|
| Thursday, June 05, 2008 |
|
Apple And Google's Awkward Mobile Marriage (AAPL, GOOG)
Hank Williams | June 3, 2008 10:45 AM
The relationship between Google (GOOG) CEO Eric Schmidt and Apple Inc., on whose board he sits, must, at times be incredibly awkward.
Google is developing Android, an open source operating system for mobile phones, which is directly competitive with Apple’s iPhone. One would think that Apple (AAPL) would not be happy with this, and would ask Schmidt to step down. Yes, I know Schmidt leaves the room when they talk about the iPhone in board meetings. But this still must feel a bit like Apple having Microsoft on its board.
But here’s the thing. Apple *needs* Google.
Like it or not, Google is -- and probably will continue to be -- Apple’s most important iPhone developer. Google “gets” mobile in an incredibly deep way. And they are passionate about it. This “getting it” thing stems, in part, from Google’s pioneering almost every great idea in online mapping since the concept was invented. Mapping is also an incredibly deep area to mine, and depending on what you are trying to do, it can be very expensive work with no immediate opportunities for monetization. Perfect for Google. Not so great for anyone else.
The interesting thing about all of this is that, if you think about it, mapping is far more important to a phone than playing music or particularly watching movies. That’s not to say that music is not important. But if you let your imagination run wild, you can see that the potential for location based services and mapping tied to a mobile Internet connected device is profound. And no one on the planet is likely to do that better than Google.
Unfortunately for Apple, this cuts both ways. Apple loves the work Google does on the iPhone. But Google is so passionate, they want every phone on the planet to be as cool as the iPhone so they can deliver great experiences (and eventually ads —Ed.) for every phone. Hence Android, a free operating system that they are expecting to license to almost every major cell phone manufacturer. For Google, volume matters, profits not so much. For Apple, that formulation is reversed. From Apple’s perspective, that has got to suck.
More unfortunate for Apple: Google is now demonstrating features that, in this arena, are far more sophisticated than anything Apple has shown or is likely to show on the iPhone without Google’s help. The most stunning concept they have demonstrated is a kind of virtual reality for cell phones.
This new feature is based on a part of their web based mapping service called Street View. Street View allows users to go to a particular location and to see the street in moving pictures. You can turn 360 degrees and even walk down the street and get a kind of animated walk through effect.
What Google demonstrated was that Street View has now been implemented inside Android so that when you move the phone it detects where you are and what direction the phone is aimed, and shows you the view of the street on your phone. As you move the phone, the view changes. The effect is stunning. Of course this is much easier to see than to read about and so I recommend viewing the video below.
Presumably, at some point, these kinds of features will find their way into the iPhone, but probably not without Google’s help. Google’s innovation in this area means that, in some respects, Apple is really beholden to Google. This is because Google is the only company with the real chops and resources to deliver mobile mapping in a truly robust way (not including Microsoft, which is unlikely to develop for the iPhone).
And what Google is demonstrating is just the first little peek into what location based mobile services will bring. Google is baking programming interfaces (APIs) into Android that will allow *all* Android application developers to leverage their mapping and location based tools. So far there is little evidence that that will happen on the iPhone to the same degree, at least in the near future.
And so, while Steve Jobs is used to having the upper hand with his partners like the now supplicating record labels, this relationship is different. Indeed, Steve may hate the whole idea of Android, but he may soon need to get used to his new Google overlords just like the rest of us.
SAI Contributor Hank Williams is a New York-based entrepreneur. He writes Why Does Everything Suck? Exploring the tech marketplace from 10,000 feet.
Source: Silicon Alley InsiderLabels: Android, Apple, Google, iPhone |
| posted by Perimbean @ 1:55 AM |
|
|
|
| Wednesday, June 04, 2008 |
|
Kingston co-founder: DRAM makers have no way out
Josephine Lien, Taipei; Esther Lam, DIGITIMES [Tuesday 3 June 2008]
 David Sun of Kingston Technology
Photo: Josephine Lien, Digitimes, June 2008
David Sun, co-founder of the number-one DRAM module company Kingston Technology, believes that no DRAM maker will voluntarily quit the market as they have no backdoor to exit. Even if more consolidation is seen in the industry, overall capacity will not reduce.
Sun commented that the recent trough in the DRAM industry is very different than the one seen in 2000. Companies such as Fujitsu, IBM, Mitsubishi Electric, OKI, Texas Instruments (TI) and Toshiba quit the DRAM market due to severe competition and shifted their focus to other products. Toshiba, for example, shifted its focus from DRAM to NAND flash and the company has a proven achievement in the NAND flash industry now, stated Sun.
The situation is different now. No DRAM maker will quit the market on their own as they are have no "backdoor" to exit, Sun stated. These players are forced to stay in the DRAM business. Taking Micron Technology as an example, Sun said the company is still having its core operation based on DRAM, despite its extended presence in CMOS image sensor and NAND flash production.
Reviewing major DRAM makers such as Elpida Memory, Nanya Technology, ProMOS Technologies, Powerchip Semiconductor Corporation (PSC) and Winbond Electronics, Sun said all of these players have no other product line to shift to if they quit DRAM production.
Even if a price plummet spurs consolidation, Sun said global DRAM capacity is not going to decline. The only solution to return the industry to discipline is a rational expansion that will eventually reduce fresh capacity, he added.
Speaking on behalf of a memory module company, Sun said Kingston thinks visibility in the DRAM industry is relatively clearer than that for NAND flash. Since the price of DDR2 is unlikely to drop below US$0.80 again, procuring in this price range and selling it when the price appreciates will still guarantee profit for the company.
Source: DigiTimesLabels: CMOS, DRAM, Fujitsu, IBM, Kingston, Mitsubishi Electric, NAND, OKI, Texas Instruments, TI |
| posted by Perimbean @ 12:25 AM |
|
|
|
| Sunday, June 01, 2008 |
|
Microsoft urges Windows users to shut down Safari
'Carpet bomb' Safari bug can be combined with unpatched IE vulnerability
By Gregg Keizer
May 31, 2008 (Computerworld) In an unusual move, Microsoft Corp. on Friday warned Windows users to swear off Apple Inc.'s Safari Web browser until a patch is available that plugs holes that could let attackers compromise computers.
One security researcher noted that Microsoft's public warning — and Apple's silence on the subject — are typical for the two rivals and illustrate their different approaches to security.
Friday, the Microsoft Security Response Center (MSRC) issued a security advisory for what it called a "blended threat" caused by combination of a bug in Apple's Safari Web browser and a vulnerability in how Windows XP and Windows Vista handle executable files placed on the desktop.
"Microsoft is investigating new public reports of a blended threat that allows remote code execution on all supported versions of Windows XP and Windows Vista when Apple's Safari for Windows has been installed," said the advisory.
The Safari bug Microsoft referred to is the same one disclosed two weeks ago by researcher Nitesh Dhanjani, which Apple declined to treat as a security issue, said Andrew Storms, director of security operations at nCircle Network Security Inc. "Clearly, that's what they're talking about," said Storms.
In mid-May, Dhanjani posted information about what he dubbed a "carpet bomb" attack made possible because Safari lacks an option to require a user's permission to download a file. Attackers, Dhanjani claimed, could populate a malicious site with rogue code that Safari would automatically download to the desktop.
Apple told Dhanjani that it did not consider the problem a security issue, but might fix it in a future Safari update. The next week, the anti-malware group Stopbadware.org criticized Apple for that position. "We encourage Apple to reconsider its stance and treat this as the security issue that it is," said the group in a statement May 19.
Then on Friday, Microsoft also fingered Safari as a problem. "Restrict use of Safari as a Web browser until an appropriate update is available from Microsoft and/or Apple," the company told users in the advisory.
But Microsoft also acknowledged that a successful attack would require not only leveraging the Safari bug, but also exploiting a vulnerability in its own software. "A combination of the default download location in Safari and how the Windows desktop handles executables creates a blended threat in which files may be downloaded to a user's machine without prompting, allowing them to be executed," said Microsoft.
In the advisory, Microsoft called out Windows XP — including SP3, the newest service pack — and Windows Vista as vulnerable, as well as Internet Explorer 6 and IE 7.
Microsoft, however, did not delve into details of the Windows or IE vulnerabilities that could be combined with the Safari bug to hack PCs.
Aviv Raff, an Israeli security researcher, filled in some of the blanks. On Saturday, Raff said a vulnerability in IE that he had reported more than a year ago was the Microsoft side of the blended threat. "The combined attack requires IE," Raff said in a e-mail, answering questions about the source of the Windows-side flaw.
He would not, however, get specific about the vulnerability. In a post to his own blog earlier Saturday, Raff said he would not publicly disclose any details until Microsoft or Apple patched the problem.
But he did ding Microsoft for telling users that they could prevent attacks by changing the default download location for files retrieved using Safari. "I can only say that Microsoft's suggestion for a workaround is not enough," said Raff in his blog post. "There are other vulnerabilities which can be combined with the Safari vulnerability to execute code," he added in the e-mail.
In the end, Raff's best advice was similar to Microsoft's: "The current best solution is to stop using Safari until Apple fixes their vulnerability," he wrote on his blog. "Even if Microsoft fixes their vulnerability, Safari users will still be vulnerable."
Odd though it is to see Microsoft issue an advisory that calls out software not of its making, the incident is a good example of the contrast between Microsoft's and Apple's approaches to security disclosures, said nCircle's Storms.
"It's not very surprising to see Microsoft in the forefront here," he said. "They're known to issue advisories without having all the information [about a vulnerability] and without a patch. Apple, on the other hand, is completely different. Until they release a patch, they say nothing, and when they patch, it's a complete surprise.
"It's two different ways to handle it," said Storms, explaining that the vastly different approaches stems from their core customer base. "Microsoft has really embraced the enterprise, and decided that disclosure and a regular patch schedule is what the enterprise needs to support and maintain its products.
"Apple, on the other hand, appeals to consumers, and believes that for the majority of consumers, issuing an advisory without a patch would probably just create FUD [fear, uncertainty and doubt]," Storms concluded.
As Storms noted, Apple has remained silent on the Safari carpet bomb problem. Last week, it did not respond to a request for comment on its security team's decision against adding a user-approval option to Safari. The company was not available Saturday.
Microsoft did say that it is working with its rival, however. "[We] are working with our colleagues at Apple to investigate the issue," said Tim Rains, a product manager in Microsoft's malware protection center, in a post to the MSRC blog.
No timetable has been set by Microsoft for patching its software to block combined Safari-IE attacks. As it often does in security advisories, the company only said that it may issue a patch.
Source: ComputerWorldLabels: Apple, Microsoft, Safari, Windows Vista, Windows XP |
| posted by Perimbean @ 10:05 AM |
|
|
|
|
|
Me.com belongs to Apple; Vegas iMac deal
By Aidan Malley
Published: 11:25 AM EST, June 1, 2008
Apple's Internet service strategy has become clearer with the transfer of Me.com to the company. Also, Apple has struck a deal for iMacs in a Las Vegas hotel.
Apple's ownership of Me.com
Lending further support to notions that Apple will soon rebrand and overhaul .Mac, Daring Fireball and fellow technology pundits on Twitter have discovered that Me.com is now in Apple's hands.
On Saturday, records show the domain transitioned from any EasyDNS registration to MarkMonitor, a company that manages domain name services for commercial brands -- including Apple and its host domain. MarkMonitor was most recently used to park various MacBook Air-related domains by the electronics maker.
And while it's possible to abuse such registrations, a site report for Me.com points to the domain administrator as Apple's Ken Eddings, the same employee responsible for MobileMe.com as well as many other Apple-owned domains, including iPod.com. Currently, a simple NSLOOKUP command also confirms Eddings' link to the site.
The reason for the acquisition remains unclear, although the switch would give Apple access to a simpler domain than MobileMe.com for any future service; the company's most recent Mail and Safari updates have deliberately left placeholders for .Mac's new name that would allow a simple change.
Recent discoveries in the iPhone SDK have suggested that the future service will include tighter integration with iPhones and iPod touch players.
iMacs to populate Vegas hotel
In a more public deal, the upcoming Fontainebleau hotel in Las Vegas will partner with Apple for a unique addition to its suites, according to a fresh tip sent to TUAW.
The hotel's official brochure (PDF) notes an "innovative relationship" with Apple where each of the 3,889 rooms will have an iMac to "share memories and encourages personal expression." It also suggests a connection between the two companies for online booking and planning as well as interactive content at the hotel itself.
Few details beyond this are available, though the Fontainebleau opens the doors to its Vegas location in the fall.
Apple has a relatively strong presence in Vegas, operating two existing retail stores as well as a planned third store at Caesar's Palace.
Source: AppleInsiderLabels: Apple, Me.com |
| posted by Perimbean @ 10:00 AM |
|
|
|
|
|
Intel X48 vs. Nvidia nForce 790i Ultra SLI
The high-end chipset offerings from Intel and Nvidia couldn’t be more similar — at least at first glance. Both were designed to power the latest Core 2 Duo, Quad or Extreme processors, both support the latest technologies such as PCI Express 2.0 and DDR3-1600 memory, and both combine these with a plethora of interfaces. Both are highly overclockable and offer proprietary features to enhance performance.
The two target different types of enthusiast customers, however. Intel is the old bull in the chipset arena, owning roughly 50% of the core logic market, but has come a long way from providing reliable business products to also catering to the enthusiast. The X series, starting with the 925X, has been the enthusiast chipset line for some time, yet the Intel enthusiast series isn’t necessarily the best choice for everyone. The mainstream chipsets, such as the P965 and P35, have provided similar performance, interfaces and overclocking features as the 975X and X38 models. The enthusiast chipsets mostly support faster system speeds or optional dual graphics configurations.
The X48 release once again introduces a new chipset that does not offer substantial advantages over the mainstream products — the main difference this time is its qualification for FSB1600 system speeds. At the same time, X48 had support for DDR2 memory and ECC DIMMs removed on the specification sheet, although DDR2 is physically still supported.
Nvidia has addressed the enthusiast audience ever since, which is obvious by its aggressive branding and graphics-centered strategy, paired with a strong feature set. Not only was Nvidia the first firm to introduce Scalable Link Interface (SLI) dual graphics with the nForce 4 family, but it also introduced SLI-ready memory with EPP in the nForce 600. EPP stands for Enhanced Performance Profiles, a technology that allows the motherboard to automatically enable the fastest memory speed and aggressive memory timings. The Nvidia chipsets also were first to introduce comfortable management tools, and the overclocking utility nTune. The latest nForce 700 chipset family focuses on DDR3 memory, two-, three-way and four-way SLI based on two x16 PCI Express 2.0 slots, plus an additional x16 PCI Express 1.0 slot.
Though Nvidia is the multi-GPU graphics specialist, supporting both dual SLI with two SLI-ready graphics cards, three-way SLI with GeForce 8800 GTX or Ultra and quad SLI using two GeForce 9800 GX2 double whoppers, Intel does in fact also support dual graphics. With the introduction of the 975X chipset, two x16 PCI Express lanes were available in X series chipsets to host two ATI Radeon graphics card in Crossfire mode.
If you favor either ATI’s Crossfire or Nvidia’s SLI technology then your chipset choice has been made. If not, then you’ll find our comparison of a Gigabyte X48T-DQ6 and an Asus Striker II Extreme useful. Look at the features, performance, overclocking abilities and power consumption of the two high-end choices before making a decision.
Some Chipset History
Despite being first to market with new interfaces, storage controllers and other motherboard-related features, Intel had always been more than conservative with regards to hardcore features. This in spite of having a hardware basis that has been powerful, flexible and usually possessing a lot of headroom for clock speed increases. Overclocking was a taboo that had to be broken carefully over time, and it wasn’t until overclocking had progressed from being an unwanted phenomenon to everybody’s favorite pastime that Intel finally acknowledged it.
Although the X48 is still a 90 nm part, it is highly overclockable, and combines maximum performance with a rich feature set. However, the X48 is only a moderate advance over the X38: the main difference is official support for FSB1600 speeds. Intel has always had a focus on strong interface subsystems, which includes many flexible USB 2.0 ports as well as a powerful storage solution, which it calls Matrix Storage Technology. Although the feature set isn’t richer than Nvidia’s storage feature lineup, Intel has provided better throughput and I/O performance (see storage benchmarks for details). Dual PCI Express interfaces were also introduced with the 975X chipset, supporting ATI’s Crossfire standard.
Nvidia has gone in an entirely different way. For them, entering the chipset business was a logical consequence in order to provide enthusiast and mainstream solutions, rather than just graphics cards. The 3D company had a hard time entering the chipset market with the first nForce chipset generation. AMD’s Athlon 64 success was a key driver for Nvidia’s success of the nForce 3 chipset family. The fourth nForce generation was the first product to also support the Intel platform, as Nvidia wanted to introduce SLI dual graphics support for the AMD and the Intel worlds at the same time.
The nForce chipset family has for some time had more sophisticated dual network controllers that allow users to combine two Gigabit Ethernet ports and to prioritize network traffic (via the DualNet and FirstPacket technologies). The storage management for the six SATA/300 ports and one UltraATA/133 channel is consolidated under what’s called MediaShield technology.
The X48 is a classic chipset that consists of two building blocks: the northbridge 82X48 MCH (Memory Controller Hub) and the southbridge ICH9 (I/O Controller Hub). Intel’s MCH includes the Front Side Bus interface to the processor, the dual channel DDR3 memory controller and two x16 PCI Express 2.0 interfaces. The memory controller supports Intel’s XMP technology, which stands for Extreme Memory Profiles. This is very similar to Nvidia’s SLI-ready memory, as it allows the motherboard to configure the memory according to its maximum capabilities. This means that memory clock speed and timings will automatically be set to ideal performance values. Other auto-configuration mechanisms rely on memory’s Serial Presence Detect (SPD) ROM data, setting workable timings that won’t deliver maximum performance. Intel’s and Nvidia’s approaches are similar, yet you have to get either XMP or SLI-ready memory if you want to use this feature. This chipset not only supports DDR3-1066 and DDR3-1333, but for the first time DDR3-1600 is supported as well.
The ICH9 southbridge is available in three different flavors: ICH9, ICH9DO (Digital Office) and ICH9R (RAID). The last of these is used for enthusiast-class motherboards, as it offers six AHCI SATA/300 ports with Native Command Queuing (NCQ) support, together with flexible RAID configuration options (RAID 0, 1, 0+1, 5). A High Definition audio sound system is mandatory today, and Intel also included a Gigabit Ethernet port (10/100/1000 Mbit/s). Although Intel’s block diagram doesn’t mention 32-bit parallel PCI anymore, it is still part of the ICH9 southbridge. However, Intel wants the industry to focus on PCI Express: the ICH9 offers six lanes, which can be turned into one x4 PCIe slot and two x1 slots, or into a larger number of x1 PCI Express slots. Finally, the ICH9 comes with 12 USB 2.0 ports; SATA and USB ports can be disabled for security.
Technically, the X48 chipset can support all Socket 775 processors, as the socket has only undergone small modifications. However, BIOS support from manufacturers will be limited for Pentium 4 or Pentium D class processors. You can be sure that all X48 motherboards will support the entire Core 2 processor lineup, though.
If you don’t want to go for an expensive X48 motherboard, you can certainly select a P35 or one of the upcoming P45 motherboards. These offer almost the same feature set, with the exception of dual x16 PCI Express dual graphics. Native FSB1600 support requires P45, but most upper-mainstream P35 motherboards will run reliably when overclocked from FSB1333 to FSB1600, if you planned on overclocking your system anyway.
Unlike Intel’s strict split between enthusiast products (X48) and the mainstream line (P35, G35), Nvidia has a more comprehensive chipset portfolio. The nForce 790i Ultra SLI, as reviewed in this article, is the top model for enthusiasts. The key feature for the Ultra version is SLI-ready memory support, which goes as high as DDR3-2000 speeds. None of the nForce 790i chipset versions supports DDR2, only DDR3. The nForce 790i SLI (non-Ultra) also supports SLI-ready memory, but only up to DDR3-1333. FSB1600 support is available with both 790i SLI versions.
Apart from the 790i flagships, the nForce models 680i and 780i also support three-way SLI (twin x16 PCI Express 2.0 plus one x16 PCI Express 1.0 slot), but these go with DDR2 memory instead of DDR3. There is SLI-ready memory support for automatic settings of ideal RAM clock speed and timings, which is limited to DDR2-1200 speed. Since there won’t be much faster DDR2 memory, this is more than adequate.
Finally, there is an nForce 750i SLI entry-level model, which is limited to two x8 PCI Express 2.0 slots, and so does not support three-way SLI. Memory support is restricted to DDR2-800, and SLI-ready memory is not supported. The nForce 750i has to live with one instead of two Gigabit LAN ports, and with only four instead of six SATA/300 plugs. While Intel is about to release single PCI Express 2.0 into the mainstream with P45 at Computex, the nForce 750i SLI is already there.
The block diagram of the 790i Ultra SLI is similar to that of Intel’s X48. The 790i Ultra SLI SPP northbridge (System Platform Processor) connects to the CPU via a 400 MHz quad data rate bus (FSB1600). There are as many as 62 PCI Express lanes; 2 x 16 of them are used by the northbridge to provide two x16 slots, the remainder are provided by the southbridge. The dual channel DDR3 memory controller supports speeds up to DDR3-2000, although SLI-ready DIMMs are necessary to have the system select ideal clock speeds and timings automatically.
While Intel connects its MCH and ICH via a 2 GB/s interface called the Direct Media Interface (DMI), Nvidia utilizes HyperTransport. This is particularly interesting since this is the interface that AMD uses to connect its processors and core logic — and it is being used on an Intel platform. However, HyperTransport is necessary, as the third x16 PCI Express slot for three-way SLI (using eight lanes) has to be provided by the Media and Communications Processor (MCP) southbridge.
The 790i Ultra SLI MCP provides two Gigabit Ethernet ports with teaming features, ten USB 2.0 ports (versus 12 ports with Intel), and six SATA/300 ports plus an UltraATA/133 channel for two legacy devices — Intel dropped support for parallel ATA with the introduction of the P35 chipset line. Nvidia also still supports up to five 32 bit PCI bus master slots for legacy expansion cards. High Definition audio is pretty much essential these days, so it doesn’t come as a surprise to find an appropriate unit on the MCP as well.
Intel X48: Gigabyte X48T-DQ6
BIOS version: F7
As mentioned in our introduction, we used a Gigabyte X48T-DQ6 to represent Intel’s X48 chipset. The DQ6 family is Gigabyte’s top model, offering a 12-phase voltage regulator. There are massive copper coolers and a copper heat pipe connecting the voltage regulators, the northbridge and the southbridge. The two x16 PCI Express 2.0 slots are blue; there are more x1 PCI Express slots, and two 32-bit PCI slots are available for add-in cards (these are black and white). Gigabyte didn’t want to offer only the six SATA/300 ports of the ICH9R southbridge, so it put an additional controller on the motherboard, which provides two more SATA ports and an UltraATA/133 channel. The latter is important for legacy hard drives or optical drives. eSATA is supported via a separate slot panel.
If you still have a non-USB keyboard or mouse then you’ll be happy to find PS/2 connectors on the back panel. There we also found optical and coaxial digital audio outputs, standard audio jacks, two Gigabit Ethernet ports, a Firewire port and as many as eight USB 2.0 ports — four more are on the motherboard, but those require a slot adapter.
The energy-saving feature called Dynamic Energy Saver (DES) enables or disables voltage regulator phases depending on the processor load. Using a large number of phases ensures reliable voltage supply at high-performance environments, but it decreases power efficiency. With DES, the system switches off up to eight of the 12 phases, even when using a Core 2 Extreme QX6850. DES, however, requires a driver and software to be installed, so it can monitor the CPU load.
The CPU Throttling Feature is supposed to help to further reduce power consumption, but it reduces system performance noticeably: we found that WinRAR took 2:40 instead of 2:10, and we didn’t detect any reduction in power consumption either. These are the power requirement levels using DES and its various settings:
Idle DES Off 12 Phase = 105W DES Level 1 Normal 6 Phase = 100W DES Level 2; Advanced Power Savings 6 Phase = 100W DES Level 3; Extreme Power Savings 4 Phase = 94W
Load (using Prime95) DES Off 12 Phase = 201W DES Level 1 Normal 6 Phase = 191W DES Level 2; Advanced Power Savings 6 Phase = 187W DES Level 3; Extreme Power Savings 4 Phase = 180W
Please be aware that you can either use the EasyTune overclocking utility, or enable the DES power saving mechanism: combining both isn’t possible, which we believe is a shame. Even when overclocked, there are ways to avoid unnecessary power draw. EasyTune is based on a graphically overloaded interface, but it allows you to tweak and overclock all the major settings in order to overclock the system without having to go into the old-fashioned BIOS.
We reached FSB2100 speed on this motherboard, which is 100 MHz faster than the maximum system speed on the nForce 790i Ultra SLI motherboard by Asus. FSB1800 can be reached without any modifications: just select the FSB speed and leave everything else on auto.
nForce 790i Ultra SLI: Asus Striker II Extreme
BIOS version: 0402
The Asus Striker II Extreme is called the “ultimate gaming/overclocking platform” on the Asus website. This seems to be the right product to represent the Nvidia nForce 790i Ultra SLI chipset, then!
It is equipped with an eight-phase voltage regulator together with Asus’s Energy Processing Unit (EPU). Depending on the power saving settings, EPU is intended to reduce power consumption by modifying the use of the voltage regulators depending on the processor load. Even the two memory channels are supplied by two independent voltage regulators, which may not be ideal for power efficiency, but for performance and stability reasons. The voltage regulators, northbridge and southbridge are connected with a massive heat pipe solution, and these components are also covered by heat sinks. The northbridge has been prepared for liquid cooling solutions by the addition of an integrated water block called Fusion; Asus also includes the necessary mounting materials.
The two x16 PCI Express 2.0 slots are blue; the third slot using eight PCI Express 1.0 lanes is white. In addition, there is one x1 PCIe slot — the second (black) one was designed for the SupremeFX II sound module, which carries all necessary 3.5 mm jacks for audio. Digital coax and optical audio outputs can be found on the motherboard back panel, together with the two Gigabit LAN ports, six USB 2.0 ports, two eSATA ports, Firewire and a PS/2 mouse port. The remaining element on the back panel is a light switch called EL I/O that illuminates the back panel connectors if you want to plug in cables under your desk, where you typically lack light.
The Q connector is a small adapter that you use to connect all the internal cables such as power, reset or HDD cables. Once this is done, you can connect the adapter to the pin panel — obviously, this is easier and in fact, almost hassle-free. Finally, the LCD poster is a little LCD display that can be used to monitor system parameters without invoking any Asus application, such as while gaming. Other motherboard features are the Extreme Tweaker overclocking and tweaking software, COP EX component monitoring to prevent damage when components run overclocked, Q-Fan Plus temperature controlled fan speed management, and an on-board power switch.
We found that memory bandwidth and Prey performance were slightly faster on the nForce 790i SLI motherboard by Asus, but the X48 solution from Gigabyte provided the better results across almost the entire benchmark suite. Please be aware that the differences are minor and far from being noticeable, but for the record, Asus and the nForce 790i Ultra SLI won the benchmark comparison. Of course, this would be different for most of the 3D graphics and game benchmarks once two GeForce 9800GX2 dual-GPU graphics cards or up to three GeForce 8800 GTX or 8800 Ultra faced two high-end ATI Radeon HD3870 cards. In this case, Nvidia would win due to its current superiority in the graphics space.
We weren’t able to update the BIOS version using the Windows BIOS update utility, as the update feature didn’t find the appropriate files — we had to look for BIOS updates on the Asus FTP instead, which is a rather cumbersome task. We reached up to FSB2000 speeds on this motherboard, while the X48 board by Gigabyte was able to show even better overclocking results.
Source: TomsHardwareLabels: 790i, GeForce, Intel, nForce, Nvidia, X48 |
| posted by Perimbean @ 2:21 AM |
|
|
|
|
|
$5K PC Takes On $4.6M Supercomputer
1:50 PM - May 30, 2008 by Wolfgang Gruener
Antwerp (Belgium) - Recent advances in general purpose GPU computing are beginning to shift perceptions in supercomputing applications. Belgian researchers have assembled a relatively simple enthusiast PC with an emphasis on graphics processing capability, which beats a multi-million dollar supercomputer in its target application.
The desktop PC, called Fastra, was built with a focus on the development of new computational methods for tomography. Tomography is a technique used in medical scanners to create three-dimensional images of the internal organs of patients, based on a large number of x-ray photos that are acquired over a range of angles. As these 3D images can be quite large, advanced reconstruction techniques can sometimes require weeks of computation time on a regular PC. Which means that supercomputers are usually required to process computer tomography (CT) images.
The research group Vision Lab at the University of Antwerp came up with a different solution and constructed a PC that integrates four GeForce 9800 GX2 graphic cards (with a total of eight GPU cores) that runs CUDA-optimized tomography applications. The specifications include a MSI K9A2 Platinum motherboard, an AMD Phenom 9850 CPU, 4 x 2 GB Corsair TWINX DDR2 PC6400 memory, a Samsung Spinpoint F1 750 GB hard drive, a Thermaltake Toughpower 1500W Modular power supply unit as well as four MSI 9800GX2 cards. The researchers said that the cost of the system was less than 4000 Euro or about $5300.
It isn’t quite a tricked-out gaming system and the 3DMark06 score is just above what your average PC can manage to come with today (12,603 points). However, it is the CUDA application where this PC really shines. Compared to the 512-processor, $4.6M CalcUA supercomputer purchased in 2005, the PC can be more than a match: The projection of image slices took 23.4 seconds on the supercomputer and 35.1 seconds on the PC. The reconstruction of the slices was displayed after 67.4 seconds on the supercomputer systems and after just 52.2 seconds on the PC. The Vision Lab crew now believes that a real-time construction is possible through GPUs and is now building a cluster of such systems.
While it is an impressive example how GPUs can be applied in non-traditional ways, there are a few notes to be added. Of course, GPUs cannot replace traditional supercomputers, which still can be applied to applications with a broader range. Also, supercomputers usually carry huge memories, often in the Terabyte range, which cannot be matched by today’s GPU clusters. When we talk to scientists working with supercomputers and GPUs, they typically believe that future supercomputers will not completely transition to GPU clusters, but may develop into systems that consist of a traditional supercomputer structure as well as GPU capability.
An interesting side note about the Fastra PC is its motherboard. Eagle-eyed readers may have noticed that the MSI K9A2 Platinum board is not an Nvidia SLI-based board, but uses AMD Crossfire (780 chipset). The simple reason to choose this board may have been cost, but it is unlikely to impact the performance of the system: CUDA does not support SLI at this time, which means that the GPUs have to communicate with each other as well as with the CPU via PCI Express. The researchers claim that they have not seen any impact on performance and the GPUs apparently are scaling well.
Source: TomsHardwareLabels: AMD, Corsair, Crossfire, GeForce, MSI, Nvidia, PC, Phenom, Samsung, SLI, Supercomputer |
| posted by Perimbean @ 2:07 AM |
|
|
|
|
|
Microsoft urges Windows users to shun 'carpet bombing' SafariRare security warning from Redmond
By Dan Goodin in San Francisco
Published Saturday 31st May 2008
Microsoft's security team is advising users to stop using Apple's Safari browser pending investigation into a quirk that allows miscreants to litter their desktop with hundreds of executable files.
Windows users who visit a booby-trapped site with Safari could be forced to download and execute malicious files with no prompting, Microsoft says. The "blended threat" is a result of the default download location in Safari and the way the Windows desktop handles executable files.
This Microsoft advisory suggests users "restrict use of Safari as a web browser until an appropriate update is available from Microsoft and/or Apple."
The recommendation comes a week after researcher Nitesh Dhanjani reported that Apple's browser doesn't seek user permission before downloading certain types of files. Even when encountering malicious iframes - a common occurrence these days even on the most trustworthy of sites - Safari obediently does what it's told to do, including downloading a file hundreds of times.
Apple's security pros, upon learning of the so-called carpet bombing vulnerability, said they didn't see it as a significant threat. A researcher in Cupertino wrote to Dhanjani that it may get fixed at some point down the road as "a further measure to raise the bar against unwanted downloads," but said it could take a quite a while, if ever, for that to happen.
Apple's unfortunate refusal probably explains why Microsoft's security arm has resorted to the unusual recommendation. We can't remember the last time Redmond counseled users to avoid installing a mainstream product for security reasons. Apple representatives didn't respond to a request to comment for this story.
And before any Mac users decide this is an issue they can safely ignore, remember this: While Microsoft's recommendation obviously is limited to Windows users, Dhanjani says the carpet bombing scenario can play out on OS X, too. ®
Source: TheRegisterLabels: Apple, Microsoft, Safari |
| posted by Perimbean @ 2:03 AM |
|
|
|
|
|
Apple store detains teens for installing iPhone gameCoppers called over demo unit download
By Austin Modine → More by this author
Published Friday 30th May 2008 21:53 GMT
--------------------------------------------------------------------------------
Four teens in Palo Alto, California, say they were detained and photographed by an Apple store after they downloaded a third-party application to an iPhone demo unit.
The youngsters told Mercury News that a manager of their local University Avenue Apple store called for police reinforcement after discovering the racing game, "Raging Thunder" had been installed on the touch screen phone. High school senior Daniel Fukuba was demonstrating the iPhone's features to his chums, when he downloaded the game.
Both the store employees and manager were at first unperturbed by the boys playing around with the demo unit. But after they left the store, a manager bolted out and demanded the teens return. The manager then called the police.
Sgt. Sandra Brown of the Palo Alto Police confirmed to Mercury News that they were called in by the Apple store, although no arrests were made.
The group claims they were also lectured by the manager about the dangers of "hacking," were all photographed, and told their pictures would be sent to surrounding Apple stores "so they'd be on the lookout for us," according to one of the teens involved, Noah Rogers. In all, they were detained for two and a half hours.
The boys also say they were issued a ban from all Apple stores, although an Apple spokesman has denied the claim.
In March, Apple opened the iPhone SDK to let third-party apps run on the platform — although apparently this is only something that should be attempted in the privacy of your own home. ®
Source: TheRegisterLabels: Apple, iPhone |
| posted by Perimbean @ 1:38 AM |
|
|
|
|
|
Mobile phone users itching to take a bite of Apple
Garry Barker
June 1, 2008
THOUSANDS of Australian mobile phone users are trembling with anticipation at the imminent arrival of Apple's revolutionary iPhone, described by commentators as "one of the most talked-about consumer products in history".
The Mac community believes that new 3G, wireless broadband iPhones will be announced by company founder, chief executive and all-round visionary Steve Jobs in San Francisco at the opening of Apple's Worldwide Developer Conference on June 9.
This will coincide (June 10 Australian time) with the opening of Apple's first Australian company store on the corner of King and George streets in Sydney.
Apple Australia is preserving its usual monastic silence on anything to do with iPhone, although Optus and Vodafone have briefly said they will be selling it in Australia later this year.
Telstra has yet to announce its intentions, but few believe it can ignore iPhone. Telstra is also about to open a new shop, on the corner of Bourke and Swanston streets in Melbourne, and iPhone could feature there. That opening, however, will be towards the end of June.
Current iPhones are 2.5G devices and work best on Telstra's original GSM network, which, alone in Australia, has the higher speed EDGE technology that iPhones can use.
Until now, Apple has avoided venturing into 3G, mainly because available chipsets were physically large and power hungry, meaning a big phone and poor battery life. But microprocessor companies have been working on the problem. The new iPhones (there may be more than one model) are expected to use the SGOLD3H chipset from Infineon. This can handle the High Speed Packet Access data transmission technology used by Australian 3G networks and run at 7.2 megabits per second, giving good video and download performance.
Having established iPhone by doing exclusive deals in the US and Europe with single carriers in each region, Apple is now aiming for mass sales, with much attention focused on the big European and Asian markets in which 3G capability is vital.
Apple's ploy now is to take advantage of the high consumer interest now generated and sell it across entire markets, says Gartner Group mobile technology analyst Robin Simpson.
"In the beginning everyone was sceptical because Apple had no mobile phone history and they thought it could not do one well. Well, everyone was proven wrong. They have done a brilliant job," he said.
Figures of sales so far are hard to come by, but the iPhone's reputation and desirability far exceeds the 6 million or so it has sold since its introduction in the US in June last year.
It went on sale in Britain, and then in Europe, in November and gained a solid place in its high-end smartphone category, especially considering it did not offer 3G capability. Australia will be a significant iPhone market, Mr Simpson says.
Industry estimates suggest there are between 50,000 and 60,000 iPhones already in use here, bootlegged from vendors in Hong Kong and openly sold in computer shops in Melbourne and other Australian cities, or bought from US and Asian sellers trading on eBay.
It is relatively simple to "hack" an iPhone so that it can be used on an Australian mobile phone network, and nobody seems to mind that doing it voids the warranty.
Source: TheAgeLabels: Apple, iPhone |
| posted by Perimbean @ 1:34 AM |
|
|
|
|
|
David Sun of Kingston Technology
