Intel launches new online dev magazine

Visual Adrenaline offers tips on multi-threading and code optimisation for games devs
Leading component and software vendor Intel has launched a brand new digital magazine targeting the games development sector.

Entitled Intel Visual Adrenaline, the mag is free to those who sign up to the Intel Software Dispatch news service. The digital publication covers the use of multi-threading and code optimisation in the rendering of games and apps, as well as information on the latest tools and development products.

The mag will be released once per quarter and takes the form of a downloadable PDF.

Resource - Develop

Intel evaluating new netbook concepts, form factors

Netbooks have been one of the major success stories of 2008, but recent comments from Intel indicate that the company isn't satisfied with current device form factors. Speaking at a Raymond James IT Supply Chain conference last week, Stu Pann, an Intel VP of sales and marketing, admitted that Intel initially miscalculated the market segments that netbooks would appeal to.

"We originally thought Netbooks would be for emerging markets and younger kids... It turns out the bulk of the Netbooks sold today are Western Europe, North America, and for people who just want to grab and go with a notebook," Pann said. "We view the Netbook as mostly incremental to our total available market."

Pann also noted (his comments can be streamed here) that netbook screen sizes don't lend themselves to long-term use, saying, "if you've ever used a Netbook and used a 10-inch screen size, it's fine for an hour. It's not something you're going to use day in and day out."

Consumer interest in netbooks definitely caught Intel off-guard, and it took the company months to ramp Atom production in order to satisfy demand. I agree with Pann, when he portrays netbooks as providing an "incremental" increase to Intel's total market, but I'm not sure he's right when he states that consumers won't use a netbook on a daily basis. It's true that the majority of us would have a massive headache after trying to work on a 10" LCD all day, but why assume that a netbook user is perpetually limited to the device's display?

A typical Western netbook user will have access to a much larger secondary display, as well as a keyboard and mouse at both work and home. Combined, these three simple devices negate the unfortunate physical aspects of netbook use. Mobile users, meanwhile, are more likely to use a netbook for short spans of time, which would fit within Pann's "fine for an hour" distinction.

Given how popular they've become, it's easy to forget that netbooks are only a year old, while the Atom itself has only been on the market for approximately six months. The device manufacturers themselves are still experimenting with features, sizes, and price points; it would be downright short-sighted for Intel to bring down the gates on experimentation and categorically lock the definition of the netbook form factor. What Pann failed to address—and it's a significant point—is that the sudden jump in netbook sales might be indicative of a long-term purchasing shift.

It's been posited that one reason netbook sales spiked in 2008 is because consumers were concerned about the economy and opted to purchase a cheaper, more portable system, which would fit Pann's "grab and go with a notebook," theory. But the continuing popularity of these smaller systems indicates that consumers like them, and have generally found that they provide an adequate amount of horsepower. Ultimately, yes, a netbook is just a smaller notebook, but a significant number of shoppers appear to want a smaller, cheaper, notebook.

Full-size notebooks aren't going to go away, anymore than full-size desktops are; there will always be a group of buyers who want more computing power and will trade portability for it. Nevertheless, I think the rise of netbook sales in 2008 is the beginning of a buying pattern rather than a trendy spike.

Second-generation netbooks, including hypothetical models with larger screens, will cast a wider net and capture groups of buyers that weren't interested in earlier models. Meanwhile, existing netbook owners already seem to be using these devices "day in and day out." If you're a current netbook owner, or are interested in buying one, what's your take on this? Are you buying it as a device you intend to use on a daily basis (even if it isn't your only system) or will it be confined to occasional portable use?

Resource - Ars Technica

SGI Designs Atom-based Supercomputer

A 3U-high package houses 90 blocks. "A few more blocks can be added because the actual package is slightly deeper," SGI said.

The substrate mounted with the Atom N330 and the block composed of two such substrates. The substrate is mounted with actual chips, but the block is a mockup.

A close-up picture of the substrate. The slender chip on the left is the Atom N330. The big chip in the center is a controller IC. The four chips on the front side are memory chips.

Silicon Graphics Inc (SGI) of the US exhibited a concept model of a novel high performance computer (HPC) at Super Computing 2008 (SC08), an international convention and exhibition on HPCs.

The computer is being developed under the code name of "Project Molecule." The aim of the project is to "create a supercomputer featuring ultrahigh-density, low power consumption and low cost using the ultimate commodity processor that can be easily programmed," SGI said.

The commodity processor used in this concept model is the "Atom N330," which Intel Corp developed for mobile devices called "MID (Mobile Internet Device)."

SGI exhibited a chassis, which corresponds to a 15cm-high (3U) rack. And, in the chassis, 90 blocks that look like building blocks are laid out with little space left. The block has several square pores dimensioned about 5 x 5mm so that air from the air-cooling fan on the back of the package can pass through the pores.

The block consists of two 5 x 5mm substrates attached to each side of a pore structure used for air cooling by resin. Each of the substrates is mounted with microprocessors, LSIs for datapath control, and eight memory chips on two sides. The dual-core Atom is incorporated as a microprocessor. This atom is a 1.3GHz model and supports "x86-64" instruction set. Each substrate has a memory capacity of 2 Gbytes.

In order to achieve ultrahigh density, more than 180 microprocessors with CPU cores exceeding 360 are packed into the 3U rack. The power consumption of this Atom is as low as 10W, making it possible to array many blocks leaving small spaces for tiny air-cooling pores. The power consumption of the whole package is less than 2kW, according SGI.

"As the next step, we are planning to employ a liquid-cooling system and further improve the density by, for example, packing more than 10,000 CPU cores in one rack," the company said.

The Atom is a low power consumption microprocessor developed by Intel for use in MIDs. Intel explains that it is not suitable for complicated processing and only good for Web browsing and checking e-mail.

In respect to this point, SGI said, "We place priority on memory bandwidth, and the Atom's memory bandwidth is not small. A supercomputer based on this model will be intended exclusively for calculation, in which memory bandwidth is important."

IBM Corp's "Blue Gene" supercomputer was developed under a similar concept. The Blue Gene uses the "PowerPC 440" embedded microprocessor. SGI admits that there are some similarities between its concept model and the Blue Gene, but said "they are not the same." The biggest difference is in that the Atom microprocessor supports the x86-64 instruction set, which is designed for general PCs and servers.

In respect to its commercialization, "It has yet to be given the go-ahead. But once its commercialization has been decided, we will be able to develop a product in 15 to 18 months," SGI said.

Resource - Tech-On

Intel Revs Compiler Suite

On Tuesday Intel announced that it has updated its compiler and cluster suites ahead of the introduction of the Core i7 (Nehalem) processors. The Fortran and C/C++ compilers have been revved to version 11, and this release includes version 3.2 of the Intel Cluster Toolkit. The releases are part of Intel's 18-24 month update strategy in its workhorse programmer tools suites, and are aimed at the wide spectrum of developers who need parallelism on scales ranging from the single socket to thousands of processors. HPCwire spoke with James Reinders, chief product evangelist and director of marketing for Intel's Developer Products Division, to find out what's new this time around.

This week's announcement follows Intel's August announcement of Parallel Studio, a massive investment aimed at enabling Windows software developers to leverage the power of multiple cores into mainstream, desktop applications. This week Intel returns the focus back to tools for programmers in HPC, with many features that will resonate with parallel application developers on the traditional high end of the performance spectrum as well as the growing audience of developers aiming at smaller scale clusters.

Intel's C++ Compiler 11.0 and and Fortran Compiler 11.0 have been released, as well as a new Math Kernel Library (10.1), Integrated Performance Primitives (6.0), and version 2.1 of Threading Building Blocks. Each of these is available immediately for Windows, Mac, and Linux. Version 3.2 of the Cluster Toolkit (Compiler Edition) for Windows and Linux includes the compilers and the math library (including ScaLAPACK), as well as Intel's MPI Library (3.2), Intel Trace Analyzer and Collector (7.2) with the new MPI correctness checker, MPI Benchmarks, and the cluster installer.

The MPI correctness checker is particularly interesting, and Reinders indicated beta users were enthusiastic about the product. Developers turn on performance analysis at compile time and link with an instrumented version of Intel's MPI library. Post-run they can analyze what's going on inside MPI calls using the Trace Analyzer, with automated support for catching errors such as mismatched send and receive buffers, and other insidious bugs that can be difficult to ferret out in development.

This version of Intel's developer tool suite also includes support for new lambda functions and other features of the C++ 0x draft standard, parallel lint for static parallel application analysis, and features from Fortran 2003. OpenMP 3.0 is included with function-level parallelism for both data and task parallel models.

The diversity of Intel's investment in developer tools reflects two key concepts: incremental investment and forward scaling. The breadth of Intel's tools allow developers to introduce parallelism into their applications gradually, using TBB or OpenMP before perhaps going "all the way" with an MPI implementation. "Forward scaling" is the conceptual handle for the idea that an investment made in an application today shouldn't turn into a dead end when the company doubles the number of cores in its hardware tomorrow. For example, the latest version of the C++ compiler takes advantage of the semantic properties of valarrays, which already existed in the language standard, to automatically parallelize operations on them for higher performance. Developers that take advantage of valarrays today can expect future versions of Intel's compilers to "do the right thing" when it comes to scaling that performance on Intel's future, more highly parallel, hardware.

There are differences between the interfaces and options that developers on different ends of the HPC spectrum need, a challenge that Intel has been dealing with for some time. "Multicore parallelism more strongly demands programmer productivity and more loosely demands efficiency from the implementation than traditional HPC," says Reinders. The practical implication of this is a more polished interface with less exposed complexity for developers in tools meant for the low-end software developer. At the very high end, however, developers demand access to every flag and option to control what the compiler is doing and how it does it. Intel provides this access through its compiler "black belt" guides that detail hundreds of switches and flags for very fine-grained control. Happily for Intel, the core technology is the same, and small teams focus on targeting the technology with interfaces and APIs or by exposing (and
documenting) the arcane options in the compilers for the two audiences.

Intel's product presentation includes about 60 slides on the performance of its compilers compared to the previous product release and to competing compilers. While admittedly very few companies present information that makes their products look bad, the sheer volume and technical detail in Intel's performance studies reveal the confidence the company has in its new products. Performance is nearly uniformly better for this release on a very wide range of benchmarks.

You can find more details of these results on Intel's developer software pages.

Reinders says that the final key piece of Intel's release this week is stability. Reinders says that "There simply isn't wiggle room to have anything about these products be unstable" for Intel's customers -- the companies that build the software we use every day.

Resource - HPCwire

Google demanding Intel's hottest chips?

When purchasing server processors directly from Intel, Google has insisted on a guarantee that the chips can operate at temperatures five degrees centigrade higher than their standard qualification, according to a former Google employee. This allowed the search giant to maintain higher temperatures within its data centers, the ex-employee says, and save millions of dollars each year in cooling costs.

If the chips failed prematurely at these higher temperatures, the former Googler says, Intel was obliged to replace them at no extra charge.

Intel denies this was ever the case. "This is NOT true," a company spokesman said in an email. Google declined to comment on its relationship with Intel. "Google invests heavily in technical facilities and has dozens of facilities around the world with many computers," reads a statement from the company. "However, we don't disclose details about our infrastructure or supplier relationships."

The ex-Google employee learned of this Intel pact a little more than a year ago, during a Google "Tech Talk" open to anyone at the company. The talk was given by a Google thermal dynamics engineer, part of a small team - perhaps no larger than two people - that oversees heat issues inside the company's data centers.

According to the same ex-employee, it is now the norm for Google to construct its data centers by piecing together intermodal shipping containers pre-packed with servers and cooling equipment. In 2003, Google filed for a patent on this sort of modular data center, and the patent was granted last October.

This modular setup - known internally as Project Will Power - allows Google to construct the building blocks for each data center at a central location and then ship them around the world as needed. Robert X Cringely first leaked word of Google's modular data center work in 2005, claiming the project began after representatives of the Internet Archive pitched the idea to Google co-founder Larry Page.

Heat Saves

It's no secret that Google builds its own data centers - and many of the servers within. It's long been said that after the top server vendors, the ad broker consumes more processors and hard disks than anyone else in the world - and Intel freely acknowledges that it provides at least some of the chips.

Intel also customizes power saving motherboards for Google - a service not afforded other customers. But our source tells us that any motherboard pact is separate from the companies' high temperature chip agreement.

If Google can leverage an extra five degrees centigrade from Intel, it can save a few penguins - and some serious cash. According to Data Center Knowledge, Google runs at least 36 data centers across the globe, and several more are under construction. One of the newer sites, in The Dalles, Oregon, includes three data center buildings, each measuring 68,680 square feet.

According to Mark Monroe, Sun Microsystems' director of sustainable computing, data center managers can save 4 per cent in energy costs for every degree (Fahrenheit) they raise the thermostat. If you run your data center at a higher temp, you use less power and spend less money on the equipment needed to cool it down.

But if you raise the thermostat, you may void your hardware warranties. "The hardware manufacturer will usually have a temperature range," says Rich Miller, editor of Data Center Knowledge. "If the equipment is running outside of the range, the manufacturer might be inclined to challenge any sort of returns or credits. That's the thinking across the industry."

Resource - The Register

Apple Migrating to NVIDIA Chipsets in new MacBooks

Appleinsider reports that they have "confirmed" that the new MacBooks will utilize NVIDIA's new MCP79 platform. While Intel will continue to supply the main processor for Apple's notebooks, the underlying support chips will be made by NVIDIA:

Kept uncharacteristically secret by NVIDIA for most of the year, the MCP79 platform is so far considered a substitute for Intel's Centrino 2 "Montevina" platform, offering support for the same 1066MHz front side bus, optional DDR3 memory and PCI Express 2.0 interfaces.

AppleInsider lists several design advantages over Intel's chipset, including:

- Smaller physical size using one chip rather than two
- DriveCache, which uses Flash storage to speed up boot times
- Hybrid SLI which switches from discrete to integrated graphics when battery is low

Apple may or may not choose to take advantage of these technologies, but the option exists. The biggest advantage in using the NVIDIA chipset, however, will be its enhanced graphics capabilities.

NVIDIA is believed to use a new set of GeForce 9300 and 9400 series integrated GPUs which "will theoretically blow past" the Intel integrated graphics chipsets that currently power Apple's MacBooks. Still, the integrated graphics chipsets will lag behind the dedicated ones that are used in Apple's higher-end notebooks. The use of NVIDIA chipsets will also be of importance when Apple releases Snow Leopard, which will be able to offload general processing onto GPUs.

While there had been speculation that Apple might adopt NVIDIA's chipsets, Mac Soda first reported confidently that the chipset would be NVIDIA's.

Resource - MacRumors

'Atom 101' - Deciphering the Intel codewords around MIDs

With the Intel Developer Forum (IDF) kicking off this week in Shanghai there's going to be a mini-avalanche of MID, Atom, and Moblin related announcements coming out. I'll try to publish a few digests here of anything I find interesting, like this video fellow Intellian Ken Kaplan shot about some of the MID devices from the Consumer Electronics Show. I'm looking forward to finally getting some of this hardware in my own hands to play with!

In my ongoing self-education on this topic, there was a valuable post on our intranet that was billed as "Atom 101". See, even Intel employees can't keep all our kooky code names and chipsets straight! Anyway, the author, Walden Kirsch, gave me the green light to post it externally so I hope it is of help to others.

Intel® Atom™ processor. The brand name for a family of low- power processors designed for a range of new market segments including mobile Internet devices and simple, low-cost Internet-focused machines such as netbooks and nettops (see definitions below). The Intel Atom processor will also be used in embedded and consumer electronics. The initial versions of the Intel Atom processor are code named "Silverthorne" and "Diamondville" (see below). The Intel Atom processor family is based on the core code named "Bonnell." The Intel Atom processor is Intel's smallest chip, built with the world's smallest transistors. Contains 47 million transistors.

Silverthorne. The code name for one version of the Intel Atom processor, optimized first for low-power then for low-cost. Silverthorne is targeted for use in mobile Internet devices. Manufactured on Intel's 45nm process technology.

Diamondville. The code name for another version of the Intel Atom processor, optimized first for low-cost then for low-power. Slightly larger packaging than Silverthorne so it can be used on less expensive 4-layer motherboards. Diamondville is targeted for low-cost platforms including nettops and netbooks. Manufactured on Intel's 45nm process technology.

Mobile Internet device (MID). A small mobile device that delivers the full Internet in your pocket. Powered by the Intel Atom processor. The best Internet experience in your pocket is enabled by Intel® Centrino® Atom™ processor technology.

Intel Centrino Atom processor technology. The brand name for a low-power platform designed for mobile Internet devices. Intel Centrino Atom processor technology includes the Intel Atom processor, a low-power companion chip with integrated graphics, a wireless radio, and other features.

Menlow. Code name for the Intel Centrino Atom processor technology platform. Menlow includes the Silverthorne processor and the Poulsbo chipset.

Poulsbo. Code name for low-power chipset for mobile Internet devices. Used with Silverthorne in Intel Centrino Atom processor technology based machines. Part of the Menlow platform. Poulsbo includes 3D graphics, HD video playback (720p and 1080i content) and the memory controller.

Embedded Menlow. Similar to Menlow with the Silverthorne processor and the Poulsbo chipset, but runs embedded operating systems and includes embedded lifecycle support of at least seven years.

Moorestown. Code name for successor to the Menlow platform, scheduled for late 2009 or early 2010. For mobile Internet devices. Intel has publicly committed that Moorestown will have at least 10 times less idle power consumption than the previous-generation Menlow platform.

Netbook. A small, low-cost, mobile computing device designed for consuming content, rather than creating new content.

Nettop. A desktop version of the netbook.

Sodaville. Codename for System on a Chip (SoC) product for consumer electronics devices, arriving in 2009. Sodaville will be used in non-battery CE devices used for digital entertainment in the home. Sodaville will be based on the same low-power core found in Intel Atom processors.

That's the list. If there is anything missing, let me know!

Resource - Intel Technology Blog