Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : C-Cube -- Ignore unavailable to you. Want to Upgrade?

To: DiViT who wrote (48513)	1/28/2000 2:45:00 PM
From: Peter V	Respond to of 50808

I actually thought of that post as I wrote to you. It's coming back a bit, maybe the buy the dippers will bring us back. I'm still holding the last of my Feb 65's in hopes they will be worth more next week.

To: DiViT who wrote (48513)	1/28/2000 4:50:00 PM
From: BillyG	Read Replies (1) \| Respond to of 50808

TI approach to MPEG-2 decoding. And perhaps encoding. In hardware. Just plug it into a DIMM memory socket... eetimes.com TI fields DSP-based media processor on a DIMM By Anthony Cataldo EE Times (01/28/00, 11:47 a.m. EST) TOKYO — Texas Instruments Inc. is quietly beating a path into what has proved to be a no man's land for the media processor. After virtually giving up on its quest to field a DSP-based media processor for the PC several years ago, the company is again exploring the notion of putting a high-performance digital signal processor on the desktop to handle hard-core programming chores that would otherwise throttle a CPU. If TI is able to reach a few more milestones in coming months, and if early customers give the green light, DSPs could find themselves saddled with a dual-in-line memory module (DIMM) as early as this summer, primed for use in such applications as medical imaging and video decoding. TI displayed its Performance Enhanced Memory Module (PEMM) at Microsoft Corp.'s Windows Hardware Engineering Conference in 1998. Since then, however, not much has been heard from the PEMM effort. Module holds surprise At TI's Tsukuba R&D Center, which is becoming the focal point of the Dallas-based chip maker's research and product development efforts in Japan, Yuji Itoh, senior member of the technical staff, is hovering over a PC, fingering a 32-Mbyte DIMM. But there's something different about this module. It's about 2 inches tall, larger than a standard DIMM. After turning it over, Itoh reveals why: There's a C6x DSP strapped to the back, along with a piece of programmable logic. A media processor on a DIMM? "We reached the idea that the DSP should handle lots of data, and data is available around the memory," Itoh said. "So we decided to put the DSP on a memory module." Put it on a PCI bus and you encounter lots of contention over the bus, which will bog down the DSP. Put it on a DIMM, along with the right drivers supporting synchronization and multiplexing with the CPU, and you have high bandwidth coupled with quick access to memory. "The DIMM approach is a brilliant way to get into an existing socket," said Will Strauss, president of the research firm Forward Concepts (Tempe, Ariz). "Perhaps the bandwidth is higher on a DIMM slot and it's much higher on memory [bandwidth]." TI isn't trying to turn the idea of a CPU-centric PC upside down. But it is making a case for what it calls "heterogeneous multiprocessing." In other words, if you really need performance, TI is proposing to throw in a DSP — maybe with its own companion hardware accelerator — to take care of math-intensive multimedia applications. The idea itself is not new. The idea of DSP-based processing gained steam during the latter part of the 1990s, as a way to solve the crush of multimedia applications invading the PC. These PC media processors came in many forms, and from a variety of companies, including Philips and Samsung, as well as Chromatic Research and its licensees, LG Semicon and STMicroelectronics. Texas Instruments bowed out early when Microsoft turned thumbs down on Intel Corp.'s ill-fated Native Signal Processing (NSP) initiative. According to analyst Strauss, Intel realized it would need help from a DSP for NSP to work, and proposed to hang a TI C30 off the ISA bus. But NSP relied on gaining access to Ring 0 at the heart of the operating system, which ran afoul of Microsoft's DirectX program. Step aside Meanwhile, Intel boosted the Pentium's speed and introduced the Accelerated Graphics Port, in large part obviating a DSP solution. "The Pentiums just got better and better," said analyst Strauss. Yet the newest spin of media processor just entering the market today "is having success in things like HDTV and high-end stereos," Strauss said. "And so it looks like there will be a market for media processors . . . but not in the main PC chassis." Faced with so many obstacles, TI is taking a cautious approach. Since it first publicly floated the idea of the media processor-based DIMM at WinHEC nearly two years ago, it has done what it can to build the software support and ensure motherboard compatibility. It even pushed through a new DIMM standard, the PEMM. Motherboards made in Taiwan, China and the United States have unique data line inversion schemes, and the PEMMs needed to work in each one. "We solved the problem ourselves by introducing a mechanism in the driver to handle the data line conversion," Itoh said. "The driver is just above the operating system." TI's prototype PEMM packs a TI C6201 running at 200 MHz. Running in a parallel configuration, it can burn 1.6 billion instructions per second, though Itoh said the actual performance would be lower. Next to the DSP is a bit of programmable logic to handle the data bus control. It could be converted into an ASIC for volume production, Itoh said. TI doesn't expect the PEMM to replace, say, a 3-D graphics card. There's already plenty of specialized IC companies fighting over that market, Itoh said. Instead, the company is aiming at specialized application such as medical imaging or accelerators for photo-lab technicians. While these uses may be considered a niche, they are stimulating the interest of DSP suppliers. Applications like volumetric imaging, where, for example, an ultrasound image can show a 3-D form as well as the internal structures, need more DSP horsepower, said Strauss. "TI has been telling me they are beginning to add some emphasis on the medical imaging market," Strauss said. "Medical imaging is not big in volume, they buy thousands of chips a month. But it is lucrative. If you have [a medical imaging system] on a PC you have a good human interface." TI is looking at other applications too, such as manipulating MPEG video images. For example, TI's PEMM can now handle MPEG-2 decoding and, with a little help, encoding.

To: DiViT who wrote (48513)	1/30/2000 7:46:00 AM
From: Maya	Read Replies (2) \| Respond to of 50808

fool.com Son of DIVX: DVD Copy Control By Rob Landley GAITHERSBURG, MD (January 27, 2000) -- Every once in a while, a Fool just has to sit back and go, "Wow, that's really dumb." I shouldn't be surprised when dumb things come out of Hollywood. Robert Heinlein once wrote a morality tale about how an overly militaristic society can be a bad thing, and based on that, Hollywood came out with the campy action movie Starship Troopers (i.e., "Bevery Hills 90210 goes to war"). But then the DVD Copy Control Authority (a name straight out of a James Bond movie if you ask me) attempted to sue the ENTIRE INTERNET. It's like a class action lawsuit in reverse. Even for Hollywood, this has to set some kind of record for sheer bulk corporate idiocy. The mind boggles. I've written in the past about how the Internet is commoditizing information and forcing information providers into a service role. And, I've mentioned how the currently dominant players in the act of being displaced are kind of unhappy about it. Personally, I'm waiting for a subscription service that lets me view 20-year-old episodes of Dr. Who (or the Dilbert episode I missed last week) through the Internet for maybe 25 cents per half hour, or even better, for a flat monthly fee. That would eliminate piracy by making it no longer worth the effort. Access to a huge library of video is a service I'd certainly be willing to pay for, and keeping video on my hard drive or passing it along to friends would then be a complete waste of my time even if it was as easy as breathing. But attempting to use laws to prevent the natural commoditization of previously proprietary products and the resulting transition to a service-based industry... It's like prohibition in the 1920s. That's not going to work. It's not even a holding action, it's just... dumb. A while back, I wrote about a bad idea called DIVX, which could best be described as "profoundly stupid." I was then asked by UberFool David Gardner to gloat in a follow-up article when DIVX was pulled from the market. In both cases, the general thrust was, "The DIVX creators have declared war on their own customers! The DVD people could never even aspire to this level of profound stupidity!" Putting the two together, it honestly never occurred to me that the DVD camp would make a concerted effort to prove me wrong by fighting the natural trends within their own industry so vehemently. A little background: DVD disks are encrypted with something called the Content Scrambling System (CSS). The DVD player decrypts the data on the fly as it's reading the disk. People who want to make a DVD player have to get an expensive license to this secret encryption algorithm from the DVD Copy Control Authority (hereforth, simply "Copy Control"). Each applicant gets their own secret key with which to unscramble the DVD so it can be played. And if one of those keys were to fall into the "wrong hands," then future DVD disks could be made so that key wouldn't work on them anymore. When DVD players started appearing in computers, no company was willing to spend thousands of dollars to get a license to write a DVD player for Linux. So the Linux people wrote their own CSS descrambling program (deCSS), and posted the source code on the Internet. They reverse engineered the encryption algorithm, and then spent the rest of the afternoon finding (and posting) every possible decryption key that would play existing DVD disks. Copy Control freaked and called for their lawyers. They sent "cease and desist" letters to everyone they could find who had the source code on their websites, or who provided links to those websites, or who provided links to people who linked to... I don't know if Yahoo! (Nasdaq: YHOO) got a letter on general principles or not. When the majority of the Internet failed to pay any attention whatsoever, Copy Control started suing people. Lots of people. Hundreds of people. They even arrested the 16-year-old Norwegian programmer who had originally posted the code, on the grounds that since he was one of the few people who had taken the code down in response to the original letter from the lawyers, he was obviously easily intimidated and they had to take their frustrations out on SOMEBODY. (How they got the Norwegian government to go along with this, I have no idea.) All the while, the DVD Copy Control people have been screaming "piracy!" But the Linux deCSS program is not a case of piracy at all. Fact is, you don't need to decrypt DVD disks to copy them. The DVD disk is supposed to contain encrypted data in order for a standard DVD player to play it. A DVD player can't tell a verbatim copy of the encrypted information from the original. Pirated DVDs have been coming out of Hong Kong for a year, long before the deCSS program was written. Ignoring this first point, pirating DVDs is uneconomic anyway. Blank home-burnable DVD disks cost $50 apiece, whereas a brand new mass-produced DVD costs in the $20-$30 range at retail. In a few years, Blockbuster Video will probably start selling "previously viewed" DVDs for about $10 each, the same way they unload their old videotapes. DVD piracy with a home DVD burner simply doesn't make economic sense, and the home-burnable DVDs have an extra precaution: The track where the encryption keys go is pre-burned with zeroes. From another viewpoint, copying a DVD that you own is perfectly legal. Giving that copy to somebody else would be illegal, but making a backup copy for your own use wouldn't. Owners of software are allowed to make a backup copy for archival purposes. Owners of CDs are allowed to copy them to audiotape to play them in the car. This is well-established case law. Let me repeat all this: The deCSS program is neither designed nor necessary for copying DVD movies, which isn't economically feasible anyway and not technically possible with the partially prewritten blank disks being sold today. In any case, a tool to copy DVDs would be legal for personal use. So why is Copy Control so completely frantic to put the genie back into the bottle? Well, two reasons. First, remember that the DVD decryption takes place in PLAYERS, not in copying disks. The Linux people made their own DVD PLAYER. The DVD encryption gave Copy Control a monopoly on DVD players until somebody made a DVD player without them, and they're mad. They made a lot of money licensing their secret recipe to DVD player makers. Now, nobody needs them anymore. The second reason is that the people pitching the DVD format to the movie industry said that the encryption protects against copying. That this wasn't technically true means that some serious posterior covering is taking place about now. The DVD Copy Control Authority's battle over deCSS isn't just to prevent what it does do, but to distract from what it DOESN'T do. Three separate lawsuits from Copy Control are now ongoing. (Each time they face a setback, they file a new lawsuit in another jurisdiction.) My favorite moment was when the defendants showed up in T-shirts bearing the deCSS source code, which Copy Control tried to enter into evidence (motion denied). The deCSS source code itself HAS been entered into the public record though, as "Hoy exhibit B," which is kind of strange if the point of the lawsuit was to suppress it. Larry, Moe, and Curly would be proud of the DVD Copy Control Authority's legal efforts thus far. On the bright side, this case is likely to clear up a whole lot of legal issues by finally challenging them in court. Is it legal to link to other people's content on the Internet? Are shrink-wrap licenses legally enforceable? How about the contract-o-matic (Click "yes" to sign over your firstborn or this program will refuse to run)? Unfortunately for the existing publishing industry, on each one of those issues, they're on what appears to be the losing side, and clinging to the tatters of old laws to support the status quo against changes in technology and the marketplace. The Scientologists tried to sue people who posted their religious doctrines on Usenet. The Recording Industry Association of America keeps suing random people trying to make portable MP3 players. Now the DVD people are explicitly suing the Internet at large, which makes about as much sense as trying to hold the Pacific Ocean legally responsible for Hurricane Andrew. Even if you win, what is it supposed to mean? By the way, if you still don't think that this is nearly as stupid as the limited-play aspect of DIVX, don't worry, somebody's working on that one too: DVDs that self-destruct. One final note from the Fool Radio folks: On this week's Motley Fool Radio Show, we're holding our annual Super Bowl Special and we want to hear from you. When did you drop the ball? Did you bet the farm on a penny stock? Did you get carried away with that credit card? If you have a financial fumble you want to share with the world, our Fool Radio team wants to hear from you. And it won't be all financial fumbles on Fool Radio, so if you're not in the confessing mood but have an investing question or comment you'd like to share, we want to hear from you as well. If you're interested in participating on this week's show, please email radio@fool.com with your confession/question and please include a daytime phone number. - Oak Relevant Links: UberGeek Eric Raymond (author of The Cathedral and the Bazaar) has written an excellent explanation of the situation. Hoy Exhibit B from the public record: The actual DeCSS source code. Wired Magazine on the technical issues: Why the DVD hack was a cinch. A great parody of some of the issues at stake: Copyrighting Fire. Several people are happy to finally see shrink-wrap licensing challenged in court. And is it legal to link to other people's content or not? The Legality of Linking. (New York Times, requires free registration) News flash: Copying is still legal. (New York Times, requires free registration) C/Net's special coverage of the DVD Copy Control Authority throwing a temper tantrum. Intimidation 101: Pick on the weakest individuals. Arresting a 16 year old (and his father) in Norway.

To: DiViT who wrote (48513)	2/1/2000 1:53:00 PM
From: BillyG	Read Replies (2) \| Respond to of 50808

Matsushita Shrinks Image from MPEG Streams nikkeibp.asiabiztech.com Matsushita Electric Industrial Co, Ltd has developed Realtime Moving Picture Coding Experts Group (MPEG) Reduction Technology, making it possible to shrink an image including multiple video, text and picture elements to a specified size directly at the MPEG data stream level. The new technology has been applied to an Internet-based teleconferencing system linking up to nine sites. Instead of decoding the data to original image, shrinking it and then encoding it again, the new technique applies shrinking process to the decoding stage, that is, to discrete cosine transform (DCT) data. For example, to halve an image in both horizontal and vertical directions, one 4-pixel x 4-pixel low-frequency component block is extracted from each of four 8-pixel x 8-pixel DCT blocks, and the four 4-pixel x 4-pixel blocks are combined to form a single 8-pixel x 8-pixel DCT block which is again quantized and coded to the variable length code. As a result, the processing volume is cut to 1/20th that of the conventional scheme, so that it is possible to reduce the size of any of the nine images being displayed simultaneously. The new approach offers much better image quality than shrinked imagery generated by simple downsampling. The firm also developed I/P Frame Conversion Technology to maintain smooth frame rates for key images when Internet throughput (effective transfer rate) is low, and Media-Specific Optimized MPEG Generation Technology to enhance decoded image clarity by allocating more encoding bits to character and line drawing regions than to video area.

To: DiViT who wrote (48513)	2/1/2000 1:54:00 PM
From: BillyG	Read Replies (2) \| Respond to of 50808

4.7GB DVD-R Using Varying Specs for Light Sources nikkeibp.asiabiztech.com The basic approach for the next-generation digital video disc record (DVD-R) standard, featuring a single-side capacity boosted to 4.7 Gbytes, has almost been finalized. The existing DVD-R disk is based on DVD-R Standard Ver.1, which was finalized in 1997, and has been most commonly used in authoring of play-only media. Its single-side 3.95-Gbyte capacity, however, makes it difficult to use on the proposed 4.7-Gbyte play-only disks, and therefore the development of a new standard was urgent. There was discussion whether to continue to use the 635-nm wavelength light source defined in Ver.1, or switch to the same 650-nm source as used in DVD random-access memory (DVD-RAM) and DVD rewritable (DVD-RW) drives. The 635-nm group claimed it would provide compatibility with existing systems, while the 650-nm side pointed out that only one firm is making the 635-nm laser diodes, practically speaking, and claimed that this is expensive and makes it difficult to develop consumer products In addition, 650-nm group explained that developing higher-output 650-nm device would make future higher recording speed possible. The DVD Forum decided to split the DVD-R standard into two parts, one for authoring and one for general use. The authoring standard maintains compatibility with existing systems, using a 635-nm source. The general-use standard uses the 650-nm light source, making it easier to achieve lower product cost and provide additional DVD-R capabilities for DVD-RAM and DVD-RW drives. In other words, compatibility has been deliberately lost between the two. Addressing and schemes to prevent unauthorized copying will also be different.