Original Link: https://www.anandtech.com/show/4583/the-sony-ericsson-xperia-play-where-do-you-want-to-take-your-gaming-today



The debate over the potential (or not, depending on your perspective) for cellular handsets to supplant dedicated portable gaming consoles was already at the 'dull roar' stage when Steve Jobs unveiled the first-generation iPhone in January 2007. Successive iPhone iterations, along with iOS ecosystem expansion to the iPod touch and iPad, have upped the argument amplification a notch or few, as have competitive offerings based on the Android, RIM, WebOS and Windows Mobile (now Windows Phone) operating systems.

Sony's approach to addressing the standalone-versus-cellphone debate is, if nothing else, intriguing. The multi-subsidiary company includes the game console division, of course, which is determined to do its utmost to maintain a lucrative dedicated-function portable hardware business. Yet Sony Computer Entertainment is also responsible for a plethora of game software titles, whose developers are likely challenged to sell as much content as possible, ideally but not necessarily exclusively running on Sony-branded hardware. And then there's Sony Ericsson, a joint venture company chartered with maintaining and expanding its status as a top-tier cellular handset manufacturer.

On one end of the strategy spectrum, Sony has to date produced four different models in its PlayStation Portable line; the original PSP-1000, PSP-2000, PSP-3000 and PSP Go. The upcoming PlayStation Vita successor, formally unveiled at June's E3 Conference with availability (beginning in Japan) slated for later this year, aspires to one-up even the most powerful current-generation smartphone with features such as a SoC containing a quad-core ARM Cortex-A9 CPU (clock speed currently unknown) and quad-core 200 MHz Imagination Technologies SGX543MP4+GPU, not to mention PS Vita's 5" OLED display. On the other end of the spectrum is Sony's PlayStation Certified program, unveiled in late January, which conceptually enables generic Android-based hardware to run PlayStation Suite content.

And in-between these two extremes is the subject of this particular writeup, Sony Ericsson's Xperia Play gaming cellphone:

The mythical 'PlayStation Phone' had been rumored for several years, but when it finally appeared in late March in 11 countries (not then including the United States), it was curiously absent any explicit 'PlayStation' branding. Sony Ericsson's initial U.S. carrier partner was Verizon, who began selling the handset in late May subsequent to its first official U.S. unveiling, a commercial which ran during February's Super Bowl. More recently, AT&T picked up the handset in mid-July. One week later, Verizon dropped the Xperia Play's contract-subsidized price to $99.99, from $199.99 at introduction. Was Verizon's action a competitive response to AT&T's entry, a reaction to poor Xperia Play sales, or some combination of these and/or other factors? Verizon's not saying, but let's see how well (or not) the handset performs to get a sense of its degree of market appeal.



The Xperia Play's form factor was the topic of an opinion-filled, entertaining and informative recent cyber-debate between Anand, fellow staffer Brian Klug, and myself. Anand, who briefly had the Xperia Play before transferring it to me, commented, "it may be light for a slider but I really wasn't pleased with the thickness or build quality for that matter, it all felt too thick and loose." I, on the other hand, was quite pleasantly surprised with its seemingly svelte shape and diminutive weight, both in an absolute sense and relative to the two other phones currently in my possession, an Apple iPhone 4 (Verizon) and a Google Nexus One (AT&T).

The table below shows how the three handsets stack up from a factor standpoint, along with a HTC Evo Shift (Sprint) that I recently had the opportunity to handle. For grins, I also included my previous work phone, a Motorola Droid whose design whose first production dated back to October 2009. And finally, I tossed in the slider that I owned prior to that; the very first Android handset, the T-Mobile G1 dating from October 2008:

Form Factor Comparison
  Weight (w/battery) Height Width Thickness
Sony Ericsson Xperia Play  (slider) 6.2 oz (175.0 g) 4.7 in (119.0 mm) 2.4 in (62.0 mm) 0.6 in (16.0 mm)
HTC Evo Shift (slider) 5.9 oz (167.3 g) 4.6 in (116.9 mm) 2.3 in (58.4 mm) 0.6 in (16.0 mm)
Apple iPhone 4 (slate bar) 4.8 oz (147 g) 4.54 in (115.2 mm) 2.3 in (58.66 mm) 0.37 in (9.3 mm)
Google Nexus One (aka HTC Passion) (slate bar) 4.6 oz (130 g) 4.7 in (119 mm) 2.35 in (59.8 mm) 0.45 in (11.5 mm)
Motorola Droid  (slider) 6 oz (169 g) 4.56 in (115.8 mm) 2.4 in (60 mm) 0.54 in (13.7 mm)
T-Mobile G1 (aka HTC Dream) (slider) 5.6 oz (158 g) 4.63 in (117.7 mm) 2.19 in (55.7 mm) 0.67 in (17.1 mm)

Here's how the Xperia Play looks when placed side-by-side with the Nexus One:

Anand's right, of course, the Xperia Play is thicker (and heavier) than either of the two phones I own. But as I note in the above table, that's largely because it's a 'slider' form factor; as such, it's identical in thickness to the HTC Evo Shift (and thinner than the geriatric G1). Whether or not Sony Ericsson was wise to devote the lower layer of the 'slider' to gaming-centric functions versus a generic physical keyboard (as with the HTC Evo Shift, Motorola Droid and T-Mobile G1) is a discussion that I'll save for later in this writeup. And to that point, keep in mind that the Xperia Play is notably thinner than a dedicated portable gaming console; the upcoming PlayStation Vita is 0.73 inches thick, for example, and the Nintendo 3DS is an even more bulbous 0.83 inches deep.

To me, part of the reason that the Xperia Play doesn't feel as bulky as its specs would otherwise suggest is because of its tapered shape, which results in its thickest-girth portions being in areas that are cradled by the palm of your hand but with a design that still allows it to lie flat when put down. Conversely, the 'industrial' design of the iPhone 4 seems bulkier to me than its specifications reveal to be the case; note that it's the thinnest of the bunch! More generally, by this point in time I tend to find that pretty much all smartphones are 'good enough' from both thickness and weight standpoints to comfortably fit not only in the hand but also in a shirt or pants pocket, with the possible exception of ultra-large-screen models such as the HTC Evo.



I scanned some Xperia Play user feedback posts (which aren't hard to find by this point, since the handset has been available for sale for several months) in the process of doing my evaluation. In some cases, the gaming phone owners were critical of either (or both) the materials used to assemble the handset or their put-together fit. Frankly, I don't see what the big fuss is about, though I realize that other folks with other feature set priorities might come to other conclusions. Glossy black may not be everyone's cup of tea, but I personally don't think it's at all bad looking, although I admit that it more quickly collects visible fingerprints than do other color schemes (as well as being highly reflective, as you may have noticed from some of the earlier product shots I took).

Yes, the silver trim is plastic, not metal, but the materials choice not only trimmed the handset's weight but also reduced its manufacturing cost (therefore price tag) and minimized the potential for interference-induced antenna sensitivity degradation. Had Sony Ericsson put a glass back on the Xperia Play, the reviewers would have grumbled that the company had made an Apple iPhone 4-like non-ruggedness mistake. And maybe it's just me, but the very first thing I do when I get a new handset is to buy a 'gel' case for it, thereby protecting it from scratches caused by keys and change in pockets, inadvertent drops, and other day-to-day usage issues. If you're not a fan of plastic assemblies, the Xperia Play may feel 'cheap' to you, but after you consider the expense, heft and other downsides of alternative materials, you might have at least a partial change of heart.

The Android four-button suite (Home, Menu, Back and Search) at the bottom of the front panel is comprised of actual physical buttons, versus being capacitive touch-implemented. They're a bit 'mushy', as are the volume toggle and power button, but perfectly adequate and preferable to the virtual button suite found on my Nexus One and other Android phones. And yes, they're in a different order than with other Android handsets, but Sony Ericsson can't be uniquely faulted for this discrepancy; Google clearly hasn't made rigid mandates in this regard, judging from the diversity of combinations found in the Android Army.

One physical button that I missed on the Xperia Play is a camera shutter, which the Nexus One implements via its trackball. With the 'stock' Xperia Play, you can only trip the shutter via the on-screen virtual 'button' icon. Enthusiast efforts have transformed the R1 'flipper' gaming button into an alternative shutter; head to the XDA Developers site for more information. Hardware upsides include integrated 'stereo' audio playback (with sound coming from between the two slider layers; the two speakers aren't visible), as well as the ability to access both the SIM and microSD slots without needing to remove the battery.

About that 'flipper'...I'll discuss the gaming-centric control set soon, but the means by which you access the bulk of it (located on the slider's lower layer) is by sliding the Xperia Play's screen up, an action which also automatically launches the Xperia Play application.

While, as noted above, Anand found the slider mechanism to be 'loose', I was quite content with it. It seemed solid, and I was confident that it would remain so through iterative use over time. I could also easily manipulate it with only a single hand; once you get the screen ~halfway through its full travel, it takes over and completes the desired open or close operation all by itself. Anand's fondled far more phones than I, so I'm inclined to defer to his comparative critique, but from an absolute standpoint I found the Xperia Play's slider to be functionally sufficient. With that said, I'll concede at least one of Anand's points; the slider design wasn't as rigid when closed as is a single-element handset...but I've yet to find a slider that can meet that particular design challenge.



After deciding whether or not I'll be able to comfortably tote a cellular handset around in my blue jeans, my next evaluation step is to test its reception capabilities...because, after all, it's first and foremost a phone. Up front, I should note that I was reliably able to notably degrade the Xperia Play's received signal strength by holding it in either hand with my fingers pressed against the bottom half of one side and my palm aligned in the same location on the other side, in a natural hand-holding position:

In my upstairs bedroom, for example, the Xperia Play read -81 dBm when I wasn't touching it, but -91 dBm or worse when I held it as shown above. Stronger-reception environments may not experience the same degree of degradation (or, for that matter, dropped-call outcome) and, as is already well documented at this point, my iPhone 4 has its own 'Antennagate' issues. The previously mentioned gel case neatly addresses the iPhone 4's design shortcomings, and I suspect that the same would be the case (pun intended) for the Xperia Play.

All of the measurements that follow were taken, therefore, with the mobile phones not in my hand. I'm based in Truckee, CA, a small town in the northwest corner of Lake Tahoe that's blessed with an atypically robust telecommunications infrastructure courtesy of the large number of Silicon Valley residents who own vacation homes in the area. The Xperia Play I'm reviewing is a Verizon model, and Verizon happens to have the most robust coverage here of any of the four carriers. AT&T coverage is somewhat weaker, and I've therefore supplemented it with a Wi-Ex zBoost YX510-PCS-CEL cellular booster.

The YX510-PCS-CEL is a dual-band model that also comprehends Verizon's voice and data frequencies, so I've provided internal measurements both with it on and off. And for comparison sake, I've also listed the received signal strength measured by my Verizon-tailored iPhone 4, which I've jailbroken and from which I've therefore obtain dBm measurements courtesy of a SBSettings switch (although labeled as being exclusively for 'GSM' use, it works equally well for CDMA):

Comparative Reception
  Xperia Play iPhone 4
Interior: upstairs bedroom (booster on) -81 dBm -78 dBm
Interior: upstairs bedroom (booster off) -81 dBm -78 dBm
Interior: downstairs living room (booster on) -75 dBm -74 dBm
Interior: downstairs living room (booster off) -75 dBm -74 dBm
Exterior: front deck -59* dBm -63 dBm

A few comments; first off, the YX510-PCS-CEL (whose retransmit antenna is also located in the downstairs living room) clearly doesn't seem to be having any positive Verizon reception effect, at least in the test locations I chose! Premium cellular boosters are intentionally designed to disable their amplification effects whenever the ambient signal is already sufficiently high, so perhaps this explains what I'm seeing (or more accurately, not seeing). And when measuring reception outdoors, I initially placed the Xperia Play to the right of and a couple of inches away from the iPhone 4 on the deck railing, wherein I received a -56 dBm result on the Sony Ericsson handset.

Since this result was notably divergent from what I'd gotten inside, I decided to swap the locations of the two handsets...wherein the Xperia Play reported -62 dBm of signal strength! Therefore, I averaged the two results for the asterisk'd -59 dBm number you'll find in the above table. This situation exemplifies the Achilles Heel of any testing of this nature, since results are dependent not only on handset location but also the presence or absence of any ambient degraders (close-proximity spectrum broadcasters, weather effects, intermediary walls and other attenuators, etc). In contrast, by the way, the iPhone 4 delivered consistent -63 dBm data in both locations.

With respect to phone functions, the Xperia Play was passable albeit not spectacular. Volume from the conventional speaker and speakerphone was sufficient for all but the highest ambient noise environments. Folks on the other end of the connection almost always claimed that they could hear me clearly, although I would have preferred that Sony Ericsson include a multi-microphone array for ambient noise suppression purposes, either done in software (aka Motorola Droid) or silicon (Nexus One and iPhone 4, via an Audience audio processor).

And regarding data transfer speeds, I returned to the front deck and captured a few screenshots of the Xperia Play running Ookla's Speedtest.net application:

The Xperia Play advertises itself as being compatible not only with the older EV-DO Rev. 0 standard but also the newer, faster Rev. A approach. Based on the data above coupled with my prior experiences with CDMA cellular data-cognizant devices, I'm inclined to believe Sony Ericsson. The randomness of the results is somewhat intriguing; I presume it reflects a time-varying network load on whatever cell tower I happened to be connected to at the time. But considering that I saw peak download speeds of over 1.5 Mbps, for example, I'm confident in the Xperia Play's EV-DO Rev. A capabilities. For comparison's sake, here are some sequential screenshots of Speedtest.net running on my Verizon iPhone 4:



The Xperia Play embeds a single rear-mounted 5.1 Mpixel image sensor alongside a LED flash. There's no dual-sensor rear array, as in the HTC EVO 3D or LG Optimus 3D, although Sony Ericsson included a VGA resolution (640x48 pixel, aka 0.3 Mpixel) front-mounted camera for video chat purposes. The Xperia Play captured passable rear-mount camera snapshots even in low ambient illumination environments, although it doesn't come close to offering the brightest LED supplemental illumination that I've encountered.

Still imaging acceptability was unfortunately coupled with video imaging downsides. For unknown reasons (though I suspect a combination of battery life concerns and CPU performance limitations), the Xperia Play's captured per-frame video resolution is only 800 x 480 pixels (aka widescreen standard definition). I'm profoundly baffled by this shortcoming; my Nexus One contains prior-generation iterations of both the CPU and GPU, yet manages 720p video capture (albeit, admittedly, via a CyanogenMod-delivered hack), as does the silicon-comparable iPhone 4 even in an unjailbroken state. Plus, I personally found the resultant video to be even softer than SD resolution capture would normally deliver, along with a sub-par soundtrack that may be in part a reflection of the non-optimal microphone location.

Anand tested both power consumption and display attributes prior to shipping the Xperia Play to me, so the data that follows comes straight from him. Here's the comparative battery life of the Xperia Play versus other handsets AnandTech has tested, when web browsing both over EV-DO:

Smartphone Web Browsing Battery Life

and over Wi-Fi:

WiFi Web Browsing Battery Life

And here's the 'talk time' battery life comparison:

3G Talk Time Battery Life

Anand then measured the Xperia Play's display brightness and contrasted it (pun intended) with other mobile phones:

Display Brightness

The default display intensity is a bit dim, perhaps reflecting a decision made by Sony Ericsson to stretch the between-charges usage, but the display acquits itself well versus other gear I've handled in the past once you bump up the setting a notch or few. Its viewing angle is also passable albeit not stellar, and it's much easier to see in daylight than is an OLED alternative. And if you do the math on its atypical 854 x 480 pixel resolution, you'll see that not only does it notably exceed that of the PlayStation Portable (480 x 242 pixels), it also works out to be an exact (to within a few decimal places) 16:9 ratio. This means no cropping, no black bars, and no squishing-or-stretching distortion when watching widescreen movies.

Apologies, by the way, for the lack of a display black level brightness measurement, therefore the associated absence of a contrast metric. Anand had mistakenly thought the Xperia Play had an AMOLED screen (it's a LED-backlit LCD), therefore only took a white level brightness measurement before sending the handset to me, and I don't currently own a colorimeter.



Gaming is at the core of the Xperia Play's "raison d'être," of course, so it deserves some focused editorial attention. Slide the screen up and underneath it you'll find the gamepad, which is conceptually reminiscent of a DualShock controller. You already saw this photo a few pages back; here it is again as a reminder:

On one end is a digital directional key suite, and on the other is a set of 'face' buttons. Below them are an Android 'start' button (on the left) and 'select' and 'start' buttons (on the right). Since Sony Ericsson desired to make the Xperia Play as thin as possible, there's no vertical room for DualShock-like D-Pad analog sticks or even the single, thinner analog joystick of a PlayStation Portable. Instead, Sony Ericsson substituted circular capacitive sliders, with capacitive buttons (providing, among other things, orientation-handy indents) in their centers. And where are the DualShock-like L1 and R1 controls? In normal portrait-orientation phone use, they're on one side of the Xperia Play, bracketing the volume toggle. And when you're ready to "get your game on" and landscape orientation-rotate the phone, they end up on the top edges of the lower slider layer, behind the screen.

Verizon bundles seven game titles with its version of the the Xperia Play: Asphalt 6, Bruce Lee: Dragon Warrior, Crash Bandicoot, Madden NFL 2011, Tetris, The Sims 3, and Star Battalion, accessible both standalone via the Android home screen and (with the exception of Tetris, for unknown reasons) within the Xperia Play application:

Additional content is available for download through the Xperia Play app, on a separate screen. I counted 43 titles, at a range of first-download (up to $9.99, from my limited sampling) and subscription (up to $3.49) prices. But what currently exists isn't hardware-exclusive; it's also available on other platforms:

To that point, one common complaint I read from Xperia Play owners is that they needed to re-purchase game titles they'd already bought for other Sony hardware; the existing licenses couldn't be expanded (or at least transferred) to include the Xperia Play. Folks also weren't particularly fond of the D-Pad-replacing capacitive circular pads, and I tend to concur; my success in convincing them to react in the way I intended was somewhat hit-and-miss, and the overall lack of tactile feedback versus a joystick left me overwhelmed. With that said, I'm more forgiving than some in understanding the design constraints that compelled Sony Ericsson to use them.

The bundled, Xperia Play-optimized games I auditioned generally provided an acceptable playback experience, albeit with occasional stutters on a few titles. Bruce Lee: Dragon Warrior was probably the most immersive title I tried, from both audio and visual standpoints; I say this as someone who's not even a particular fan of the action fighter genre. Flight simulation and racing games are more my cup of tea; as such, Asphalt 6 and Star Battalion didn't disappoint, either, though I generally found myself underwhelmed by the inferior richness of the Xperia Play titles (which were reminiscent to me of PlayStation 1 or Nintendo 64 content of days past) in comparison to games intended for a dedicated-function portable console.

As you'll read in detail in the sections to come, the Xperia Play's hardware allotment is modest, but leading-edge graphics drivers from Qualcomm provide the phone with more robust performance than that of its similar-component peers. Nonetheless, the dearth of compelling Xperia Play-optimized content is disappointing, and dulls the overall enthusiasm for the system. Hopefully, AT&T's recent embrace of the Xperia Play is indicative of better news to (soon) come in this regard. And speaking of hardware, a mini-HDMI output that would enable optional tethering of the Xperia Play to a television or other larger display for a more immersive gaming experience is also a notable 'miss'; perhaps if a second-generation design ever appears, Sony Ericsson will see fit to include such a capability.

Other games can be downloaded from Google's Android Market, the Amazon Market for Android, or the Verizon VCast Store. I frankly don't recommend the VCast Store, due both due to its comparatively limited selection and higher prices. Gun Bros, for example, is available for free from the Android Market; the very same title will cost you $6.99 for first download from the VCast Store, or $2.99 for re-download. These generic Android titles don't leverage the Xperia Play gamepad, of course, no matter that they play smoothly on the hardware. Fortunately, many of them harness the handset's accelerometer, as do (optionally, versus the mechanical controls) the Xperia Play-optimized titles.



UBM TechInsights conducted a teardown of the Xperia Play in late May, based on a Canadian-sourced unit and slightly ahead of the gaming phone's U.S. production launch. The companion coverage by UBM sibling publication EE Times goes into more photographic detail than does TechInsight's web page (but then again, TechInsights wants to sell reports, so all is forgiven), and I commend EE Times' writeup to your inspection. In it, you'll garner an exhaustive parts list inside the Xperia Play, of which I want to focus on only a few key components.

First and foremost, consider the applications processor, Qualcomm's MSM8x55 1 GHz 'Snapdragon' SoC, which befitting its 'MSM' prefix contains an integrated cellular modem. In the GSM-tailored variant of the Xperia Play dissected by UBM TechInsights; that SoC was the MSM8255. My Verizon CDMA-targeted Xperia Play instead includes the MSM8655. Qualcomm's website neatly spells out the differences between them, also including the modem-less (and GPS-less) APQ8055, all S2-class SoCs per Qualcomm's recently unveiled rebranding campaign.

Also embedded on the MSM8x55 die is the Adreno 205 graphics core. A notable percentage of the negative feedback I've seen on the Xperia Play concerns the SoC; specifically, cellphone enthusiasts are disappointed in Sony Ericsson's seeming 'trailing-edge' component selection. They expected to see a dual-core processor from Qualcomm, Nvidia or another supplier, and/or a SoC based on Qualcomm's latest Krait microarchitecture (S4 SoC). Sorry, but I don't buy that argument for a second, no matter that the MSM8x55 archaically dates from last fall's HTC Desire HD.

I'm an engineer, by training and by trade. As such, I know that there's no such thing as a black-and-white decision, only shades-of-grey discernment, and that component selection made for a particular project will likely not apply to the next design in the pipeline. Absolute performance is not the sole criteria for picking an IC; cost, power consumption, board space, sourcing options and volume availability, and development tools maturity are often equal in importance...if not greater, as long as performance is 'good enough'. Take a look at the benchmarking section that follows, and I think you'll agree that the MSM8x55 is an adequate candidate partcularly given the system's target screen resolution, form factor (therefore battery size), price tag and other criteria.

I'm less sanguine about the internal memories' capacities. 512 MBytes of RAM seems scant, particularly given the memory-intensive games that the Xperia Play is chartered with tackling. However, RAM deficiences will largely only impact the handset's ability to simultaneously multitask-juggle multiple concurrently running applications. 1 GByte of embedded flash memory, on the other hand, is a far less acceptable allocation, particularly considering that only 400 MBytes' worth of it is user-accessible.

Theoretically, at least, the local nonvolatile storage capacity can be supplemented by a microSD card (up to 32 GBytes in size on the Xperia Play), an ability that came with Android 2.2 'Froyo' and its encrypted external-storage support. To wit, the Xperia Play bundles a 4 GByte microSD card. However, to date I remain unimpressed by both Google's and third party developers' embrace of the potential for installing and moving programs to external storage.

Insufficient local storage capacity is perhaps my biggest beef with my Nexus One (512 MBytes total, 190 MBytes user accessible, in that particular case). Through multiple upgrade iterations of both O/S and Google-branded and -included applications, local storage has been slowly whittled away to the point that I can only install a few third-party programs before available memory dips to 20 MBytes or less, the O/S starts complaining that 'phone storage space is getting low', the phone abruptly stops receiving new emails, etc. Although the Xperia Play has a bit more than twice the user-accessible local storage of the Nexus One, I fear that many Xperia Play owners will sooner-or-later suffer similar frustrations.



At Brian Klug's recommendation, I ran the Xperia Play through a battery of common benchmarking tests. Below you'll find the alphabetically ordered results, in comparison to similar-featured and similar-age Android handsets from other suppliers:

Basemark (aka 3DMarkMobile) ES 2.0 V1

RightWare Basemark ES 2.0 V1 - Hoverjet

RightWare Basemark ES 2.0 V1 - Taiji

BrowserMark

Rightware BrowserMark

GLBenchmark 2.0 ('Standard' results in both cases)

GLBenchmark 2.0 - Egypt

GLBenchmark 2.0 - PRO

kwaak3 aka Quake 3 (note that this particular test was run with lightmaps off, per longstanding AnandTech practice)

Quake 3

Linpack

Linpack - Single-threaded

Linpack - Multi-threaded

And SunSpider 0.9

SunSpider Javascript Benchmark 0.9

The results are largely predictable, but some encouraging surprises also emerged. Performance differences between platforms based on the first-generation 65 nm-based QSD8x50 (aka S1-class SoC) and second-generation 45 nm-based MSM8x50 SoCs predominantly derive from two factors, since the processor cores are fundamentally the same save for the process they're built on:

·       Clock speed differences (800 MHz versus 1 GHz, for example), and

·       Graphics core differences (Adreno 200 with the QSD8x50, Adreno 205 with the MSM8x55)

Slight variances between similarly clocked systems in CPU-centric benchmarks are likely the result of benchmark run-to-run differences, coupled with minor evolutionary Dalvik virtual machine enhancements in successive Android versions. Perhaps the most significant Dalvik improvement came with the transition to Android 2.2 'Froyo', which added a JIT (just-in-time) compiler to the mix with a resultant claimed 2x–5x performance boost with CPU-heavy code. The Xperia Play still notably outperforms its fellow MSM8x55- and Adreno 205-based peers on graphics-heavy benchmarks. This result surprised me, until Brian Klug explained that the Xperia Play represented one of the first production systems containing improved graphics drivers from Qualcomm.

Unlike with a PC, where you can regularly install updated drivers from AMD, Intel or Nvidia, with a handset you need to rely on Google and its hardware and carrier partners to bundle the updated drivers with an operating system upgrade for your hardware...or in this particular case, for an OEM like Sony Ericsson to include them from the get-go. Note, for example, the frame rate improvements of the Xperia Play on Basemark ES 2.0's 'Taiji' and Hoverjet tests, versus the seemingly identical (at least from CPU and GPU standpoints) HTC Thunderbolt. Similarly, the Xperia Play had notably higher frame rates in the Quake 3-based kwaak3 test, versus the Thunderbolt.

One other result was also intriguing, until I thought about it for a bit. Linpack recently added a multi-threaded benchmark option, so I decided to run both it and the legacy single-threaded test. The multi-threaded MFLOPs (millions of floating-point operations per second) was lower than that for the single-threaded test variant; consider, however, that the MSM8x55 is a single-core CPU. I believe that the decrease is the result of multi-thread management thread-switching latency overhead, in spite of the Cortex-A8 superscalar dual-issue microarchecture foundation upon which Qualcomm built its Scorpion approach. Conversely, if you look at the Linpack data for a system such as the HTC Sensation 4G, based on a true dual-core CPU such as Qualcomm's MSM8x60 (S3-class SoC), you'll see dramatically higher multi-threaded Linpack results in comparison to those for the traditional single-threaded test.

FYI I also tested the Xperia Play using Qualcomm's recently introduced Vellamo benchmarking suite, but AnandTech isn't quite ready yet to start publishing Vellamo results. Stay tuned for more in this regard. Similarly, I ran the Xperia Play through the newer SunSpider 0.9.1, whose results we'll start publishing once we have a sufficient-sized system sample set.



I'll admit upfront that I'm not a fan of user interface eye candy, which in my experience hampers more than helps me to use a handset (specifically, to get to the programs and settings I need to access in as straightforward and speedy a manner as possible), not to mention the incremental processing burden it puts on the CPU and GPU, and the consequent decrease in battery life. I'm one of those folks who immediately reverts any Windows XP or Vista system I build or inherit to the entirety of the 'Classic' interface theme, for example. With that said, I acknowledge that I may be in the minority; that HTC's 'Sense' and Motorola's 'MotoBlur' interfaces are viewed as desirable by the bulk of Android users.

Speaking of 'skins', I had previous experience with the Xperia X10 mini, which employed Sony Ericsson's 'Timescape' homescreen and application UI tweaks. Judging from the above comments, you can probably imagine what I thought of it. And based on my perusal of an Engadget review of the Xperia Play published in late March, I was bracing myself for more of the same (although there was some encouraging news; Engadget indicated that Timescape use was now a user-selectable option, not a default configuration). Yet, when my unit arrived from Anand and I powered it up, I was pleasantly surprised to be greeted by a generic Android homescreen, with no Timescape launch icon to be found anywhere:

Engadget had tested an international version of the Xperia Play, which seemingly includes a different software build than the U.S.-targeted hardware which started showing up a couple of months later. Apparently, I'm not the only one who's loath of 'skins', after all.

The Xperia Play supposedly embeds DLNA capabilities, as a conceptual alternative to having a HDMI output, but I was unsuccessful at wirelessly streaming content from it to my PlayStation 3 (which couldn't find it on the network). It turns out, as a bit of Internet research uncovered, that the initial production firmware didn't enable DLNA capabilities; a recently released update had added DLNA support along with a few other features, such as stereo soundtracks for video and a landscape-orientation home screen display option. However, when I checked for updates within the Xperia Play's settings menu, none was reported as available:



Already familiar with Sony Ericsson's upgrade software from my previously mentioned X10 mini experiences, my next strategy was to hit up the company's support website. There, I found four different software packages available for download, three of them Windows-based. The PC Companion utility was the one that I'd already used before:

it includes (among other things) firmware upgrade capabilities, as does the seeming functionally redundant Update Service program. Also offered were a multimedia sync program called Media Go, which again seemed redundant with that same feature in PC Companion, and a Mac OS X-tailored functional twin of Media Go called Media Sync.

I tried PC Companion first. After installing it on my Windows Vista Ultimate-based Dell XPS M133 laptop, a tediously long process which also installs a suite of device drivers, I plugged in the Xperia Play (after first ensuring that the handset was not in USB Debugging mode) and all seemed to go well at first:

I was then prompted to change the phone's connection mode:

by first unmounting the phone:

and then switching it in to MTP (media transfer protocol) mode...an option that the phone didn't even offer to me:

The only option that the phone did give me was to remount it, an operation that inevitably led to a connection failure notification, and an encouragement to (fruitlessly) repeat the multi-step attempt:

Internet research suggested that I might have better success if I manually dismounted the phone's microSD card before connecting the Xperia Play to my laptop, but that attempt only led to an endless circle-cursor on the phone's display upon USB tether, until I pulled the plug (manually broke the USB connection).

Sigh. Next step; Update Service:

which also explicitly listed the Xperia Play as a supported handset:

In this case, I needed to first power off the phone, then hold down the Android 'back' button while connecting it to the laptop over USB:

However, after doing so, Update Service reported to me that no firmware updates were available:

Sigh, again. And, as it turns out, the firmware version situation is even worse than I originally thought. Not only was the latest software version unavailable to me, my Xperia Play was still running its original Android v2.3.2 build, whereas handset owners in other geographies had started receiving an Android 2.3.3 upgrade (complete with extensive Facebook integration) in early June. Verizon's qualification of the upgrades is presumably the availability bottleneck here.



A few months back, I attended Qualcomm's Uplinq conference in San Diego, where among other things I met AnandTech's Brian Klug for the first time. Sony Ericsson Mobile Communication's Rikko Sakaguchi, the company's Executive Vice President and "Chief Creation Officer", delivered one of the conference keynotes. Within his pitch, Sakaguchi emphasized that his company's aspiration was to not be another 'me-too' smartphone manufacturer but to instead deliver unique products with tangible value to target customers.

I strongly resonate with that corporate goal, with the qualifier that I've also heard it plenty of times before from other companies, and that repeated past-history case studies suggest that it's 'easy to say, but hard to do'. And I also assume (and hope) that Sony Ericsson did at least a modicum of market research before rolling out the Xperia Play. But some amount of doubt admittedly begins to creep into my consciousness when I think back to Sakaguchi showcasing the Xperia Play as a 'no compromises' platform in his pitch...and compare his claim against my hands-on findings during this evaluation.

As such, after spending a week-plus with the Xperia Play, I'm admittedly skeptical of its chances. Granted, standalone handheld game consoles aren't doing a stellar job of holding off the gaming-on-mobile phone onslaught. The very day I finished the first draft of this writeup, for example, Nintendo slashed the U.S. price of the latest-generation 3DS from $250 to $170, in conjunction with announcing a quarterly operating loss. And Sony was atypically aggressive with the pricing on the upcoming PlayStation Vita ($249 for the Wi-Fi-only model, $299 for a 3G cellular data-inclusive variant), which the company unveiled at the early-June E3 Conference.

But I'm not sure that a gaming-tailored cellphone is the solution. Hard-core gamers will still buy and tote around a dedicated gaming console (or few), in addition to a generic cell phone. Casual gamers will stick with a mainstream cellular handheld, unwilling to accept the incremental price, weight and thickness of the Xperia Play (especially when, for the increased heft of a 'slider' design, they only get game controls on the lower layer, versus a more versatile-function physical keyboard). Is there a sufficient-sized middle ground between those two user extremes to cultivate a fiscally profitable customer base for Sony Ericsson? Maybe, but early market-embrace indications suggest otherwise.

I'm admittedly not a hard-core gamer. I am, on the other hand, beyond the other end of the spectrum; someone who occasional indulges in nothing more challenging than a bit of Angry Birds while waiting in line at the grocery store or gas pump. I own both iOS- and Android-based handsets, along with an iPod touch, a first-generation iPad and several Android tablets, all loaded up with an assortment of game titles...not to mention two jailbroken Sony Playstation Portables, both a Nintendo DS and a 3DS, and an assortment of living room gaming consoles. I'm arguably smack-dab in the middle of the Xperia Play target demographic.

So would I buy the 'PlayStation Phone'? Honestly, probably not. Generic titles for Android and iOS are less expensive than their Xperia Play-optimized counterparts, not to mention more numerous; the developers are making less money per sale but in exchange have access to a much larger potential sales audience. And many of the broad-base Android and iOS titles are at least as engaging as their Xperia Play counterparts, some of which I've already played on other Sony hardware and therefore wouldn't buy again, if not more. Heck, I've even tried a few homebrew PSP games that are the equal of their Xperia Play counterparts in overall entertainment value. The Xperia Play delivers a decent gaming experience, but it makes too many hardware tradeoffs (weight, thickness, etc) and its content is too expensive and limited in variety to justify its presence in my particular gaming stable.

Log in

Don't have an account? Sign up now