Yesterday Microsoft finally took the covers off the new Xbox, what it hopes will last for many years to come. While Microsoft was light on technical details, I believe we have enough to put together some decent analysis. Let’s get to it.

Last year AMD was hemorrhaging talent. While we don't have a good indication of the extent of the talent exodus, Qualcomm seemed to benefit quite a bit from the tough times that had fallen on AMD. AMD PR mounted the beginnings of a turnaround with the announcement that Jim Keller, former K8 architect and chip-head at Apple had rejoined the company. Then came John Gustafson and last month, we got word that Raja Koduri rejoined as well - also after a multi-year stint at Apple. 

Today I just got word from a couple of very accurate and trusted sources that my old friend Sean Pelletier will be joining AMD as well. Sean will abandon his role as Senior Technical Marketing Manager at NVIDIA to assume a similar role at AMD, initially focusing on GPUs. Reporting on individual hires doesn't actually tell you a lot about talent within a company, but it can give insight into whether or not a company is viable. Not too long ago, leaving Apple, NVIDIA or pretty much any other tech company to join AMD sounded like a career death sentence. The fact that smart folks from all paths are considering AMD as an option for long term employment tells us a lot about how things have changed.

Google I/O 2013 recently wrapped up, and I’ve been spending some time ingesting all the information from sessions during the event, including ones I couldn’t directly attend. While Google didn’t announce a new version of Android at the event, something nearly everyone speculated would happen, it did directly allude to new upcoming features that will be implemented in this still unnamed upcoming version.

Google allegedly assigns version numbers late in the process, but what is known is that this next release will implement upcoming API level 18. If you follow the progression there’s a likelihood this will be Jelly Bean MR2 (Management Release 2), where Android 4.2 was management release 1. Based on some other discussions and sources I also suspect this might be Jelly Bean MR2. All of that is really just semantics however, what really matters are what specific features are coming and which of those Google touched on during I/O.

Bluetooth 4.0 LE Support

In a session on day one, Google announced upcoming support for Bluetooth 4.0 Low Energy. Bluetooth low energy has been branded as Bluetooth Smart for low energy only devices, and Smart Ready for devices which support low energy in addition to classic (such as a mobile phone). Bluetooth LE implements a completely different physical layer compared to Bluetooth classic, with lower duty cycles that enable deployment in lower energy devices like proximity tags, sensors, pedometers, and watches which rely on small batteries that go longer than a day between charges. There’s a misnomer that implementing Bluetooth LE will magically result in lower power consumption, it simply enables a completely different architecture optimized for different use cases. The Nexus 4 recently went through the Bluetooth SIG with Bluetooth 4.0 certification, the WCN3660 Qualcomm WLAN+BT combo chip inside has in fact always had compatibility for Bluetooth 4.0, it was just a matter of adding the APIs to Android to make it useful. 

With Android 4.2, Google changed both NFC and Bluetooth stacks. In the case of Bluetooth, from BlueZ to BlueDroid with an open source disclosure by Broadcom. It’s now becoming clear this was done partially with the forward-looking benefit of enabling support for Bluetooth LE and addition to enabling the Android team to rapidly add more features and profiles.

API level 18 Bluetooth LE features will be added to the Android Compatibility Test Suite as well, which means OEMs who have already implemented Bluetooth LE features via their own APIs will have to support their third party APIs and the canonical Google ones in this future version of Android. The API will support the central profile role only, which means both transmit, receive, and the ability to initiate connections plus serve as the master. Peripheral role is not supported.

API level 18 also adds support for AVRCP 1.3 (Audio / Video Remote Control Profile) which essentially enables compatibility with things like a car head unit, AV receiver, or so forth. This enables the device to control commands like play or pause, as well as metadata such as album artwork, artist, name, and status of the music. This enables much better compatibility with car audio and so on. Interestingly on the last slide from this session the date for API level 18 is given as coming in "a few short months."

Graphics

Perhaps the most revealing session was the one I was most interested in on Android Graphics Performance. It was here that we essentially were given a few glimpses of the new platform features which improve the performance of the hardware accelerated 2D rendering pipeline running on a device as well.

First up is intelligent reordering and merging of draw commands for given UI elements. Like elements are reordered and then issued together to take advantage of the GPU in an optimal manner without incurring a change in the shader state to render bitmaps, text, or nine-patches for example. This also minimizes the number of draw calls issued to the GPU for the same equivalent UI.

Google showed an example before and after with the same Google+ UI going from 88 draw calls to 39 after this feature was enabled.

Second is multithreading of additional parts of the hardware accelerated 2D rendering pipeline for some tasks. Rendering operations will now happen automatically on multiple cores if present.

Third, hardware acceleration for non-rectangular clipping was added, previously this was not hardware accelerated. This includes clipping around paths and transformed rects.

There are new developer tools present in this new version as well. In Android 4.2 Google added an on-device overdraw visualization, similar on-device functionality is coming to the rendering profile tool previously added in 4.1 which required a longer tedious workflow. 

Toggling the profile GPU rendering option now gives the option to draw frame time (display list, rendering, then buffer swap) with a bar graph or line graph at the bottom of the screen in a persistent fashion instead of off-device in a spreadsheet. There’s a handy green line which corresponds to 16ms (60 FPS).

This is a major boon for developers wanting to debug frame render time or occasional hitching. We saw this demonstrated on a device running the new tool and profiling parts of the Android UI, this was the first glimpse of the new Android version running on a device in public I'm aware of.

Systrace also gets a handful of improvements in this unnamed upcoming version of Android with an easier to run command line script invoking it, and the ability to trace each OpenGL call. Systrace is a very powerful tool for looking at what an Android device’s underlying hardware is doing during a trace.

Conclusions

Even though we weren’t explicitly told there’s a new version of Android coming, nor its nickname or version number, there were repeated direct allusions and references to it throughout this year’s I/O. Although many lamented it not being directly announced, it was there if you looked for it, including a few direct glimpses.

It’s clear that the new version will implement API level 18 and bring further improvements to 2D rendering performance throughout Android as well as support for Bluetooth Smart (LE). These are both things closer to hardware and system which require changing the platform software entirely as opposed to pushing an update out to Google Play Services.

Just two days ago, Google announced a variant of the Galaxy S 4 packing the "Nexus user experience." Although not an I/O giveaway, Google chose I/O to announce the new variant. I've spent a lot of time at I/O tracking down what information I could about this SGS4 with Nexus experience device, but there's not a whole lot to be known beyond what was announced in the one title slide and blog post. At a high level the details are known — it's an SGS4 with 16 GB of internal storage, unlocked modem and bootloader, with support for T-Mobile and AT&T LTE and WCDMA band support. The price is what you'd pay for any other SGS4 unlocked without a subsidy, $649 through Google Play starting June 26. 

First, the SGS4 with Nexus Experience doesn't really have an official name yet. At present it's just being called an SGS4, and doesn't appear to be getting tagged with the Nexus brand. Next, by process of FCC ID elimination and lots of asking around, I am now fairly certain that the rumors about this being the T-Mobile USA SGS4 hardware variant, specifically SGH-M919, are true.

SGH-M919 has always included support for LTE on Bands 2, 4, 5, and 17 (that's 1900 PCS, 1700/2100 AWS, 850 Cellular, and 700 Lower B and C) and WCDMA on Bands 2, 4, and 5. At another level, this is the same hardware as the AT&T variant but without the arbitrary RAT (Radio Access Technology) locking that AT&T has put in place to restrict use of Band 4 WCDMA which T-Mobile needs for a good experience. This translates to that support for AT&T and T-Mobile LTE and WCDMA. That also means Snapdragon 600 (APQ8064AB) and no Exynos 5. 

Lastly, I'm fairly certain that Google is using the "Nexus user experience" in place of the previous "Google experience device" branding which it used previously. This does however mean updates are indeed delivered by Google and not some other entity. Matias Duarte referred to this work with Samsung to bring an SGS4 with Nexus experience directly during the later Android fireside chat. We'll have a review of the SGS4 with Nexus experience in due time as well. For enthusiasts who like the SGS4 and don't want the TouchWiz software bloat, this might be the device to get, if they haven't gotten one of the skinned, carrier-approved bundles already.

The only other interesting device related news is tangential and unrelated. I had a chance to hold Taylor Wimberly's (of Android and Me) White Nexus 4 which he picked up from an unnamed source at I/O. The white Nexus 4 looks stunning in person, it's stark white on the back and has a slightly cream colored tinge on the lip. There's still the laser etched pattern behind the glass as well, but obviously white. The front of that Nexus 4 remains black however. There's a chrome lip around the camera aperture as well instead of the black ring. Hardware inside remains unchanged of course, there's none of that LTE goodness we hoped might get enabled, but it's still a striking aesthetic change. 

Source: Android and Me, Google

In response to buzz about NVIDIA's Shield console, NVIDIA has moved up the preorder date a few days, May 20th to today, May 17th. Shield will be available for the previously stated $349 price for shoppers in the US. If you're not familiar with NVIDIA's Shield, it's a portable handheld gaming console including a 1.9 GHz Tegra 4 SoC and running Android 4.2.1. This is the first time NVIDIA has sold an Android device directly to consumers and will be supporting it directly with TegraZone marketplace alongside Google Play. 

The following sites will be taking preorders for Shield:

• ·         NVIDIA – store.nvidia.com/buyshield
• ·         Newegg – www.newegg.com/buyshield
• ·         Gamestop – www.gamestop.com/android/consoles/nvidia-shield-16gb/109517
• ·         Canada Computer – canadacomputers.com/nvidia/nvidiashield.php

Source: NVIDIA

It’s been a couple of weeks since our last “best laptop” article, mostly because I had concerns about recommending anything with AMD’s Richland and Intel’s Haswell processors right around the corner. At this point, it looks like Haswell is going to be a relatively small portion of Intel’s shipments for the near term, with a heavy focus on higher performance quad-core parts and Ultrabooks—basically, a focus on the more lucrative markets. AMD’s Richland APU on the other hand will be a drop-in replacement for Trinity that offers slightly faster CPU and GPU performance, but nothing earth shattering. That leaves plenty of room for laptop recommendations for the next several months, and today we’re going to look at the business sector.

Let’s first start by discussing what it means to be a business laptop. My take is that a business laptop needs to offer three things. First, it should be easy to service, so anything with an integrated battery or with memory and/or storage that can’t be upgraded is out. Second, warranty support needs to be there, and this mostly means that I’m going to look at larger OEMs, specifically Dell, HP, and Lenovo—sorry, but in my experience as an IT professional, those three brands account for roughly 90% of all business laptops, and for the most part that’s with good reason. (Most of the remaining 10% comes from businesses either using Apple products or else trying to cheap out and get by with consumer laptops; since I already cover those areas elsewhere with these guides, we should be okay.) Finally, features and build quality are important—this can include security features like TPM, remote management solutions, better materials, etc. All things being equal, the laptop with the better features/quality (and/or the lower price) wins out.

Just to quickly go over each OEM, Dell has five different “business” laptop lines: Vostro, Latitude, Precision, Inspiron, and XPS; however, the last two (Inspiron and XPS) are really just their consumer lines so I’m going to leave those out of the running, and the Precision line is more for Enterprise and mobile workstation users—that’s basically limited to the very high end and could almost be a topic for another laptop guide—leaving us with the Vostro and Latitude lines. Similar to Dell, HP has quite a few laptop product lines; the Pavilion, Envy, Spectre, and Essential Home are all more or less consumer brands, so we’ll simply bypass those and focus on the ProBook and EliteBook offerings. There are mobile workstation EliteBook options as well, but like the Dell Precision lines, those are mostly for the highest end users. Finally, Lenovo has IdeaPad and ThinkPad laptops, and our focus here will be on ThinkPad offerings as IdeaPads are mostly for consumers. The ThinkPad consists of several different lines as well, however, with the L-series being more budget friendly and the T, W, and X series focusing on higher build quality and materials.

And with that out of the way, here’s a look at what the three primary OEMs have to offer in the way of budget ($500 or so), midrange ($800-$1000), and high-end ($1200+) laptops. While many will shop directly at the OEM site, in general you can find better prices and/or better configurations elsewhere, so I’ve done my best to search for good prices as well as configurations. I’ll pick what I feel is the best overall laptop for each price range, along with mentioning some alternatives as appropriate.

Business on a Budget

When it comes to budget business laptops, the base level of features for around $500 is similar among all three OEMs: Core i3, 4GB RAM, 320GB HDD, 1366x768 LCD, and the various other typical accessories. I ended up narrowing down the field to three choices, the Dell Vostro 3460, HP’s ProBook 4440s, and the Lenovo ThinkPad Edge E430 (even the budget L-series from Lenovo ends up starting at closer to $700). There are 15.6” options that are priced similarly as well (Vostro 3560 and ProBook 4540s), but I’m biased towards 14” business laptops at this price point.

Budget Business Laptop: Vostro 3460 ($539)

Ultimately, my pick goes to the Dell Vostro 3460, because I like the “boring” aesthetic it offers, but the ProBook 4440s is a very close second. The Vostro 3460 gives you 4GB RAM, i3-3120M CPU, and a 500GB 7200RPM hard drive for $539—nothing spectacular, but not too bad all considered. Other features include a fingerprint reader, four USB 3.0 ports, a DVDRW, and an 8-in-1 flash card reader. The 3460 isn’t quite what I would call a “thin and light”, but it weight just under five pounds and measures 13.6”x9.61”x1.1” so it’s at least somewhat close. While Dell ships the 3460 with 4GB RAM on most models in a single-channel configuration and the specs page lists a single DIMM slot, but there are in fact two SO-DIMM slots so upgrading to 8GB total RAM is easily accomplished (and maximum RAM support tops out at the usual 16GB). You can also upgrade to a larger/faster HDD/SSD without trouble, and the internals are quickly accessed by removing one screw and the pulling off the back plate. Sadly, only the most expensive 3460 allows you to choose a different color, as I’d otherwise like the red Vostro!

Mainstream Business Users

This is really the sweet spot for business laptops, and not surprisingly it's what most companies buy for their employees. For my midrange business laptop recommendation, I really wanted to get a better display—at least 1600x900, if not 1080p. Almost all of the OEMs fail to deliver a compelling price for such a configuration if you shop directly at their respective web sites (e.g. I think the Dell Vostro 3560 $999 model is the only option with such a display priced under $1000 that I could find), but a bit of searching via Google turns up plenty of retail outlets with preconfigured models at reasonable prices. (Update: Dell does have 900p and 1080p panels on their Latitude E5430, E5530, E6430, and E6530, if you look at the right model. In all four cases, the least expensive base configuration lists LCD upgrade options, so feel free to shop accordingly.) 1080p tended to be a bit too expensive in most cases, so I settled on 1600x900 LCDs and came up with the Dell Latitude E6430 and Latitude E6530, the HP EliteBook 8470p and EliteBook 8570p, and the Lenovo ThinkPad T430 and ThinkPad T530. A quick look at those prices should tell you why I ended up with another alternative that ends up being from the Mobile Workstation crowd: the EliteBook 8570w.

Midrange Business Laptop: HP EliteBook 8570w ($950)

Dustin reviewed a higher-end model of the 8570w last December, and the biggest concerns were the cost as well as throttling under heavy sustained loads. Well, without the DreamColor LCD and some other pricey upgrades, the 8570w can be had for $950; what’s more, the dual-core i5-3360M isn’t likely to trigger thermal throttling in a chassis designed for quad-core CPUs, and you still get a Quadro K1000M GPU as a bonus—all for less than any of the other “midrange” business options that include a decent LCD. There are a few drawbacks, however, most notably the size and weight: measuring 15”x10.1”x1.5” and weighing 6.8 lbs, the 8570w is substantially larger than other alternatives.

So, if you don’t need the extra power or girth, my secondary pick would be the ThinkPad T430, which for some will be their primary pick. Whichever way you go, the Latitude, EliteBook, and ThinkPad T lines are all high quality machines that are built to last. I think today’s models may actually be less durable than those of a few years back, thanks to concessions towards aesthetics and weight, but they’re still far more durable than the typical consumer laptop.

Straddling the line between the mainstream class of laptops and the Ultrabook and high-end categories, there is one other laptop that deserves a mention: the Lenovo ThinkPad X230. It's a bit too thick to qualify as an Ultrabook, and it has a lower resolution 1366x768 LCD, but that LCD is IPS at least, and that counts for a lot. I'd love to pick a model that comes standard with 8GB RAM and an SSD, but pricing for such builds ends up being in the $1500+ range, and you can perform those upgrades on your own. For $1030, you can get a 4GB/500GB HDD laptop with an i5-3210M processor (faster than any similar size Ultrabook), then upgrade the RAM and storage for another $300 and you're still coming in below the price of similarly equipped business Ultrabooks. 1366x768 isn't actually a bad fit for a 12.5" LCD, but I still wish we could get at least 1440x900 in its place; otherwise the X230 is a strong offering, particularly for those that enjoy a more svelte laptop experience.

Ultra-Business with an Ultrabook

Given the problems with getting good screens and all the other extras that I’d like to see in a decent laptop, it’s no surprise that at the high-end I had to continue sifting through the options before I could find something I really liked. First things first, along with a good LCD, at the high-end having a dedicated SSD for your storage device is absolutely critical. Few users will really notice the difference between a Core i7 and a Core i5 processor, but if you give them a chance to experience an HDD vs. SSD in a laptop it’s like night and day. Naturally, that brings Ultrabooks into the equation, and while I don’t normally consider Ultrabooks in the non-Ultrabook guides, business Ultrabooks are a bit of a special case. I decided that a 256GB SSD and 8GB RAM was a good target, so let’s take the contestants in turn.

In the Ultrabook camp, Dell has the Latitude 6430u, HP has the Folio 9470m, and Lenovo has their ThinkPad Helix and ThinkPad X1 Carbon (the Yoga 13 ends up falling into the “consumer” classification for me while the ThinkPad T430u lacks a decent LCD, if you’re wondering). Dustin recently looked at the Folio 9470m and the ThinkPad X1 Carbon, and found both were lacking in areas. Thankfully, I’m able to eliminate the dire 1366x768 LCD on the Folio from the equation, but it’s still not without concerns. The ThinkPad Helix is the most revolutionary of the options, but it’s also the most expensive by a sizeable amount, so that’s a tough call to make. In the end, I went with the one option that we haven’t reviewed, mostly because it looks to have a solid feature set and it's priced well (considering).

Business Ultrabook: Latitude 6430u Ultrabook ($1449)

While most Ultrabooks are often ill suited for IT departments, the 6430u does have most of what they would want: it has a user replaceable battery, two SO-DIMM slots, and only two screws secure the cover that provides access to the internals. Getting such features does make it a bit chunkier, however, as it weighs in at 3.7 lbs (1.69kg) and is 20.9mm thick—so this is as close to the limit as you can get while still being an Ultrabook. In truth, the specs of the 6430u and the HP 9470m have a ton of overlap, but considering the best price I can find on the HP is a full $200 higher, Dell wins out (again). Keep in mind that HP's Folio does give you an mSATA slot along with a 2.5" drive bay, so for some that can still be the better buy, but unless you absolutely need the extra storage space the cost is pretty steep.

For $1449, you probably expect a lot of computer, and Dell does its best to deliver with a 1600x900 LCD, 8GB RAM, and a 256GB mSATA SSD. The CPU is still at the middle of the ULV spectrum (i5-3437U), but you do get Windows 7 Professional and a TPM (the fingerprint scanner is optional on the 6430u, but the top-end model includes it). The 6430u also comes standard with a 3-year basic on-site warranty, which helps make the price a bit more palatable. I’ve found the build quality of Dell’s Latitude line to be quite good, and it’s definitely part of the reason why Latitude costs more than Vostro, but you pay for the privilege.

Let's See the Mobile Workstations

For the high-end, I really wanted something that could handle everything I could possibly want, which means it needs a quad-core CPU, discrete graphics, 16GB RAM, and a 512GB (or 480GB) SSD. Guess what? Only Dell actually let’s you easily configure such a laptop. [Insert frustrated scream here.] HP’s EliteBook 8570w is out as we had throttling issues with the quad-core configuration, leaving the EliteBook 8770w as the only possibility—and it’s a notebook with a beautiful LCD option at least; the largest SSD HP will install is only 256GB, unfortunately, though you could put two of them in RAID 0 at a rather astronomical price. Lenovo is in a similar situation with their ThinkPad W530. Thus, Dell basically wins by default, though you still have to decide between the M4700 and the M6700; I opted for the former, mostly because I don’t really need or want a 17.3” notebook. If you're not opposed to doing upgrades on your own (and you have an employer that allows such things), I'm inclined to go with the Lenovo W530 over the Precision, but opinions on such matters are definitely up for debate. Anyway, let's talk about the M4700.

High-End Business Laptop: Dell Precision M4700 ($2658 on sale)

So how would I configure this bad boy? While some might be tempted to simply max out every component, I’m still reasonable, so the I7-3740QM is more than enough CPU. I’d stick with Windows 7 Professional 64-bit, while on the memory front 16GB of DDR3-1600 in two DIMMs leaves room for an additional 16GB down the road if I get ambitious. The 1920x1080 Premium Panel is absolutely required (and interestingly only add $34 relative to the default 1366x768 junker panel), and for the GPU I’d go for the Quadro K2000M. The SSD is a large expense, and you could potentially stick with a cheap HDD and upgrade on your own to save money, but for simplicity’s sake we’ll grab the 512GB SSD option. (You can also do a 256GB mSATA SSD and still have room for two 2.5” drives if you forego the optical drive.) Add in the backlit keyboard, Centrino 6300 WiFi, and a 9-cell battery and we’re done. The “starting price” comes to over $3900, but Dell has a big sale going right now that brings the final price down to $2658—not too bad considering the hardware, though still way more than I’d actually be able to spend. (That’s what your employer is for, right?)

Here’s the catch: even if I were in a position to buy a high-end business laptop right now, I’d suggest holding off and seeing what Haswell can do next month. It may not matter all that much, and business laptops using Haswell are likely still several months off, but with just a few weeks left to wait there’s no sense in jumping the gun at this stage. Dustin also pointed this out in his M6700 review, but the aesthetics of the Dell Precision line are also falling behind—one more reason to wait and see. For the midrange and budget offerings, Haswell won’t have such an immediate impact, since it sounds like we’ll see quad-core first and dual-core will come later (similar to what we saw with the Sandy Bridge and Ivy Bridge launches). Ultrabooks are the exception here, as Haswell Ultrabooks are apparently going to roll out quickly, so that’s another category where I’d suggest a “wait and see” approach.

Even without any new hardware announcements, Google's opening keynote at I/O '13 this morning was an absolute treat on the software front, and even more so for developers. Notwithstanding the enhancements to Google Maps on the mobile front, the massive update to Google Maps for non-mobile devices was definitely the highlight of the morning.

Most of us have probably seen the gradual progression of Google Maps ever since it launched, but today's update brings to the table an entirely new design paradigm, promising a holistic user interface, allowing users to focus on tasks at hand and not be distracted by traditional UI elements; the map is the UI.

I was never really a big fan of the current iteration of the Google Maps UI, not because it was confusing at times or the sidebar wasted valuable screen space, but because the UI often turned out to be unintuitive, progressively increasing my reliance on more constrained mobile mapping solutions. Google's re-imagination of their entire mapping experience this morinng really ticks the box for most of my gripes with Google Maps, if not all of them. The new Google Maps is built using the latest web standards, leveraging vector maps support and promising an extremely snappy user experience in the browser environment, with no plugins required.

The user interface is uncluttered, almost pristine, allowing users to focus on the content, with just the search bar on the top left corner. All search results are displayed on the map simultaneously, with finer descriptions for top results, allowing users to delineate them easily. Integration with Google+ pulls up places that your friends have reviewed or liked, making it that much more easier to pick your perfect destination. The search bar is dynamic, providing contextual options, based on the search results. Once you have decided on the place, you can also access indoor imagery (if available) of the establishment to resolve any outstanding doubts, if any. And since the results are sourced from other Google services, they are dynamic and ever-evolving, only getting better as time progresses. Another neat feature highlights similar and related places on the map once a desired place has been selected, making it easier to plan your itinerary for places you are visiting. 

Getting directions to a place has also been streamlined, by highlighting the relevant routes and automatically displaying available public transit information using an intuitive UI, allowing users to compare driving times versus public transportation times, to plan trips with granular precision. This feature will be very useful in cities like Chicago with well established public transportation systems, where the new public transit scheduling UI will add a whole new level of precision when it comes to planning your trips.

On an international front, the new Maps experience integrates stunning panoramic views of available locations, including relevant crowd-sourced imagery and viewpoints, including Photo Sphere integration. Google has also promised support for the Oculus Rift and the Leapmotion Controller, for those of us slightly ahead of the technological curve. Google has also added other interstellar features, where zooming out a bit further presents a blissfully accurate rendering of our planet, including accurate real-time cloud and planetary data.

All in all, this is unequivocally the biggest upgrade to Google Maps in recent times and most likely helps Google leapfrog any advances by Microsoft or Apple to their mapping platforms, at least in the near future. 

So go on and sign up for the preview now. Google has promised invites to start rolling out tomorrow morning.

Google Maps Preview

Google’s Chris Yerga took the stage this morning at the company's sixth annual I/O Conference to announce Google's own flavor of a subscription-based music service called Google Play Music All Access. All Access promises to leverage Google’s deep understanding of your preferences, listening habits and social circles to deliver a highly tailored music discovery experience.

Google aims to one up the competition by enabling users to get to their own and potentially new music (that they’ll hopefully like) as quickly and intuitively as possible. Google showed off a brief demo of the app running on Android, but the service can be accessed via a traditional web-based interface as well. The orange and white colored UI was slick and aesthetically pleasing, allowing users to instantly queue up songs and listen to personalized radio stations. The demo did however leave a lot of unanswered questions such as mobile cross-platform compatibility, support for traditional desktop operating systems and other features such as offline syncing, but we can definitely expect tight integration across all of Google's services, especially Google+.

Google Play Music All Access launches today in the United States for $9.99/month, with the promise of a gradual international rollout. There’s also a 30-day free trial and users that sign up before 30 June 2013 get a discounted rate of $7.99/month. It is unclear whether this is a lifetime or a limited time offer, but Google’s been known to be notoriously generous in the past, so we can always hope. 

Although the subscription-based music streaming market is by no means nascent with heavyweights like Spotify and Rdio, Google’s undoubtedly beaten Apple to the punch here, which is long rumored to be developing its own subscription-based music service, presumably for launch later this year. 

For the time being though, if you are in the US, head on over to Google Music and try All Access out for yourself.

Google Music

Hot on the heels of last month’s expansion of AMD’s Never Settle Reloaded game bundle, which saw Far Cry 3 Blood Dragon added across the board, AMD is back once again with another expansion of their game bundle program.

The latest rendition of AMD’s program will see AMD launching a new, smaller game bundle to run alongside the Never Settle Reloaded bundle. The new bundle is to be called the Level Up bundle, with the complete package being dubbed the Level Up with Never Settle Reloaded bundle. Level Up won’t add any new games to AMD’s bundles – the game industry is between major launch periods anyhow – but instead Level Up will see AMD offering a greater number of their bundle games with most of their video cards.

With the exception of the 7990 at the high-end and 7770 at the low-end, most AMD bundles are composed of a subset of four games: Crysis 3, Bioshock Infinite, Blood Dragon, and Tomb Raider. Level Up will see most AMD bundles expanding by one more game, going from 3 to 4 and 2 to 3 respectively, and generally resolving the oddites of some cards including Tomb Raider while other cards include Crysis 3. Specifically the 7900 series and 7870 will become complete sets, with the 7900 series adding Tomb Raider while the 7870 will add Crysis 3. Meanwhile the 7850 remains unchanged, and the 7790 will get a third game with the addition of Tomb Raider.

Notably, unlike past AMD bundle expansions, Level Up doesn’t completely replace Never Settle Reloaded. Instead it is a bundle that is running concurrent to Never Settle Reloaded. Technically Never Settle Reloaded is unchanged, with the additional games coming from the Level Up bundle. The significance of this separation being that these are two bundles with two different vouchers, making it possible to start and end each program separately while cleanly avoiding the issue of changing a bundle in the middle of its run. At the same time AMD is being especially specific this time around that Level Up is a “while supplies last” offer (though this is technically true for NSR too), so all signs are pointing to Level Up being a more temporary offer, at least for now.

Throughout all of this it’s worth pointing out that game bundles are traditionally based around buying large number of copies of a title in advance. So besides the more straightforward benefit of increasing the value of their cards, the separate and temporary nature of Level Up may be AMD using the opportunity to burn off excess Crysis 3 and Tomb Raider keys.

All things considered the expansion of AMD’s game bundle is not unexpected, as AMD’s success with Never Settle programs means that they will continue using game bundles to increase the value of their products and offset at least some of the need for price cuts. But even this was admittedly earlier than we figured any additional expansion would arrive, so this comes as a pleasant surprise. More interesting perhaps, the 7790 which launched only a bit more than a month ago is already down $10 on average and has seen both Blood Dragon and Tomb Raider added to its game bundle, so we’re still seeing some price cuts alongside these larger bundles. Overall with 7790s going for as little as $110 after rebate, and 7870s for as little as $220, game bundles are clearly becoming an increasingly large part of the complete package for AMD video cards, with all of the benefits and drawbacks thereof.

Earlier this week I noted that AMD’s Radeon 8970M and Richland A10-5750M are both available in MSI’s new GX70 3BE gaming notebook, but at the time I couldn't fully disclose the specs on the Radeon 8970M. Today, AMD is officially launching the 8970M, so we can talk specifics. Cutting right to the chase, here’s what the 8970M brings to the table, with the rest of the 8000M family details to help put things in perspective:

Obviously, the 8970M is going to be AMD’s fastest current mobile GPU, but what we’re missing from the above slide are the specs of the previous generation 7970M. There’s not much to add, unfortunately: the 7970M is identical to the 8970M except that it lacks the 900MHz Boost clock. Both use the same Pitcairn core (codename Neptune for the 8970M and codename Wimbledon for 7970M), with 1200MHz GDDR5 (4.8GHz effective). I’ve even got an Alienware M17x with a 7970M that overclocks without issue to 900MHz/1300MHz, so really the 50MHz Boost Clock is just a software tweak to the existing chip to improve performance by up to 6%.

While there’s not a whole lot to discuss in terms of the hardware or architecture, there’s a separate story that does warrant some discussion. For the past several years, NVIDIA has basically been on a roll as far as gaming titles are concerned, with a large number of the AAA titles sporting an NVIDIA logo. Our last two mobile gaming suites have had quite a few NVIDIA titles, and there have been other major releases with NVIDIA branding (both Batman games, Borderlands 2, Mafia II, and Metro 2033 to name a few). There are also a few recent and upcoming releases from the green team, like the just launched Metro: Last Light and Star Trek, with ARMA 3 coming down the pipeline.

That’s all well and good, and up until the past nine months or so I’d say NVIDIA had an advantage in terms of game developer support. However, check out the list of games with ATI branding from the last year or so: Far Cry 3 (and the spin-off FC3 Blood Dragon), SimCity, Bioshock Infinite, Tomb Raider, Crysis 3, DmC (Devil May Cry), Hitman: Absolution, Sleeping Dogs, and Deus Ex: Human Revolution are all on the list, and outside of DmC and Sleeping Dogs I’d say all of those are names that any gamer would recognize and most of those games have been good if not excellent. If AMD can keep up the pace of AAA releases, they might regain some lost ground. The Never Setting and Never Settle: Reloaded bundles are possibly the best gaming suites ever given away for free with a new GPU purchase. AMD is hungry, I’ll give them that.

Not surprisingly, many of the AMD branded titles perform better on AMD hardware, and in general the GTX 680M and HD 7970M (nee 8970M) are basically on equal footing. With the appropriate settings and drivers, AMD shows a pretty significant performance advantage in their slides relative to NVIDIA, but as always we would take those numbers with a grain of salt. Like NVIDIA, AMD cherry picks the games and settings to put their GPUs in the best light; we’re working to provide some independent testing of a larger variety of games with branding from both vendors along with a larger spread of settings, but right now we don’t have all the required hardware in hand for such an article.

The one item that still generates some concern is AMD’s Enduro switchable graphics. My earlier encounters with the technology (especially prior to the Enduro branding) were less than acceptable, and even with Enduro I can’t say AMD has caught up to NVIDIA’s Optimus. In particular, support for older GPUs and iGPUs is sketchy at best (Sandy Bridge and Llano generally require modded drivers for support, and even then they don’t always work). There are also certain applications that present difficulties; for example, I’ve tried Bitcoin and Litecoin mining on Enduro laptops, with less than stellar results. I’m not suggesting those are critical applications for most people, but if GPU compute doesn’t fare any better it would certainly be a problem. Still, at least the last few driver releases from AMD have all had mobile versions that installed with Ivy Bridge and Trinity Enduro solutions, so that’s definitely an improvement.

The other thing to consider is that where AMD’s 8970M is basically just a rebranding and minor clock speed increase from 7970M, NVIDIA has yet to announce their GTX 780M. Oh, we know it’s coming, and I would assume we’ll hear about it sooner rather than later (the Haswell launch practically demands a new high-end NVIDIA GPU); the question is whether GTX 780M will be a rebranding/higher clocked GK104 part, or if NVIDIA will do something more—like, say, getting the higher performance of the GTX 680MX into a 100W power envelope. (GK110 in a notebook isn’t going to happen any time soon, I suspect.)

The first notebook to ship with HD 8970M is MSI’s already discussed GX70 3BE, but more will certainly follow. At the very least, Alienware and Clevo notebooks are basically a gimme, considering they already support HD 7970M without issue. Pricing will also likely continue to be in AMD’s favor—NVIDIA hasn’t made their highest performing mobile GPU competitive in price for a while, mostly because they apparently don’t need to. We’ll see if anything changes with the status quo this round, but outside of the software story the Radeon HD 8970M doesn’t bring anything truly noteworthy to the table. We’ll probably have to wait for 20nm mobile GPUs before we see major improvement in performance, but will have to wait until the next round of GPU updates.

As usual, the full deck of AMD slides is provided in the gallery below, lest anyone think we’re intentionally skipping over important information.

Gallery: AMD Launches Radeon HD 8970M

For many users who have gone to a large LAN event, taking stock of all the crazy PC builds is part of the experience.  Some users modify the case for a particular styling, and then insert a water cooling system with just the right amount of LED lighting.  Last year at Computex, Lian Li showed us a mini-ITX chassis in the form of a train, and even put an engine inside it instead of a PC to show it could actually function as a train.  This time around, we get a case to show off all the mini-ITX goodness without too much effort.

The PC-Q30 is a little different to say the least.  We are dealing with a curved aluminum chassis adorned with an acrylic window.  Lian Li believes the styling befits any PC location – as a HTPC in the living room, as a small gaming machine, or just as an under the desk work machine.

Aside from what is visible from the window, the chassis can hold a dual slot GPU up to 7.8 inches (200mm, or just above mini-ITX spec), making it ideal for the ASUS GTX670 Mini as shown above, or something like a stock GTX460.  The space below the motherboard is for up to four 2.5” hard drives in a removable cage, but there is no space for any optical drive.

Ventilation comes in the form of a rear 140mm exhaust fan, along with apertures at the top and sides for intake vents.  Power supply options are limited to those of 125mm in length, so we are really looking at something like the Silverstone SFX or Seasonic SFX series PSUs.

Other features on board include a power button that lights blue at idle and red while loading, as well as the IO panel to the left hand side.  This includes a pair of USB 3.0 and front panel audio, although the motherboards IO does stick out of the top just in case you need any more ports. 

Retail price is $149, with expecting shipping in June.  The website for the case can be found at this link.

Gallery: Lian Li PC-Q30

The Haswell launch gets just a little more complicated each day, with the latest wrinkle being that standard ATX power supplies that conform to the ATX 2.3 standard may not actually be capable of supporting Haswell's extremely low power sleep states. The result has been power supply manufacturers scrambling to see if their power supplies can handle Haswell's requested 0.05A on the 12V rail as opposed to the ATX 2.3 minimum's 0.5A.

Both Corsair and Sea Sonic have been rigorously testing their power supplies to determine which ones definitely can handle Haswell's C6/C7 states, and both have a list of power supplies which are certified to run Haswell, and in the case of Corsair and their substantial portfolio, which power supplies should be able to handle Haswell with no problems.

Starting with Sea Sonic, the following power supplies are certified Haswell ready:

Corsair is a little bit more complicated. Certification testing is ongoing, but the following power supplies have been guaranteed to be fully compliant with Haswell:

Corsair's validation efforts continue on the following models (predominately legacy), which they say are "likely compatible" but are not confirmed yet:

If you're looking to make the jump to Haswell next month, the best thing you can do is likely going to be to watch the homepage and support page of your vendor of choice. Haswell will still work just fine with most power supplies, but you may have to disable these lower power sleep states to maintain stability.

In the desktop form factor NAS space, it is quite difficult to find units with more than two network ports and / or
PCIe expansion slots. These features are usually reserved for rackmount units targeting high end SMB users. Synology is bringing some of these high-end SMB / enterprise features to the desktop form factor with the launch of the DS1513+. The enclosure has been redesigned and the drive bays are now screwless.

Gallery: Synology DS1513+

To differentiate itself from the DS1512+ (with which it shares the same computing platform, a 2.13 GHz dual core Atom D2700), the unit has four GbE network ports. The unit also brings multi-path IO (MPIO) to the desktop form factor. SHA (Synology High Availability) is also part of the enabled features. This provides for seamless failover transition by maintaining identical contents in real time across two Synology NAS units.

Synology claims read and write throughputs of 350 MBps and 200 MBps respectively when all four ports are link aggregated. The unit is now shipping globally, but availability may vary across regions.

It has been a while since we last heard from NVIDIA about Project Shield. Today, NVIDIA is dropping "Project" from the name, making it formally just Shield, and simultaneously announcing pricing and availability information about its Tegra 4-packing handheld gaming console. The specs for Shield remain the same as what we saw at CES 2013 — 1.9 GHz Tegra 4 SoC, 5-inch 720p display, and Android 4.2 Jelly Bean. What's nice about getting a device straight from the SoC vendor in this case is that NVIDIA promises it will be able to push out OTA updates with the latest and greatest Android version basically in lock step with its own official Tegra 4 BSP (Board Support Package) software offerings, which makes it essentially the Tegra 4 reference platform.

As for availability, the big news is pricing, which will be $349 in the US, with preorders starting on May 20th from vendors familiar to everyone. NVIDIA called out Newegg, GameStop, Micro Center, and Canada Computers explicitly as preorder vendors, with others to follow after the preorder period. As for ship date, NVIDIA is aiming for late June for fulfillment. At $349 the Shield is more expensive than the major first party handheld gaming consoles like the Sony PS Vita or Nintendo 3DS, but an impressive middle ground and price point nonetheless for basically what boils down to a higher-end smartphone sans cellular stack but with a built in gamepad. We're excited to get hands on with Shield in its final form with the final tuning of its joysticks, triggers, and D-Pad.

In conjunction with the launch of Shield will be availability of the PC game streaming functionality as well, initially in beta form. NVIDIA has a set of recommended titles which have been optimized for the Shield controller scheme, and as we experienced at CES likely include UI tweaks to make 720p handheld gaming a reality. 

Source: NVIDIA Blog (Shield), (5 Games) 

This is one of the more interesting pieces of hardware to keep an eye on if you're looking at a reasonably priced gaming notebook. I've toyed around with MSI's GX60 notebook a bit, and while I won't have a full review (it's too late to be useful), I will be posting benchmarks in the near future. Well, GX60 is about to become the old news, as the MSI GX70 3BE specs are now posted at MSI's website. Here's the quick overview:

So, first the good news: this will be the fastest possible AMD APU and AMD dGPU for the next generation of notebooks. We can't share the specifications of the 8970M right now, but all appearances are that it is basically a clock speed increase from 7970M. As for the APU, Richland is likewise a higher clocked version of Trinity, only in this case we have raw numbers: its base clock is 200MHz higher than the A10-4600M while the maximum Turbo Core speed is 300MHz higher. In general that means the A10-5750M should be around 10% faster than A10-4600M, which will certainly help in CPU limited situations.

The bad news then is that a 10% clock speed increase from Trinity isn't going to be enough to close the gap in many titles, depending on the resolution and quality settings. Looking at Trinity vs. Ivy Bridge with 7970M, I've seen Intel outperform AMD by 50% or more, particularly in titles that pound the CPU (e.g. Skyrim and StarCraft II); on other games, however, it's basically a wash at high quality 1080p settings, so as a more budget-friendly gaming notebook the GX70 has potential.

The other bad news is that my continuing experience with Enduro is that it's not all that it's cracked up to be, but going pure AMD helps quite a bit. Getting updated drivers with an AMD APU and dGPU is easier, and AMD dGPUs simply cooperate with AMD iGPUs better it seems. I've done some testing with the latest 13.5 Beta2 mobile drivers issue on several other laptops (including the MSI GX60), and for most mainstream applications and games they have been fine. However, there are still times when everything doesn't work quite as smoothly as I'd like.

We don't have an MSRP on the MSI GX70 3BE yet, and there will be a variety of models for the various markets. At the lower end of the spectrum, I expect we'll see pure HDD models with 8GB RAM sell in the neighborhood of $1200-$1300, while higher end models with SSDs, Blu-Ray, and 16GB may push into the $1500+ range. The chassis design appears unchanged from the existing GX70, which isn't too surprising, so basically we're getting faster hardware. ETA for the MSI GX70 3BE is June 2013.

Gallery: MSI GX70 3BE: Richland A10 APU and Neptune 8970M Gaming Notebook

Yesterday Western Digital and SanDisk announced their collaboration for hybrid hard drives (or SSHDs as they are now called). The idea behind the move is that SanDisk will supply Western Digital with iSSDs (I’ll explain what these are in a bit), which WD will then integrate with their hard drives to act as a read/write cache. The new 2.5" WD Black that was first showcased at IDF last year will be the first fruit of the collaboration, and it will also be WD’s first SSHD.

WD’s approach with SSHDs is slightly different from Seagate’s. Seagate simply put a NAND package on the PCB and used their own controller manage it. SanDisk’s iSSD, on the other hand, is a standalone SSD with a SATA 6Gbps interface in BGA-156 form factor (sometimes called µSSD). Both implementations obviously have their pros and cons: Seagate has total control over the NAND (garbage collection, etc.) but it also means the NAND performance is up to Seagate’s engineers, whereas WD can rely on SanDisk’s expertise on the NAND frontier and concentrate on caching and hard drive technologies.

Now, before we get too excited, putting an iSSD inside a hard drive won’t magically solve the biggest problems that SSHDs have. The first generation WD Black will only have 8GB-32GB of NAND, which is better (or the same) than Seagate’s SSHD (or Momentus XT as it was called previously) but it’s still not enough to provide performance that is even close to the SSD-only experience. SanDisk promises pretty impressive numbers for the iSSD (450/350MBps sequential read/write, 9K/1K IOPS random read/write), but those are for the largest 128GB SKU. Remember that one of the key elements of SSD performance is high parallelism: The more NAND you have, the better the performance is because you can access multiple dies simultaneously. As the Black SSHD will only have up to 32GB of NAND, it won’t be able to achieve a similar level of performance as the 4x larger 128GB model.

With increasing NAND densities the performance difference between capacities has become an even bigger issue because performance decreases with every process node, yet capacity per die goes up. This is a double-whammy that results in lower parallelism (and hence performance) at the smallest capacities. With SLC NAND you could somewhat dodge the performance issue since SLC NAND is faster to begin with, but unfortunately the iSSD is MLC based (which shouldn’t surprise anyone given the pricing of SLC NAND).

When you combine the very limited amount of NAND with a low-power integrated controller, it’s simply impossible to get performance that’s anywhere close to a decent standalone SSD. Of course there’s the caching side too because only a small portion of your data can be stored in the NAND, so in most cases you will still be limited by the spinning platters. With 32GB it should finally be possible to cache Windows in full, although the hurdle of hardware-level caching is that you have no say in what goes where.

What’s really special about this announcement is the timing as Seagate introduced their first consumer SSDs only a day earlier. I have a feeling that WD and SanDisk had not planned to go public with their partnership yet but Seagate’s announcement changed their plans. With WD’s biggest rival entering the SSD market, it’s clear that shareholders want to know WD’s strategy in order to maintain credit on the company.

Aside from keeping investors happy, there are also concrete reasons for the partnership. By far the most important one is the fact that SanDisk is a fab owner (a joint-operation with Toshiba where SanDisk gets 49% of the NAND output). Nowadays if you want to do something that requires NAND, there’s no other way to guarantee a steady NAND supply than to partner up with a NAND fabricator. There have already been several NAND shortages in the market (and it's only going to get tougher this year) and the brutal fact is that the ones without a fab or partnership are the last ones in the supply chain.

I’ve already heard from several fab-less SSD OEMs that they have not been able to keep up with demand because there’s not enough NAND in the market. For someone like WD a steady NAND supply is even more important because at least in the beginning the WD Black SSHD is aimed towards OEMs (there's a custom connector so it doesn’t work in regular systems without an adapter). If your production is dependent on the fluctuations of the NAND market, OEMs will likely not choose your product because they don’t want to take the risk of halting their own production due to the lack of drives. It’s not a coincidence that for example Apple sources their SSDs from Samsung, Toshiba, and SanDisk, which are all NAND fabricators.

If WD ever decides to re-enter the SSD market, the partnership will obviously be even more important. I wouldn’t be surprised if WD and SanDisk were actually working on an SSD together. SanDisk has controller IP thanks to the acquisition of Pliant in 2011, but that arm of SanDisk has mainly been focused on the enterprise segment. The consumer market has much lower profits and it’s usually not profitable to design a consumer-grade controller on your own, but with a partner like WD it can turn out to be a good investment as R&D costs can be shared and WD has an enormous distribution channel for providing the product to the market.

All in all, I have a feeling that the real fruits of this partnership won’t be seen today or tomorrow. The WD Black SSHD is definitely an interesting product and we will try to get one in for reviewing as soon as possible, but it's likely that you will still be better off with a small-ish SSD accompanied by a hard drive for storage. My gut is saying that this is more of a transitory product as WD gets ready to re-enter the SSD market. That doesn’t mean it’s the end of story for SSHDs, but this announcement should have happened two, preferably three, years ago.

Our next set of motherboard recommendations sit at the feet of Sandy Bridge-E, the current de-facto platform for consumer throughput performance.  If you have a non-GPU accelerated workload and a consumer budget, X79 offers a platform with quad channel memory as well as two 6-core processor SKUs with hyperthreading enabled.  The only way to get more is to jump to Xeon processors, which can be 2-3x more expensive, or 2P/4P systems from both AMD and Intel, which offer a mix of benefits depending on how the software is programmed.

X79 and Sandy Bridge-E were released in November 2011, with a cavalcade of motherboards in the first 12 months of launch.  Now in 2013, new motherboard production has been slow, with only a few models to look at before the next high-end installment for Ivy Bridge-E, which most media and users alike are expecting sometime later this year.  We would assume that the socket will be the same, if Sandy Bridge to Ivy Bridge is any indication, though details on ‘X89’ are few and far between.  Motherboard manufacturers should be in there right now designing with products ready for launch, hopefully beta testing the first or second batches of ES boards.

But on X79 today, there is still plenty of choice.  Since launch I covered 13 different motherboards (which doesn't sound like much), including the high end models from ASUS, ASRock, MSI and ECS.  Here are a few of our favorites that are worth considering.

Pick of the Range: ASUS Rampage IV Extreme ($430) – AnandTech Silver Award

Our pick of the best falls at the feet of the most feature packed and almost the most expensive model -  the ASUS RIVE, part of the ROG range.  When testing this motherboard, it did almost everything right for overclockers and would keep a lot of users very happy.  It comes with MCT at stock but also overclocks like a beast, being one of the top boards for competitive overclockers due to its stability, memory compatibility, and features like the OC Key which you do not get in other motherboards.  The board supports 8 DIMM slots for up to 64 GB of memory, an extended heatpipe arrangement across the power delivery, overclock tools such as voltage reading pads, VGA hotwire, an LN2 switch, a SLOW mode switch, Subzero Sense, PCIe disable switches, and other features such as extra SATA 6 Gbps, extra USB 3.0, USB BIOS Flashback, as well as enough PCIe lanes to abuse 4-way GPU gaming.

The software stack and BIOS from ASUS are great pieces of kit, and there is also the ROG forums if you have any concerns and issues – the admins are there to help you push the limits of your system and get the best out of the top line purchase.

Read our review of the ASUS Rampage IV Extreme here.

Gallery: ASUS Rampage IV Extreme

AnandTech Bronze Award: MSI Big Bang XPower II ($370)

The MSI Big Bang XPower II (BBXP2) did not win an award because of the novelty heatsinks – if anything that was more of a detractor.  What it does score highly on is performance, as well as functionality.  We get a triple BIOS system, with one BIOS being easily replaceable – these are alongside 10 SATA ports, 8 USB 3.0, a full array of full-length PCIe slots, dual Intel gigabit Ethernet and bundled in the box is a USB 3.0 front/rear panel, an eSATA back panel, eight SATA cables and a variety of SLI connectors.

For overclocking there is the OC Genie button for a one-button OC, or the Direct OC buttons allow overclocking on the fly.  Like the RIVE, we get PCIe switches to disable PCIe slots, voltage read points, and power/reset buttons with a two digit debug.  There are a few areas for polish (fan controls), but if you picked one off the shelf I doubt you would be disappointed.  It is longer than other boards, such that users will need to check their case dimensions.

Read our review of the MSI Big Bang XPower II here.

Gallery: MSI Big Bang XPower II

AnandTech Bronze Award: ASUS Rampage IV Gene ($270)

With the power and heat output of Sandy Bridge-E, smaller motherboards might come across as more of a novelty – it would be understandable that with quad channel memory and 40 PCIe lanes that a mini-ITX board might not exist outside of ODMs, but there is room for micro-ATX.  At the top of the list of microATX boards is the Rampage IV Gene from ASUS.  This motherboard is designed to be the entry point into the ROG world for X79, and features the same software and BIOS pack like the Rampage IV Extreme above.  By being a smaller board however, we have some adjustments to make – it is primarily a gaming board, but still overclocks really well.  We get SupremeFX audio rather than an ALC898 for example.  It costs a bit more than the ASRock X79 Extreme4-M while having similar outward facing hardware (the beauty is in the underlying details and component choices), but the Gene takes a crown for performance and feature set.  Gene users want that small gaming system that rockets rings around other setups.  If Bitfenix make a micro-ATX prodigy, I would want this build in a red case.

Read our review of the ASUS Rampage IV Gene here.

Gallery: ASUS Rampage IV Gene

AnandTech Bronze Award: ASRock X79 Extreme11 ($600)

As a Sandy Bridge-E owner, what would you like on your motherboard?  How does space for 64 GB memory, four-way GPUs at x16/x16/x16/x16 for gaming, dual Broadcom NICs for teaming, a Creative Core3D audio codec, an integrated LSI 2308 SAS/SATA RAID chip capable of 8-way RAID-0 and peak speeds of 4 GB/s sound?  If that excites you, the ASRock X79 Extreme11 is the board to have.  As a technical exercise, fitting all the obscene power features a user may want onto a motherboard sounds like a recipe for disaster, and a mountain on a wallet.  In the end bundling two PLX chips and an LSI SAS controller on board does make it expensive, but ASRock are improving steadily with their software and BIOS bundling, making the board an attractive choice if you need one of its usage scenarios.

With our testing, and the price range of this motherboard, it is safe to say that this product is more aimed at workstation projects, such as an 8-core Xeon with ECC, rather than a product for gamers or overclockers.  Tool it up with eight SAS drives, seven single slot GPUs (or four dual slot), and away you go with a nice number crunching machine.

Read our review of the ASRock X79 Extreme11 here.

Gallery: ASRock X79 Extreme11

The Future for X79

While playing around with X79 is fun, the entry point for the platform is quite high.  The cheapest X79 motherboard on sale today is $190, and the cheapest processor is $300, giving a sub $500 entry point.  But that does not show the whole picture – while the price of the cheapest motherboards is not bad (considering the main bulk of Z77 is around or just below that price) the good motherboards will cost around double that.  Also the cheapest X79 processor, the i7-3820, is like an i7-2600K with a couple of extra memory channels – if you are not hitting the memory hard enough, it will not feel like an upgrade.  As a result, you might end up looking towards the i7-3930K which is $570-ish new.  Combine that with the Rampage IV Gene and we have an $840 entry point, before we consider memory, GPUs, SSDs, PSUs, etc.  Most users could take the top end Ivy Bridge today with that money and get a nice boost on the rest of the system, leaving those that need Sandy Bridge-E to actually go out and get it.

With all that being said, the future for X79 is unclear.  It is still the choice for pure multithreaded performance, even though Ivy Bridge surpasses it (at stock) in single threaded performance and Haswell will make the jump a little bit more.  With Ivy Bridge-E assumed to take socket 2011 like SB-E (if you believe recent leaked screenshots), then current X79 motherboards will get updated to support the new processors and perhaps a new batch of motherboards under the name X89 will be released.  We do not have details on that yet though, unfortunately - I want full SATA 6 Gbps and PCIe 3.0 for a start.  To put it into perspective, I use an X79 system as my general gaming/work system, and when dealing with the multithreaded workloads, I would not have anything else – except perhaps a 2P or 4P machine, but those are not inside my budget.

After entering the PC memory market about a year and a half ago, things have for the most part been quiet for AMD’s fledgling memory operations. With DDR3 enjoying a long, prosperous reign as the memory of choice for PCs, memory is not a fast-moving field that has the kind of rapid innovation and quick product turnovers as AMD’s GPU and CPU businesses do. But with that said, even as a commodity product there’s still periodic bumps in capacity and performance, and that’s what AMD is announcing today.

Joining AMD’s existing 1600MHz (RE1600) and 1866MHz (RP1866) Radeon Memory products today is a new SKU, the 2133MHz (RG2133) Gamer Series. The RG2133 is a middle of the road 2133MHz part, sporting a CAS 10 latency (10-11-11-30) and a voltage of 1.65v. In keeping with AMD’s other memory products they aren’t using any flamboyant heatsinks here, so these are simple DIMMs enclosed in a heatspreader, keeping the DIMM height at a low profile of 30mm.

As with AMD’s other memory products RG2133 memory is being manufactured and sold in 16GB kit form. The DIMMs are 4GB each – unlike AMD’s other speed grades which are 8GB these days – so the full 16GB kit is 4x4GB rather than 2x8GB. Meanwhile AMD continues to contract out the actual manufacturing of their memory products, and while they haven’t named the manufacturer of the RG2133 DIMMs they’ve said that they’re keeping the same manufacturer as with their other products, which would mean it’s once again Patriot doing AMD’s manufacturing.

For AMD the primary benefit of introducing another speed grade of RAM is not only to keep themselves competitive in the memory market, but also to work the fact that AMD’s APUs greatly benefit from increased memory bandwidth. Even though the GPUs in AMD’s APUs are lower performing than AMD’s discrete GPUs they’re still memory bandwidth starved to a very large degree, and as we’ve seen even back in 2011 with Llano most games benefit from additional memory bandwidth. So by pairing RG2133 memory with Trinity/Richland, AMD can significantly improve their GPU performance in many cases even with the same silicon.

The catch as always will be pricing. 2133MHz memory carries a distinct premium, so although faster memory will improve gaming performance it’s not a “free” upgrade. Builders and buyers will be looking at a $155 price tag for a 16GB RG2133 kit, $30 more than the RP1866 kit. However short of adding a discrete GPU, this is going to be the best way of improving AMD’s iGPU performance for the time being.

On a final note, AMD will once more be bundling licenses for their branded version of Dataram’s RAMDisk software with their memory. RG2133 kits will come with a license good for a 64GB RAM disk (note that you can’t actually get 64GB of RG2133 into a system at the moment). All other AMD memory products will continue to ship with a license for a 6GB RAM disk. The usefulness of AMD’s RAMDisk software remains questionable, but as AMD’s pricing is generally competitive it’s essentially a freebie with an otherwise solid memory package.

Today Sony is launching their newest line of laptops, the VAIO Fit series. There will be two different lines, the Fit and the Fit E, and all of the new laptops are "thin and light" and feature aluminum exteriors. Sony didn't provide detailed specifications, but it sounds like the VAIO Fit series is going after users that like the idea of an Ultrabook but aren't willing to pay the higher price premiums. That likely means entry level models will come with conventional HDD storage rather than SSDs or hybrid solutions, though Sony does note that SSDs and hybrid options are available on higher spec models.

Noteworthy features of the VAIO Fit laptops include 1600x900 HD+ LCDs on the 14" model and 1080p displays on the 15.6" models, with both sizes offering optional capacitive touchscreens. Sony also touts improved webcam functionality and high quality audio and states, with the 14E and 15E including "big box speakers" as well as a subwoofer on the 15E. All of the laptops also feature full-size backlit keyboards, with a numeric keypad on the 15" models. Finally, the VAIO Fit laptops will also feature Near Field Communication (NFC) technology, allowing users to share website URLs and other data with compatible NFC devices.

As with other recent announcements, the timing is enough to let you know that the models being discussed today are using Ivy Bridge (3rd Generation Intel Core processors), but we may see updates after the Haswell launch. Processor support tops out at Core i7 (presumably dual-core), with Core i3 and i5 also available. NVIDIA Graphics will also be an option, though the specific GPUs aren't listed.

The Fit 14 and 15 will be available in mid-May and will start at $649 and $699 respectively. The Fit 14E and 15E are apparently the lower-spec models and will launch at prices of $549 and $579.

Gallery: Sony Announces VAIO Fit Series of Laptops

Alongside the Seagate 600 and 600 Pro SATA SSDs, Seagate is also announcing its 1200 SSD aimed squarely at the enterprise market. While the 600/600 Pro use Link A Media's LM87800 controller with some degree of Seagate firmware customization, the 1200 apparently uses a fully custom Seagate designed controller.

Seagate is planning on using a mixture of custom designed and 3rd party controllers with Seagate custom firmware depending on the application. As I hinted at in our review of the Seagate 600/600 Pro SSDs, I don't believe the Link A Media relationship is one that will last - eventually I see Seagate transitioning to its own controllers. The 1200 is an indication of just that. 

We don't know a whole lot about the 1200 other than it is a 12Gbps 2.5" SAS drive. The 1200 will be available this quarter.

Seagate also unveiled a Virident FlashMAX II based PCIe SSD simply called the X8. The X8 features a FusionIO-like architecture that moves all NAND management from the SSD to the host CPU cores in the server. Given how readily available (and sometimes underutilized) host CPU resources can be, the tradeoff is sometimes worth it. The X8 will be available starting on May 27th. 

Like most converging markets (in this case, storage + NAND), the SSD industry hasn’t been dominated by players in the market that came before it. Instead, SSDs attracted newcomers to the client/enterprise storage business. Not unlike DRAM, owning a NAND foundry has its benefits when building a profitable SSD business. It’s no surprise that Intel, Micron and Samsung are some of the more frequently discussed SSD vendors - all of them own (either partially or fully) NAND foundries.

Whether or not ownership in a foundry will be a requirement for building a sustainable SSD business is still unclear, but until that question gets answered there’s room for everyone to play in the quickly growing SSD market. This year, Seagate re-enters the SSD market with a serious portfolio. Today it not only announces two 2.5” SATA drives, including its first client-focused SSD, but also a 2.5” SAS product and a PCIe SSD solution.

The products that we’re focusing on today are the two 2.5” SATA drives: Seagate’s 600 and 600 Pro.