Seagate recently rebooted their NAS offerings, completely revamping their 2013 Business Storage lineup and dropping the old software platform altogether. In its place, they adopted the Debian-based NAS OS,  development of which was started by LaCie prior to their acquisition by Seagate. In their 2014 lineup, Seagate has two classes of products, the NAS and the NAS Pro. While the former is suitable for workgroups of 1 to 25 clients, the Pro version pushes that up to 50. Read on for results from our evaluation of the 4-bay NAS offering.

The 2014 Commonwealth Games have officially opened at a ceremony in the Scottish city of Glasgow

One of ASUS’ many releases during Computex was for their new ROG Swift PG278Q monitor that boasted a number of impressive specifications all at once. The PG278Q combines a 2560x1440 panel capable of 120/144 Hz operation with support for NVIDIA G-Sync and 3D Vision, putting it firmly in the region of gaming and hence the ROG moniker.

Aside from NVIDIA G-Sync, the PG278Q comes with a Turbo Key on the rear for quick selection between 60 Hz, 120 Hz and 144 Hz depending on user preference. The GamePlus hotkey gives a crosshair overlay to enhance the gaming environment (useful in games that do not offer steady central crosshairs), as well as timer functions. The OSD is navigated by a joystick-like nub behind the side of the monitor.

Response time is listed as 1ms GTG, with 16.7M colors and 160-170º viewing angles. Connectivity is via DisplayPort 1.2 only, with a USB 3.0 pass-through hub also in the electronics. VESA support is for 100x100mm, and the monitor is listed at 7.0 kg (15.4 lbs). The PR gives a bezel dimension of 6 mm.

Due to the high refresh rate and inclusion of G-Sync, the Swift comes in as one of the most expensive TN panels on the market. Pricing will start at $799, varying by region, and the monitor should be available in Taiwan, APAC and EU today, with China in mid-August and North America by the end of August.

ASUS ROG Swift PG278Q Display 27-inch (68.5cm) widescreen with 16:9 aspect ratio Resolution 2D mode: 2560 x 1440 (up to 144 Hz)
3D mode: 2560 x 1440 (up to 120 Hz)
2D/3D surround: 7680 x 1440 (2D up to 144 Hz / 3D up to 120 Hz) Pixel pitch 0.233mm / 109 PPI Colors (max) 16.7M Viewing angles 170-degree (H) / 160-degree (V) Brightness (max) 350cd/m² Response time 1ms (GTG) ASUS-exclusive
technologies ASUS GamePlus Technology (Crosshair / Timer)
ASUS Refresh Rate Turbo Key (60 Hz /120 Hz/ 144Hz Overclocking)
ASUS 5-way OSD Navigation Joystick NVIDIA® 
technologies NVIDIA® G-SYNC™ Technology
NVIDIA® 3D Vision™ Ready
NVIDIA® Ultra Low Motion Blur Technology Input/output 1 x DisplayPort 1.2
2 x USB 3.0 (Upstream x 1, Downstream x 2) Stand Tilt: +20°~-5°, Swivel: ±60°, Pivot: 90° clockwise
Height adjustment: 0~120mm
VESA wall mount: 100 x 100mm Size 619.7 x 362.96 x 65.98mm         Weight (est.) 7.0kg

Gallery: ASUS ROG Swift PG278Q Monitor Released in APAC/EU, North America Coming September

Source: ASUS

Version:next-20140724 (linux-next) Released:2014-07-24

As the Commonwealth Games kick off in Glasgow, the BBC explains the background to the action - in 60 seconds.

The BBC's Yolande Knell visits a family in Gaza struggling to maintain their day-to-day lives as the conflict between Hamas and Israel rages.

Natalia Antelava gains access to a village where one woman apparently died, and another was badly injured, when it was allegedly shelled by Ukrainian Grad rockets.

China detains five people as part of its probe into a firm that allegedly supplied out of date meat to fast food chains, including McDonald's and KFC.

Rudy Kurniawan, who manufactured phony vintage wine in his kitchen, will be sentenced for fraud on Thursday.

Lyse Doucet reports from Gaza on a rescue mission to evacuate wounded civilians from Khan Younis and Shejaiya.

The leader of Pro-Russian rebels in eastern Ukraine, Alexander Borodai, has admitted contacts with Russian security services.

While communication from Microsoft about its layoffs and reorganization lacks a certain amount of clarity, one statement made in its earnings call yesterday did appear to be straightforward: "We will streamline the next version of Windows from three operating systems into one single converged operating system for screens of all sizes" said CEO Satya Nadella.

The immediate reaction was twofold. From some parties, there were congratulatory noises, praising Nadella for this new strategy that moved away from the Ballmer-era multiple operating system. From others, there was glee that the "confusing" line-up of Windows, Windows RT, and Windows Phone would soon be gone and that in the future users would no longer need to worry about what their devices were using. Some are even cheering the "fact" that this means that Windows RT will be killed off forever.

That Nadella's remarks provoked headlines and column inches is ever so surprising, however, because what he said isn't new, isn't really being interpreted properly, and wasn't really his idea.

Read 13 remaining paragraphs | Comments

Once in a while someone points out a POSIX violation in Linux. Often the answer is to fix the violation, but sometimes Linus Torvalds decides that the POSIX behavior is broken, in which case they keep the Linux behavior, but they might build an additional POSIX compatibility layer, even if that layer is slower and less efficient. more>>

Antana

Heritage Auctions, the Texas company orchestrating the Bitcoins.com sale, pulled the auction listing on Wednesday afternoon, stating: "This lot has been withdrawn from this auction. Bids are no longer accepted and previous bids are cancelled."

The move comes as the result of a federal judicial order issued on Tuesday that put an immediate halt to the sale of Bitcoins.com, the domain name owned by embattled Mt. Gox CEO Mark Karpeles.

"The lot is being held for now so we can get this sorted out one way or the other," Noah Fleisher, a Heritage Auctions spokesman, told Ars. "I haven't heard from [Karpeles] at all."

Read 9 remaining paragraphs | Comments

Today, NASA announced that it's issuing a Request for Information that seeks parties, either academic or commercial, who are willing to set up a communications relay orbiting Mars. Should the agency like the information it gets, it could extend its current fee-for-service approach well beyond Earth's orbit.

Because of weight and power restrictions, the hardware that we've landed on Mars can't carry high-bandwidth communication devices that can reach Earth (it does, however, carry lower-bandwidth hardware that can establish a direct connection). Instead, missions like the Mars Reconnaissance Orbiter, which has its own science instruments, also carry communications hardware that lets them receive high volumes of data from the planet's surface and quickly send it back to Earth.

MRO is the most recent hardware that serves this purpose, but it's already nearly a decade old; Odyssey, its fellow relay, is even older. Fortunately, the MAVEN mission, which arrives this year, will also have relay capabilities, as will the ESA's ExoMars orbiter, which should arrive in 2016.

Read 1 remaining paragraphs | Comments

NASA

With all the attention given to every nuance of climate data, areas of research that would have never attracted much public interest sometimes find themselves in the spotlight. So it is with the process of measuring sea ice cover. People pay careful attention because it appears to be a leading indicator of climate change. In the Arctic, where the warming has been most intense, sea ice is retreating rapidly, with record lows having been set every few years over the past decades.

But at the other pole, Antarctic sea ice has been steadily expanding, creating a bit of a conundrum for scientists. They've come up with a variety of explanations for why the two poles might be behaving differently but, in the mean time, people have latched on to the difference to question our understanding of climate change.

Now, a paper has come out questioning whether the difference between the poles is as dramatic as it seemed. The reason for the potential difference? Measuring sea ice is remarkably hard.

Read 13 remaining paragraphs | Comments

The Asus RT-AC87.

If you've been waiting for a Wi-Fi router that supports multi-user beamforming, there's now one for sale. The Asus RT-AC87 advertises support for 1.73Gbps throughput, and it implements one of the most highly anticipated features of 802.11ac Wi-Fi: MU-MIMO (multi-user, multiple-input, and multiple-output).

As we explained in the feature, "Wi-Fi networks are wasting a gigabit—but multi-user beamforming will save the day," MU-MIMO relies on multi-user beamforming to send data streams of up to 433Mbps to three or more users simultaneously. This is an improvement over single-user beamforming, which could send multiple streams of data, but only to one device at a time.

The Asus RT-AC87 uses a Quantenna MU-MIMO chipset and sends data over four streams to get up to 1.73Gbps on the 5 GHz band. It can send another 600Mbps over 2.4 GHz.

Read 5 remaining paragraphs | Comments

Earlier this Summer Microsoft released Surface Pro 3, a fully equipped Haswell ULT based tablet in a chassis that was much more in line with what you'd expect from a tablet. From a hardware standpoint, Surface Pro 3 is the chassis design that Microsoft needed from the very start. If you've ever been tempted by the Surface Pro story, the latest model was bound to push you over the edge.

Unlike previous generations however, Microsoft delivered multiple CPU/GPU options with Surface Pro 3. The device's price range extends both lower and higher than any prior iteration. While the original Surface Pro launched at $899 and $999, the second version offered increased storage options that drove the max price up to $1799. Surface Pro 3 starts at just $799 and can be configured at up to $1949. In our review we were sampled one of the launch versions of the tablet, a $1299 configuration featuring Intel's Core i5-4300U. Until next month the Core i5 models are the only ones available, however starting on August 1st you'll be able to purchase cheaper Surface Pro 3s with a Core i3-4020Y or more expensive versions with a Core i7-4650U.

We just got our hands on the $799 entry level Core i3 configuration and are in the midst of a full review of the device. Battery life testing alone will take a few days, but I wanted to share some of my initial performance data since there is a pretty substantial difference between the Core i3 and i5 models. For a quick refresher, this is what the Surface Pro 3 lineup looks like:

Microsoft Surface Pro 3 Configuration Options Configuration $799 $999 $1299 $1549 $1949 CPU Intel Core i3-4020Y Intel Core i5-4300U Intel Core i5-4300U Intel Core i7-4650U Intel Core i7-4650U SDP/TDP 6W/11.5W -/15W -/15W -/15W -/15W Cores/Threads 2/4 2/4 2/4 2/4 2/4 Frequency Base/Max Turbo 1.5GHz/- 1.9/2.9GHz 1.9/2.9GHz 1.7/3.3GHz 1.7/3.3GHz GPU Intel HD 4200 Intel HD 4400 Intel HD 4400 Intel HD 5000 Intel HD 5000 GPU EUs 20 20 20 40 40 GPU Frequency Base/Max Turbo 200/850MHz 200/1100MHz 200/1100MHz 200/1100MHz 200/1100MHz Storage 64GB SSD 128GB SSD 256GB SSD 256GB SSD 512GB SSD RAM 4GB 4GB 8GB 8GB 8GB

To hit the $799 price point Microsoft went to a 64GB M.2 SATA SSD. Interestingly enough, the CPU doesn't actually save Microsoft any money. The Core i3-4020Y carries the same 1000Ku price as the i5-4300U ($281). The Y at the end of the part number implies a Haswell ULx SKU, the only Haswell type to ship with a SDP as well as a TDP rating. Even using the same workload, the Y-series parts weigh in with a lower TDP than the more typically used U-series parts (11.5W vs. 15W). Given the substantial reduction in chassis thickness, I suspect the Y-series parts may have been a better fit for the design to begin with.

You do give up performance in the process however. Since this is a Core i3 processor Intel disables turbo boost in addition to lowering the base frequency of the chip in order to hit its 11.5W TDP. At 1.5GHz without turbo, peak single threaded performance could be as little as half of the Core i5 models. The GPU performance story is arguably more interesting. Although there's a decrease in max GPU turbo, I wonder if the lower TDP may help improve sustained performance in games. I ran a few tests to see how things change with the move to a Core i3. Let's start with a look at the CPU.

Single/Multithreaded Performance - Cinebench 11.5

One of the benefits of a Surface Pro is the ability to run heavy PC workloads just as well as lighter tablet workloads. The device can behave as a tablet, notebook or even a desktop. We'll start with Cinebench 11.5, an offline 3D renderer that is largely CPU bound. The single threaded incarnation of this benchmark should give us a good look at the worst case scenario performance delta between the Core i3 and Core i5 Surface Pro 3s:

Without the ability to boost above its 1.5GHz base clock, the Core i3-4020Y comes in substantially slower than the Core i5-4300U model. It turns out a 1.5GHz Haswell core is around the performance of a 2.3GHz AMD Piledriver core.

With both cores (and all four threads) active, the gap between the Core i3 and Core i5 parts doesn't really shrink any. It's clear that if you're after running heavier PC workloads on your Surface Pro 3, you'll want to spring for the Core i5.

General Usage Performance - PCMark 8 v2

Cinebench may overstate the difference you'll see between the $799 and $999/$1299 models in regular usage, so we turn to PCMark 8 v2 and PCMark 7 to give us a better look at real world performance differences when running lighter tasks. PCMark 8 v2 runs through a bunch of usage driven application scenarios grouped into home, creative and work usage flows.

From Futuremark's description of the Home test:

"The PCMark 8 Home benchmark includes workloads that reflect common tasks for a typical home user. These workloads have low computational requirements making PCMark 8 Home suitable for testing the performance of low-cost tablets, notebooks and desktops. Home includes workloads for web browsing, writing, gaming, photo editing, and video chat. The results are combined to give a PCMark 8 Home score for your system."

A 20% increase in performance is what the Home suite shows for the i3->i5 upgrade, and that seems fairly reasonable. The Core i3 model ends up performing quite similarly to the original Surface Pro, but in a much more appealing chassis.

Next up is the Creative test:

"The PCMark 8 Creative benchmark includes workloads typical of enthusiasts and professionals who work with media and entertainment content. With more demanding requirements than the Home benchmark, this benchmark is suitable for mid-range computer systems. PCMark 8 Creative includes web browsing, photo editing, video editing, group video chat, media transcoding, and gaming workloads."

The creative suite shows a larger 34% advantage for the Core i5 vs. Core i3. Once again the i3 performs within 10% of the original Surface Pro, but for heavier use cases the Core i5 is necessary if you want Ultrabook/Macbook Air-class performance in a Surface Pro 3.

Finally we have the Work test:

"The PCMark 8 Work benchmark test measures your system's ability to perform basic office work tasks, such as writing documents, browsing websites, creating spreadsheets and using video chat. The Work benchmark is suitable for measuring the performance of typical office PC systems that lack media capabilities. The results from each workload are combined to give an overall PCMark 8 Work score for your system."

Here we see another 33% advantage for the Core i5, and surprisingly enough a pretty similar gap between the Core i3 SP3 and the original Surface Pro. It's really the lighter workloads that will understandably show less of a gap between the i3 and i5.

PCMark 7

I'm also including PCMark 7 results which tend to fall in between the PCMark 8 v2 results, here showing a 26% advantage for the Core i5 model.

GPU Performance - 3DMarks

I'll split our look at PC GPU performance into two sections, first let's look at peak theoretical performance using 3DMark. The Core i5-4300U has a 29% higher peak GPU clock compared to the Core i3-4020Y so we could see gains of up to that in any GPU bound scenario.

The heaviest 3DMark 2013 test here is Fire Strike and we see a 27% advantage for the Core i5 vs. Core i3 model. Note that the i3 is still substantially quicker than the original Surface Pro.

In lighter graphics workloads the gap narrows, with the Core i5 model holding a 21% performance advantage over the Core i3. The entry level Surface Pro 3 still delivers substantially better GPU performance than the original Surface Pro.

The Ice Storm test seems to provide a good mix of CPU and GPU bound work for the platforms, which contributes to the largest gap we've seen thus far in our GPU tests of 30%.

Looking at 3DMark 11 there's virtually no performance difference between the two platforms. This ends up being a bit more strenuous of a test, likely forcing the Surface Pro 3 into a thermally constrained situation where GPU clocks can't remain at max for long.

PC Gaming Performance - Dota 2

In our original Surface Pro 3 review I asked Ryan Smith to put together a reasonable Dota 2 benchmark to better illustrate what a prolonged light graphics workload would do to the system. In the end I found that the thermal constraints and default fan profile of Surface Pro 3 can contribute to lower sustained performance vs. Surface Pro 2. Our Dota 2 benchmark did a great job of showcasing a real world scenario where SP3 could be appreciably slower than SP2.

With the Core i3 version I was curious as to whether the better binning and overall power reduction would result in a device that would be able to sustain a higher level of performance for longer than the Core i5 version. I ran the Core i3 model through the exact same Dota 2 workload and came away with some interesting results:

Whereas the Core i5 model couldn't make it through a single run of our test without stepping down in performance, the first run on the Core i3 system came very close to the performance of the Surface Pro 2. Subsequent runs however did drop to performance similar to the Core i5 model, although the i3 is able to sustain higher performance.

SSD Performance

My original review sample featured a 256GB Samsung PM851 (an OEM version of the SSD 840 EVO) SSD. The $799 Core i3 model I received features a 64GB SSD, and in my case it's an SK Hynix drive. SK Hynix is a fully vertically integrated NAND and controller manufacturer thanks to its acquisition of Link A Media. The HFS064G3AMNB-2200A in my Core i3 based Surface Pro 3 uses a LAMD 87800 controller like the Corsair Neutron/Neutron GTX and Seagate SSD 600 series of SSDs. Both of those drives proved to be highly consistent performers, even when running in a full state. I'm not sure if all 64GB models will use the Hynix/LAMD drive, but given how little free space is available on the 64GB drive I would hope as many of them are this configuration as possible. After installing my relatively limited suite of tests for this review I had 10GB free on the device, with Windows telling me that another 10.5GB was taken up by things I'd installed.

PCMark 8 v2's storage test gives us a good look at high level performance, and here there's no real difference between the 64GB and 256GB drives:

However this is a bit of an optimistic look, I turned to Crystal Disk Mark to give us a quick idea of peak sequential and random performance:

 

64GB Hynix SSD (left) vs. 256GB Samsung SSD (right)

Testing over a 4GB span there's clearly a reduction in sequential and random IO performance but both are quite solid on the 64GB model. I don't expect users to store large files on a 64GB Surface Pro 3 so the reduction in sequential read/write speed isn't a huge deal. Overall I'm quite pleased with the performance of the 64GB drive in my i3 sample.

Tablet Performance

Last but not least, I wanted to look at how the Core i3 would perform in a lighter set of benchmarks we typically run our ARM based tablets through. Surface Pro has always ended up at the top of these tests, but we've never had a Surface Pro clocked as low as the Core i3 SP3. At 1.5GHz this also gives us an interesting look into the IPC comparison between Haswell and Apple's Cyclone cores as the iPad Air runs at a similar frequency (1.4GHz). Do keep in mind that there's a fairly large difference in platform power between a Surface Pro 3 and an iPad Air however.

The light workload js benchmarks end up being great measures of single threaded performance, which is why we see such a substantial difference between the Core i5 and i3 models here. That being said, the i3 based Surface Pro 3 still ends up being faster than any of the traditional ARM tablets. The iPad Air comes dangerously close though.

Kraken tells a similar story, although here we see a larger gap between the i3 based Surface Pro 3 and the iPad Air.

The GPU performance story is no different than in our earlier tests. The i3 model takes a hit, but GPU performance is understandably much better than any other tablet you'd come across.

Final Words

Subjectively, the entry level Surface Pro 3 feels pretty quick. Even running at only 1.5GHz, a pair of Haswell cores are plenty fast. The real question is whether or not the extra $200 is worth the increase in performance. For anyone looking to use the Surface Pro 3 like a real PC and less like a tablet, the $200 Core i5 upgrade is a wise investment. Lighter tasks and more tablet oriented usage models however may not merit the extra expense. In a lot of lighter tasks we're looking at a 20 - 30% advantage to the Core i5 for a 25% increase in system cost. Ultimately I feel like the increase in storage capacity in addition to CPU performance may be what really justifies the larger expenditure for users who don't necessarily need the extra CPU performance.

The big unanswered question at this point is whether there are any gains in battery life. I'm working on producing that data now and I'll post a follow up once it's complete.

As many as 50,000 websites have been remotely commandeered by attackers exploiting a recently patched vulnerability in a popular plugin for the WordPress content management system, security researchers said Wednesday.

As Ars reported in early July, the vulnerability in MailPoet, a WordPress plugin with more than 1.7 million downloads, allows attackers to upload any file of their choice to vulnerable servers. In the three weeks since then, attackers have exploited the bug to install a backdoor on an estimated 30,000 to 50,000 websites, some that don't even run WordPress software or that don't have MailPoet enabled, according to Daniel Cid, CTO of security firm Sucuri.

"To be clear, the MailPoet vulnerability is the entry point," he wrote in a blog post. "It doesn't mean your website has to have it enabled or that you have it on the website; if it resides on the server, in a neighboring website, it can still affect your website." In an e-mail to Ars, he elaborated:

Read 2 remaining paragraphs | Comments

This image, confirmed by Ars Technica to be buried in the latest Steam client beta, could show a potential new design for a Steam Controller with a traditional analog stick. Valve

When Valve first unveiled its prototype for a handheld video game controller last September, the most striking thing about it, from a modern design perspective, was the complete lack of analog joysticks. It's an omission that remained even after Valve updated the controller prototype to include more traditional digital button placement.

So it's quite interesting that the latest version of the official Steam beta client includes the above image, showing a version of the Steam controller with an analog stick where the directional buttons used to be.

The file seems to have been first spotted by an enterprising member of the FacePunch.com forums, but we've confirmed that anyone with access to the PC version of Steam's latest beta client update should have this file on their hard drive (if you've updated the beta, it should be in [Steam directory]\tenfoot\resource\images\library\alpha_conroller_lines_d0g.png in case you want to confirm for yourself; we haven't checked the Mac and Linux clients yet). The file on our system, which appears to be an overlay for some sort of controller configuration or help menu, was created on May 19 and modified to the current analog-stick-sporting version on July 16.

Read 5 remaining paragraphs | Comments