and activating the $50 upgrade card, a full 4MB of L3 cache is enabled along with HyperThreading.

Compatible processors:

Intel® Core™ i3-2102 Processor
Intel® Pentium® G622 Processor
Intel® Core™ i3-2312M Processor
Intel® Pentium® G6951 Processor

As with all forms of DRM and software unlocks, it’s only a matter of time before modders/hackers find a way to circumvent Intel’s $50 upgrade card.

We write latest and greatest in Tech Guides, Apple, iPhone, Tablets, Android,  Open Source, Latest in Tech, subscribe to us@geeknizer on Twitter OR on Facebook Fanpage:

The company believes that Intel’s x86 architecture is built to work for generic applications that are mostly limited to single cores and few threads, they cannot scale well beyond to levels where Intel cannot. Tilera Gx family targets high-throughput applications.

Tile Gx family has 3 flavors 36-core Gx3036, the 64-core GX3064 and the 100-core Gx3100. All of them target high performance multi-core computing designed for high-end web servers & applicatons including vast databases and heavy frameworks.

If the company Tilera is to be believed, the Gx family is at least 10 times power efficient, per-watt. With 100 cores on a single die, the processor consumes less than 48 watts, which is way lower than Sandybridge. Each core rated at less than 0.5 watt is the among the most efficeient processors ever designed for servers.

Tile Gx Family Specs:

• An array of 16 to 100 general-purpose processor cores (tiles)
• 64-bit VLIW processors with a 64-bit instruction bundle
• Three-way pipeline with up to three instructions per cycle
• 32K L1i cache, 32K L1d cache, 256K L2 cache per tile
• Up to 750 billion operations per second (BOPS)
• Up to 200 Tbps of on-chip mesh interconnect
• Over 500 Gbps memory bandwidth with four 64-bit DDR3 controllers

Sandy bridge consumes upto 95W, and Xeon goes upto 130W.

Tilera goes further in criticizing Intel’s upcoming architecture “Knights Ferry n 2012”. They argue that

“an x86 is an x86, whether it’s Atom cores, Pentium cores — the overhead that you’re carrying is the same. You need to carry the overhead to be binary compatible,” Bishara said. “What [Intel] do, even with Atom, is lower the frequency to get lower power and to get more cores in the same chip.”

Tilera’s Gx3036 would become available around July, followed by Gx3064 and Gx3100 in Q1 2012 and stratton 28nm chip in 2012 going upto 200 cores. With the big plans on its line, its already in talks with big cloud players like Amazon, Google, Microsoft.

Tilera is non-x86, but Dev Toolchain would be available

Gx chips are compatible with 2.6.36 kernel release of Linux showing love for CentOS. The applications can be written in any of the popular languages: C and C++, Java, PHP, Perl and Python, which are implemented via a GCC compiler that has been ported to Tilera’s architecture.

Tilera’s Architecture – The secret recipe

The DNA of the multicore super-performance is the way it communicated among multiple-cores. Gx chips pass information between cores via a mesh network, which distributes the cache throughout the processor and allows for multiple connections between individual cores. Applications can access the on-chip network, and can fully use it to communicate between cores, the result is a superior throughput in the IO among the chips, and better cache management as compared to sandybridge.

We do believe that Tilera is bringing some breakthrough improvements in the CPU architecture especially on the server sides. The ultimate question that still remains unanswered is “Will the customers (big cloud players) port apps to a new architecture?, would it be easy enough?”

We write latest and greatest in Tech Guides, Apple, iPhone, Tablets, Android,  Open Source, Latest in Tech, subscribe to us@geeknizeron Twitter OR on Facebook Fanpage:

Today’s announcement brought a new promise to MeeGo tablets. The announcement came from Tencent and Intel launching the new renewed vigour and Tencents dedication to bring games to MeeGo tablets.

China is a perfect breeding ground for MeeGo because of their new grown hatred for Android. The Chinese journalist confirm that the lack of marketplace support leaves a bad taste for consumers.

Meego Fair is the new Tablet hardware based on Oaktrail. Its a 10.1 inch tablet tat is just 1cm thick (10mm) powered by the Z670 1.50GHz processor and 1GB of RAM and 16GB and up SSD drive. It has all the stuff you would fin in a Honeycomb or iPad tablet : Micro SD, wifi, accelerometer, etc.

The best part of this Tablet is the all new UI that hates vertical stuff, and sports horizontal scroll to navigate among app widgets. Checkout the UI, you will be amazed with the number of improvements:

We write latest and greatest in Tech Guides, Apple, iPhone, Tablets, Android,  Open Source, Latest in Tech, subscribe to us@taranfx on Twitter OR on Facebook Fanpage:

If you know about HDMI, USB 3.0, you might actually feel the inferiority of those technologies, as LightPeak takes over.

LightPeak is Intel proprietary Optical cable interface to connect peripheral bus giving throughputs upto 10Gigabit/s and have a potential of extending to 100Gb/s over a period of next few years.  At 10Gb/s, you could transfer a full-length Blu-Ray movie in less than 30 seconds. Optical technology also allows for smaller connectors and longer, thinner, and more flexible cables than currently possible. Light Peak also has the ability to run multiple protocols simultaneously over a single cable, enabling the technology to connect devices such as peripherals, displays, disk drives, docking stations, and more.

A typical computer system involves a large number of interfaces namely SATA, SCSI, USB, Firewire, PCI Express – each one dedicated to its scope with limited bandwidth capabilities. LightPeak would clear of this mess by providing one unified interface to connect almost anything – whether it’s a Display panel/Monitor or a external/internal harddrive, they all work on LightPeak, and they work much faster than before.

Apart from unifying communications, it would make interfaces cheaper Light Peak, unlike most fiber ICs, was designed to be lithographically created directly on a CMOS IC, which can reduce size and manufacturing costs. However, the LightPeak will not replace existing HDMI, DisplayPort technologies immediately, it would be gradual.

The Light Peak cable contains a pair of optical fibers (each 62.5 micron wide) that are used for upstream and downstream traffic. This means that Light Peak offers a maximum of 10 Gbit/s in each direction at the same time i.e. 10Gbps Full duplex. The maximum length of the cable can run upto 100meters, requiring repeaters beyond that.

Since LightPeak cables has decent max length, they have also been looked as 10Gigabit Ethernet killers. 10Gigabit ethernet is way more costly, but its hard to actually transition all future networks to a new physical layer.

Apple’s Thunderbolt Technology

Apple is doing a rebranding for the LightPeak technology to Thunderbolt High Speed I/O. The rumor gears are really starting to churn with regard Apple’s upcoming MacBook Pro refresh and now we have a purported leak of the upcoming spec sheet for the new 13-inch model. The Thunderbolt port would look much like the Mini DisplayPort but has a tidy little lightning logo next to it.

Update: Apple has announced new MacBook Pros with Thunderbolt technology.

“The new MacBook Pro brings next generation dual and quad Core processors, high performance graphics, Thunderbolt technology and FaceTime HD to the great design loved by our pro customers,” said Philip Schiller, Apple’s senior vice president of Worldwide Product Marketing. “Thunderbolt is a revolutionary new I/O technology that delivers an amazing 10 gigabits per second and can support every important I/O standard which is ideal for the new MacBook Pro.”

Imagine accessing multiple streams of uncompressed HD video — from your notebook — at speeds that let you edit an HD feature film in real time. That’s how Thunderbolt technology will connect the next generation of high-performance peripherals to the next generation of computers — starting with MacBook Pro. Ultrafast and ultraflexible, the Thunderbolt pipeline is more than 12 times faster than FireWire 800 and up to 20 times faster than USB 2.0, and it offers unprecedented expansion capabilities. It changes what you can do on a notebook.

The Thunderbolt port will give you plug-and-play performance with a whole new world of Thunderbolt peripherals, as well as with the Apple LED Cinema Display and other Mini DisplayPort peripherals. You can daisy-chain as many as six devices, including your display. And with support for video and eight-channel audio, it’s easy to connect HDMI-compatible devices — like your TV and home stereo — using the HDMI adapter you already have. Current VGA, DVI, and DisplayPort adapters are also supported.

For latest Hardware, iPhone, android, Tech news @taranfx on Twitter and Facebook Fanpage you can also subscribe to use below:

The new Core i3, i5, and i7 processors, will come in 29 new flavors with Integrated graphics options. The new Core processor range is an upgrade from existing Westmere 32nm processor, essentially falling into Intel’s “Tock” development phase in which they improve the performance, features, efficiency on the existing Chip fabrication technique i.e. 32nm.

Whats New in Core 2011 Processors

The new family of processors are all based on Intel’s 32nm microarchitecture and are the first to put the processor, memory controller, and graphics on the same die. This means you will have a smaller package and parts get to take advantage of each other better, e.g. by dynamically clocking both the CPU cores and graphics to match whatever workload you throw at it, and giving them up to 1MB of shared cache. Speaking of those graphics, while they may still not be on par with a discrete video card, they’re more powerful than ever before. According to Intel, the new HD 2000 and 3000 processor graphics provide 2x the performance of Capella-based systems.

Intel has improved its Turbo Boost and Hyper-threading technologies such that the new chips enable higher levels of CPU performance as well — up to a 60 percent improvement with quad-core mobile CPUs.

The performance is not at the cost of efficiency. The new integrated chips reduce power consumption by using a smaller integrated package and can actually switch off components its not using , to improve battery life.

The processor details

Intel is this time looking to satisfy Overclockers with a set of specially-marked overclockable ‘K’ series processors which lets them individually set the clockspeed ratios for each individual core, provided they run on a premium P67 Express motherboard.

New Features in Core 2011 Processors:

Geek part aside, these are the few features every user can understand:

• Intel Quick Sync Video -  Intel’s promising way faster video transcoding with it’s new integrated Quick Sync feature, which does encoding in hardware — it says it will be 17x faster than older generations of integrated graphics. Intel’s partnered with media software companies like CyberLink, Corel, and ArcSoft to enable this hardware-accelerated H.264 and MPEG-2 video conversion.

• Intel InTru 3D / Clear Video HD – 3D Blu-ray playback over HDMI 1.4? Not a problem for Intel’s Core 2011 platform. This doesn’t mean Intel’s providing the 3D technology for laptops or desktops, but it’s promising that you can play stereoscopic 3D and HD content on your TV using HDMI.

• WiDi 2.0 – Our biggest complaints about the original WiDi was the lack of support for 1080p. Well, Intel’s finally enabled streaming of full HD with its latest processors. There’s still lag and you will need to pick up a new TV receiver, most likely a new one from Netgear, to take advantage of the new full HD abilities. We’ve got more details on this and Intel’s new Insider video content service here.

Additional Details:

Related: BIOS death Nears with Sandybridge, UEFI is the Future

We write on Hardware, Open Source, Programming, Tech news, gadgets,  Get them @taranfx on Twitter or Subscribe to us below:

He said, current Antivirus efforts focus on building up a library of known threats against which they protect a user, but Intel would love to move to a world where only code from known and trusted parties runs on x86 systems. Run only the code that has been blessed as safe by some trusted authority.

Alright, sounds good, but is there a downside?

This sounds something like Apple’s App Store where each and every developed gets approved by an authority. Regardless of what you think of the idea, its success would have at least two unmitigated upsides: 1) everyone will get vPro by default , and 2) it would put every security company, out of business.

Inttel plans to have two roles as a security provider: a component provider role, and an end-to-end platform/software/services provider role.  Its possible that Intel could build its own secure app store ecosystem, where developers send code to McAfee for approval and distribution. In this model, McAfee would essentially act as the “Apple” for everyone making, say, MeeGo apps.

But, the truth is,  walled garden approach sounds very attractive for the desktop, and they’ll probably be rejected outright by many Linux and open-source users. So what looks more doable is that Intel would set up a number of trusted signing authorities for x86 code, and developers could approach any one of them to get their code signed for distribution. Something very similar to how https web pages work on the web. This distributed approach seems to work well enough online, and I wonder why it can’t work offline on PCs too.

Security would be embedded, not a choice.

We write latest and greatest in Google, Firefox, Chrome, Tech Guides, Apple, iPhone, Tablets, Android,Open Source, Latest in Tech, subscribe to us @taranfx on Twitter OR:

Intel, the biggest chip-hardware player, is buying software/security maker McAfee for $7.7 billion, a huge 60% premium  over yesterday’s market price.

Acquisitions like this one are Really rare in the Tech industry, here’s why:

• Intel bought a company which is almost as big as itself.
• A core Hardware company buying a core security software company.

However, Intel believes that security has become an integral part of computing, which makes it their business, too.

“With the rapid expansion of growth across a vast array of Internet-connected devices, more and more of the elements of our lives have moved online. In the past, energy-efficient performance and connectivity have defined computing requirements. Looking forward, security will join those as a third pillar of what people demand from all computing experiences,” said Paul Otellini, Intel president and CEO.

McAfee will operate as a wholly-owned subsidiary, and it will be reporting to Intel’s Software and Services Group.

The full press release is here.

Here’s the rationale for Intel:

• Acquisition enables a combination of security software and hardware from one company to ultimately better protect consumers, corporations and governments as billions of devices – and the server and cloud networks that manage them – go online.
• Intel elevates focus on security on par with energy-efficient performance and connectivity. The acquisition augments Intel’s mobile wireless strategy, helping to better assure customer and consumer security concerns as these billions of devices connect.
• Intel has made a number of software-related acquisitions of leaders in their respective industries that also rely on great silicon, including Wind River, Havok and now McAfee.

We write latest and greatest in Google, Firefox, Chrome, Tech Guides, Apple, iPhone, Tablets, Android,Open Source, Latest in Tech, subscribe to us @taranfx on Twitter OR:

The Light Peak optical interconnect can transfer data at 10Gbits/sec in both directions, and is touted as an all-in-one replacement for USB, DisplayPort and HDMI and all others in competition. At such speeds, You can transfer a Full HD 1080p BluRay movie in a second!

Intel has demonstrated working Light Peak models twice this year, as the technology moves ever closer to launch. But today it pulled a surprise by announcing an even faster potential successor to Light Peak.

The new interface uses an indium phosphide hybrid laser inside the controller chip. By encoding data at 12.5Gbits/sec across four laser beams of differing wavelengths, the connector yields a total bandwidth of 50Gbits/sec, five times that offered by Light Peak.

“50Gbits/sec is just the beginning,” says Intel, as they plan to scale the interface up to 1Tbit/sec in the coming future.

The new interface should also be cheaper to produce than original Light Peak concept, thanks to its simple integrated design. With low-cost optics, in the future everything can be connected by fibre.

Optical data is seeing its first practical application in Light Peak, which tops out at just 10Gbps and is intended primarily to speed up peripherals in the coming 2-3 years.

We write about Google, Twitter, Open Source, Programming, Web, Apple, iPhone,Android and latest in Tech @taranfx on Twitter or by subscribing below:

Intel is building the car at its research headquarters which would record Car’s dynamic characteristics like speed, steering and braking along with video footage from inside and outside the vehicle and then like it opr not, will send the data to police and insurance companies in the event of an accident making it easier to determine the cause of car crashes and identify the person responsible.

Apart from the objectional side, Intel is building the technology that will transform cars into smart vehicles that use AI to detect dangers on the road their criticality and even Override the Driver in case of Risky driving. The cars will also be able to track the location of surrounding vehicles and alert drivers if they get too close or try to change lanes when another vehicle is in their blind spot. Using the LED rear and head lights found in most modern cars, the car is able to track the location of other vehicles and display them on a satellite navigation map.

The concept is not new, its been there for a decade, but Intel could be the first one to make it commercial by developing cars that are permanently connected to the internet and vehicles talk to each other over wireless connections thereby smelling dangers and take appropriate actions on both parties.

Taking it further, camera systems can recogonise street signs, consider the situation and take a smart move e.g. it can take over control of a car if the motorist tries to drive the wrong way up a one-way street, or tries to drive fast on a hilly road that has lower speed limit. If you think the drivers are the sole beneficiaries, you are  probably wrong. Onboard sensors will also be able to detect pot holes in the road and report their location to road maintenance authorities as the car is moving.

In addition, Motorists will also be able to use their mobile phone or computer lock and unlock their car remotely, turn on the alarm and even start the engine to warm it up in the morning.

When ?

No doubt, we’ve seen so many of such projects failing in the past. Will this one work? If yes, How soon can we see them on roads?

Intel has been in discussions with car manufacturers about putting the technology into new vehicles. Justin Ratner, the director of Intel Laboratories and chief technology officer, said:

“We are looking at a whole range of enhancements that will improve the driving experience, safety and security of vehicles.The intelligent vehicle is what we are talking about here. Once a car is connected, more or less on a continuous basis, all sorts of interesting possibilities present themselves. We have talked to highway maintenance departments about using sensors that are already in cars to report the GPS coordinates for pot holes in the road to the maintenance department.”

“Insurance companies are always looking at new in car technology. A system like this could certainly help speed up the process of determining the cause and responsibility of an accident. Any system would have to not increase the cost of repairing vehicles though, to ensure that motorists see a fall in their insurance bills.”

We write about Latest in tech, cars,  Android, iPhone, Tablets, Gadgets, Open Source, Programming. Grab them@taranfx on Twitter or below:

ARM currently controls a big piece of the mobile processing market, but Intel is attempting to make a comeback.

To give you an estimate of the power of the new processor, Intel demonstrated Quake 3 running lightning fast (100fps) on one of their custom smartphones powered by Atom-Z (Moorestone), a 45nm chip.

Without a doubt, such performance can also be used to offer flawless smooth HD video output, multitasking apps, and media-rich web browsing giving a desktop-equivalent experience.

Strong on Specs

Chip Power: They run from upto 2.0 GHz and have between 0.65 and 2.0 W TDP rating respectively that can dip down to 0.01 W when idle.

Multithreading, Caches, FSB: It features a 2-issue simultaneous multithreading, 16 stage in-order pipeline with 32 KB instruction L1 and 24 KB data L1 caches, integer and floating point execution units, x86 front end, a 512 KB L2 cache and data transferred at 533 MHz FSB (front-side bus).

Fabrication: The design is manufactured in 9M 45 nm high-k metal-gate CMOS and housed in a 441-ball µFCBGA package

Intel Atom Z vs. ARM Cortex

Intel claims that its latest Atom Z is a better option than popular ARM processors. As per the chip giant, Power consumption is 50x less than current Atom chips, and added advantage being for developers: ” No porting required”, all x86 apps would run as is.

Initially, the Atom Z will be seen in two variants: A 1.5GHz for smartphones, and a 1.9GHz model for Tablet devices, and both platforms run PowerVR 3D graphics acceleration.

Atom-Z Availability

Intel says devices will appear on the market in the 2ndhalf of 2010.

For latest Open Source, iPhone, Android, Hardware, Tech news @taranfx on Twitter or subscribe below:

Currently, the Android OS is exclusively available for ARM based architecture, however, Since it’s open source, we’ve seen few x86 ports for Netbooks. One of them was Acer, who ported Android to netbooks based on Atom microprocessors last year.

Intel now has done it officially(porting from ARM RISC to x86), and already has Android running on Atom-based smartphones.

Why Porting My Intel Matters? The reason is obvious, Intel would write the best optimizations for Android source code so that it can leverage the true power of x86 architecture. All the Special instruction sets of x86 would be used so as to make Android run even better.

Intel had long dreamt of entering MID, Smartphones with it’s Intel Atom processor but so far they are limited to netbooks. Currently, ARM dominates the smartphone industry. However, porting Android to Intel Atom would help them power the future mobile devices (smartphones) with more promising Atom 1.6Ghz+ PineTrail chips, which would  be a big step ahead of current generation snapdragon 1Ghz processor.

So in the future, its highly possible that most Android devices like : Android TV, Android tablets: NotionInk, WePad, Dell Mini5,  Android SIP Desk Phone, Android based Cars, Photoframes, smartphones would be using Intel Atom or even high end processors.

For latest Open Source, iPhone, Android, Tech news @taranfx on Twitter or subscribe below:

The processor will be targeted at four-socket servers with each physical core will run two threads simultaneously, giving the chip 64 virtual processing cores on servers and can even go upto 128. (but if this is good old Intel Hyperthreading (HT), it ain’t as good as anticipated).

Nehalem-EX is Intel’s fastest processor to date, however, clock speeds are still mystery but the company has said it will include 24MB of cache, and 2.3 billion transistors. The new processor will be targetted for high-end servers especially to Cloud computing farms and one of the first implementers would be IBM.

Nehalem architecture overcomes the limitations older Intel chips had:  integrates the memory controller and improves system speed by cutting data bottlenecks. Apart from that, the new architecture have some advanced Error-recovery mechanisms like MCA As far as the fabrication is concerned, it would use 45-nanometer process, however, 32nm is very much work-in-progress.

Advanced Error Recovery: Intel is also including new technologies like  Machine Check Architecture (MCA) recoveryerror correction that could make servers more fault tolerant and provide greater uptime. This added capability is brand new to the Nehalem product family and will help businesses lower their total cost of ownership. With MCA Recovery, errors can be isolated on a virtual machine and the other machines can still run uninterrupted while that one is being restarted. The processor will be able to detect system errors originating in the CPU or system memory and correct them by working with the operating system.
Apart of these error recovery were seen on Intel’s Itanium processors, which are totally different architecture.

Buffered Memory Chips: The new processor will also contain separate buffered memory chips that can store data temporarily alongside the main memory for faster task execution. No this is not L2 or L3 cache but something different.

Advanced Memory channelling: Intel will also offer four memory channels per processor, simply said more channels provide more memory bandwidth to run programs faster. AMD‘s 12-core Opteron server processors also offer same memory channels. But, Intel will build large amounts of cache inside Nehalem-EX that could help the processor deliver faster performance, but AMD‘s Magny Cours beats Nehalem-EX when it comes to more physical cores per chip.

Intel claims that Nehalem-EX will give a serious challenge to AMD’s Magny cours, but the results have to been before being commented on.

Intel claims, “The results could vary depending on the type of task assigned, but performance is just one part of the story. Nehalem-EX could reach out to new markets with additional features like fault-tolerance. It could tread into territory of high-end servers dominated by Intel’s Itanium and chips based on RISC (reduced instruction set computer) architecture, which includes IBM’s Power and Sun’s Sparc processors.”

Here’s the video from last year’s Intel’s channel talking about the Processor:

The 32nm horizon:

Sure, it was a great innovation but several years later was copied and re-branded as Hyper Threading (HT) by Intel #mce_temp_url#in it’s x86 processors. Intel claims it to produce two parallel threads without loss in performance. Well, it’s not true.

AMD has recently talked about why it doesn’t opt for Hyper threading, rather lays multicores as it’s core business strategy.

To understand this, lets go back to basics.

Threads vs. Cores

Cores are physical hardware blocks in the processor that can run applications serially. Threads, on the other hand, aren’t physical –  they are software-generated tasks that can execute
independently. A well threaded program will run itself across multiple cores.

As Intel quotes it:

HT adds circuitry and functionality into a traditional processor to enable one physical processor to appear as two separate logical processors. The added circuitry enables the processor to maintain two separate architectural states and separate APIC which provides multi-processor interrupt management and incorporates both static and dynamic symmetric interrupt distribution across all processors. The shared resources include items such as cache, registers, and execution units to execute two separate programs or two threads simultaneously.

As AMD quotes it:

So – cores are like bikes, threads are the riders. Running more threads increases throughput for applications as long as you have available cores. If you have threads waiting to be scheduled and no available cores – you have a bottleneck.

Think of SMT and HT as a tandem bike.  Yes it can move two riders, but not as quickly or efficiently as two separate bikes.

The hard thing for SMT & HT is that, it forces two threads to share a single  physical core.  When one thread pauses or stops due to a wait, then the second thread can jump in and take over.  But, the second thread needs to unload the cache, load its data, then, when finished, reload the original data for the first thread – this is what bring a big overhead.  This  helped with processors that had long pipelines of request.  Todays’ generation and next gen CPU have shorter pipelines make this type of effort  less important.

To be precise, HT is nothing more than a layer of abstraction added to the “processor->kernel” interaction that lets CPU decide how to handle multitasking rather than letting Operating system doing that.

All in all, HT can give applications as much as an extra 10-25% increase in performance by making better use of CPU, which is like a poor option over adding a new core.

Hyper Threading in the Real world

These, limitations were just theoretical. Practically, it’s even worse. Leaving desktop applications aside, it almost fails everywhere. In the production environments I’ve seen HT causing serious performance degrades and often admins just have to switch it off.

Hyper-Threading will increase the latency of individual requests and trade that for increased throughput; any individual request takes longer to complete, but more requests can be completed in a given time period. Going from uni-processor to multi-processor also adds cache contention and the increased locking complexity also slows things down; so performance will vary depending on what you are doing.

Larger multi-processor/Hyper-Threading systems with more than one physical processor tend to do better because their Hyper-Threading companion processors are more likely to be idle and halted. But as the load increases, those processors wake up and the overall latency in the system increases.

• When your bottleneck is device I/O, whether LAN or Disk, Hyper-Threading does not increase performance.
• When you have a single operation/thread generating high CPU utilization, Hyper-Threading does not increase performance.

I remember the time when we deployed my company’s Supplychain J2ee based applications on unix sandboxes, Java performance degrade of  upto 30% was observed. That was the first time I learned the myth about hyper threading.

Today, almost every production environment has mandated to stay away from it. Here are few examples:

• Novell calls Disabling HT a cool solution.
• Cognos, a leading BI software by IBM, recommends  disabling HyperThreading for better performance, stablity.
• Microsoft recommends turning off HyperThreading when running PeopleSoft applications because “our lab testing has shown little or no improvement.”
• A Microsoft TechNet article recommends disabling Hyper-threading for production Exchange servers and “only enabled if absolutely necessary as a temporary measure to increase CPU capacity until additional hardware can be obtained.”
• Advanced Clustering found when running High Performance Linpack (HPL) that “Using HT on the other hand causes a ~10% drop in performance compared to HT not being used.”

Ofcourse HT can provide a cheaper and power efficient solution for desktop by giving 10-20% performance benefit but as desktop apps and Operating Systems are becoming more capable, intelligent, demanding and hence more Multithreaded, HT loses it’s purpose.

We need more cores. Sun SPARC has 8 cores, AMD and Intel too will have them soon. We need to continue this innovation more than looking at the cheaper alternatives that add little value.

We write on Hardware, Open Source, Programming, Tech news, gadgets,  Get them @taranfx on Twitter

Intel claims that Nvidia Ion is overkill for Netbooks battery. But practically, Intel’s Pine Trail Atom n450 chips are pretty bad in almost everything. Their power management is not as good as Nvidia’s and when it comes to performance, there’s no comparison.

Luckily, the idea created a new Niche though was exaggerated by several big brands. Watching the growing success, Intel now upgrades existing line of Atom Processors. Intel’s latest offering is N450 processor and NM10 Express chipset also referred to as “Pine Trail”. Intel Plans to debuts this in CES 2010 January. Side-by-side HP, Acer, Dell, Asus, Toshiba, Lenovo and others are expected to announce new systems with the latest chips.

The single-core Atom N450 chip is 60 percent smaller than existing Atom processors, and consumes close to 20 percent less power. The resultant chip draws about 5.5 watts of power, that’s crazy low!

The Atom N450 will run at a clock speed of 1.66GHz, which is the same as an existing Atom N280 netbook chip. However, the improvements in the N450 come from the smaller chip size, achieved by integrating the graphics and memory controller into the CPU, which makes it one step closer to System on a Chip.The new processor is supposed to process multimedia faster and free up bandwidth for the processor to communicate with other components and have 512KB L2 cache, Hyperthreading, and 64-bit support.

The graphics improvement will come as relief to netbook users who have criticized the chipset in current Atom netbooks for its limited graphics compared to Nvidia’s Ion platform, which plugs a GeForce graphics core into an Atom chip to deliver full 1080p graphics.

The integrated graphics processor in N450 is capable of 720p high-definition graphics natively, but Intel claims that the N450 processor is meant to consume Internet content, not to play graphics-intensive games or view high-definition movies. However, different vendors will integrate broadcom cards to get FullHD and BluRay playback.

Netbooks powered by Atom N450 will run Windows 7 out of the box.

Without any disruption, Intel has revealed its plans for new products to be unveiled at next month’s Consumer Electronics Show in Las Vegas.

Intel plans to debut 27 new components, including 17 new processors. They include two notable integrated processors for desktops and notebooks PCs. Both use a 45 nm graphics and memory controller die paired with a 32 nm dual-core x86 die in a single package. The new chips are designed to cover the entire spectrum and bring the Nehalem architecture throughout the company’s product line. Among the new products will be new Core i5 and Core i3 chips to complement the high end Core i7 line.

The Core i3 chips will provide the basic advantages of the Nehalem architecture for low-end users, while the Core i5 chips will feature the Turbo Boost feature and focus on users who perform more demanding tasks such as video or photo editing. On the Graphics side, Intel is also planning to update its onboard graphics and sound platforms. The GMA graphics platform will be rebranded as Intel HD Graphics, and will be offered on all of the company’s new dual-core systems along with improved onboard sound systems. Intel expects to have OEMs showing systems using its latest 32 nm processors. It will also provide details on 27 new chips for 2010 including an upgraded Wi-Fi/WiMax combo chip and several chip sets.

The graphics core is an upgrade of Intel’s 4 series, a 65 nm chip used in Intel’s north bridge PC chip sets.

“We have two 32 nm fabs running and are in high volume production to supply OEMs,” said Stephen Smith, director of PC client operations at Intel. “That’s our single largest investment in the company,” he added, noting Intel has two more fabs gearing up for 32 nm technology.

Intel Onboard graphics systems have long been declared “LAME” by users for their low performance and limited abilities, but Intel claims that the latest systems are far more advanced, and capable of running more demanding tasks and supporting new standards such as Blu-ray playback.

The graphics are definitely going to be  lame for Gamers/professionals, on the contrary, it might serve a “Normal user” well.

As per the latest publication Intel’s Pittsburgh lab aims to develop brain implants that can control all sorts of gadgets directly via brain waves by 2020. The purpose is to eliminate and out-date the usage of keyboard and mouse. Instead, you will use your brain waves to  control your computer, says Intel Corp. researchers. So what exactly they mean is, users will be able to open documents and surf the Web using nothing more than their brain waves.

“I think human beings are remarkable adaptive,” said Andrew Chien, vice president of research and director of future technologies research at Intel Labs. “If you told people 20 years ago that they would be carrying computers all the time, they would have said, ‘I don’t want that. I don’t need that.’ Now you can’t get them to stop carrying devices. There are a lot of things that have to be done first but I think implanting chips into human brains is well within the scope of possibility.”

“We’re trying to prove you can do interesting things with brain waves,” said Pomerleau. “Eventually people may be willing to be more committed … to brain implants. Imagine being able to surf the Web with the power of your thoughts.”

To get to that point, Intel is heavily researching on decoding human brain activity.

The team has used Functional Magnetic Resonance Imaging (FMRI) machines to determine that blood flow changes in specific areas of the brain based on what word or image someone is thinking of. People tend to show the same brain patterns for similar thoughts.

For instance, if two people think of the image of a bear or hear the word bear or even hear a bear growl, a neuroimage would show similar brain activity. Basically, there are standard patterns that show up in the brain for different words or images.

Human brain is a big processing powerhouse, once they decode the signals, they may be able to achieve a fasater and better supercomputers, like human brain.

Of couse, there’s nothing to be scared of, they won’t be planting chips in your brain against your will. Researchers expect that consumers will want the freedom they will gain by using the implant.

What do you think, would you get one and become a cyborg?

The moment of joy was expressed by AMD CEO Dirk Meyer. He  said the settlement leads both companies will bring a  “new era”  in the chip industry.

The two companies also sealed a five-year cross license deal and said they would give up any claims of breach from their previous license agreement.

“While the relationship between the two companies has been difficult in the past, this agreement ends the legal disputes and enables the companies to focus all of our efforts on product innovation and development,” AMD and Intel said in a joint statement.

Under the terms of the settlement, Intel will pay AMD $1.25 billion. Also:

• AMD and Intel both get patent rights in a cross-licensing pact;
• Intel will give up any patent claims against AMD;
• Intel will agree to adhere to business practice provisions;
• And AMD drops all pending litigation against Intel.

Today marks the beginning of a new era… one that confirms that the game has changed for AMD. It is an important milestone for us, for our customers, our partners, and most important – for consumers and businesses worldwide. In addition, it represents the culmination many years of litigation and regulatory engagement. And we are optimistic that it will usher a new era for our industry.

We look forward to healthy competition with the mutual respect one would expect between world-class competitors.

It remains to be seen if this settlement marks a new chip era, but the “healthy competition” Meyer refers to will determine AMD’s fate.

Shares of AMD jumped 23 per cent to $6.55 in early trading on the New York Stock Exchange this afternoon. Intel shares were halted. Intel now expects spending in the quarter to be $4.2 billion, up from $2.9 billion.

So the “new Era” is about to begin where chip manufacturers respect the competition? What do you think?

Subscribe to Twitter updates, or RSS, join Facebook fanpage for more Snow Leopard, iPhone updates.

Ajay Bhatt is more than what you know: USB, AGP, PCIe, PPMA, etc. were all co-invented by him.

Probably every one has seen the Intel’s commercial where they show that their RockStars are different from ours.

The Ad features Ajay as the co-inventor of the USB port. In reality, the guy who films int he ad isn’t the real Ajay Bhatt.

Ajay Bhatt is an Intel Fellow and Chief Client Platform Architect in a Mobile Platforms Group. Bhatt leads definition and development of the next-generation Client Platform architecture.

In addition, Bhatt continues to hold a position of Intel’s Chief I/O architect. In his role as a Chief I/O architect, he is responsible for the platform and I/O interconnects directions for Intel

Bhatt is an industry-recognized expert in the area of I/O technologies. At Intel, Bhatt had been working on the technologies which are now wide-spread like USB, Accelerated Graphics Port AGP, PCI Express (PCIe), Platform Power management architecture and various chipset enhancements. Bhatt joined Intel in 1990 as a senior staff architect on the chipset architecture team in Folsom.

He holds nine U.S. patents with several in various stages of filing.

Recently, Conan from NBC interviewed Ajay and as usual, added humor to the whole story while appreciating his efforts. Watch the video:

The Experiment started on lighter Notebooks, then started to be known as Netbooks. Luckily, the idea prevailed and sales grew in number, and finally found a space in market as  new niche segment.

Now that Netbooks are popular, every chip manufacturer wants to jump-in. We’ve seen Nvidia’s initiative with Nvidia Tegra, and of course large no. of Atom based machines. So, why should ARM, most popular chip manufacturer for smartphones, stay back?

ARM announced that it will begin offering its customers ARM’s Cortex A9 processor on its 28nm process. This extends the reach of ARM’s top-end core, which is taking direct aim at Intel’s Atom.

GlobalFoundries, AMD’s subsidiary,  would work with ARM on a 28nm A9 implementation, and it’s a win for ARM because it makes any GlobalFoundries customers into potential ARM customers.

A bit of background: an system-on-a-chip (SoC) provider who is already using GlobalFoundries to produce its SoCs can more easily mix its own technology with ARM’s to produce SoCs based on Cortex A9. This is the same reason that Intel ported Atom to TSMC—so that existing TSMC customers can also mix Atom with their own IP to make Atom-based SoCs. (Of course, what was a shocking move for Intel is standard procedure for ARM, which is a fabless semi company whose total revenues are less than what Intel spends to develop one processor.)

The most interesting aspect of today’s announcement surrounds AMD as a possible Cortex A9 user, because AMD is still GlobalFoundries’ main customer.

Given that AMD was just in the ARM business and got out of it just 9 months ago, it’s seems unlikely that it will jump back in with another ARM-based SoC courtesy of GlobalFoundries. This is too bad, though, because there’s a ton of enthusiasm around A9-based “smartbooks,” and an A9-based Imageon would’ve provided a great basis for a high-powered, low-cost, long battery-life Linux portable. But AMD has gotten out of the SoC game just as it’s getting interesting.

Given the scope of GlobalFoundries’ ambition and the recent uptick in global semiconductor demand, this ARM announcement seems logical and timely.

Intel announced that initial production of 32nm processors has begun. It’s taken many years (and a $7 billion investment) to get these Westmere chips onto silicon, with the Clarkdale (desktop) and Arrandale (notebook) lines expected to start showing up in over-sized cardboard boxes before the end of the year.

What’s next for 45nm chips, like those new Lynnfields? A regular fate that one sees in hardware industry: fade till obsolescence.

Check out the video from Intel’s clean room:

Now, sources say that Google approached Intel a while back to work together closely, to design Chrome OS suited for Intel architecture. The reason is but obvious, nearly 80% of the PCs/Notebooks/Netbooks are equipped with Intel processors.

Google has plans for release of Chrome Os in later half of 2010, with the first phase targetting to capture Netbook market, and eventually they will target NoTebooks and desktop PCs. On Netbooks, Intel Atom processor is a clear majority, capturing 97% of the market. So its pretty justified why Google chose Intel. Though there is a twist over this. Intel already had been working on a Portable computer OS called the Moblinfor it’s so called MIDs. Intel had been working over this OS for the last 6 months atleast. Intel wants to target Mobile devices. They are serious on getting Atom processor to the mobile platform.

Google made no mention of Intel in a blog posting earlier this week when it named companies already working with the Chrome OS. The list includes PC vendors HP, Acer and Lenovo and mobile phone chip makers Qualcomm, Texas Instruments and Freescale Semiconductor. Chrome OS  is being  developed as Linux-based operating system for heavy Internet users.

Today, most of the smartphones run on ARM architecture. ARM is good on battery power but limited in the processing side. With Intel taking Atom processors to Mobiles, they can pump the capacity of the mobile devices to run Typical “Desktop applications” on their phones.

Strategy from Intel and Google seems to be converging, and It’s quiet evident that their vision of MIDs can converge innovation that can lead to much more powerful portable devices for the future. What would we call it ChroIntel or Googintel ?