Friday, 19 August 2011

Engadget Primed: all mobile displays are not created equal

The quality of a mobile phone's display is arguably the most important factor to consider when you establish a relationship with a handset. It's inescapable, really. Whether you're playing a rousing game of Robot Unicorn Attack or (regrettably) drunk-dialing an ex, it's the one interface element that you're consistently interacting with. It's your window to the world and your canvas for creation, and if it's lousy, it's going to negatively influence everything you see and do. Today, we're delving into the world of mobile displays, where we're aiming to entertain and edify, and hopefully save you from making regrettable decisions -- when it comes to purchasing new phones, anyway.

In this edition of Primed, we'll be examining the different qualities and underlying technologies of several displays, starting with the ubiquitous TFT-LCD and moving through the nascent realm of glasses-free 3D and beyond. We'll also be addressing the importance of resolution and pixel density. Finally, we'll be scoping out a handful of upcoming technologies -- while some are thoroughly intriguing, others are just plain wacky. Go ahead... buy the ticket, take the ride, and join us after the break. It's Primed time.

The Engadget Interview: HP's Stephen DeWitt discusses the state of webOS

The last time we spoke with Stephen DeWitt, the head of HP's webOS global business unit was on the defensive about the TouchPad. In spite of -- or perhaps because of -- the anticipation of the device, the first webOS tablet received a lukewarm reception at hands of reviewers. DeWitt vigorously defended the slate against the critics, suggesting that technology writers had been approaching the product the wrong way. A lot has happened since that conversation, of course, including reports of unmoved stock and, more significantly, yesterday's news that HP would effectively be discontinuing production on its webOS devices, the TouchPad included.

There's been a good deal of confusion around precisely what yesterday's announcement means for both the company and the mobile operating system that it picked up with its purchase of Palm back in April of last year. In spite of his understandably packed schedule, DeWitt sat down with us to set the record straight and shed some light on the future of webOS -- a future both he and the company remain rather optimistic about.

Sprint's Epic Touch 4G may be first US-bound Galaxy S II, arriving September 9th?

Samsung is expected to unveil its US Galaxy S II lineup on August 29th, and if you have that date highlighted on your calendar for that very reason, this rumor will likely turn you ancy for the next ten days. We're hearing from SprintFeed that the Samsung Epic Touch 4G, Sprint's variant of the upcoming device, may win the race to market by launching on September 9th. It's no surprise that the Now Network would beat T-Mobile, given whisperings that the Samsung Hercules won't make it to stores until October 26th, but we're still unsure of when to expect Big Red and Ma Bell. Regardless of if this report is accurate, the next two months are still gearing up to be rather exciting, wouldn't you say?

Retina Display-equipped iPad 3 looking more and more likely for early 2012

iPad WITH A VENGEANCE
Well, the Wall Street Journal's headline certainly isn't going to surprise you -- Apple Developing New iPad -- that has been a safe assumption since the iPad 2 was announced. Rumors are circulating though (some, for quite a while) that are painting an increasingly convincing picture of a Retina Display-equipped slate shipping in early 2012. Reports are in that suppliers have already received Q4 orders for parts that may eventually find their way into Apple's next tablet. Can we say for certain that the iPad 3 will hit shelves in Q1 of next year sporting a 2048 x 1536 screen? No, but it's starting to look more and more likely.

Elgato HDHomeRun iPad app brings (some) cable channels to the tablet

Whether or not your cable TV provider has decided to offer live TV streaming to your mobile devices, you can do it yourself with the new HDHomeRun iPad app from Elgato. Paired with one of SiliconDust's new HDHomeRun Prime CableCARD tuners, the $17.99 app can tune into cable channels that are sent without encyption or are marked copy freely (varies from provider to provider) and even allows users to record them right on the app. Restrictions include that it only works on the iPad 2, with standard definition MPEG-2 channels and (for now) only the most recent HDHomeRun hardware is supported. Check out the video demo for more information on exactly how it works or click the source link to get to iTunes and buy -- in a market suddenly flooded by CableCARD tuners maybe this extra functionality is just what's needed to tip the balance between one device or another.

Andy Pad Pro makes film debut, shows off bird slinging skills (video)

Still waiting to get your affordable Android tablet on? You've got a little over a week to go, but here attempting to tide you over is a montage of the Andy Pad Pro. First made official in July (alongside a video of its lesser sibling), the Gingerbread tablet still rocks a 1.2GHz Cortex A8, a "powerful 3D GPU" and 1080p HDMI output. And at £179 (or about $290) the 7-inch capacitive slate won't break the bank. Ready to see it and its more affordable brother shred through a bevy of Android games? You know the drill: hop on past the break, bud.

Let the liquidation begin -- HP's 16GB TouchPad on sale for $99

Well, that didn't take long. It's only been a day since HP announced its intention to discontinue production of webOS devices, and it looks like the outfit's already throwing its flagship tablet in the bargain bin. PreCentral reports that HP sent out a memo to its affiliates asking them to cut the price of the ill-fated TouchPad to $99 and $149 for the 16GB and 32GB editions, respectively. The date apparently put forward by HP is August 20th, but at least two Canadian retailers (Best Buy and Future Shop) have already made the cut -- though both outlets show the sale ending August 22nd. As proof that being fashionably late isn't always a good thing, the white 64GB edition has surfaced on HP's US site, sporting a $600 asking price, leaving us to wonder, well: why now? If all of that isn't enough to turn your smile upside down, webOSroundup is playing the bearer of bad rumors, with what they say is an internal memo from AT&T stating that the carrier's launch of the Pre 3 has been "completely cancelled."

Update: Looks like the SlickDeals forums folks have discovered a way to bring those Canadian prices down south, and are reporting that they will be official in the US on Saturday morning. Hit up the link below for details.

Top Stories Let the liquidation begin -- HP's 16GB TouchPad on sale for $99 4 hours ago The Engadget Show is live, here at 6:00PM ET! 7 hours ago The Engadget Interview: HP's Stephen DeWitt discusses the state of webOS BREAKING NEWS 8 hours ago Engadget Primed: all mobile displays are not created equal 10 hours ago Engadget's back to school guide 2011: fun stuff! 13 hours ago All News Reviews Reviews Mobiado's Grand Touch phones ditch the Nexus S plastic for a precious metal finish

Nothing quite screams money to burn like last year's tech gilded in pricey, CNC-machined metals. Purveyor of mobile extravagance, Mobiado, has recently rolled out the red carpet for two new additions to its Grand Line -- the Grand Touch and Grand Touch GCB. These high-end exercises in wireless hauteur are merely gussied up versions of Samsung's Nexus S, sporting the same ol' 1GHz Hummingbird processor, 16GB of storage, 512MB RAM, 4-inch Super LCD display and front-facing VGA / 5 megapixel rear camera. Available in either anodized aluminum or solid brass with gold plating (take a guess which is costlier), the sapphire crystal-backed duo comes unlocked with quad-band GSM, and tri-band UMTS / HSDPA radios, so you can stay connected aboard that yacht in St. Tropez.
Bandwidth caps are all the rage these days, in North America and around the world. We're talking about the notorious ceilings on how much broadband data you can use before your ISP starts charging you extra, or slowing down your connection, or shaping your traffic use (eg, blocking your access to certain applications).
AT&T's shiny new caps started on May 2: 150GB a month for DSL and 250GB for UVerse. AT&T customers will get a $10 charge if they cross the line, with that ding repeated for every 50GB thereafter. Comcast set a 250GB cap in 2008. Canadian ISPs have had them for a while, too.
Ars readers sometimes write in asking us how they can monitor their bandwidth use to avoid extra charges. Here are some links and pictures from major ISPs, which are broadly representative of the tools used by most Internet providers.

USA

AT&T offers a monitoring site here. Login with your user name and password, and you'll get a report like this.
An AT&T DSL data use report
This is from the AT&T DSL account retained at my end of the Ars Orbiting HQ. Our typical monthly usage has averaged out to 43GB a month over the last four. That includes plenty of Apple TV Netflix viewing, Skype video conferencing, Pandora listening, and YouTubing on top of the usual low bandwidth consuming activities, like website viewing and e-mail.
Comcast has a similar meter, which offers the same basic features: the metric for your cap (250GB), how much of the allotment you have used, and how much you have left to consume.
Comcast's usage meter

Canada, UK, Australia

If you are a Rogers Cable customer in Canada, you've got to log into your account, register, and then you can track your usage on a fairly detailed basis (at least according to the instructions), including use by day.
Rogers usage meter
The process is similar for the United Kingdom's BT broadband service. Log in here. Go to "manage your services," hit the "broadband" tab, click "view your broadband usage," and you'll see something like this:
BT usage meter
Needless to say, there are too many ISP services around the world for us to document. Last on our tour is Australia's BigPond ISP. Instructions here. And here's what the monitor looks like:
BigPond usage meter

News Guides Reviews Upgrade to a Premier Subscription Customize ▾ OpenForum Login/Join One Microsoft Way Has Microsoft just ruined Windows Home Server?

Microsoft's Windows Home Server is a funny little product. The company's ambition when developing the product was to have us all run little home servers: small, low-power, appliance-like machines with some network connectivity and gobs of storage. We'd use these home servers as a place to back up our PCs, share files and printers across our home networks, stream media to our Xboxes, and gain remote access to our files when away from home.
In practice, most of these things can be done perfectly well with a normal desktop version of Windows. Windows Home Server does have some advantages—it had a management front-end that let the server be easily controlled remotely, and it is based on Windows Server 2003 to slim down its own hardware demands—but for the most part, it isn't doing anything too unusual. As a result, Windows Home Server has remained a niche product. Much loved by its users, but never really making it as a mass-market success.
It does, however, have one special feature, a feature without any real equivalent in any other version of Windows, whether for desktop or for server. That feature is called Drive Extender. Conceptually, Drive Extender is quite simple: it allows multiple hard disks (regardless of interface or size) to be aggregated to provide a single large pool of storage. Folders on the pooled storage could also be selectively replicated, meaning that Drive Extender would ensure that copies of the files were found on multiple physical disks.
For a home fileserver, this is obviously a very handy capability. It allows simple ad hoc expansion of storage—no RAID rebuilding, no need to match disk capacities, no need to stick to any drive interface—and does so without the inconvenience of multiple drives, each of which has to have its free space managed manually.
For Windows Home Server's small, but vocal, following, Drive Extender is arguably the most important feature. It's what makes Windows Home Server something better than simply plugging some USB disks into a PC. Without Drive Extender, Windows Home Server as a product makes a lot less sense.
Indeed, Drive Extender was fundamental to the home server concept. A home server as originally envisaged by the Windows Home Server team should have, in essence, infinite storage, and storage that should be transparently extensible. This is why RAID, though common in corporate environments, isn't an viable solution for most home users. With RAID, you can't simply add another disk whenever you need more space. RAID requires management and planning.

Enter "Vail"

Windows Home Server version 2, codenamed "Vail," is currently in development. It is intended to give Windows Home Server richer backup capabilities, and to make it more extensible and simpler to set up and use.
Vail betas have, until now, also included a new and improved Drive Extender—one designed to make this key feature more robust and reliable. "Until now," because Microsoft has just announced that Vail will no longer include Drive Extender. Arguably the most significant feature of Windows Home Server, now gone.
The response from Vail testers and Windows Home Server users alike has been astonishment and outrage. A bug raised against Vail demanding the feature's restoration was filed almost immediately after the announcement, and has rapidly accrued hundreds of votes and many angry comments. This is not a popular decision at all, with many of the complainants saying that Drive Extender was the only reason they used Windows Home Server in the first place.
Drive Extender was also set to be a feature of the forthcoming Small Business Server 2011 Essentials (codename "Aurora") and Windows Storage Server 2008 R2 Essentials (codename "Breckenridge") products, giving it reach beyond the consumer market. Microsoft, for its part, says that after soliciting feedback from both consumers and small businesses, it no longer saw a need for the feature. Large hard drives are now readily available, and Microsoft is saying that these represent an adequate alternative to the Drive Extender technology.
With the removal of Drive Extender, it's hard to see what the point of Windows Home Server really is. A Drobo is probably a better solution for most file storage needs, and tasks like printer sharing and remote access are either built in to Windows anyway or available with free Windows software like Windows Live Mesh. Without the expandable storage, Windows Home Server's role as a file server, surely its raison d'être, is severely compromised.
The only winners from this decision are hardware companies. Aurora and Breckenridge customers will just buy RAID systems instead, and customers after Vail-like functionality will buy Drobo units.

The real reason for its cancellation

The current Windows Home Server's Drive Extender implementation is actually pretty peculiar. Disks pooled together in Windows Home Server are just regular NTFS-formatted disks. You can take them out of the server, plug them into another machine, and see all the files. There's nothing fancy going on with the disks themselves. Instead, Drive Extender provided a fileshare that gave a unified view of all these pooled disks. Behind the scenes, the software would periodically copy files between the physical disks in order to ensure that any folders that were meant to be duplicated indeed were.
This worked, more or less, but it wasn't entirely robust. Data loss bugs cropped up when, for example, using Office documents on the pooled share afflicted Windows Home Server, and these persisted long after the product's initial release. To this day, Drive Extender can change the timestamps of files that it duplicates. These flaws are likely the reason that the technology was never used in any other products.
With Vail, Drive Extender was completely rewritten in a manner that should make it both more flexible and more reliable. Instead of using regular NTFS disks, Vail inserts a layer underneath the filesystem. This layer was responsible for distributing blocks of data between disks, replicating them to ensure fault tolerance, and de-duplicating them to make the system more efficient (something especially valuable for a backup server). The problems with the old system—corruption of in-use files, delays before files are duplicated, incompatibility with EFS encryption, errors in timestamps—were banished for good with this new layer.
These changes caused some backlash among the Windows Home Server community, as they meant forfeiting the ability to just take a drive out of a Windows Home Server system and stick it in another machine to access its files. Many users felt this to be a significant step backwards.
They are wrong. The new layer resolved many fundamental flaws with the previous mechanism, and could have served as a foundation for building even richer capabilities in the future, in a way that the old (current) Drive Extender just can't. Giving up the ability to use the disks outside of the server is, in the long run, a price worth paying.
Unfortunately, the new block layer in Vail doesn't quite work right. Just like Drive Extender in Windows Home Server, there have been bugs. Different bugs, but bugs all the same. Microsoft hasn't gone into explicit detail about what these problems are, but there were some issues with its ability to correct errors, and some Small Business Server testers reported application compatibility problems.
So instead of fixing the flaws, and potentially delaying the three products dependent on Drive Extender, Microsoft is killing the feature altogether.

Broader implications

Windows Home Server is being ruined because of issues that affected Small Business Server workloads, and as a result Vail will be, for most or all users, basically pointless. Windows Storage Server 2008 R2 Essentials, which is essentially Windows Home Server for small businesses, is also badly hurt by the decision; perhaps not quite as badly, due to greater use of RAID hardware in its target market, but it nonetheless loses a lot of flexibility.
The impact on Small Business Server 2011 Essentials and Windows Storage Server 2008 R2 Essentials is smaller, partly because it's not a regression (there is no pre-existing user base expecting this functionality), and partly because they still do other things that justifies their existence. Small Business Server 2011 Essentials should be the cornerstone of a small company's office network; while enriching its file server abilities is useful, the loss isn't catastrophic.
The bigger problem is that this decision means that Windows is still no closer to having the kind of flexible, powerful storage subsystem that is a feature of competing platforms like Linux and Solaris, as well as dedicated hardware devices like the various Drobo boxes. Advanced filesystems like the Sun-developed ZFS include this kind of flexible block storage, and the new Drive Extender was the first step towards giving Windows similar capabilities.
The current Drive Extender was clearly never going to be able to deliver this kind of flexibility, which is why it never saw any usage beyond the current Windows Home Server version. But the new Drive Extender was intended for more than just Windows Home Server: its design and features were believed to be good enough for real server workloads. These are features Windows should have, and something that Microsoft needs to get working. Even without Vail, this is work that Microsoft should be doing.
If Microsoft is going to stick with its decision and remove Drive Extender across the board, the company might as well cancel Windows Home Server altogether. I think, however, this is a bad decision.
A better solution would be to take the time to make the new-style Drive Extender work. Ship Aurora according to the current schedule (it's due some time next year), but delay Vail and Breckenridge until the feature works. Once it's robust, retrofit the feature back into Aurora as a downloadable extra, and make it a built-in component of Windows 8's corresponding server version.
Windows Home Server needs Drive Extender. The Windows platform as a whole needs Drive Extender. Hobbling Vail, and removing Drive Extendor, cannot be the right thing to do.

1Gbps fiber for $70—in America? Yup.

1Gbps fiber for $70—in America? Yup.
Installing aerial fiber optic lines for the Sonic.net trial project
American ISPs have convinced us that Internet access is expensive—getting speeds of 100Mbps will set most people back by more than $100 a month, assuming the service is even available. Where I live in Chicago, Comcast's 105Mbps service goes for a whopping $199.95 ("premium installation" and cable modem not included). Which is why it was so refreshing to see the scrappy California ISP Sonic.net this week roll out its new 1Gbps, fiber-to-the-home service… for $69.99 a month.
Sonic.net has been around since 1994, selling DSL service in California, but it has recently expanded into fiber; the company has even secured the contract to manage Google's own 1Gbps fiber network that will connect 800+ faculty homes at Stanford University.
Sonic.net's new approach to broadband involves stringing its own fiber lines to homes and offering bargain-basement pricing; indeed, the new 1Gbps offering is the same price as the company's earlier bonded 40Mbps DSL offering (in which two phones lines each provide 20Mbps of bandwidth to a home). The price even includes home phone service.
Is this really a sustainable model? After all, Comcast offers 1.5Mbps service for a list price of $40; Sonic.net's new offering is more than 600x faster at only twice the price.
Dane Jasper, Sonic.net's CEO, tells me that the new fiber-to-the-home deployment is a trial and will reach about 700 homes when complete. "Honestly, only as those wrap up will we have a complete picture of the economic model," he says. "But I believe that fast service for a low cost is possible."
If the pilot in Sebastopol, California goes well, Sonic.net hopes to expand the service across the region.
Jasper doesn't think like a typical US Internet exec; in an interview last year, he made clear that his company tries to avoid artificial limits as a way to make more money. "The natural model when you have a simple duopoly capturing the majority of the market is segmentation: maximize ARPU [average revenue per user] by artificially limiting service in order to drive additional monthly spending. But fundamentally this is the wrong model for a service provider like us, and we have looked to Europe for inspiration… I believe that removing the artificial limits on speed, and including home phone with the product are both very exciting."
Though the current trial is small-scale, Sonic.net's pricing reminds us just how much room there is in the US Internet market for truly disruptive pricing of the kind that Google has been promising—but on a much larger scale—with its 1Gbps fiber builds in Kansas City, Missouri and Kansas City, Kansas.

It didn't have to end this way: what HP should have done with webOS

It didn't have to end this way: what HP should have done with webOS
HP has missed out on a fantastic opportunity with webOS. The company was in a position, by hook or by crook, to give webOS the kind of wide distribution that even Apple would be impressed by. It just had to spend some cash to do it. As an operating system, webOS has what it takes to be a success. The operating system's user interface was well-received, and it has strong concepts, such as unified messaging and card-based multitasking, that rival platforms are only starting to compete with. It also had a development model familiar to millions of Web developers.
What it needed was a bit of momentum. A reason for those Web developers to start developing for it, a reason for mobile operators to start caring about it and promoting it. Palm's advertising, with the creepy girl, was lackluster. HP took that to the next level. I was excited about, and interested in, the Pre3 when it was announced earlier in the year. But I didn't even realize it had launched a few days ago, such was HP's total unwillingness to promote the thing. Bring back the creepy girl—at least it's something.
And yet it seems to me that HP had a perfect opportunity to get webOS into people's hands across the world, thanks to its enormous PC business. Granted, it's now obvious that HP doesn't like its PC business very much. But its reach and penetration, across both the corporate and consumer worlds, is substantial.
So here's what I would have done if I ran HP. I would recognize that a successful mobile and tablet platform can earn a boatload of cash. It's a market that's absolutely worth going after. I would recognize that with webOS I had the fundamentals of this platform, but lacked the critical mass of users that is required to create a sustainable platform.
So I would have done two things. One cheap and easy, one expensive and easy.

Paying the price, whatever it is

The cheap and easy one is to port webOS to Windows. Or rather, to port the webOS runtime environment. Not the Linux-y bits, but the JavaScript-y, HTML5-y bits. There's probably not even a huge amount to do here; WebKit already runs on Windows, and underlying technology like node.js is gaining a first-class Windows port. I'd then develop a bunch of neat applications, plumb the entire thing into webOS's App Catalog, and preinstall them on every HP desktop and laptop that the company sells. I'd also make it a free download with a simple user-friendly installer.
This wouldn't be enough to save webOS. But it would be cheap and easy to do, and it would put the technology in front of people. It would make the webOS development stack something that developers knew about, and thought about, and could trivially use to make their own applications. It's all about making developers recognize that webOS exists and think "oh hey, this is neat."
Now for the expensive part. Give away the hardware. You buy a cheap HP consumer laptop or netbook? You get a Veer. Buy a more expensive one? You get a Pre3 or a TouchPad, your choice. Run the promo for a quarter or two. Advertise all over the place. Pay whatever it takes to get 30-day SIMs with data plans from mobile networks and preinstall them in the phones. Preload them with Angry Birds. Make sure that anyone buying an HP PC will get this cute little gizmo that works and does all sorts of neat stuff out-of-the-box. The PCs should also change a bit. The TouchPad and webOS phones can sync webpages with each other just by dint of being in close proximity; the PC should be able to play the same game. Encourage people who might not otherwise be interested in the tablet or the smartphone to experiment with them.
Not everybody will care about their free phone or free tablet—that's a given. So those folks could give it to their kids, friends, or neighbors. Some devices would no doubt end up languishing in drawers or attics. But plenty of them are going to get used.
No doubt about it, this will cost a big chunk of money. But writing down 250,000 unsold TouchPads also costs a big chunk of money. Buying Palm for $1 billion and then doing nothing with its really neat technology costs a big chunk of money. Your share price dropping 20 percent after you announce the plans to kill webOS hardware and sell off your PC division isn't exactly great financial news either.
And there's still no guarantee of success. Perhaps everyone would try webOS and hate it. Perhaps developers would ignore several million new webOS users, leaving it still desperately in need of applications.
But if they liked it? HP would have had a thriving ecosystem on its hands. Millions of people trying webOS, liking it, recommending it to their friends and family. A development framework that millions of developers already know and understand. An app store with customers in abundance. HP would have been well on its way to smartphone success.
If HP had primed the pump only to find that success still didn't flow, then sure, kill it off, because plainly it's never going to make it. HP, however, never even gave webOS a fighting chance. This wasn't a great big spectacular failure from a company that tried hard, gave it its all, but still couldn't succeed. It was half-hearted capitulation. That's a missed opportunity, if ever there was one.