Sony Ericsson Xperia Play review

Decent hardware, but in current format doesn't cut it for gaming

 


After a while in the mobile wilderness, Sony Ericsson is back with a splash with its new line-up of use-focused Xperia handsets. The Arc is its flashy line leader, impossibly thin and sporting a gorgeous Bravia-powered screen, the Pro its business-focused model and the Xperia Neo (read our review here) a good-quality mid-range option.


But all the hype has been about the Play, the long-mooted 'PlayStation Phone' that comes promising "real" console gaming on a mobile for the first time. Make no mistake, in the wake of the iPhone's App Store success, this is the selling point that Sony Ericsson is focusing on, and for good reason, as the Play is no more than competent in most other areas.

Sony Ericsson Xperia Play: Specifications


It has a perfectly decent 4-inch capacitive multi-touch LCD touch screen with a 480x854 pixel resolution and 16.7 million colours, and a 5.1-megapixel camera-cum-video recorder with autofocus, flash, image stabilizer and geotagging, which knocks the socks off the iPhone but doesn't set the world on fire outside Apple's camera void. It comes fully Gingerbreaded up with Android 2.3.2, which is swift and intuitive, with the usual media player offerings, customisable interface and a very nice cover flow-style integrated Twitter feed. Connectivity-wise it has the expected HSDPA-backed Wi-Fi for internet access, 3G and Bluetooth functionality, as well as A-GPS. There is 400MB of internal memory, but an 8GB microSD card is included and this can be expanded up to 32GB.






So far, so standard. But let's face it, with the frankly gorgeous and similarly priced Xperia Arc rocking a much more impressive rig, not even Sony Ericsson is expecting this to sell on these stats. This is meant to be the first Android handset to kick Apple in the joypads. So what of the all-important gaming inclusion?

Sony Ericsson Xperia Play: Controls

Well, the PlayStation controls hidden within make the handset a little on the chunky side (119 x 62 x 16mm), but less intrusive than we'd imagined. When closed, it's reassuringly solid and expensive looking, unlike the slightly plasticky Neo. At 175 grams it's by no means light, but not as pocket-denting as you might think, so its passes first impressions with flying colours. Get one out on the Tube and people will be impressed, if that's your thing.






The familiar PlayStation controls of a D-pad, four face buttons, two shoulder buttons, select and start, and the not so familiar flat analogue thumb areas, slide down on two runners, but once freed of their casing the build begins to feel a little flimsy. The Play clearly has its design foundations in the reworked PSP Go, although it's noticeable that beyond the iconic square/triangle/circle/cross buttons there's no PlayStation branding anywhere on the handset, and the slide mechanism doesn't feel like it will stand up to much punishment. It's had to be made impossibly thin so that the overall handset isn't too bulky, but the result is it appears eminently breakable under intense gaming.


When playing with the controls slid down, the handset also feels very top-heavy and not particularly comfortable to hold. We didn’t get much gaming joy from the slightly bizarre analogue thumb pads, but we can confirm they do kind of work. The shoulder pads are hidden behind the screen and not at all ergonomically designed, while the D-pad feel plasticky and not firm like we've come to expect on dedicated gaming systems, distancing you slightly from the experience. It's functional but not much more. Importantly, it doesn't really feel like you're playing a PlayStation. Or a PSP, for that matter.

Sony Ericsson Xperia Play: Games

Then there are the games. As advertised, there are more than 60 titles available for the Xperia Play at launch. But only six of these are actually purpose-built, PlayStation-certified ones, which you access through the PlayStation Pocket hub, with the rest standard Android games reconfigured to the Play's controls. But unfortunately this lack of conformity – some would say the defining issue between Apple and Android – means the standard and speed of experience varies wildly, and the button reconfiguration even more so. At present, the term 'PlayStation Phone' seems very much a misnomer.


Qualcomm's 1Ghz Snapdragon CPU and Adreno GPU processor promises smooth 60 frames per second, although this doesn't seem to be being utilised at the moment and it’s a shame it’s not dualcore. While FIFA (10, not 11, unfortunately) is palpably faster than the iPhone version, the graphics are poor and controls so unresponsive, we found it more effective to use the touch-screen's virtual stick instead - although the screen slides around a lot on the control-board's runners, so screen-based gaming is far from ideal. The quoted gaming battery life of five and a half hours also seems slightly optimistic.






Next up, the intergalactic shooter Star Battalion takes a mammoth two and a half minutes to get from app press to actual gaming. After much bashing of all the buttons trying to skip the stuttering loading screens, we found that a simple tap of the touch screen at least skipped the relentless intros, despite this not being flagged up anywhere and you being in PlayStation control mode at the time. Confusing.






The one pre-installed PlayStation-certified game, colourful 3D platformer Crash Bandicoot, is fun but the graphics disappointingly don't fill the widescreen display, with big ugly black borders either side, and while it's initially fun, it hasn't aged well over the 15 years since release. This could be a big problem, as while this is undoubtedly a PlayStation experience, it's a really, really old one. The upcoming NGP will boast PS3-quality games, while the PSP already does a good impression of PS2 titles – a focus on PS1 titles, as Sony’s PlayStation Suite intends to do, feels very regressive, especially when both Apple and Android are doing such a good job of bringing new and dedicated games to market.




Funnily enough, there's a video output that lets you play your games on a big screen if you so wish, effectively turning it into a portable PlayStation One, but you can pick the original PS1 and a bundle of games up off eBay for less than a tenner, so this seems of little consequence. It also puts the £5-£7 retail price for Play games into perspective.




The Play retails at £460 off-contract but if you shop around online, you can get the handset for free on a £25 a month contract from several providers if you're prepared to put faith in the slide mechanism lasting 24 months, though some companies, including O2, Vodafone and 3, have delayed its release for a variety of reasons, from stock delivery issues to software bugs. T3 was sent and reviewed a final boxed version of the Play, but in the light of three of the biggest operators postponing release, possible future updates may change user experience.

Sony Ericsson Xperia Play: Verdict

Despite it being rumoured for nigh-on half a decade, the 'PlayStation Phone' actually feels like it's been rush-released, and as such its compromises mean it falls between two stools. There are better and cheaper handsets for everything other than gaming, and at present there just aren't enough dedicated, optimised games worthy of people's attention for it to survive as a handheld that happens to make calls. If you’re someone who just wants to play Android games with a proper control pad, this is the best there is, but bear in mind the current crop of games weren’t designed with this configuration in mind and play as such. We'd like to reassess the Xperia Play once the PlayStation Suite service is properly up and running, and there are a range of purpose-made games to test, but right now we'd struggle to recommend it for either of the fronts it’s fighting on, although we’re sure this won’t prevent it from finding an audience.




Sony Ericsson Xperia Play launch date: Out now, link Sony Ericsson


Sony Ericsson Xperia Play price: £460-£599 online

Sony Ericsson Xperia Play Specifications

* OS: Android 2.3

* Storage: 400MB int, 8GB or 16GB card supplied Micro SD

* Screen: 4-inch 854x480

* Connectivity: WiFi 802.11 b/g/n Bluetooth 2.1, microUSB, A-GPS

* Camera: 5MP main, VGA front

* Video: 720p 30fps

* 385 minutes

* Dimensions: 119 x 62 x 16 mm

* Weight: 175g

Responsive web design

With sites being accessed by an increasing array of devices and browsers, users deserve a quality experience no matter the size of their display. Ethan Marcotte explains how our designs can become more responsive

A professor of mine once told me that every artistic movement was a response to the one before it. Bebop threw off the melodic strictures of swing, while film noir countered the glossy sheen of big studio comedies with gritty antiheroes. George Lucas watched a few too many Kurosawa films and a bunch of old Westerns, and Star Wars was the result.

Influence and inheritance: it’s an old song. The still-young field of web design is no exception, and is very much influenced by its predecessor, print.

Unlike print, though, much is out of our control: browser incompatibilities, a reliance on locally installed fonts, an ever-changing browser window: it’s enough to drive a poor designer mad. So in the face of every inconsistency, we impose constraints on our medium to better establish a sense of control. Fixed-width layouts and minimum screen resolution are two such constraints: they better help us isolate the problems we feel we can solve from those we can’t. Divide, quarantine and conquer.

But the web is becoming more unpredictable, not less. There are more devices, more browsers than ever before. The explosive adoption of excellent small-screen browsers such as Mobile WebKit has further compounded the effect.

Unfortunately, our early attempts at designing beyond the desktop have felt all too similar to our divide-and-quarantine approach from the past. Here’s a quick example: I was working on my laptop when a friend emailed me a link to an article she’d just read on her phone:

Note the webkit subdomain: the site’s owners had quarantined the ‘mobile experience’ on a separate subdomain, fixing the width to 320px. Once that link was shared (via email, or saved to Delicious or Instapaper), readers were locked into that context, regardless of browser. As you might guess, the reading experience was less than optimal (read: “awful”) on a desktop browser.

Fragmented content

You could argue that this example simply suffers from poor execution, that some simple browser sniffing could easily redirect users to the “full” article. But I think it’s symptomatic of our general approach to designing beyond the desktop: in the face of the many uncertainties, we place constraints on our designs, such as a minimum screen resolution, to better isolate the problems we can fix from those we feel we can’t. But fragmenting our content across different “device-optimised” domains is a losing proposition. We simply can’t compete with the pace of technology, updating our sites each time a new, competent browser hits the market.           

Rather than creating disconnected designs, each tailored to a particular device or browser, we need to treat them as facets of the same experience. Responsive web design means creating an adaptive design that’s aware of the context it’s viewed in and optimises its display accordingly. And we can do so simply by embedding standards-based technologies in our work. By starting with a flexible grid (with flexible media), we can incorporate media queries to create that responsive design.

To illustrate this, I’ve built a simple little (fake) site called ‘Robot or Not’ (see above). It’s the definitive answer to who is and isn’t a robot. Your cat? Not a robot. K-9? Totally a robot.

Find our case study at robot-or-not.com

The design is straightforward, draped across an eight-column grid (see below). Our first challenge: how do we take a fixed-width mockup built in a graphics editor, and translate it into a fluid grid that can expand and contract along with the viewport?

First ingredient: the fluid grid

If you’ve ever sized text with ems, you’re well prepared to answer that question. Let’s assume my body element has a font-size of 100% – which, by default, is equivalent to 16px. Further down in the document, I have a headline that’s supposed to be sized at 22px in the comp, but I want to set it in nice, relative ems. To do so, we simply take the target value for the headline’s font-size in pixels, and divide it by the font-size of its container – ie, its context. The result is our desired font-size, expressed in proportional, em-ready terms. In other words: target ÷ context = result.
We can take our pixel values and plug them into this formula. To properly set our headline in relative terms, divide the target value (22px) by the font-size of its context (16px): 22 ÷ 16 = 1.375.

The design for our robot site is modest, and so is the eight column grid that it’s based on

There we are: our headline is 1.375 times the default body size, or 1.375em, which we can copy directly into our headline’s font-size property.
We've just described a font size not as pixels, but as proportions measured against an element’s container. But every element of our grid – and the elements laid upon it – can be expressed proportionally in exactly the same way. In other words, we’re looking not just at the desired size of a particular element, but the relationship of that size to the element’s container. Doing so will allow us to construct a grid whose pixel-based width might change, but whose intrinsic proportions will remain intact as it resizes. All by reusing our trusty type formula: target ÷ context = result.
Looking at the design (see below), we can see that the content area is designed to be 637px wide, with two significant elements inside it: the main content, floated to the right, is 420px wide; the metadata is floated to the left, and sized at 204px. And if we were content with pixels, we’d be done with our job by now:

1. #blog .main {

2. float: right;

3. width: 420px;

4. }

5. #blog .aside {

6. float: left;

7. width: 204px;

8. }

Of course, we can do much better. The pixel values from the comp are simply our target values, which we can express in relative terms. Let’s start with the main content area, whose target value is 420px. The element’s contained, however, within the designed width of 637px – the width of the entire blog module. So all we need to do is divide the two: 420 ÷ 637 = 0.659340659. Move the decimal over two places to the right, and we get 65.9340659%, a width we can drop directly into our CSS (resist the urge to round that number; browsers are actually adept at taking care of that for you):

1. #blog .main {

2. float: right;

3. width: 65.9340659%;      /* 420px / 637px */

4. }

A visual inventory of the blog, which we’ll use to construct a simple, flexible grid

Let’s see if we can do the same with the metadata’s target value of 204px. Using the same context (637px) as before, we get 204 ÷ 637 = .32025117. As before, we can take that 0.32025117, or 32.025117%, and apply it directly to the metadata’s width:

1. #blog .aside {

2. float: left;

3. width: 32.025117%;      /* 204px / 637px */

4. }

So we’ve completed the large, macro-level columns. But we’re not quite done; there’s a pullquote embedded in the main content block, floated to the right. According to the comp, it’s the same width as our metadata (204px) as each occupies exactly one column in our grid. But while we were proportionally describing .main and .metadata in relation to the full width of our design (637px), our blockquote is nested inside of .main. Our context has changed.

Without any CSS constraints, images will just overflow out of their flexible containers

Rather than dividing by 637px – 204 ÷ 637 = .32025117, we’ll need to divide it by the width of .main, its parent element, and our new context: 204 ÷ 420 = .48571. With that, we’ve finished our little mockup. Our new width is .48571, which beautifully describes the width of our pullquote in relation to its container – all in flexible, proportional terms.

1. #blog blockquote {

2. float: right;

3. width: 32.025117%;      /* 204px / 637px */

4. }

Constructing a fluid grid is that simple: dividing your target pixel value by context results in a rational, flexible width. As the viewport expands and contracts, the pixel widths of individual columns change but the proportions of our grid stay intact.

With max-width: 100% applied, images work perfectly in a flexible environment

Of course, a number of CSS frameworks have recently stepped up to the challenge of creating flexible grids since I came up with the initial approach. Nicole Sullivan’s OOCSS grids module is phenomenal; designinfluences.com/fluid960gshttp://designinfluences.com/fluid960gs has a ‘fluid port’ that might be of interest to users of that framework. But if you’re working with irregular grids, or you aren’t interested in frameworks, the target ÷ context = result formula is all you need to build your own flexible, grid-based designs.

Second ingredient: flexible images

With our grid in place and resizing beautifully, let’s focus on incorporating some fixed-width elements. After all, it’s painfully simple for text to reflow in a flexible container; what happens when we introduce fixed-width elements? Well, I hear images are going to be just huge on the web, so let’s start there.
What happens when we drop in an oversized graphic we know to be larger than its container? Awful, unspeakable things. Our image is over a thousand pixels wide, and there aren’t any constraints on its width. It’s simply overflowing its container, and breaking our page pretty thoroughly. But what if we could say that our image should reflow and resize itself proportionally, and should never exceed the width of its container? With a quick dab of CSS, we can get exactly that:

1. img {

2. max-width: 100%;

3. }

First discovered by Clearleft’s Richard Rutter, the max-width: 100% technique ensures an image never exceeds the width of its containing element. If the width of its container changes, the image resizes proportionally in nearly all modern browsers. The max-width: 100% treatment also applies pretty well to other embedded media, such as video:

1. img, video, object {

2. max-width: 100%;

3. }

Of course, IE doesn’t support max-width pre-IE7. If that’s a concern, workarounds are available. Some include JavaScript-based patches to implement max-width-like behaviour; others use proprietary, Microsoft-only CSS expressions to simulate the effect. I advise a simpler approach: just drop this rule into your IE-specific stylesheet:

1. img {

2. width: 100%;

3. }

Our first media query can repair our broken masthead, optimising the layout for a smaller display

Note this is a drastically different rule: whereas max-width: 100% says that the element should never exceed the width of its container, width: 100% says that the element should always match the width of its container. As a result, I recommend scoping the IE-specific rule to elements that you can predict will be oversized in relation to their containers, like so:

1. img.full, .entry img, .entry object {

2. width: 100%;

3. }

We’ve now completed a design that expands and contracts proportionally with the browser window. But responsively speaking, our job’s not quite done.

Third ingredient: media queries

Flexible or fixed, no design scales well beyond the context for which it was originally designed. When made smaller, the background sprite used for the navigation becomes clipped at smaller resolutions, and our content becomes illegibly narrow. Conversely, when viewed on a widescreen display, the images grow freakishly large.
Enter the media query. Part of the CSS3 spec, a media query enables us to inspect the physical characteristics of the device rendering our design. If the device passes the test outlined in the query, then the CSS contained within is loaded. As our flexible layout resizes, we can apply media queries at different resolution ranges to surgically correct issues as they appear – whether in a widening desktop browser window or on mobile devices.
For example, let’s tackle some of the smaller screen displays. Since issues crop up when my browser window is 660px or lower, I can create a media query that targets that range:

1. @media screen and (max-width: 660px) {

2. …

3. }

Our widescreen display is a drastic shift, but one responsively optimised for that reading context

With that in place, I’ll include a few rules to reorient some of the masthead elements, optimising their display in a smaller viewport. For example, I could isolate the logo from the navigation by placing each on its own row, and have each primary navigation item occupy roughly a quarter of the screen.

1. @media screen and (max-width: 660px) {

2. #logo a {

3. display: block;

4. float: none;

5. margin: 0 auto;

6. }

7. #nav {

8. clear: both;

9. padding-top: 0;

10. }

11. #nav li {

12. margin-right: 1.5053763%;

13. width: 23.87096774%;

14. }

15. }

But what if it’s viewed on a smaller resolution, or just reoriented in portrait mode? No problem: I can easily shift the navigation to a stacked, two-column layout, all by adding another media query:

1. @media screen and (max-width: 340px) {

2. #nav li {

3. width: 49.247311827956989247%;           

4. /* 458px / 930px */

5. }

6. }

It’s not just about ‘designing down’, though: we can use media queries to optimise for the higher end of the resolution spectrum as well. By including an @media screen and (min-width: 1100px) block, I can introduce an alternate layout that optimises my content for larger displays. And with these three media queries in place, our site’s finally finished.

Working together

It’s tempting to reduce the concept of responsive web design to one of its component parts, that it’s simply about ‘triggering alternate layouts with media queries’, or perhaps ‘designing for mobile.’ In fact, the three components are inseparable: devices or browsers without support for media queries can still avail themselves of a flexible layout that affords a measure of resolution independence.
Of course, that’s not the end of the challenges before us. How does a highly art directed layout adapt to a changing browser window? How do we rethink complex, interaction-heavy interfaces on a smaller display? Responsive web design offers us a foundation, a framework for tackling these problems – both now, and in the years to come.

Wireframe tools: the top 10 solutions

Create your next prototype the easy way with our guide to the best wireframing tools out there

Every website design has to kick off somewhere, and in most cases, this initial starting point comes with the creation of wireframes. Choosing the right wireframe software can be a complex business, depending on your needs. There are plenty of straightforward and user-friendly online wireframing tools that will sort you out with a quick prototype in no time at all, but if you want to create a more detailed site prototype, then you may be looking at shelling out a hefty wedge of cash to fulfil your needs. Whatever you desire from your wireframes, there is more than enough choice to provide you with what you're after. So, here are 10 of our current favourites for your delectation.

1. Omnigraffle ($99.95)

A longstanding favourite with the Mac community (it's not available for Windows), Omnigraffle makes it easy to create interactive wireframes using its intuitive WYSIWYG system. Although originally designed as a diagram creation tool, Omnigraffle's wide range of free stencils from the likes of Yahoo makes it an ideal solution for prototyping. A 14-day trial is available.
Download the Omnigraffle free trial

2. Axure RP ($589)

Boasting a client list that encompasses many of the most successful companies around, Axure RP is a one of the most comprehensive pieces of wireframe software around. Available for both Mac and PC, Axure makes use of widgets, custom libraries and master documents  to streamline the process of creating wireframes, while annotating tools enable you to insert descriptions wherever necessary.

3. Balsamiq Mockups ($79)

If Omnigraffle and Axure RP aim to be one-stop shops for your wireframes from start to finish, then Balsamiq Mockups is unabashedly for people who want a quick and dirty layout. Built on Adobe AIR, so you'll need to have this installed beforehand, Balsamiq Mockups places a variety of hand-drawn pre-built objects at your disposal, with which you can fashion the beginnings of a mighty web presence.
Check out the Balsamiq Mockups web demo

4. Mockingbird (Free)

What could be easier than dragging your intended page elements onto a blank page to generate your shiny new wireframe? How about doing it all online, without having to sign up first, and with the option to share your creations with a simple click? Mockingbird is still in Beta, so expect the odd glitch.

5. Jumpchart (from Free)

Any tool with a website as gorgeous as this one has to be good, and Jumpchart thankfully delivers on its promises. It uses a different input method to any other wireframe software we've tried; instead of dragging and dropping on a WYSIWYG interface, you chose elements from a drop-down menu. In the editing view, things can get quite confusing, but when it comes to previewing your work, everything is as it should be.

6. ProtoShare (from $29/month)

ProtoShare is the professional option when it comes to wireframe software, with names like Sony and Motorola plastered all over their site. Collaboration is key, with discussion fields present in all wireframes, not to mention the lure of real-time collaboration. Throw in the usual drag-and-drop components, as well as the ability to use CSS and Javascript, and you have a monster of a wireframer. It'll cost ya, though.
Download the free ProtoShare trial

7. Lumzy (Free)

We know what you're thinking: what the world really needs is an online wireframing tool called Lumzy. Despite the baffling name, Lumzy actually delivers the goods, offering a totally free service that comes across as a much, much smaller Axure RP. It might not be one for the design purists, but the results speak for themselves.

8. HotGloo (from $7/month)

If online help, a straightforward interface and a nice, cosy atmosphere were the markers to perfect wireframe software, then HotGloo would romp home with every title. In fact, HotGloo delivers on every front, with object selection about as easy as you could imagine, as well as a host of other features which should make your wireframing life a whole lot easier.
Check out HotGloo

9. FlairBuilder ($24/month or $99/life)

The aims of FlairBuilder should be familiar to you now: give users the chance to quickly and easily build wireframes and prototypes. With a host of features and a very speedy interface, FlairBuilder does the job, and gains even more marks by allowing you to assign a vast array of interactions to any element on the page.

10. iPhone Mockup (Free)

No wireframe software round-up would be complete these days without a nod to the iPhone at some point. This dedicated online service allows you to create basic layouts for your iPhone apps or sites and, while it doesn't do much else, its iPhone wireframe tools are superb, with all of the standard interface elements at your disposal, as well as a choice between illustration or pencil effect.

Actually A Fantastic Idea: Hasbro’s My3DNow 3D iPhone ViewMaster

Most people over the age of 20 will remember the ViewMaster toys that let you see slideshows from your favorite movies and cartoons in 3D. Shabby, static 3D, but still. Well, Hasbro has put out a product that, while it sounds ridiculous at first, is actually sheer genius. Replace the slide wheels with an iPhone! It’s brilliant!
What you do is download an app and some content, then attach the iPhone to the end of the viewer goggle-thing by means of a special bracket (there’s one for every iPhone or iPod touch model). It sits securely in there and shows the two stereo images, which the goggles make into one awesome, totally self-contained 3D content.
Add headphones and your kids will do this for days. Right now there are no brand names, and the programs (Shark Simulator, Santa Monica Pier Simulator, Starfish Simulator) might not actually be enough to hold a kid’s attention for more than 15 or 20 minutes, but hey, it’s just the launch titles. The idea is solid as hell. And it only costs $35! More for the apps, of course, but hey!
It’ll be available at Target on April 3rd, and you’ll probably be able to order it online shortly after.

Google Wants To Make Your Face Searchable (Updated)

Google is planning to introduce a smartphone app that can identify faces. If you capture someone’s head shot on your smartphone, Google will search their database of photos to try name them. We were pretty excited when Google Goggles came out. But this is a bit excessive.

Google, not wanting this news to contribute to their already-precarious reputation regarding privacy, set some requirements to become searchable. People have to first agree to be identifiable by the app; by doing so, they agree to give Google permission to access their pictures or profile information.

This strikes me as bit strange. After all, if you are close enough to someone to obtain a clear headshot — on your smartphone — isn’t it likely you already know them, and therefore don’t need to search for them? A simple introduction is all that’s required, if you don’t already know the person. I guess it could help you remember who the people are in group photos.

Way to be creepy, Google!

Update: It turns out CNN was running with a pretty thin story. Google says they won’t release a mobile app with these capabilities unless a good privacy model is in place. “As we’ve said for more than a year, we will not add facial recognition to Goggles unless we have strong privacy protections in place. We’re still working on them. We have nothing to announce at this time,” said a Google spokesperson.