Some Intel IDF Highlights

Intel will unveil a multi-company initiative to support a 480Mbit/s wireless USB technology over 10 meters based on UWB. Backers will includeDolby Labs and many of the companies currently supporting the wired USB 2.0specification, said an Intel spokesman. The wireless USB initiative and theIntel chip will be based on the multiband OFDM technology backed by an industryalliance that includes Intel. It will also be supported by the WiMedia Alliancethat is helping define UWB software interoperability. The wireless USB chips arenot expected to ship in products until 2005.

The Intel spokesman would not confirm or deny reports that the company willdemonstrate a 64-bit x86 at IDF. However, he did say "the subject will bebrokered," probably in the keynote address of Craig Barrett, Intel's chiefexecutive. Intel will also announce its first 90-nm communications processors atIDF with plans to apply the in-house technology across its portfolio of wiredand wireless parts. In total, Intel will make as many as 16 new announcements atIDF. Third parties are planning about a dozen more. Other news at IDF willinclude new tools to ensure interoperability of digital home products, new NORflash technology and the creation of a memory implementers' forum to addressissues with DDR II and a front-side bus in-line memory module.

Sapphire's Z68 unveiled at CeBIT

Its easy to forget that Sapphire is back in the motherboard business, as so far the company has only launched three boards, but at CeBIT Sapphire was showing off two additional models for Intels Sandy Bridge platform, one full size ATX board with the Z68 chipset and one mini-ITX board with the H67 chipset. Neither board appeared to be retail ready, but wed expect the mini-ITX board to be available before the Z68 board.

The Z68 appears to be going under the name of PI-CI7S33Z68 as for now, but Sapphire tends to have much better retail names for their boards, so well have to wait until the launch to find out what itll be. Were not quite followings Sapphires design either, as the board has three x16 PCI Express slots of which the second one only work in dual x8 mode while the third one is limited to four lanes worth of bandwidth. This is all fine and good, but the only other expansion slots are three PCI slots, of which two will be unusable in CrossFireX or SLI and there doesnt appear to be a reason why the topmost slot is missing.

Sapphire has added two additional SATA 6Gbps ports via a third party controller for a total of four SATA 6GBps ports and four SATA 3Gbps ports. Were not quite sure why Sapphire put the additional two ports so far away from the six connected to the Intel chipset, but at least it should reduce confusion as to which ports goes where. There are also headers for eight USB 2.0 ports and as with previous models Sapphire is using its special kind of chokes and the board is also equipped with a user selectable dual BIOS and a POST80 debug LED display.

One other interesting feature we spotted is that Sapphire has fitted a Proadlizer to the PWM, a feature that so far only appears to be popular on high-end graphics cards. Around the back we find a PS/2 port, four USB 2.0 ports, two USB 3.0 ports, a Gigabit Ethernet port, a Bluetooth dongle, DisplayPort, HDMI, DVI and D-sub connectors and 7.1-channel audio with optical S/PDIF out. Overall were not sold on this board, it has some interesting features, but the slot layout isnt what wed hope to see on a Z68 board.

The mini-ITX H67 board is as far as we can tell currently called IPC-CI7S1DH67 but again, were expecting a much better retail name. Sapphire decided to go with a single x16 slot for this design, which is pretty much all you can fit to a mini-ITX board, although hidden at the bottom of the board is a mini card PCI Express slot into which something like a Wi-Fi card can be installed. You also get two SATA 6Gbps and two SATA 3Gbps ports and two headers for an additional four USB 2.0 ports. Around the back Sapphire has managed to squeeze in four USB 2.0 ports, two USB 3.0 ports, a Gigabit Ethernet port, a Bluetooth dongle, DisplayPort, HDMI, DVI and a D-sub connector as well as 7.1-channel audio with optical S/PDIF out.

Sapphires boards are still looking a bit rough and ready, but given some time were confident that Sapphire will come out with some excellent boards as long as the company listens to customer feedback. Were quite eager to find out what Sapphire is working on for AMDs upcoming platforms as well, since Sapphire is a strong AMD partner on the graphics card side.

Source: Computer Base

TwinMOS Unveiled TwiSTER DDR3-1800 Modules

The DDR3 has become more popular in the market since Intel’s introduction theP35 and X38 chipsets. After introducing the high-level DDR3-1333 module fordesktops and the high-level DDR3-1333 SO-DIMM for laptops, TwinMOS would like tointroduce the TwiSTER over-clocking series, which includes DDR3-1600 andDDR3-1800 models, to fulfill the hardware requirements of professional gameplayers and overclockers. The TwiSTER series is the highest quality product linefrom TwinMOS, featuring reliable data access at ultrafast speeds. Every moduleis made with selected original dies and high-standard production processes. Moreimportantly, every module is tested to comply with the published specifications,and quality is controlled according to the highest standards.

DDR3-1600MHz and DDR3-1800MHz memory modules of the TwiSTER series are made withthe original dies from Micron Technology Inc in a 128Mx8 layout and aredelivered in 1GB capacity. These dies are packaged with the latest FBGA(Fine-pitch Ball Grid Array) technology. Other specifications of these modulesinclude 240-pin design, working voltage at 1.8V, CL= 8-8-8-21, clock frequencyat 1600MHz and 1800MHz, and data transfer bandwidth at 12800MB/sec (DDR3-1600)and 14400MB/sec (DDR3-1800). Therefore, it is real good news to overclockers.The high-performance heat-sink equipped on the module provides perfect coolingand electrical characteristics to enhance the work efficiency of the module. Thecircuit design layout conforms completely to the JEDEC circuit specification toprevent noise interference in order to ensure multimedia operation stability athigh speeds and the reliability of huge data storage for full compatibility toPCs.

TwinMOS provides global, lifelong warranties for all memory products. Inaddition to quality assurance, TwinMOS provides permanent warranty service withits global operations as a world leading memory supplier. Moreover, as TwinMOShas been a major memory module supplier for world leading system suppliers, tocope with the lead-free environmental process promoted internationally, allTwinMOS production lines comply with the RoHS environmental directive of the EUin order to make greater corporate contributions to a greener earth.

Key Features

8-bit pre-fetch designUltra high frequency at 1600MHz(DDR3-1600)/1800MHz(DDR3-1800) with the240-pin designData transfer bandwidth at 12800MB/sec (DDR3-1600) and 14400MB/sec(DDR3-1800)Working voltage : 1.8VHigh compatibility and reliability

Shuttle Launches NVIDIA SLI Ready XPC

Shuttle Inc - the world’s leading manufacturer of small form factor PCs andaccessories - has today released final details of its much anticipated gamingbarebone, the XPC SN26P. Previewed by a number of hardware review websites andmagazines earlier in the year, the XPC SN26P has, even before today’s launch,earned itself numerous awards including “Most Innovative Product” awards fromwebsites HardwareZone and VR-Zone.

“The response we got to our XPC SN26P preview release was amazing,” saidShuttle Product Manager Troy Yang, “We knew we had a great product, but we alsoreceived some great feedback on where we could improve. The XPC SN26P we arereleasing today is our response to both the praise and the criticisms. We thinkeveryone will be pleased with the results.”

At its core the XPC SN26P is powered by NVIDIA’s most advanced chipset, thenForce4 SLI. This gives the system enthusiast level performance featuresincluding full support for AMD socket 939 processors, SATA 3Gb/s hard driveswith RAID, and gigabit Ethernet with hardware acceleration and firewall. 7.1channel surround sound is enabled by the popular Via Envy24PT audio processor.The barebone has been approved to be compatible with NVIDIA GeForce 7 seriesgraphics cards in SLI configuration, a list of compatible cards is available onthe Shuttle website. The barebone will not require the specially designed SLIheat-pipe cooler as seen in prerelease models.The XPC SN26P has begun shippingaround the world and should be available in stores in less than one week.

EVGA releases Z68 FTW, Z68 SLI and Z68 SLI Micro

Common to all of EVGAs Z68 boards is the red and black color scheme, support for Intels Socket LGA 1155 processors, SATA 6Gbps, USB 3.0, voltage read points and even dual cooler mounting holes to fit both todays LGA 1155 / 1156 heat sinks and the older LGA 775 ones.

To provide extremely clean power to the processor, the EVGA Z68 FTW flagship features a processor socket with 300% more gold content, a 12-phase voltage regulation module (VRM) and even a low-ESR, high current capacity Prodalizer film capacitor. There are two upright 8-pin ATX12V connectors for CPU power and an angled ATX 24-pin connector, possibly done so for the sake of better cable management.

While all three Z68 boards are capable of NVIDIA SLI and AMD CrossFireX, NVIDIAs NF200 PCI Express switch chip on the Z68 FTW enables support for 3-way NVIDIA SLI plus PhysX. The internal SATA 6Gbps (two) and SATA 3Gbps (four) storage ports are all routed to the Intel Z68 Express chipset, while the two eSATA ports at the rear are handled by Marvells 88SE6121. The dual Gigabit Ethernet ports are also handled by another Marvell controller, the 88E8057. There is no lack of connectivity options on this board, with a total of ten USB 2.0 ports, four USB 3.0 ports and Firewire.The VRM and chipset heatsinks are huge, which should provide more than adequate cooling.

The slightly watered-down EVGA Z68 SLI features an 8-phase VRM and three PCI Express x16 slots. The board drops the NF200 chip, thus suporting only 2-way SLI plus PhysX. There are five internal SATA ports (two SATA 6Gbps and three SATA 3Gbps) and the last SATA 3Gbps going to the rear panel as eSATA. Two less USB 3.0 ports on the Z68 SLI as well.

Being smaller doesnt mean cost-cutting; the microATX form factor EVGA Z68 SLI Micro retains most of the features found on its older SLI sibling, such as dual BIOS support, USB 3.0 and Firewire.

The EVGA Z68 FTW retails for US$264.99. Addingthe EVGauge and ECP V4 overclocking tools to the FTWwill set you back another US$35. The EVGA Z68 SLI retails for US$209.99, and the Z68 SLI Micro for US$159.99.

Justice Department Investigating Cable Providers, Net Neutrality

The U.S. Justice Department has begun conducting a large, wide-ranging antitrust investigation into cable companies and whether or not they are attempting to stifle upcoming competition from online video providers. Officials have spoken to several online companies, including Netflix and Hulu, as well as cable providers like Comcast and Time Warner Cable, about issues such as data caps. No comment was given about the investigation.

The problem arises from the fact that cable companies, which provide bundles of television channels as well as high-speed internet access, are having their traditional channel bundles as well as video rental on demand services bypassed by customers using online services such as Netflix, Hulu, and Amazon. Companies, having invested billions of dollars into the construction and maintenance of their networks, have little to no incentive to get out of the business of providing television channels, and major entertainment groups with multiple television stations are highly invested in the current model of television viewing, because by requiring cable companies to license packages of channels rather than picking individual channels, they ensure that all their channels, no matter how popular, make it into homes, and are able to receive fees for stations that would otherwise wither and die from lack of viewership. The probe from the Justice Department serves as a highlight into how shifts in patterns of television viewership are shaking and reverberating through the television entertainment industry.

The Justice Department has already shaken the eBook publishing industry with a lawsuit that alleges publishers colluded with Apple to fix prices on eBooks, rather than letting the market set the prices. Several publishers settled in April, but Apple and the remaining publishers continue to fight the suit. In the cable probe, the Justice Department is taking a close look at data caps that have been used by TV providers like Comcast and Time Warner have used to try and deal with the rapid growth of online video traffic. The limits are needed, the companies claim, in order to stop heavy users from overwhelming their networks. Online video providers like Netflix, however, have accused cable companies of using the limits at stopping consumers from dropping their television services and switching completely to online video services. There is also concern that the cable companies’ online video offerings, many of which have begun service in just the past year, will get an unfair advantage by not affecting data cap limits or receiving priority traffic on the networks. Comcast did nothing to alleviate those accusations when they announced in March that its Xfinity app on the Xbox 360 would not be counted against their subscribers’ data caps, unlike every other video service available on the system.

Because of this announcement, the Justice Department is examining whether or not Comcast violated legal requirements made of the company in 2011 in order for them to secure approval for the takeover of NBCUniversal. Under those legal terms, Comcast cannot “unreasonably discriminate” against other companies transmitting data over its network or treat its own content differently than it treats theirs. Comcast claims that they have not, in fact, violated those requirements, because they are not treating data from the Internet any differently; they are claiming that they have a loophole in that Xfinity uses data only from Comcast’s internal network, not the public Internet. “We have consistently treated all video carried over the public Internet the same whether it comes from our sites or anywhere else on the public Internet,” the company said.

Another issue being investigated are paywalls around certain content providers that require users to have an active cable subscription, such as ESPN’s WatchESPN service, which can only be accessed in mobile form after entering your cable subscription details. The Justice Department is investigating whether these paywalls are anticompetitive, as well as whether contracts signed by providers and entertainment groups are anti-competitive by requiring subscribers to purchase packages of channels, whether or not they actually watch those channels.

Judge Rules Oracle APIs Not Copyrightable

It has been more than seven weeks since the lawsuit between Oracle and Google began, but the question at the heart of that case, whether or not the structure, sequence, and organization (SSO) of the 37 Java APIs that were used could be copyrighted, has finally been answered. The presiding judge has concluded that a partial verdict that Google infringed on the SSO is invalid as the SSO is not copyrightable, though this decision could be overturned on appeal. This means that Google doesn’t have to worry about an injunction, or ban, against their Android operating system for at least a couple years. If this decision is overturned on appeal, then the question of whether or not Google’s use of the APIs is protected under the fair use defense will have to be determined either by a new jury or by a judicial ruling; Google argues that a new jury being selected means that they would have to reevaluate the infringement, which would cause an even longer delay. For now, Google’s seemingly rash decision to defend itself in court has paid off; it gets away scot-free (for a few years, at least). Few if any other companies would have taken on the risks Google did with this decision. The proverbial jury is still out on this case, at least until Oracle appeals the decision, but Google has more immediate IP litigation worries relating to a long list of Android-related lawsuits against Google, Motorola (a newly-acquired subsidiary), and other device makers, such as Samsung and HTC. Two weeks ago, a US trade agency ordered an import ban on Motorola’s (now Google’s) Android devices, and less than a week later a German court ordered a permanent injunction on the same devices.

It would be understandable for the jurors, who just spent seven weeks of their life missing work and being reimbursed far below the minimum wage, to ask why the judge couldn’t have decided the copyrightable issue firest, and saved them all the hassle. The reason Judge Alsup (the presiding judge) waited until the infringement trial was over was because the judge knew from the start that any decision on the copyrightability of the API SSO would face a huge risk for appeal and overturning. He had already looked into the issue last year when he denied Google a motion for summary judgment, and he certainly wouldn’t have put the question before a jury had he been completely certain of the ruling he was planning to give. Throughout the trial, Judge Alsup showed nothing but consideration for the jury, even telling Google and Oracle that, should they settle, they should do so before the weekend so as to minimize the impact on the jurors’ lives.

Clearly, given that the judge himself is unsure of the defensibility of his ruling in the Federal Circuit Court of Appeals, the next stop for the trial should Oracle pursue it, those who are overjoyed by this ruling should be a bit more cautious. Google built key components of the Android operating system on a highly controversial legal theory that will ultimately be decided by either the Federal Circuit or the Supreme Court of the United States. A couple of years might seem like an eternity in a field like software development and mobile operating systems, but it’s obvious that Google has much longer-term interests in the Android operating system. And to be frank, a ruling that the judge himself was unsure of in no way means that there was no case. This decision is unprecedented; no comparable amount of software code (400 class definitions and thousands of methods) has been held uncopyrightable despite satisfying the originality requirement. Both sides had reason to appeal should the case have gone against them.

“This action was the first of the so-called ‘smartphone war’ cases tried to a jury,” began the order on copyrightability, which, although technically true, is not the first case to be resolved; a number of other cases were filed after the Oracle v. Google which have been adjudicated without juries. More importantly, this case has unique characteristics that other smartphone-related cases don’t, instead dealing with patents instead of copyrights, or without the parties in question being in the middle of negotiating a licensing deal prior to the case. Even among copyright lawsuits, this is a rare case, because it deals not only with the infringement and fair use, but whether or not infringement is possible in the first place (i.e. can this material be copyrighted?). The judge, realizing the import of this decision and just how many people would be analyzing it, clarified the intended limitation of the scope of his holding at the end, saying: “This order does not hold that Java API packages are free for all to use without license. It does not hold that the structure, sequence, and organization of all computer programs may be stolen. Rather, it holds on the specific facts of this case, the particular elements replicated by Google were free for all to use under the Copyright Act.”

Judge Alsup, rather than attempt to dictate how an evolving marketplace can operate, instead decided to say that method definitions are so inherently functional that they can’t be copyrighted, pointing to interoperability considerations to back up his ruling. The example of a method definition he gives is:

public static int max (int x, int y)

The method described here is a public (accessible by all the other code) and static (it gets initialized immediately when the program runs, and only one copy is needed since it never changes) function called “max” that returns an “int” (or integer) result based on the interaction of two variables, x and y, which are also integers. The problem is that programmers like myself know that this is a fairly simple, broad function; object-oriented languages like Java (and even non-object-oriented languages like C) allow for far more complex function names, which can deal with highly specific, even proprietary, formats for variables, returned data, or even project- or class-specific formats, data types, or highly specific types of functions designed to do a particular task within the project, and Oracle never claimed a monopoly over a generic function like “max.” The API part of the lawsuit was about a far larger and more expansive body of work than just basic mathematical functions.

It is now up to the higher courts to decide on whether or not the specific code is copyrightable or not. Should the side with Judge Alsup, then Google is cleared and a new precedent is set in copyright law when dealing with writing software. Should they side with Oracle, then the question becomes whether or not the way Google used the code constituted fair use. Either way, this case is going to be very interesting to watch as time goes on. Rest assured, we will continue to cover this case as it makes its way through the courts. In many ways, it’s just getting started.