[bbmf]

sa: http://www.zunescene.com/

[bbmf]

Introduction
Like a creeping fog, DRM smothers more and more media in its clammy embrace, but the sun still shines down on isolated patches of the landscape. This isn't always due to the decisions of corporate executives; often it's the work of hackers who devote considerable skill to cracking the digital locks that guard everything from DVDs to e-books. Their reasons are complicated and range from the philosophical to the criminal, but their goals are the same: no more DRM.

We're going to revisit the history of the most famous DRM cracks. While the stories themselves are fascinating, one of the merits of such an exercise is to use the lessons of the past to consider the challenges of the future. Along the way, we'll address the following important questions:
Will DRM someday be unbreakable? Do content companies care if it is?
Who or what is a "Beale Screamer"?
What does the history of DRM mean for new technologies such as Blu-ray discs and HDCP links?
Can a marker violate the DMCA?
What's more important: technology, Congress, or the market?
Will a Stalin statue make a brief cameo appearance in the conclusion of this article?
We'll start our survey with one of the most-used DRM schemes in the country, Apple Computer's FairPlay.

It's all about the music
FairPlay
The FairPlay system, despite the professional effort that went into it, turned out to be surprisingly hackable, given Apple's reputation for robustness and general platform security. When Apple decided to take a bite out of the digital music market several years ago, it marshaled its internal technical resources to develop a home-grown DRM scheme dubbed FairPlay. By the middle of 2004, the encryption scheme had been cracked more times than a toppled Humpty Dumpty. Jon Johansen, the Norwegian hacker partly responsible for cracking the encryption on DVDs (see below), released the primitive QTFairUse, which attempted to bypass (rather than break) the FairPlay encryption. QTFairUse relied on Apple's software to decrypt the protected song files and then grabbed the unencrypted music from RAM. It then wrote this data to an unencrypted AAC file that turned out not be readable by most music players.
QTFairUse would not be the program to bring unencumbered iTunes downloads to the mainstream user, but it did represent one possible line of attack. Another approach was provided by playfair, a little program capable of stripping the DRM from iTunes files. Instead of grabbing the unencrypted data, playfair relied on grabbing the key FairPlay encryption uses. This key was stored on the iPod and was also easily accessible on Windows systems; once it was grabbed, songs could be decrypted and written to disk (Mac systems initially required the iPod to be attached to the computer).
This approach meant that you could only decrypt songs to which you had the rights anyway, but Apple was still unhappy about it. They modified their software to make the key much harder to grab. They also leaned on the web hosting company that playfair used; the project was also pulled from SourceForge. Showing just how hard it is to stuff the code genie back in the bottle, though, playfair development continued. The project was renamed "Hymn" and new versions are still being released, though all still have problems with certain versions of iTunes.
A third approach came from PyMusique, software originally written so that Linux users could access the iTunes Music Store. The software took advantage of the fact that iTMS transmits DRM-free songs to its customers and relies on iTunes to add that gooey layer of DRM goodness at the client end. PyMusique emulates iTunes and serves as a front end to the store, allowing users to browse and purchase music. When songs are downloaded, however, the program "neglects" to apply the FairPlay DRM. (A variant of PyMusique, called SharpMusique, has been developed and maintained by Johansen, though it has not been updated in 10 months).
The attacks on FairPlay have been enlightening because of what they illustrate about the current state of DRM. They show, for instance, that modern DRM schemes are difficult to bypass, ignore, or strip out with a few lines of code. In contrast to older "patches" of computer software (what you would generally bypass a program's authorization routine), the encryption on modern media files is pervasive. All of the software mentioned has still required Apple's decoding technology to unscramble the song files; there is no simple hack that can simply strip the files clean without help, and the ciphers are complex enough to make brute-force cracks difficult.
Apple's response has also been a reminder that cracking an encryption scheme once will no longer be enough in the networked era. Each time that its DRM has been bypassed, Apple has been able to push out updates to its customers that render the hacks useless (or at least make them more difficult to achieve). The resulting cat-and-mouse game between the company and its users will no doubt become a familiar feature of future DRM schemes, most of which are now built with the ability to update themselves even after deployment.
On the other side, it's also more difficult than ever to shut down a cracking project. Coders simply move their work from server to server until they find a spot where the long arm of Apple Legal cannot reach.
Windows Media
When it comes to music, Microsoft's own DRM system comes second only to Apple's FairPlay—though that's not saying much in the US, where Apple has an 80 market share. Still, it's a solid achievement for the Redmond, WA-based company, which beat out earlier contenders like Liquid Audio and Real. Windows Media Audio has been hacked less frequently than Apple's competing system, but as security analyst Bruce Schneier puts it, this is due to "market share, nothing more." Because relatively few music downloads are sold in the format, hackers devote less energy to cracking it.
Windows Media Audio has stood secure for several years now, but it was memorably hacked back in late 2001, when version two of the encryption was in use. A hacker using the pseudonym "Beale Screamer" posted a long message to the sci.crypt newsgroup in which he described how to circumvent the DRM protections of Windows Media Audio and provided code to do so. His program, dubbed "freeme," worked, but with limitations; users needed to have a license for the song before the decryption would work. It also was designed only for version 2 of Microsoft's DRM, the one included with Media Player 7.
Beale Screamer actually had some complimentary words for the Microsoft engineers who designed the scheme:
You guys have put together a pretty good piece of software. Really. ... My real beef is with the media publishers' use of this software, not the technology itself. However, it's easy to see where software bloat and inefficiency comes from when this code is examined: every main DLL has a separate copy of the elliptic curve and other basic crypto routines, and parameters passed back and forth between modules are encrypted giving unnecessary overhead, not to mention all the checks of the code integrity, checks for a debugger running, code encryption and decryption. Perhaps you felt this was necessary for the "security through obscurity" aspect, but I've got to tell you that this really doesn't make a bit of difference. Make lean and mean code, because the obscurity doesn't work as well as you think it does.
Mr. Screamer ("Beale" to his friends) made a point of asking users not to use freeme as a tool for violating copyrights, because such use would violate the serious point he was trying to make about DRM's dangers to fair use. Such principled messages are actually quite common among DRM crackers; whether users heed them is another story altogether.
The news created headaches for Microsoft's digital media group, but ultimately the hack was not of great concern. Microsoft was able to update their DRM, which has not been widely breached since that time.
Discs and e-books
"Compact discs"
Unlike DVDs, which had encryption built into the specification, compact discs were unencrypted and uncompressed, which made them the obvious go-to location for pristine digital copies. With the rise of broadband and the introduction of file-swapping systems like Napster in the late 1990s, record labels began to worry about the security of the venerable CD. When online music stores began selling songs in earnest, the problem was truly brought home to the labels. Why should they invest the time and money into selling DRM-laden songs that only irritated customers when those same customers could simply drive to the local store and pick up a better-quality, unencrypted version of the same music?
The labels were faced with an obvious decision: either forego DRM altogether or find some way of backporting it to CDs. Since the first option was never seriously considered, the second one prevailed, and an assortment of techniques appeared that promised to copy protect "compact discs" (copy-protected discs generally do not conform to the official Red Book CD standard maintained by Philips and therefore cannot legally use the "compact disc" logo). These fell into two groups: active and passive protection, and they worked (or failed to work) quite differently.
Passive protection introduced deliberate errors into the compact disc format. These errors were designed to exploit the subtle differences that initially existed between home CD players and computer CD drives. Solutions like the Cactus Data Shield initially promised that they could deliver near-total compatibility and excellent protection from ripping. In practice, things did not work out nearly so well.
For one thing, compatibility was an issue from the start. Despite minor differences, CD players and computer drives are similar enough that many computers could read the discs without problems, while some home CD players could not. One of the obvious virtues of the CD format is that you can bring it home from a store anywhere in the world, pop it into your player, and have it work without hassles. Copy-controlled CDs broke this long-standing compatibility for the sake of modestly increased security.
The other approach was to include active protection. This took the form of software installed on the disk that tried to install itself onto a user's computer in order to control the user's access to the drive. Such software was written only for Windows, which meant that sticking the disc into a Mac or a Linux box bypassed the code entirely. Even on Windows machines, the security provided was rudimentary at best. SunnComm's MediaMax 3 DRM system could be easily bypassed by using a black marker (a revelation that got security researcher John Halderman into hot water before SunnComm saw the light and declined to prosecute him).
Not everyone was eager to take a Sharpie to their newest albums, however, but an even simpler solution soon emerged. Most active protection could be bypassed simply by disabling Autorun in Windows (or holding down the shift key when the CD is inserted, which has the same effect).
No discussion of CD protection schemes would be complete without a mention of Sony BMG's notorious rootkit software, which was formally known as "XCP." XCP, created by UK firm First4Internet, created security vulnerabilities on a user's computer that were only made worse by the initial uninstall program. After a series of class action lawsuits, the company eventually pulled from store shelves all CDs containing the program and made restitution to affected customers.
Adobe e-books
While most attacks end simply with the company releasing an update to its software, American authorities took a harder line against Russian programmer Dmitry Sklyarov. Sklyarov had worked on password cracking techniques for his dissertation, and after leaving school was employed by Russian software firm Elcomsoft, where he was a part of the team that developed a crack for Adobe e-books. Although Elcomsoft marketed their Advanced eBook Processor as a way to make PDF files out of legally purchased e-books (something that was not allowed by the Adobe software), Adobe saw it as a piracy tool.
Back in 2001, when the Skylarov story broke, Adobe and Microsoft were both pushing hard to establish their e-book formats in the marketplace. An insecure format could potentially lead to lower profits for publishers, authors, and online bookstores, all of whom would request whatever technology was most secure. In this atmosphere, a highly public announcement of the security flaws in Adobe software was unwelcome. Skylarov attended the Las Vegas "Def Con" conference and presented a paper on e-book security (see the slides).
Adobe's system used a 40-bit key but proved difficult to crack by brute force methods (unlike CSS; see below). Skylarov estimated that a 450 Mhz Pentium III could take anywhere from 16 hours to 960 hours to break the code, but then showed how various companies (including Adobe) had made poor implementation decisions that made the system for more susceptible to hacking.
The next morning, he found himself on the receiving end of an FBI interrogation. Elcomsoft president Alexander Katalov described Skylarov's arrest this way:
On Monday morning, at around nine AM, he and another one of our employees, Andrei, were leaving the hotel for the airport. At the exit of the hotel they were approached by two men. They showed their FBI IDs. Dmitryi was immediately hand-cuffed. And they immediately took him and Andrei to different rooms. With Andrei, they simply talked—for about a half an hour, they talked with him about what's what, after which they let him go. He tried calling me several times, but the calls didn't get through. Later he called the Moscow office, where it's about nine thirty, and from there we got mail with the notification of the arrest.
Controversy over the arrest exploded across the Internet. To many, the case exemplified the problems with the Digital Millennium Copyright Act (DMCA), which made most encryption circumvention illegal. (Intriguingly, Elcomsoft's president also claimed that his company routinely sold password cracking software to the FBI and other law-enforcement agencies.) It wasn't long before even Adobe had cold feet about the entire situation, and the company withdrew its request for prosecution.
"We strongly support the DMCA and the enforcement of copyright protection of digital content," said Colleen Pouliot, Senior Vice President and General Counsel for Adobe. "However, the prosecution of this individual in this particular case is not conducive to the best interests of any of the parties involved or the industry. ElcomSoft's Advanced eBook Processor software is no longer available in the United States, and from that perspective the DMCA worked. Adobe will continue to protect its copyright interests and those of its customers."
The government later dropped charges against Skylarov, but only under the condition that he cooperate with their case against his employer. On December 17, 2002, a California jury found Elcomsoft not guilty on all charges; Skylarov was free to return to his wife and two children in Russia (read an interview with him that ran just after the trial ended). Skylarov continued his work at Elcomsoft, which no longer publishes the Advanced eBook Processor, and has authored a book on software security.
At the movies
CSS
When the consortium behind the DVD standard drew up the format's specification, they included a basic DRM system called the Content Scramble System. CSS was designed to deter piracy, not by being particularly robust, but by being obscure (the DVD Copy Control Association provides no information on the subject until you take out a license from them). Quickly proving the adage that security through obscurity does not provide reliable protection, the encryption scheme was broken wide open in 1999. Exactly who cracked the code, however, remains something of a mystery.
Though Norwegian coder Jon Johansen is the name most often associated with the crack, a text file distributed with later versions of his software points out that Johansen was not the original code breaker (that honor goes to an anonymous German hacker). Johansen did make a name for himself, though, by helping to develop the DeCSS software as member of the group MoRE (Masters of Reverse Engineering).
DeCSS allowed PC users to strip their DVDs of the CSS encryption and then save the resulting files to their hard drives, and it landed Johansen in trouble with the Norwegian authorities. Despite multiple trials, the Norwegian judicial system ultimately ruled that Johansen's work was legal, in part because he was simply using it to exercise his fair use rights to watch DVDs under Linux.
Although the CSS algorithm was first broken by obtaining a key from a commercial DVD player, analysis of the cipher showed that it could be broken without trouble by brute force approaches. The cipher's key length was only 40 bits, a number dictated by US export regulations in place when the DVD specification was created. Due to the cipher's design, though, it proved much easier to crack than any 40-bit cipher should be.
DVD players are factory-built with a set of keys. When a DVD is inserted, the player runs through every key it knows until one unlocks the disc. Once this disc key is known, the player uses it to retrieve a title key from the disc. This title key actually allows the player to unscramble the disc's contents.
The decryption process might have been formidable when first drawn up, but it had begun to look weak even by 1999. Frank Stevenson, who published a good breakdown of the technology, estimated at that time that a 450Mhz Pentium III could crack the code in only 18 seconds—and that's without even having a player key in the first place. In other, words a simple brute force attack could crack the code at runtime, assuming that users were patient enough to wait up to 18 seconds. With today's technology, of course, the same crack would be trivial.
Once the code was cracked, the genie was out of the bottle. CSS descramblers proliferated, spawning a true alpha geek contest to break the encryption in the fewest lines of code; 434 bytes appears to be the current record.

#define m(i)(x[i]^s[i+84])<<
unsigned char x[5],y,s[2048];main(n){for(read(0,x,5);read(0,s,n=2048);
write(1,s,n))if(s[y=s[13]%8+20]/16%4==1){int i=m(1)17^256+m(0)8,k=m(2)0,j=m(4)17^m(3)9^k
*2-k%8^8,a=0,c=26;for(s[y]-=16;--c;j*=2)a=a*2^i&1,i=i/2^j&1<<24;for(j=127;++j<n
;c=c>y)c+=y=i^i/8^i>>4^i>>12,i=i>>8^y<<17,a^=a>>14,y=a^a*8^a<<6,a=a>>8^y<<9,k=s
[j],k="7Wo~'G_\216"[k&7]+2^"cr3sfw6v;*k+>/n."[k>>4]*2^k*257/8,s[j]=k^(k&k*2&34)
*6^c+~y;}}

Because the CSS system could not be updated once in the field, the entire system was all but broken. Attempts to patch the system (such as Macrovision's "RipGuard") met with limited success, and DVDs today remain easy to copy using a multitude of freely available tools.
In the end, CSS may have halted piracy of the most casual sort, but the general result was to make criminals out of consumers who wanted to back up their disc collections or remove the annoying but unskippable segments at the beginning of movies. Even though it's now simple to crack, most consumers don't know how to do it or do not own the proper equipment.The point is as simple one: DRM schemes do not need to be uncrackable to control people's behavior.
AACS attacks!!!
DVD encryption proved easy enough to break, but we'll soon have the chance to see how robust the next generation of video encryption schemes proves to be. AACS, the basic encryption present on both Blu-ray and HD DVD titles, is supposed to be a major step up from CSS. The scheme was put together by some of the biggest names in the industry, including Intel, Microsoft, IBM, Sony, and Disney. The spec is administered by the AACS Licensing Authority, which will attempt to control the format by licensing equipment manufacturers.
AACS relies on the well-established AES (with 128-bit keys) to safeguard the disc data. Just like DVD players, HD DVD and Blu-ray drives will come with a set of Device Keys handed out to the manufacturers by AACS LA. Unlike the CSS encryption used in DVDs, though, AACS has a built-in method for revoking sets of keys that are cracked and made public. AACS-encrypted discs will feature a Media Key Block that all players need to access in order to get the key needed to decrypt the video files on the disc. The MKB can be updated by AACS LA to prevent certain sets of Device Keys from functioning with future titles—a feature that AACS dubs "revocation."
This decision certainly makes it harder for manufacturers to design insecure devices or to purposely allow consumer workarounds (which was common for DVD region coding). Now, companies who do such things could potentially find their players unable to access new titles, something sure to infuriate buyers. As AACS drily notes, a device maker had better "treat its Device Keys as highly confidential, as defined in the license agreement."
AACS also supports a new feature called the Image Constraint Token. When set, the ICT will force video output to be degraded over analog connections. ICT has so far gone unused, though this could change at any time.
A more consumer-friendly feature is Managed Copy, which studios can use in order to let consumers make protected digital rips of their media for use on HTPCs and home media servers. These will still be encrypted and locked down, but at least people will be able to back up their HD DVD collection without breaking the DMCA.
While AACS is used by both HD disc formats, the Blu-ray Disc Association (BDA) has added some features of its own to make the format "more secure" than HD DVD. The additions are BD+ and ROM Mark; though both are designed to thwart pirates, they work quite differently.
While the generic AACS spec includes key revocation, BD+ actually allows the BDA to update the entire encryption system once players have already shipped. Should encryption be cracked, new discs will include information that will alter the players' decryption code. The BDA describes the feature this way:
The BDA also adopted "BD+", a Blu-ray Disc specific programmable renewability enhancement that gives content providers an additional means to respond to organized attacks on the security system by allowing dynamic updates of compromised code. With these enhancements, content providers have a number of methods to choose from to combat hacks on Blu-ray players. Moreover, BD+ affects only players that have been attacked, as opposed to those that are vulnerable but haven't been attacked and therefore continue to operate properly.
The other new technology, ROM Mark, affects the manufacturing of Blu-ray discs. All Blu-ray mastering equipment must be licensed by the BDA, and they will ensure that all of it carries ROM Mark technology. Whenever a legitimate disc is created, it is given a "unique and undetectable identifier." It's not undetectable to the player, though, and players can refuse to play discs without a ROM Mark. The BDA has the optimistic hope that this will keep industrial-scale piracy at bay. We'll see.
Although HD DVD and Blu-ray are brand-new, trash talk has already begun. Jon Johansen has already claimed that he will crack AACS by 2007, and has registered the deaacs.com domain. Stay tuned
The front door's not the only way into a house
Looking back over a few of the best-known hacks of the past decade, a few key points stand out. First, DRM schemes are like houses: they have many ways in. If the front door is locked, there might be an open window. If the windows are bolted, check for a house key under the fake rock by the geraniums. If the main floor is secure, do the unexpected—slide down the chimney.
The ingenuity shown by the various hackers is a testament to the creativity of the human spirit (and to the desire for unencumbered media). The CSS algorithm can be cracked by a brute force attack in a couple of days, but it's the exception here. Most ciphers no longer succumb to brute force attacks in any reasonable amount of time, so hackers have looked for other ways to get at the content encrypted inside. Grabbing the key works well, but can be terribly complicated to do; sometimes it's just easier to let the decryption code do all the work, then grab the unencrypted stream from memory.
Technical measures aren't the only way to "hack" a DRM scheme. Our survey has highlighted the role that public pressure can play in having the locks removed and the house thrown open (consider Adobe and SunnComm, for instance). Superior encryption is worthless to a company if it ends up compromising their bottom line and brand reputation. Finally, some of the simplest attacks require nothing more than a marker or a finger. Both techniques show that a good hacker knows her technology, but she also knows when not to overengineer things.
The point here is that there are many ways to bypass DRM. Whether or not any particular method counts as a true "hack" hardly seems to matter, since they all accomplish the same goal. As Bruce Schneier tells Ars,
Does it matter if someone cracks the algorithm or the system? All the person cares about is if he can bypass the DRM system. If he can do that, the system is broken. Which particular piece broke is academically interesting, but operationally irrelevant.
The future: "lawsuit DRM"
A statue of Stalin flanked by grateful peasants used to tower over Prague. Perched above the Vltava on the Letna plain, it was the largest group statue in all of Europe at the time of its construction. It took years of manpower to build, and when complete, the beaming visage of "Papa Josef" dominated the skyline. Several short years after its completion, however, Stalin fell from grace in Communist circles and the massive monstrosity was torn down.
Many anti-DRM zealots talk about DRM technology as though convinced it will follow the same trajectory—massive spending and labor, years of careful construction, then ripped apart a few days later because no one wanted it after all. It's possible that this will be the case, but such a sanguine prediction could also be dead wrong. Though DRM's brief history has shown that formats can be cracked, few of the tools to do so have entered the popular consciousness, and the ones which have are illegal in the US. The force of law (and the risk of lawsuits) combined with the obscurity of most cracking tools means that even DRM solutions which are easily cracked can be effective at preventing casual piracy. Such technologies end up controlling only the behavior of legitimate users; those who want free copies of Dude, Where's My Car? from BitTorrent won't be deterred.
Although it doesn't stop piracy, this level of DRM may be good enough for most labels. It helps to instill a belief in the public that they can only do with their media what their media allows them to do (as opposed to exercising fair use rights, which may prove more broad than DRM restrictions). This type of thinking will enable media companies to better monetize their core products by selling them multiple times—movies on DVD, then again on Blu-ray, then once more for PSP, maybe a fourth time for your iPod, and a couple bucks for the privilege of time-shifting.
DRM need not be complicated to accomplish this. Even simple and highly-breakable encryption schemes have thrown up a legal barrier (courtesy of the DMCA) that will deter many average Americans from backing up their DVD collection (and then buying Finding Nemo a second time when the toddler sticks the disc in the toaster). HDCP, which is a form of "link encryption" that will guard the signal between HD players and televisions (or monitors), functions this way. Ed Felten, a computer science professor at Princeton, has analyzed the encryption and concluded that "the bottom line is clear. In HDCP, 'security' technologies serve not to disable pirates but to enable lawsuits. When you buy an HDCP-enabled TV or player, you are paying for this—your device will cost more and do less."
Sometimes "good enough" is good enough
DRM's not going away anytime soon, and newer techniques such as BD+ promise to make future technologies even more difficult to hack for long periods of time. Does that mean that hacking such systems will soon be pointless? Bruce Schneier says no.
"Do systems that allow you to change locks on your front door make burglary pointless?" he asks. "Of course not. The goal is to break the scheme and produce copies of the movie without DRM. Hacking DRM schemes will be fruitful as long as DRM schemes exist."
Given the time and money that developing strong DRM can take (and the fact that pirates will crack the format anyway), media companies may simply settle for more basic systems that still exert "good enough" control. After all, if "hacking" can take diverse forms, so can DRM, and the content industry's victory with the DMCA shows that effective control is not simply designed in the laboratory, but in the halls of power as well.

sa: http://arstechnica.com/articles/culture/drmhacks.ars

[bbmf]

AMD to Demonstrate Quad-Core Chips by Year End
Advanced Micro Devices said during its second quarter of fiscal 2006 conference call that it would not launch quad-core processors till the middle of next year, but will showcase such chips by the end of 2006. This confirms earlier unofficial information that there are no quad-core chips in AMD’s roadmap till 2H 2007.
“Leveraging the scalability of the Direct Connect architecture, we also plan to demonstrate our next generation microprocessor core in a native quad-core implementation before the end of the year,” said Dirk Meyer, AMD’s president and chief operating officer.
Earlier it was reported that the first quad-core chip from AMD is code-named Deerhound, which will be intended for socket F infrastructure and will have shared level-two cache along with dual-channel registered DDR2 memory controller. The new chip is expected to use the new code-named K8L design, which features some improvements, according to Mr. Meyer.
“We have a new core under development. The first instantiation of which will be in a quad-core form, to be launched roughly mid ‘07. What I said is we will demonstrate that by the end of the year,” said Dirk Meyer, when asked to clarify the production schedule of the quad-core chips.
Earlier the company planned to showcase dual-core microprocessors in the middle of the year, particularly, in August, when it is expected to introduce its new server platform and socket F infrastructure.
When AMD rolls out dual-core processors with built-in virtualization hooks midyear, the company also aims to demo quad-core processors running on its current server platform,” Marty Seyer, a senior vice president of AMD, said in an interview.
The chipmaker now claims that the chips will be showcased by the end of the year without giving any particular timeframes, which may signal that the company currently has no quad-core chip samples mature enough for public showcase.
According to AMD, K8L includes a quad-core design for servers, workstations and high-end desktops, and a dual-core design intended for mainstream desktop markets. These next-generation processors will be built using AMD’s 65nm Silicon-on-Insulator (SOI) fabrication process, and include a broad range of functionality and micro-architectural improvements, including a new ability to dynamically alter the frequency of each core on the chip to match application workloads and thereby reduce overall power consumption.
AMD’s larger rival Intel Corp. recently said it would commercialize its first quad-core microprocessors as early as by the end of this year, about six months ahead of AMD, which is likely to add competitive pressure on the world’s second largest supplier of x86 microprocessors. But AMD believes that Intel’s approach to put two dice on a single slice of substrate to build a quad-core processor is inefficient and AMD’s “native quad-core” design will provide better performance and scalability for servers.

sa: http://www.xbitlabs.com/news/cpu/dis...721230935.html

[bbmf]

From mainframes to microblades, farewell to GHz CPUs
Max Fomitchev
Life goes in circles, or in spirals to be more precise. Thus to get a glimpse of the future you perhaps should look in the past. Computing has become an inseparable component of our lives and therefore is a subject to the same law of cycles and spirals. So let's start in the 1940s, at the dawn of electronic computing when ENIAC was a pinnacle of scientific engineering.
Initially occupying whole buildings, then scaling down to individual rooms and towering boxes computers of 1950s, '60s, and '70s were essentially mainframes: Large and powerful, special-purpose, accessible to few. This centralized approach to computing changed dramatically in the late '70s and early '80s with the introduction of microcomputers such as Apple II and IBM PC. All of a sudden computers ceased to be shared resources built for a particular purpose and instead became personal tools for facilitating general-purpose tasks and throughout '80s and '90s began to occupy our desktops, bedrooms and closets.
This trend of decentralized computing met a subtle reverse in '90s when the Internet and World Wide Web provided a means for integrating decentralized computational resources into a unified client-server environment. In reality, what seems like 60 years of technological advancement represents a full evolutionary cycle: We started with shared computational resources occupying rooms of equipment and through brief desktop detour arrived at shared computational resource model built on the backbone of intranet/Internet. There is unquestionable numerical difference between what we 30 years ago and what we have now in the sense that computers now are used by much larger population and for a far wider range of tasks. There is a characteristic difference too: we kept our desktop PCs and we use them for more then mere terminals. In fact computing did not just come a full circle; it came a loop of spiral: We have not really come back to good old centralized computing but rather to arrived at distributed computing model. Although a bulk of work may be done by centralized resources such as servers providing computational services, our desktop PCs and client workstations handle independently multitude of tasks.
Our equipment rooms are different too from what we had decades ago: instead of one large computer modern day data centers are filled with hundreds and thousands of servers both rack mount and blades. Such evolutionary change parallels that of the evolution of electronic components: Discreet elements were gradually replaced with integrated circuits just as individual mainframes are now replaced with blade server racks. Extrapolating the parallel further we may expect even tighter-integrated "microblades" to arrive in the near future when we master computer integration to the same degree as we have mastered integrated circuits.

Evolutionary Changes
Internally, computers are undergoing cyclical evolutionary changes as well: CPUs gradually evolved from spending tens if not hundreds of cycles on individual instruction to just one cycle per instruction (scalar architecture, for example). Then introduction of additional execution units allowed CPUs processing several instructions per clock cycle thus exploiting instruction level parallelism (superscalar architecture, for instance). Later several CPUs were crammed on motherboard (multi-processor architecture). Now several CPUs are fused together in a single multi-core package, and several such multi-core chips can be installed on a single motherboard. So in the end a data center is filled with clusters of stacks of server blades with each blade sporting one or more multi-core super-scalar CPUs. So we have at least five different levels of integration:

Cluster
Server
CPU
Core
Execution unit

with four levels typically found on desktop PCs. But why do we need this complexity and how did it come into being?
As computer clock speed increased from kilohertz to gigahertz so did out imagination and understanding of what can be done with this computational power to serve our needs; for example, provide entertainment (at home) and boost productivity (which is a practical reason for computers in the workplace). Originally computers were designed to serve a clearly defined special purpose and therefore were meant to perform specific single task. When computer power grew beyond immediate needs multi-tasking was invented to allow multiple users access to the spare computational resources. But when the hardware costs were reduced to consumer level personal computers came out, and they quite naturally were designed to be single-tasking.
The first mass-produced personal computers were quite slow and the need for performance increase of microcomputer processors was justified at first: We wanted good response from desktop applications and occasionally we wanted to play arcade games. And 4.77 MHz of PC XT was not always good enough for the purpose. So as CPU power grew to meet specific tasks we wanted our PCs to perform it became too much for general tasks such as text editing or spread-sheeting. That extra power just as in the case of old mainframes led to the adoption of multi-tasking operating systems on desktop and personal computers. We had extra power and we wanted to do something with it.
Ironically, mass adoption of multi-tasking operating system on PCs (Microsoft Windows, for instance) coincided with the introduction of graphical user interface. Thus formerly more or less satisfactory CPU performance became vastly inadequate and spurred a race to increase CPU performance necessary to compensate for inefficient software. The whole transition from single-threading text-based DOS programs to graphic interfaces and multithreading Windows resulted in unprecedented bloating of software code and general system slow down due to lacking graphic and disk I/O performance. Erroneous programming paradigms such as dynamically loaded libraries, dynamic memory allocation, shared components, inefficient object-oriented programming, and multi-layered libraries also greatly contributed to the slow down. All these inefficiencies instantly justified the need for further CPU performance increases. Now we needed faster computers just to run our operating system and new versions of old software burdened with graphic user interfaces.
Thus paradoxically desktop computers of '80s initiated a major leap in Wirth's Law which states that software is getting slower faster than computers are getting faster. Perhaps the first loop of Wirth's Law spiral was objective: Initial CGA and EGA hardware and CPU performance of 12-16 MHz was barely enough for running programs with complicated graphic interfaces. However, further unraveling of Wirth's Law was completely subjective in the sense that slowdown of software that occurred further resulted from our attempts to boost programmer productivity by employing various "coding techniques" that promised simplicity at the cost of efficiency.

Software Performance Versus Developer Productivity
Indeed the first mainframes were programmed directly in machine code and a bit later in assembler. Severe memory limitations and simplicity of original instruction set (PDP 11 is a classical example) and relative simplicity programming tasks at hand resulted in highly efficient if not bare bone code that, unfortunately, was difficult to write. When in late 1950s computers became fast enough to relieve some of the coding burden from the shoulders of programmers high level languages were developed such as Ada, Algol, Fortran and C. While sacrificing code efficiency big time these high level languages allowed us to write code faster and thus extract more productivity gains from computers.
As time passed we kept sacrificing software performance in favor of developer productivity gains first by adopting object-oriented languages and more recently settling with garbage-collected memory, runtime interpreted languages and 'managed' execution. It is these "developer productivity" gains that kept the pressure on hardware developers to come up with faster and faster performing processors. So one may say that part of the reason why we ended up with gigahertz-fast CPUs was "dumb" (lazy, uneducated, expensive -- pick your favorite epithet) developers. Even now major OS release (such as upcoming issue of Windows Vista) seems to be main reason for computer upgrades because new software runs slower doing the same tasks as the older software it replaces. Of course, the new software usually does much more than the old one. So an objective reason for faster computers is higher expectations and expanded feature set of the new software. After all there are some mission-critical applications that really demand performance. Database applications on server side and games on desktop size are good examples of such apps that drive hardware development towards faster performance objectively.
Still, if you look at your desktop OS now it is unbelievably bloated. For instance, when running Windows XP under normal circumstances you will easily count 50 processes, 500 threads and only about 5-10 percent CPU utilization. This is what I get when I type this article in Word 2003 running on Athlon XP 3200 under Windows XP. Thus a 10 times less powerful CPU would have equally well satisfied my requirements for browsing and typing... Yet computers in general and CPUs in particular keep getting faster and faster driven both by developer productivity needs and the requirements of mission-critical applications.
Incredibly a new factor kicked in that is threatening to curb raw clock speed increases -- runaway power consumption. It is not unusual for a modern CPU to dissipate in excess of 100 Watts, which in the case of data centers translate into tens of millions of direct power and cooling costs. So on one had we have a habit (but rarely a need) for higher performance and on the other hand we have a looming fossil fuel crisis, global warming and rising energy prices. Shall we finally stop racing the clock speed?
Apparently, the trend for higher clock speed has already been reversed when cooler yet efficiently running AMD's Athlon processor managed to win sizeable market share from hotter and faster by clock speed Intel's Pentium 4. So how are we going to keep up with performance demands without liberal increases in CPU clock frequency?

Maintaining Performance
Well, there are many ways to maintain performance. The first one -- exploitation of instruction level parallelism -- resulted in creation of super-scalar processors that we see today. Theoretically any modern CPU whether from Intel, AMD, IBM, or Sun can process and retire multiple instruction per cycle due to multiple parallel internal execution units. Funny enough, instruction-level parallelism does not yet allow sustained performance of substantially more that 1 instruction per cycle (IPC) on general benchmarks due to memory latency and branch misprediction penalty that stalls even the fastest CPUs more than half the time (source:Intel). Only highly-optimized tests or special-purpose code is capable of 3x to 5x performance boost warranted by multiple execution units. Practical gains due to architectural improvements of cache coherency or branch prediction amount to mere 5 percent in general. Long multi-stage execution pipelines that were developed to achieve higher clock speeds and inadequate memory performance created a situation when CPU can process data faster than the data can be supplied. So the trend for higher clock speed has already reversed in favor of shorter pipelines and better memory throughput. The best example of pipeline shortening is UltraSparc T1 processor with its six stage pipeline as opposed to 31-stage Pentium 4 models (Athlon XP has 10-stage pipeline and Intel's new "Woodcrest" server chip as only 14). Extrapolating the trend it is reasonable to expect CPU frequency to roughly remain the same while the CPU performance will increase due to pipeline shortening and emphasis on memory subsystem performance improvements.
Still, there is a hard limit for instruction-level parallelism, which makes it difficult in practice to keep individual execution units inside a CPU busy. Thus to improve CPU efficiency two alternative approaches are currently being pursued. One approach is super-threading (or Hyper-threading if we use Intel's terms), which allows CPU to process several parallel threads simultaneously switching from one thread to another when a stall occurs. UltraSparc T1 takes this approach to extreme by executing four threads on each core (with 32 threads on 8-core chip), switching threads in round-robin manner and when a stall occurs. While super-threading certainly boosts performance of multi-threaded applications speculative threading is pursued for improving performance of critical single-threaded applications. Intel is highly involved in speculative threading research and offers a Mitosis technology that with the help of compilers designates threads most suitable for speculative execution. AMD is developing similar technology, although the company is more tight-lipped about it. Still many rumors are circulating about AMD's clandestine "inverse hyper-threading" technology allegedly capable of uniting two individual CPU cores into a single CPU super-core CPU that would crunch single-threaded applications with a considerable performance boost. Yet the only piece of evidence on AMD's involvement with speculative threading that so far surfaced is infamous U.S. patent # 6,574,725 that looks like hardware support for speculative threading in the vane of to Intel's Mitosis. So with clock-speed increases effectively curbed by power consumption concerns most likely performance gains on upcoming CPUs would be due to super-threading (server chips) and speculative-threading (desktop chips).
There is another approach for boosting instruction-level parallelism, which has been pursued on and off by various commercial and government entities. I mean very-large instruction word (VLIW) or explicitly-parallel instruction set (EPIC) computing. First successful application of VLIW concept can be tracked back to early 1980s when a group of Russian engineers lead by Boris Babayan (who is now an Intel fellow) development a series of Elbrus supercomputers that were produced as a part of the anti-ballistic missile defense system deployed around Moscow. Massive performance gains warranted by proper application of VLIW concept allowed Elbrus machines to overcome manufacturing and technological limitations and beautifully serve their purpose. Remember that these were a special-purpose computers running hand-optimized code.

VLIW
Commercial applications of VLIW concept in the U.S. were less successful: Multiflow Computer went down in 1990 and Intel's EPIC/Itanium adventure of late 90s and today proved to be far from successful. The reason for VLIW failure on general purpose computers is the lack of compilers, cross-compilers and automatic code optimization techniques. Intel is still heavily involved in honing EPIC compilers for Itanium (with Babayan's current team and Intel's Israeli's office heavily involved). Yet the state of current technology is such that current VLIW/EPIC compilers are not yet good enough for general purposes and therefore theoretically possible performance gains are almost never achieved (VLIW processors can execute as many as 32 instructions in parallel if a compiler can find and schedule that many). More recent attempt by Transmeta was also unsuccessful and for the same reason, although it's new Efficieon CPU looks more promising than flopped Crusoe. Still, with Itanium disappointment tarnishing commercial VLIW prospects perhaps permanently we are unlikely to see more general-purpose VLIW computers, but instead are likely to seem them in niece markets employed for solving a very limited set of special-purpose tasks.
Quite another alternative to VLIW that is already sprouting profusely is multi-core CPUs. Both Intel and AMD have been shipping dual-core chips for quite some time now with quad-core chips promised in 2007. Sun is already shipping 8-core UltraSparc T1 chips, while Rapport Inc. and IBM have already announced development of Kilocore technology that allows combining as many as 1,024 8-bit processors with a PowerPC core on a single low-cost chip. Thus extrapolating current trends we are likely to see further profusion of multi-core CPUs from all leading manufacturers, especially for server markets. Chances are that as number of on-chip cores grow the cores itself would become more simple and less-deeply pipelined (kind of like UltraSparc T1 is doing already). We are also likely to see some dedicated co-processor-like cores suitable for performing SIMD/multimedia instructions while other cores might be deprived of such capacity in favor of improved energy efficiency and increased overall number of cores.
Perhaps the most noteworthy point is that we are unlikely to see dramatic single-threaded code performance improvements unless a way of frequency increases is found that does not result in the market increase in power consumption (for example, new manufacturing technology in the vane of IBM's recent report of experimental SiGe chips running at 350 GHz at room temperature and at 500 GHz when chilled by liquid helium).
And the truth is that there is no compelling need for further raw CPU speed increases for the following key reasons:
Computers are already much more powerful than most common tasks require.
Code efficiency is at all time low and potentially hide at least a order of magnitude performance boost if we just optimize the code.
Memory and I/O bottlenecks are most common causes of slow-down.
What is amazing is that for a long time we have been using only a handful of CPU models under the aegis of general purpose computing. Further more we thought that a better CPU makes a better computer, which is no longer so. What seems to be more important now is overall system design rather than just CPU design, and we are likely to see more system and CPU specialization (and models) targeting different application areas.

Emerging Processor Lines
We already see three major lines of processors targeting mobile, desktop, and server markets. This trend is likely to continue and result in appearance of even more processor lines optimized not only for various segments but for various applications or intended uses as well. For instance, for application servers we may see Intel and AMD delivering vastly multi-core CPU with good integer capacity and dedicated encryption/decryption hardware in the vane of Sun's UltraSparc, while in mobile market we may see stripped-down extra-low-power CPUs that ensure very long batter life, perhaps with finer frequency scaling similar to what coarse-grained AMD's PowerNow! technology does now. There certainly seems to be a room for lower-performance CPUs for ultra-mobile computers since most of them are used for reading, browsing and other simple tasks that do not require much of CPU power (specific tasks such as multimedia encoding/decoding and 3D graphics are already partially offloaded to dedicated hardware and are likely to be even more confined to specialized chips in the future).
So focusing on mobile CPU market it is clear that power efficiency and that not only of CPU but of the entire system is likely to be much more important than raw processor speed. After all most mobile users are not likely to exploit potential CPU performance to the fullest extent unless we throw at them really bad code. Almost commodity pricing on computational power today is such that consumers can afford buying more and more specialized hardware that is better suited for a particular purpose thus fulfilling Bill Gates' vision of computers in every pocket. This is in fact already happening as we all are grabbing iPods, cell phones, PDAs, and BlackBerry devices to complement our laptops and desktop PCs. No more one-size fits all. This is the most certain prediction that one can make about future CPUs. We shall see more and more specialized models and not necessarily more powerful ones. Thus as far as mobile market is concerned we might see CPUs with more finely grained frequency control that responds to idle time, variable rotation rate hard drives and possibly stripped out of some advanced features such as enhanced multimedia processing instructions in favor of dedicated hardware performing the latter tasks.
In fact AMD is already making some steps in this direction with its upcoming 4x4 platform and open specification enabling 3rd party co-processor design. In the long term it makes little sense to burden CPU with DVD playback or SSL encryption. These and similar tasks should and with time will be handled completely by dedicated hardware that is going to be far more efficient (power and performance-wise) than CPU. Further variety of coprocessors will allow enhanced physics and environmental effect experience for gaming enthusiasts and improved performance for scientific/multimedia applications. Thus the role of CPU is likely to diminish with time living little reason for further clock-speed improvement.
Frankly the role of CPU as a jack of all trades started to wane with the advent of GPUs. 3D graphics was the most compelling reason to boost CPU power. Now PCs typically have a dedicated processor (or two in the case of AMD's 4x4 platform) that is far better suited for the task. Similarly most music/multimedia hardware relies on its own expansion boards outfitted with custom logic/DSP processors (take ProTools or Creamware products, for example). And with time we are likely to end up with a motherboard design that would contain numerous specialized chips or co-processors designed with a single task in mind. So in this respect we are back to the single-purpose computing we have started with, although such return is a mere new loop in the spiral.
Ironically, return to special-purpose computing results in further relaxing of requirements for higher processor performance: special-purpose code is usually better optimized and thus can perform equally well on much slower CPUs. In reality most hand-held devices are powered with few hundred MHz CPUs that are capable of providing similar experience (save for small screen and tight keypad) we have with our gigahertz-fast desktop PCs. Similarly specially-designed DSPs are far better for MPEG playback or sound processing than general-purpose CPU that can do the same running at high GHz.
In other words, what is likely to happen is that CPU frequency increases will become very modest in the near future. As hardware manufacturers compete for the markets we are likely to see less and less general-purpose and more and more specialized hardware for various purposes. Perhaps in 10 years today's Athlon and Xeon CPUs would seem like dinosaurs, hot, big, and less than bright, with the role of CPU in the computer reduced from the do-it-all-yourself to coordinate-the-work-of-others.

Conclusion
There is another compelling reason to believe that big, hot, and insanely fast CPUs will die out due to natural selection. As people become more and more aware of "green" concepts and conscious of power consumption our eyes will finally open to extremely bloated code that out GHz-rated CPUs execute at the same rate as MHz-rated processors in specialized devices. The proper question to ask would be "How much power does your software require?", where power means electricity with the implication of the high energy cost. Indeed that would mean that slow and bloated software is expensive software for it requires CPU to run at full blast. To make this point more clear think of a datacenter with a thousand blade servers with each server sporting several CPUs and hard disks. Bloated and slow software that we have today implies that the operating cost of the datacenter is high for it needs a thousand blade servers, thousand terabyte disks, and gigabytes and gigabytes of memory with cooling and power cost of 10 million a year. Now let's say if we are to optimize our software to reduce RAM, disk, and CPU performance requirements by an order of magnitude (which is easily achieved if we scrap interpreted and otherwise "managed" code with inefficient memory management model and multi-layered libraries and invest in compiler and optimizer development) and reduce the number of servers 10 times? Or instead replacing huge blade servers with gigahertz CPUs with compact pocket-size microblades outfitted with megahertz-rated CPUs, few megabytes of RAM and a microdrive?
Needless to say, there is amble room for software optimization that has been ignored for decades since the increases in CPU performance allowed us to neglect it. Yet now the situation with energy resources is such that slow and bloated software means higher costs both in direct electric power required by CPU to process it and indirectly in power consumed by RAM, enormous hard drives and cumulative cooling costs. Furthermore recent tendency to aggregate multiple software components running on shared computational resource (that is, a server) under control of a multitasking OS should be reversed in favor of completely isolated software components running on low-power dedicated hardware. Thus if we are to begin optimizing our code we are likely to see blade server racks replaced with microblade server racks where each microblade is performing a dedicated task, consuming less power; and where the total number of microblades is much greater than the number of initial "macro" blades.
Indeed such complete isolation of software components (database instances, web applications, network services, and the like) that are currently squeezed together on the same server should greatly improve system robustness due to the possibility of real-time component hot-swap or upgrade and completely eliminating software installation, deployment and patch conflicts that plague large servers of today.
When and if that happens depends on two factors energy costs and code optimization efficiency. The former drives the latter. Therefore further increase in energy prices is likely to result in gradual reduction of the role of CPU in computer system, more optimized code and return towards single-processor/single-task special-purpose computing paradigm. On the other hand this vision may never materialize if a technological breakthrough occurs on manufacturing side that would allow further CPU speed increases without the increased energy dissipation (quantum computing, advances in superconductors, photonics, and so on). However, one thing is clear -- the role of CPU performance is definitely waning, and if a radical new technology fails to materialize quickly we will be compelled to write more efficient code for power consumption costs and reasons.

sa: http://www.ddj.com/dept/64bit/190400539?pgno=1

[bbmf]

But there's still a chance Vista will be delayed againMicrosoft announced that it will spend roughly $900 million in sales and marketing initiatives to drive Windows Vista, Office 2007. During its financial results call, Microsoft CFO Chris Liddell said that his company will be spending roughly $450 million to market the three key products and another $450 to develop its sales team.
According to the announcement, Microsoft received a solid number of renewals for long term licensing of its Windows and Office software products. This in turn, injected a good sum of long term revenue for the software giant. The extra revenue said reports, contribute to the decision to invest heavily in sales and marketing. Microsoft announced early on in Windows Vista's development that it would be the largest project the company ever undertook.
Microsoft Office is already well under way and is practically just waiting to be released but many analysts are looking to Windows Vista to kick start the 2007 year. Microsoft delayed Windows Vista several times over the course of the development process and despite its projected release date of Q1'07, Bill Gates hinted that the company would not hesitate to announce further delays if it meant that Microsoft would be able to further improve the OS.
Liddell said "the shift to annuity signals stronger customer demand for Office 2007. We had strong quarter for volume licensing. Renewals were on the high end of our range of 66 to 77 percent, there's growing customer demand for new products." Despite this, Liddell admitted that next quarter revenues won't be as strong because Microsoft is at the end of its major renewal cycle.

SA: http://www.dailytech.com/article.aspx?newsid=3439

[bbmf]

Core 2 Duo processors will bring dual-core capabilities to desktops and laptops this week
Intel will this week officially launch its Conroe and Merom dual-core chips, for desktop and laptop systems respectively. The chip giant also said it plans to bring forward the release of its first quad-core processors to the end of 2006 instead of the first half of 2007.
The Conroe and Merom parts, both officially called Core 2 Duo, complete the roll-out of Intel’s next-generation dual-core microarchitecture, following the launch of the Xeon 5100 Woodcrest server chip in June.
Conroe is available immediately, and Merom is set to ship in August. The releases also signal the beginning of the end of the Pentium brand, which will be cut in price to target lower-priced PCs, Intel said.
Earlier this month, Intel disclosed clock speeds and pricing for its E6700, E6600, E6400 and E6300 Conroe chips. Similar details for Merom have yet to be disclosed, but it is expected to ship in three versions with 4MB of L2 cache, clocked at up to 2.33GHz.
The dual-core chips will soon be followed by quad-core designs featuring four processors on a single chip. Intel last week said it was bringing forward the release of its Kentsfield and Clovertown chips to the end of 2006. Kentsfield will be aimed at desktop systems, while Clovertown will join Intel’s Xeon server line-up.
Meanwhile, rival processor vendor AMD expects to release a quad-core Opteron in the first half of 2007. The chip will use the same socket as the current dual-core part. “It’s the same power envelope and the same thermal characteristics,” said Richard Baker, AMD’s marketing manager.
IBM is next Tuesday expected to respond to Intel’s Montecito dual-core Itanium with the release of new Power5+ processors and systems. The IT giant will detail plans for a refresh of its Power5+ processor – using four cores and very large memory caches – for Series p Unix servers.
And last week, Sun revealed that it has successfully booted a server running the Niagara II next-generation UltraSparc processor. Niagara II runs eight cores, each capable of processing eight threads. Commercial availability is scheduled for next year.

sa: http://www.itweek.co.uk/itweek/news/...ntium-era-core

[bbmf]

There is a lot of news surrounding ATI at the moment, especially with the AMD buy out confirmed earlier today.
Intel has pulled ATI's chipset license, meaning that there will be no more ATI (or rather, AMD) chipsets for Intel processors after the end of the year.
There will still be time for one more though, namely the upcoming RD600 chipset, which doesn't look like it will be cancelled, despite the earlier news. However, it is unclear whether the RD600 project will see the light of day though.
Intel has pushed the fact that its 975X chipset supports CrossFire quite heavily, especially with the Core 2 Duo launch. I think it is a fairly safe bet to say that Intel will contine to support CrossFire on its high end chipsets for the foreseeable future.
After all, Intel is keen to sell its own products, especially now it has the fastest chip on the market. Also, I don't believe that AMD would choose to lock CrossFire out on Intel's chipsets either, because it just wouldn't make financial sense - AMD/ATI will still profit from the video cards sold for use with Intel products.
Intel has already dropped support for CrossFire on its 965-series chipsets and it is unclear whether Intel will continue to design chipsets that support multi-GPU technologies. The company has been hinting at moving GPU tasks back onto the CPU for a few months now.

sa: http://www.bit-tech.net/news/2006/07...ipset_license/

[bbmf]

Let's keep the presses rolling: First, AMD and ATI formally announced their merger intents. Then, Intel ripped the carpet out from under ATI on the chipset front.
Now, AMD and ATI have announced their intentions for unified development by 2008 - this is quite forward thinking, considering that AMD and ATI are not one until ATI's share holders approve the deal.
The discussion all hinges from a slide AMD and ATI made about the benefits of the merger, which illustrates the move towards a "unified system" where chips can just be plugged in to meet various needs. The slide is a bit ambiguous, but it illustrates CPUs and GPUs being able to talk to one another very closely, and in differing quantities depending on needs.
This architecture design is very reminiscent of AMD's Torrenza technology proposed last year, which stressed very low latencies between core components like CPU, chipset, and RAM. The idea was to make the board able to handle other chips as coprocessors, like a GPU or maybe even physics. Initially, these coprocessors were to be built right into CPUs, though the possibility of seperate add-on chips has not ever been ruled out.
Speaking of GPUs, AMD have recently made their first GPU related patent, a move that helped to send up some flags for the merger weeks ago. The patent is for CPU and GPU on the same chip with an integrated, shared cache. This is a bold step much in line with the Torrenzo plan, and with ATI now able to work on the project openly, we can only guess that it will work pretty well.
There is no word as to the target market for the product at the moment, and we can expect that it will take a little while before the tech gets powerful enough to handle demanding graphics like games without heat issues. But even so, it's nice to dream of a graphics setup that removes any latency of the northbridge from the equation...

sa: http://www.bit-tech.net/news/2006/07...itecture_2008/

[bbmf]

With the debate over genetic cloning in full swing, hackers could not have cared less at a conference in New York City, where two presenters demonstrated the electronic equivalent of making a copy of an implanted RFID or radio frequency ID chip.
The point was to show just how easy it is to fool a detection device that purports to uniquely identify any individual.
Annalee Newitz (left) and Jonathan Westhues (right) presented their experimentations at the HOPE Number 6 conference in New York City in front of a crowd of hackers, tweakers and phone phreakers.
“This is the first time someone has cloned an human-implanted RFID chip,” Newitz said. “Since I have been chipped Jonathan refers to me as an implanted pet.”
Newitz said she has an RFID chip implanted in her right arm manufactured by VeriChip Corp., a subsidiary of Applied Digital.
“Their Web site claims that it cannot be counterfeited — that is something that Jonathan and I have shown to be untrue.”
The pair demonstrated the cloning process: Westhues held a standard RFID reader against Newitz’s arm to register the chip’s unique identification number.
Next, Westhues used a home-built antenna connected to his laptop to read Newitz’s arm again and record the signal off her implanted chip.
Westhues then takes the standard RFID reader and waves it past his laptop’s antenna. The reader beeps, showing Newitz’s until then “unique” ID. “It actually has no security devices what-so-ever,” Newitz said of VeriChip’s claims that its RFID chips can not be counterfeited.
VeriChip spokesman John Procter said in a phone interview that he had read about Newitz and Westhues work, but the company had not been able to review the evidence. He had no specific comment regarding their “cloning” project.
“We can’t verify what they may or may not have done,” Procter said, adding that: “We haven’t seen any first-hand evidence other than what’s been reported in the media.”
“It’s very difficult to steal a VeriChip … it’ s much more secure than anything you’d carry around in your wallet,” he added.

sa: http://blogs.reuters.com/author/nicfulton/

[bbmf]

We first heard about Australia/Silicon Valley based Omnidrive, an online storage company, in late 2005. Since that time we’ve extensively reviewed various online storage services, including rumored products from Google and Microsoft, as well as Amazon’s storage API solution (update here) for application developers. This space continues to heat up, to say the least.
Fast forward six months. Omnidrive is yet to launch, but they’ve continued to build out their service. Last week they invited in a new round of beta testers for their product, which includes an online and client interface (Windows only, Mac still in development). I’ve tried out the service and am posting a few screen shots.
The online interface (which is all I have tested so far) works very well although there are still a few bugs. One feature that I really like is the ability to set up a special kind of folder, called a “live folder” that is associated with a URL that contains a RSS feed. Any enclosures in that feed (images, sound files, whatever) are automatically uploaded to that omnidrive folder. To test this, I uploaded the URL to my flickr page, and the images contained in the feed (the last 20 pictures uploaded) were now copied automatically to my Omnidrive account. As I add more pictures to flickr these images will automatically sync with Omnidrive. This will work just as well with podcast and videocast sites, etc.
Omnidrive plans on sending hundreds of invites out per day until they open it up to the public, which they say will be sometime in September (expect delays).
They have also released a web services API along with toolkits and example projects for developers to build applications that access Omnidrive storage. With the API a developer can either build applications that existing Omnidrive users can use, or they can create their own users and use Omnidrive purely as a backend. The API extends to being more than just saving and retrieving a file with user management, payment management, media handling and the ability for the users of a partner application to use their desktop tools to store, retrieve and access files.
Pricing for the API has not yet been announced publicly but “will be competitive against S3 and other offerings” with a basic API account being free. Competitors Streamload and Mark Cuban-backed Box.net also have API offerings, and we plan a post in the near future comparing all four storage API solutions.
Sign up for the Omnidrive beta on their home page. The release notes for the latest version are on their blog.

sa: http://www.techcrunch.com/2006/07/25...-of-omnidrive/

[bbmf]

ARM has developed a system generator tool that it has already supplied to lead partners and will introduce generally in September that it says will allow platform providers to rapidly generate, by themselves, instruction-accurate virtual prototypes that are scaleable, reusable and fast enough to interact with in real time.
Part of the companys RealView series of tools, they will be demonstrated at this weeks Design Automation Conference in San Francisco. The tools are said to give both hardware and software engineers the power to create and modify their own system models, which keeps NRE costs under control and protects unique IP by dispensing with third-party consultancies.
ARM (Cambridge, England) says the speed of the generated virtual prototypes produced by the System Generator is comparable to the speed of currently available mobile devices, making it possible to test application software as it will appear on the end device, months before hardware is available.
For a highly differentiated consumer electronics product to succeed, electronics vendors need a way to generate application software, perform testing of hardware and application software integration and validate the end-user experience as early as possible, said Bryn Parry, general manager, Development Systems, ARM. He adds the tool provides a target early in the design cycle that enables these tasks.
The company says by using the System Generator silicon vendors, OS or software vendors or system OEMs working with SoC- or ASIC-based systems would be able to work on application software development in parallel with hardware development, reducing integration time.
The first version of the System Generator includes fast processor core models for the ARM926EJ-S, ARM1136JF-S and ARM1176JZF-S processors, as well as for a number of ARM PrimeCell peripherals.

sa: http://www.eetimes.com/news/latest/s...leID=191000176

[bbmf]

Samsung Electronics Co. Ltd. on Monday (July 24) announced the qualification of its 65-nm, low-power process technology for foundry customers.
The process will be ramped up within Samsung's 300-mm fab. The fab, dubbed the S1 Line, is located in Giheung, South Korea. The process is said to be compatible with Samsung's technology partners, IBM Corp. and Chartered Semiconductor Manufacturing Pte. Ltd.
Samsung's nine-metal-layer, copper-based process is said to consist of a family of transistors ranging from low-leakage to high-performance for low-power applications.
Complementing its new 65-nm low-power process technology, Samsung will offer its foundry customers full access to a portfolio of internal and licensed IP, which includes ARM cores, high speed interfaces and various mixed signal components.
"Dynamics in the semiconductor market are moving more companies to rely on high-end contract manufacturing," said Ana Molnar Hunter, vice president of technology for Samsung Semiconductor's foundry business, in a statement. "Our production ready advanced process technology offers our foundry customers the ability to integrate more features while maintaining very low power consumption and a fast ramp into high volume production."
After dabbling in the foundry business for several years, South Korea's Samsung is gearing up for a major thrust in the arena. This poses a threat for providers in China, Singapore and Taiwan.

sa: http://www.eetimes.com/news/semi/sho...leID=191000407

[bbmf]

Freescale has claimed that it has made a "breakthrough" in integrated circuit packaging with the announcement that its Redistributed Chip Packaging (RCP) method has allowed it to make a component that contains all the electronics for a 3G phone in a package measuring 25 millimeters by 25 millimeters.
Freescale (Austin, Texas) claimed that RCP can help produce a packaged chip component that is 30 percent smaller than a traditional ball grid array (BGA) component and added that RCP could replace BGA and flip-chip packaging as the dominant packaging and assembly approach for highly integrated chips.
Speaking at a technology forum in Orlando, Fla., Freescale Semiconductor chief executive Michel Mayer said that Redistributed Chip Packaging could eliminate wire bonds and blind vias in the dense circuit board implementations found in handsets and gaming platforms.
RCP uses a batch-processing photolith and plating method to place metallization steps on embedded die as the subsystem is defined and implemented. While the method can be used with a BGA-style package, the routing layer is placed above the die, and links directly to BGA pads. It is compatible with ultra-low-K dielectrics, and can be used with single-chip or multichip modules.
Between now and the end of 2008, Freescale plans to implement RCP in PowerQuicc, DSP, baseband processor, and power amp families, as well as in special demonstration modules.
Freescale said RCP is compatible with System in Package (SiP), with Package on Package (PoP), and with integrated cavity packages. And RCP can be adapted for 3G mobile phones and a range of consumer, industrial, transportation and networking devices, the company said.
As an example of the the technology's application, a 3G mobile phone was made in a package measuring 25 millimeters by 25 millimeters using a mixture of RCP and PoP techniques, Freescale said. The package contains all the electronics required for a 3G mobile phone including memory, power management, baseband, transceivers and RF front end modules. No mention was made of a display or keypad for the component, nor how the radio-in-package module was put together.
RCP is lead-free and RoHS compliant and meets reliability standards for commercial and industrial applications, Freescale said, meanwhile development and tests are in progress to make the technique suitable for automotive applications.
"Its unique capabilities will allow customers to create the smaller, sleeker and more efficient multifunction devices that the marketplace demands," said Sumit Sadana, chief technology officer of Freescale, in a statement.

sa: http://mooblog.blogstation.fr/images...logo-192.3.gif

[bbmf]

Intel insiders have confirmed to DailyTech that the license for ATI to manufacture chipsets for the Intel bus has not been revoked. ATI's board of directors recently approved a takeover bid from AMD which would make ATI a division of AMD if approved.

*A report on The Inquirer yesterday claims "ATI had its chipset bus license pulled, or at least not renewed by Intel." Aside from the fact that it is probably not legal for Intel to retract the agreement, the cross-license agreement between Intel and ATI has not come up for review either. It's not impossible to fathom that Intel would opt to not renew the ATI cross-license agreement, but at this time the agreement is still working and will continue to function until an announcement is made by Intel and ATI.
ATI's upcoming RD600 chipset is slated to be one of the most ambitious ATI chipsets to date, bringing Intel's Conroe and ATI's Triple Play physics to the table. However, the most recent ATI launch schedule still puts this chipset in for a 2006 launch, likely before the AMD-ATI merger would be finalized. Even then, it is difficult to ascertain if Intel would opt to not renew the bus license as it may take ATI years to fully assimilate into AMD. AMD president Dirk Meyer outlined that ATI and AMD would begin co-development of products in 2008.
Specifically with regard to RD600, Cameron Wilmot at TweakTown has confirmed that DFI's LanParty RD600 motherboard is still slated for launch. When DailyTech approached DFI San Jose General Manager Jacky Huang with regard to RD600, Huang replied "RD600 LANParty is still in development, and there have been no announcements to stop development." ATI's RS700 chipset, on the other hand, would be a chipset definitely in jeopardy if the license agreement was to terminate.
Intel is also dependent on ATI's core logic division, as Intel even has some Intel-branded ATI motherboards in its portfolio. However, there is no replacement for the ATI Xpress 1100 Grant Country motherboard with the Bearlake generation of Intel motherboards.

sa: http://www.dailytech.com/article.aspx?newsid=3494

*I, bbmf also repeated this erroneous report and apologize for not indicating that the report was unconfirmed.

[bbmf]