AMD R9 Nano confirmed: launching in September for $650(Kshs.67,500)

Last week I reported on some leaked information about the AMD R9 Nano graphics card. Now AMD has unveiled the official specifications, along with a release date and pricing information.

The six-inch R9 Nano will be going on sale on September 10 for $650(Kshs.67,500), the same price as its full-sized R9 Fury X.

Surprisingly, the R9 Nano looks set to be very similar to the Fury X, despite being smaller and having a lower TDP at 175W. The Nano has 64 compute units, 4,096 stream processors, 256 texture units, 4GB of 4,096-bit high bandwidth memory, and 64 ROPs, all of which are identical to the bigger card. The core clock speed is the only difference, which AMD lists as “up to 1,000 MHz”, compared to the 1,050 MHz of the Fury X. It’s expected that the clock speed will be lower because of the lower power consumption. Our colleagues at Maximum PC believes that demanding workloads like Furmark that hit the thermal throttle could push the clock speed as low as 600 MHz, although games should be running at around 850-950 MHz.

The only other difference between the Nano and the Fury X is the cooling solution, with the Nano using air cooling rather than the water cooling of the Fury X. AMD is targeting an operating temperature of 75 degrees celsius, which it says is 20 degrees cooler than the R9 290X. The card’s thermal throttle point is 85 degrees, but AMD says that it should never reach that point.

AMD claims that performance is better in the R9 Nano than the Radeon R9 290X by 30 percent, and says that it is both the fastest and most power efficient Mini-ITX card available, which certainly looks to be the case. AMD has also released numbers which say the Nano has a 50 percent performance per watt advantage over the Fury X, thanks to its lower power draw.

So, it seems the Nano and the Fury X are essentially the same card, except the Nano is smaller and will use less power, while the Fury X strives for maximum performance. It will be interesting to see how they line up when we get our hands on one in September.

Build of the week: F3 Scout 2 Advanced.Build one Like This

Builder David Cathey figured he’d only do what was natural and combine his love for fast vehicles and fast computers in the design for the F3. So he gutted and warped and spit-shined a case to channel his love for sports bikes. And even though it won’t let anyone tear through the streets at frightening speeds, the F3 will certainly layer on a spiritual dose of speed-demonry while tearing through Los Santos.

For those unfamiliar with the F3's titular inspiration, Cathey breaks it down: "Probably the most in-your-face feature is the top half of the case being slanted forward to resemble the look of a sportbike. I even put the temperature gauge where an instrument cluster might go and covered the top grill in seat material to carry the look further. This necessitated a new frame, which I modeled after the tube frame of the MV Agusta. The final touch was covering everything in silver and red."
And the result is a build that channels the most distinct aspects of the bike it's named after. While the parts don't necessarily channel the same energy, they get the job done. Check out the build log for more step-by-step pics and commentary.

F3 Scout 2 Advanced parts list

GPU: Asus Strix GTX 980
CPU: Intel 4690K
MB: Asus Z97-PRO
SSD: Samsung 850 EVO 500GB, Samsung 850 EVO 250GB
RAM: 8GB Corsair Vengence Pro 2133MHz
PSU: Cooler Master V850
Cooling: EK Supremacy EVO CPU BLOCK, EK GTX Strix GPU Block and Backplate, XSPC Photon 170 Reservoir and Pump Combo, XSPC EX140 Radiators, Monsoon fittings and tubing, Cooler Master Scout 2 Advanced

Don't head out on a wild goose chase when buying a GPU to play Witcher 3

[H]ard|OCP has taken the guesswork out of GPU performance on the current version of Witcher 3 in this round up featuring 10 GPUs, five from each company.  Of course only NVIDIA supports lips occluded by PhysX powered mustachios but not everyone is obsessed with perfect hair.  Indeed when it takes a $1000(Kshs.103,500) video card just to enable the lowest options on HairWorks at 1440p without disabling every other feature one wonders why HairWorks had gamers tied up in knots.  Check out the full review for performance comparisons and even some HairWorks nitpicking.
This weekend also marks the 11th Fragging Frogs VLAN party, which kicks off on Saturday August 29 10:00 AM ET and will go until the last frog has been fragged.  Sign up in this thread if you haven't already and if you are new to the Fragging Frogs follow the links to the FAQ threads for information on which patches or mods you will need to apply to your games to get playing as soon as possible.

"We take The Witcher 3: Wild Hunt, using the 1.08.2 patch and latest drivers, find the highest playable settings and examine apples-to-apples performance with and without GameWorks across 10 video cards. We put a focus on NVIDIA HairWorks and how it impacts performance and find out which video cards provide the best gaming value."

Hands-on: Corsair Gaming's new K70 is the first RGB-backlit Cherry MX keyboard

Hot on the heels of Logitech's G910 Orion Spark keyboard announcement last week, Corsair's re-branding its entire peripheral depart to "Corsair Gaming" and releasing its own RGB-enabled keyboard: the K70 RGB. The difference? Corsair gets to keep those sweet, sweet Cherry MX switches. It's a back lit mechanical keyboard arms race, and I'm all too happy to watch it play out.

See, Logitech's Orion Spark uses exclusive Romer-G switches. You can read a bit more about them here, but basically it's a mod to the no-longer-under-patent Cherry MX design. Instead of the iconic Cherry MX stem, the center of the key is hollow allowing for smooth and even back lighting. The cost? At least to my fingers, the Romer-G switches feel a bit more rubber dome-esque than I prefer.

But there's a reason for Logitech's exodus—up until recently there was no RGB Cherry switch, and not due to lack of demand. According to Corsair, "Due to the current design of the Cherry MX switch, there is only room for a single 3mm LED that could mount directly onto the key switch." An RGB LED requires five millimeters of space, by contrast.

 Corsair really wanted RGB back-lighting though, so it worked with Cherry to solve the problem. The pair found that if they mounted the LEDs directly to the circuit board and used a lens to refract light, they could keep the iconic stem design and feel of the Cherry MX switches while still enabling full RGB back-lighting.

Corsair's built some intuitive software to go along with this device, and it's easy to set up anything from a solid background color to a WASD/Shooter control scheme that pulses quietly to a setup that sends waves for each key press to...well, complete visual chaos. You can just keep adding layers of lights until your eyes bleed, though you can only have a maximum of sixteen distinct colors on the board at a time.

Too lazy to make your own lighting scheme? Corsair allows you to import/export whatever you'd like, and is partnering with a few e-sports teams to, for example, make a keyboard set-up in that team's colors.

Similar to Logitech, Corsair is also putting out an SDK for the keyboard so developers can integrate the new lighting features into games—for instance, turning your entire keyboard into a health bar that fades from green to red as you take damage.

Sounds like a fantastic keyboard, right? Just like Logitech's Orion Spark, however, I expect the price to be the point where you choke: $170(Kshs.17,600) for the K70 and $190(Kshs.19,600) for the K95 (which adds eighteen dedicated macro keys to the left side, though both the K70 and K95 allow you to set a macro on any key, both on press and release)

We'll have a more extensive review sometime soon, but at the end of the day Corsair's K70 is a Cherry MX keyboard with full RGB back lighting, and it's the only such keyboard on the market. For now, Corsair's your only option if both those features are important to you.

Detailed Photos of AMD Radeon R9 Nano Surface (Confirmed)

New detailed photos of the upcoming Radeon R9 Nano have surfaced, and Ryan has confirmed with AMD that these are in fact real.

We've seen the outside of the card before, but for the first time we are provided a detailed look under the hood.

The cooler is quite compact and has copper heatpipes for both core and VRM

The R9 Fury is a very small card and it will be powered with a single 8-pin power connector directed toward the back.

  

Connectivity is provided via three DisplayPort outputs and a single HDMI port

And fans of backplates will need to seek 3rd-party offerings as it looks like this will have a bare PCB around back.

Connectivity is provided via three DisplayPort outputs and a single HDMI port

And fans of backplates will need to seek 3rd-party offerings as it looks like this will have a bare PCB around back.

NZXT Introduces Razer Edition of Source S340 Enclosure

NZXT has created another modified enclosure in conjunction with Razer gaming, and this time it's a new take on the excellent Source S340 mid-tower

As expected given the Razer branding this is a matte black enclosure with no shortage of green lighting, including a green underglow light. It's a look those familiar with the Razer edition of the H440 will be quite familiar with.

"Forged to match your Razer arsenal, the new custom design features a backlit Triple-Headed Snake logo, tinted window, illuminated LED power button, underglow, and green USB ports."

The special Designed by Razer edition of the S340 is up for pre-order today at the NZXT store, and this version is listed at $99.99(Kshs.10,400) - a considerable jump from the standard S340's $69.99(Kshs.7300) price tag.

MSI Releases Intel Z170 Motherboards Featuring XPower Titanium

We've heard news of Z170 motherboards from a couple of vendors today, but how many of them can claim a silver motherboard? If you guessed none you'd be correct - until this moment, that is.
Feast your eyes on the all new MSI XPower Gaming Titanium motherboard, a silver ATX design featuring the new Intel Z170 chipset. While this board is obviously rather new it already has the distinction of holding the highest DDR4 overclock to date (a fact which seems not to have escaped MSI's attention).

In addition to apparently overclocking memory rather well the XPower Titanium also features:

    OC DASHBOARD: On-the-fly overclocking
    Twin Turbo M.2 64Gb/s + Turbo U.2 ready
    GAMING LAN with LAN Protect, powered by Intel
    Audio Boost 3, Nahimic Audio Enhancer
    Game Boost: 8 levels of easy overclocking
    GAMING Hotkey: Assign macros, launch your favorite games or do real-time overclocking using a single button
    XSplit Gamecaster v2.5: 1 year free premium license
    Military Class 5: The latest evolution in high quality components featuring the brand new Titanium Chokes
    MULTI-GPU with Steel Armor: Steel Armor PCI-E slots

The name of the new motherboard might have something to do with the Military Class 5 titanium chokes, but there's no mistaking the look of this board which will provide a fresh option for new builds.

Phanteks Announces Enthoo EVOLV ITX SE Enclosure

Phanteks today announced the second edition of the mini-ITX member of the EVOLV enclosure family, the new Enthoo EVOLV ITX SE.

 The new Mini-ITX variant retains the aesthetic from the prior version, and supports full-size graphics cards up to 13 inches and dual-width liquid CPU coolers via the removable upper radiator bracket.There is certainly an added dose of style with this new edition as it is being offered with two interesting color combos, with your choice of either a white exterior with black interior, or a black exterior with red interior.
The enclosures will carry a 5-year warranty and retail pricing has been announced as $69.99(Kshs,7,300) for the black/red version and $79.99(Kshs.8,300) for white/black. The Enthoo EVOLV ITX SE will be available in October.

4GB, 8GB Or 16GB; How Much RAM Do You Need?

Buying a computer is not as easy task as you would have thought. Given a choice among the plethora of configurations available for each brand, confusion is bound to creep in. The most frequent questions that tend to bewilder the regular users are about the processors i5 vs i7, the RAM, and the graphics.
How much RAM would be enough for your computer? This is always a tricky one. According to a comparative study by TechSpot for the system performance with 4GB, 8GB, and 16 GB RAM, more the merrier concept doesn’t suitably apply here with the RAM.
The Windows 10 desktop machines with separate RAMs were tested by TechSpot to check each one’s performance. Each desktop running two browsers, playing super heavy games and 65 tabs open in the background. The machines were then subjected to even further test with Postbox, Photoshop, Microsoft Word and Excel, Dropbox, and other programs running in the background to find the differences between using 4GB, 8GB, or 16GB of RAM.
Well, you can the see the results for yourself. Th 16 GB RAM doesn’t do your computer much good over its other counterparts. Other tests showed that even the 4 GB RAM would be sufficient for the regular buyers.16 GB RAM gives you a slight performance boost if you are a power user and intend to perform heavy video editing or running other virtual machines. 16 GB RAM was also the advantageous while compressing huge files with 7-zip.
Conclusion: So, if you are planning to buy a new computer or upgrading your system, then 8 GB RAM should be preferred. 16 GB RAM is only advised if you are dealing with specific high-demand scenarios.

Intel Core i7 Skylake Processor Reaches Record-breaking 7GHz Clock Speeds

If you are skeptical about the upcoming 6th generation Intel processors and upgrading to the same, here’s a news that will clear all your doubts. A professional from Honk Kong has over clocked the Intel Core i7-6700K chip and made a new record for the micorprocessor frequncy. This speed acheived by Intel Core i7 Skylake processors is among the highest ever acheived by any microprocessor of Intel Core i-series.
According to Kitguru, Kui Lam, a professional overclocker from HKEPC website from Hong Kong, has overclocked an engineering sample of Intel Core i7-6700K Skylake microprocessor to whopping 6998.88MHz. This clock speed is about 74.9.7 per cent more that its normal stock frequency.
To achieve this, Kui Lam had to disable three cores out if four and the Hyper-Threading technology, and the voltage of the processor was increased from 1.2V to 1.888V.
Other components used to make this record included Asrock Z170 OC Formula mainboard, a G. Skill Ripjaws 4 DDR4 RAM along with a 1300W power supply unit. Added to these, the processor was cooled using liquid Nitrogen (LN2).



While this professional overclocker achiever speeds up to 7GHz using Intel Core i7 Skylake processors, it would be interesting to notice if the retail versions of these microprocessor chip will be able to kiss even 5GHz mark with the commercially available cooling systems. Another point that needs to be looked upon is the use of single core instead of all four. This extreme overclocking is surely interesting, but with changing times, the value of single-core CPU is diminishing.

AMD Launches the A8-7670K APU

AMD’s Kaveri lineup has seen its fair share of success since its launch about a year ago. Both entry level and mid-end systems have incorporated this architecture in an effort to combine both respectable x86 performance an excellent graphics capabilities into affordable PCs. However, with Intel’s Skylake right around the corner and lackluster financial results dogging their heels AMD needed something, anything, to spark a bit of interest in their product stack. This is where the A8-7670K steps into the equation.

Like the so-called “Godavari” A10-7870K, the A8-7670K utilizes the same Kaveri base architecture as its predecessors but leverages the inherent advantages of a very mature 28nm manufacturing process to boost performance without negatively affecting efficiency. Yes, we realize this is the same old story AMD has been using for its rebranded 300-series graphics cards and other APUs but it’s all they’ve got for now. Carrizo will remain a notebook-oriented architecture for the time being while next generation parts are still a long while off. Meanwhile Intel has successively increased their lead with Haswell and Broadwell.

Like the A10-7870K Godavari APU before it, AMD’s A8-7650K simply utilizes the same layout and features of the xx50-series product below it and increases clock speeds by a slight amount. In this case the base frequency has increased by 300MHz while the maximum Turbo speed sees a boost of a lackluster 100MHz. Meanwhile, the 384-core R7-series graphics core gets a bump up to 757MHz. All of this has been accomplished while keeping the chip’s TDP at 95W.

Pricing shouldn’t come as any particular surprise given how AMD’s current lineup has seen its relative costs gradually decrease over the last few months. For the time being the A8-7670K hits the $118 mark or about two bucks less expensive than Intel’s i3 4160. That’s actually not too bad considering Intel’s budget-focused chip only has two cores and its integrated graphics processor tends to fall behind quite drastically. In addition AMD’s APU boasts the K-series designation so it comes with an unlocked multiplier, potentially boosting overclocking potential.

In applications that play to AMD’s relative strengths of processor graphics like gaming and GPU compute, the A8-7670K naturally pulls ahead of Intel’s offering. While Broadwell CPUs narrowed the gap between Intel and AMD chips within this respect, the Kaveri architecture has managed to remain ahead as it leverages the Radeon lineup’s substantially superior driver package.

As AMD has struggled to retain some form of presence within the processor market, certain aspects of their APUs have been brought into sharp contrast. After years of towing the same Heterogeneous System Architecture talking points it is becoming abundantly evident that development of supporting applications is nowhere near where it needs to be for their dreams to become reality. For example, hUMA and Heterogeneous Queuing optimizations have yet to make it into any mass-market programs and likely won’t anytime soon. Even the use of GPU acceleration and its abundant potential for performance speedups has been largely ignored (other than a few cases) by developers and programmers alike. Granted, the programming space is slowly dragging itself forward but it is looking more and more like Kaveri and even Carrizo will be long gone before GPU compute integration becomes mainstream.

This causes a problem for AMD since they’ve been forced to fight a battle with architectures that have awesome graphics potential but wholly underpowered x86 processing stages. The A8-7670K will face those exact same challenges as its predecessors but it shouldn’t be written off either. It can pack an almighty punch for anyone that wants graphics-first throughput for a light gaming or HTPC machine.

With all of that being said, this is an abundantly convenient launch time for AMD since they are still able to compare their APUs to the older Haswell chips. Intel’s Skylake series CPUs promise to turn these metrics on their collective heads with significantly better graphics specifications and much better performance per watt numbers. Even upcoming Broadwell CPUs will likely cascade down into lower price brackets eventually. So while the A8-7670K may staunch the bleeding in the mid-range for now, AMD will soon need a ground-up refresh to effectively compete with what’s coming down the pipeline

AMD's Radeon market share continues to collapse, now resides at a troubling 18 percent

Back in Q4 2014, we caught word that Nvidia was ahead in the discrete GPU market with 76% market share, heavily outweighing AMD's contributions. At the time, Nvidia's highest performing card was the GTX 980. This was long before Titan X came out in March of this year, and the GTX 980 Ti in June, both of which boasted improved support for 4K gaming, no sweat.
The R9 390X was supposed to introduce a positive change for the Sunnyvale, California-based chip maker, but on the contrary, it was anything but revolutionary. Instead, it was just another reskin of the 300 series video cards.
Now, Nvidia has unsurprisingly taken yet another leap, pouncing AMD once again at a sudden boost from 76% market share in Q4 2014 to 82% in Q2 2015, acording to data acquired by Mercury Research. Even after issuing a slew of new products, AMD has failed to redeem itself assigning it a worrying label of 18% market share.
Curiously, even its futuristic R9 Fury X card couldn't save it from a disastrous year in sales. Tweaktown reports that this is due to a low manufacturing rate of the HBM1 modules, with only 30,000 units made over an entire year. Also problematic is the concern over performance comparisons with Nvidia's GeForce GTX 980 Ti, which has proved to be a better value in raw price vs. performance, especially if you overclock.

Mid-Range Gaming PC Build Using Intel i5 & GTX 970 for $784 (Kshs.82,000)

 

 

Graphics Card 

EVGA Super Clocked 970 ($320)(Kshs.33000): The video card is one of the most critical build components for gaming. Something powerful is needed, but there has to be a reasonable balance between its power and the other components. In this case, we’ve chosen the EVGA Super Clocked GTX 970 for $320. It will perform well at around high and ultra settings for most games at 1440p and 1080p.
Our most recent benchmarks including the GTX 970 can be seen here. As a reference point, it generally performs a bit below a 290X/390, a bit above a 770, and decently below the GTX 980. The dual-fan design helps ensure that the card remains reasonably cooled during load -- although this isn’t incredibly hard due to the efficiency of Maxwell -- and retains silence. The GTX 970 comes with 4GB of VRAM, a core clock of 1.2GHz, and support for 4k.

CPU

Intel i5 4460 ($190)(Kshs.19700): For this build, we’ve chosen a $190 i5-4460 CPU. Despite the recent release of the Skylake K-SKU CPUs (i5-6600K, for instance), there is short supply of Skylake CPUs and the locked Skylake CPUs are yet to be released. An unlocked CPU doesn’t work well in this budget, so we’ve instead chosen the 3.2GHz i5-4460. It may not have an unlocked multiplier (used for overclocking) or the highest base clock around, but it’ll perform admirably for gaming. In certain cases it can turbo up to 3.4GHz to provide a bit extra speed.
With it being an i5, the CPU has four cores and four threads (one per core). It comes with a stock heatsink that will adequately cool it, but for users wanting cooler and/or quieter operation, the Cryorig H7 is a good addition for $35. This will also improve longevity.

Motherboard 

ASRock H97M Pro4 ($68)(Kshs.7000): The CPU needs a place to sit – life is demanding for a silicon die. The Asrock H97M Pro4 supports up to a Haswell i7, 4xDDR3 slots, 1xPCIe 3.0 x16 slot, 1xPCIe 2.0 x16 slot, 2xPCI slots, 6xSATA 6Gb/s slots, 2xUSB3.0 slots, and 4xUSB 2.0 slots.
It should be noted that the lower x16 PCIe slot is actually wired for x4. The board also has 4xUSB 3.0, 4xUSB 2.0, 1xPS/2, 1xDVI-D out, and 1xOptical on the I/O. The Pro4 is a good budget motherboard that will work fine for this micro-ATX build.

Memory

G.Skill RipJaws X Series 8GB DDR3 ($47)(Kshs.4700): This build features G.Skill RipJaws Series 1x8GB DDR3 1600 RAM – quite a mouthful. This RAM is fairly standard in that it is DDR3-1600, has a CAS latency of 9, and runs at 1.5v. 8GB is enough for gaming currently, and having one stick allows for easy upgrading to larger RAM capacities if needed down the road. Currently, the G.Skill RipJaws Series 8GB RAM is available for $40 at Newegg. It also has a lifetime warranty.
Using a single stick of memory means that dual-channel operation is lost, but this is something we’ve found to have minimal impact on dGPU gaming performance.

Power Supply

SeaSonic S12II 520W ($55)(Kshs.5700): Next up is the power supply, for which we are going with the SeaSonic S12II 520W. 520 Watts is definitely enough for this build. This allows the PSU to be used in future builds if needed/possible. It has 1x6 and 1x6+2 pin connectors, which fits the EVGA GTX 970’s requirements. The SeaSonic S12II is 80 Plus Bronze Certified, so it is reasonably efficient. To top all this off, SeaSonic has a 5 year warranty for this PSU.

HDD & Optional SSD

Western Digital 1TB Blue ($52)(Kshs.5400): For the regular user, this hard drive will provide adequate speed and capacity to start with. It’s not winning any speed awards, but the price per GB is still advantaged over SSDs, although read, write speeds, and seek times are worse. The WD Blue is a standard 7200 RPM HDD.
Samsung 850 Evo 250GB ($90)(Kshs.9300): For users wanting a faster drive for the operating system, games, and commonly used programs, the Samsung 850 Evo 250GB SSD is a good addition to this build. SSDs provide substantially better read, write, and seek speeds compared to SSDs, and also uses less power. This SSD comes with a five-year warranty backing it, granting it a longer warranty compared to much of its competition.

Case

Cooler Master N200 mATX ($45)(Kshs.4600) This build will use the Cooler Master N200 to house the components. It is an mATX case, making it a smaller and more compact case than many more  common ATX cases. The N200 supports graphics cards up to 355mm, so even long GPUs can fit with ease. The N200 has reasonably good watercooling support, which is nice addition to a case at this price point. It supports 3x3.5” bays, and 4x2.5” bays. Of course, as with any modern case, the N200 also has front USB 3.0 ports.
While the stock Intel cooler is pretty short and isn’t near this tall, the N200 supports CPU coolers up to 160mm. The Cooler Master N200 comes with 2xCooler Master XtraFlo fans.
While Skylake was recently released, the unlocked CPUs -- which are the only ones currently out -- aren’t a good fit for lower-budgeted builds. For this most recent ~$770 gaming PC build, we’ve gone for a locked Haswell i5 and GTX 970 combo to achieve satisfactory performance at 1080p and (to a slightly lesser extent) 1440p.

Google data center loses data following four lightning strikes

Google must have done something to anger the gods, for they have blasted one of Google’s European data centers with lightning not once, not twice, not thrice, but four times. The incident was serious enough that the data center actually lost some data, which is exactly what data centers are supposed to avoid.
This incident occurred at Google’s europe-west1-b data center in Belgium on August 17th. This data center houses a variety of content, but the affected disks were handling Google Compute Engine (GCE) instances. The GCE service allows businesses to store data and run virtual computers in the cloud. After the four lightning strikes, some of these drives started returning I/O errors in their GCE instances. At the height of the calamity, about 5% of the disks in the data center were experiencing I/O errors. Google was able to restore many of the drives to working condition and salvage the data, but 0.000001% of the data in europe-west1-b was irrecoverably lost.
Big data centers have systems in place to prevent data loss in the event of electrical interference, and Google is obviously no exception. However, four successive lightning strikes on the electrical systems of its data center pushed the buffering and backups to their limits. The servers have battery backups, and the building itself has a full auxiliary power system. Google says these both flipped on as expected to prevent damage to the disks. However, some recently written data was stored on systems that were more susceptible to power failure or repeated battery drain. This would be the 5% of originally affected storage.

 

Google says it is already in the process of transitioning all its storage hardware away from the configuration that made this failure possible, and most of it is already running on the new system. That’s why only a small fraction of GCE instances were affected. So I guess the good news is that even four lightning strikes on the data center’s power system wasn’t enough to affect most of the disks google is running. If this had happened a few months down the line, there might not have been any negative impact.
That might not be comforting to the handful of customers who permanently lost data in their GCE instances. While Google accepts full responsibility for the failure, it also points out that GCE is by its nature tied to a single data center. Customers who are particularly worried might want to use GCE snapshots and Google Cloud Storage for geographically independent systems.

Intel Core i5 vs. Core i7: Which processor should you buy?

Whether you’re building your next PC or shopping for a new computer online, one of the questions that comes up from time to time is whether the Intel Core i5 or Core i7 is a better bargain. The short answer, “It depends,” isn’t all that helpful, so we’ve broken the data out in more detail and for both mobile and desktop processors. Here’s what you need to know.

Mobile Core i5 vs. Mobile Core i7

Because Intel has launched a reasonable set of SKUs for its 5th generation (Broadwell) hardware, we’ll use those chips for comparing the Core i7 vs. Core i5 stack. All of Intel’s mobile Core i5 processors are dual-core chips with Hyper-Threading, going right back to Sandy Bridge. What distinguishes modern 5th generation Core i5s is their TDPs — Intel offers Broadwell in 15W, 28W, and a single 47W SKU. That chip, the Core i5-5350H is also capable of addressing 32GB of RAM and offers Intel’s Iris Pro 6200 solution based on Crystal Well.

Intel’s Core i5 mobile lineup (Broadwell)

Here are the features that separate mobile Core i5 and Core i7 processors:
More cores: Many of Intel’s Core i7 processors are quad-core chips with Hyper-Threading enabled. This isn’t universal, however, and the company does offer five dual-core + Hyper-Threading SKUs.

Higher clocks: Intel’s dual-core mobile Core i7 chips typically have higher clock speeds than their Core i5 counterparts, even at the same TDP. The fastest dual-core Core i7 at 15W TDP is the Core i7-5600U, which runs at 2.6GHz base and 3.2GHz standard. The fastest Core i5 at 15W is the Core i5-5300U, which has a base clock of 2.3GHz and a max clock of 2.9GHz. The GPU on the Core i7-5600U is clocked 50MHz faster as well.

More cache: Core i7 chips carry either 6MB or 4MB of cache. Core i5 chips have 3MB, except for the Core i5-5350H, which has 4MB.

More addressable memory: Most of the Core i7 chips are limited to 16GB, like the Core i5, but there are four exceptions. The Core i7-5850HQ, 5950HQ, 5750HQ, and 5700HQ are all capable of addressing up to 32GB of RAM.

One important thing to note, however, is that how much performance you get out of a Core i5 vs. a Core i7 will depend a great deal on your laptop’s cooling solution and whether or not the chip can handle its own heat output. We’ve previously discussed how Intel gave OEMs more freedom to define their own TDP targets and skin temperatures, but this creates scenarios in which buying a faster Core M can actually result in worse performance as the chip hits its thermal trip point and down-throttles to keep cool.

Intel’s highest-end Iris Pro graphics are available in both Core i5 and Core i7 mobile SKUs

We can’t point to specific instances where this has tilted performance between Core i5 and Core i7 chips, but it’s likely to create at least a little “slosh” between the two core families. Generally speaking, if you truly want to emphasize low power, pick the CPU that has lower base and turbo clocks. Intel historically defined TDP as “The CPU’s average power consumption when running typical workloads over a period of time.” A chip with more headroom is a chip that’ll hit its throttle point faster.

Desktop Core i5 vs. Desktop Core i7

Because Intel has only launched a handful of Core i5 and i7 desktop parts, for now we’ll focus on the 4th generation cores (Haswell). One significant difference between the desktop and mobile markets is that all but one of Intel’s 23 Core i5 SKUs offer four cores, but no HyperThreading. Intel’s Core i5-4570T is the single exception to this — it’s a dual-core + HyperThreading chip. Another difference between the mobile lineup vs. the desktop cores is that the desktop versions of both the Core i5 and Core i7 officially top out at 32GB of addressable memory.
In fact, many of the differences we observed between mobile parts don’t apply to desktop hardware. The i7 family still carries more L3 cache (8MB, save for the Core i7-4770R, which has 6MB), as compared to the Core i5 family, which has 6MB of cache on every chip except for the Core i5-4570R, 4570 and 4670R, which have just 4MB. Clock speeds are a mixed bag — while the fastest Haswell Intel manufacturers is the Core i7-4790K at 4GHz base clock and 4.4GHz Turbo, the Core i5-4690K is clocked at 3.5GHz base and 3.9GHz Turbo — essentially equal to the Core i7-4770K.
Both the Core i3 and Core i5 families use Intel HD Graphics 4600 solution with the same 350MHz base clock and 1.2GHz maximum clock. Apart from Hyper-Threading and some small cache differences, the Core i5 and Core i7 are quite similar, with overlapping frequency ranges and feature sets. Support for Intel’s TSX-NI instructions is only available on one Core i5 SKU, but only two Core i7’s offer the capability.
Similarly, Intel only offers vPro on 13 of the 23 Core i5 SKUs, compared to 9 of the 12 Core i7 SKUs — but if you want vPro on a Core i5, you can get it.

Brandywine.Build One Like This!

Case modder and painter Hanoverfist has several amazing builds under his belt—each in an eye-popping shade—but it was the Brandywine build on Overclock.net that caught our attention first. (It also put the Looking Glass song, Brandi, in our heads for about three days straight). The Brandywine build was focused on painting techniques, specifically spraying the House of Kolor Kandy paint; no metal fabrication or fancy hardware for this beauty. Hanoverfist told us that it took a lot of patience to get the paint to turn out as impeccably as it did. In order to achieve the look, several layers of paint need to be applied after the metal was painstakingly prepared—probably the most time consuming part of the build.
In addition to a killer paint job, the Brandywine features an AMD CPU + GPU, a Gigbyte GA-F2A88XM-D3H motherboard, and tons of storage (Western Digital WD Black 2TB, Western Digital 4TB W Red SATA III, a 2TB Western Digital Red NAS Hard Drive, and Samsung 840 EVO 120GB SSD in RAID0).
Hanoverfist says the entire project began out of a need for having a case to house storage drives. “The Fractal Arc Mini R2 case had a good look about it and was perfect for housing drives, but the reason for modding it was to create a work of art. Not a power build. I wanted something that sits in my living room and just looks amazing.”
For modders considering a similarly ambitious paint job, Hanoverfist recommends taking your time and not rushing anything. Congrats on that work of art Hanoverfist, and thanks for sharing the build!

Brandywine parts list

CPU: AMD A8-5600K Trinity Quad-Core 3.6GHz Turbo Socket FM2 100W Desktop APU (CPU + GPU)
Motherboard: Gigabyte GA-F2A88XM-D3H FM2+/ FM2 AMD A88X
Memory: G. Skill Ripjaws X Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 2133
Power Supply: Silverstone Strider series ST60F-PS 600W ATX12V/ EPS12V SLI Ready CrossFire Ready 80 Plus Silver Certified Full
Storage: 2 x Western Digital WD Black 2TB
Storage: 2 x Samsung 840 EVO 120GB (SSD) RAID0
Storage: 2 x Western Digital Red NAS Hard Drive WD20EFRX 2TB
Storage: 2 x Western Digital 4TB W Red SATA III
Case: Fractal Design Arc-Mini-R2-BL-W Black Steel MicroATX Mini Tower Computer Case
Cooling: Cooler Master Hyper 212 EVO
Other: Lamptron CW611 and CM615 controllers

Contact +Kevin Kimz and I'll build one for you

Asus Lays Claim to First USB 3.1 Gen 2 Certified Motherboard

Asus is in full brag mode over the claim that its TUF Sabertooth Z97 Mark 2/USB 3.1 is the world's first motherboard to achieve SuperSpeed USB 10Gbps certification for USB 3.1 Gen 2 data transfer speeds.
According to Asus, its Z97 Sabertooth board had to pass a series of tests set forth by the USB Implementers Forum (USB-IF). It's not just a raw speed thing -- the certification process also includes signal quality checks, stress tests, and sustained transmission capabilities.
"USB-IF certification is a proven testing and evaluation process that is absolutely necessary for product interoperability and reliability," said Jeff Ravencraft, USB-IF President and COO. "As the next generation of USB technology hits the market, we strongly encourage consumers to look for USB-IF-certified products, such as the ASUS TUF Sabertooth Z97 Mark 2/USB 3.1 motherboard, to ensure the best possible user experience."
It's a Generational thing
When shopping a motherboard or system with USB 3.1 listed on the spec sheet, be sure that it's a Gen 2 spec if you want the fastest speed available. It's an important distinction, and somewhat confusing, as USB 3.0 and USB 3.1 Gen 1 are basically the same thing. Or in USB-IF's words (PDF), "USB 3.1 Gen 1 and USB 3.0 terms are synonymous" -- both top out at 5Gbps. That's still a huge gain over USB 2.0 (480Mbps), but only half of USB 3.1 Gen 2 (10Gbps), which is what many people think of when they see USB 3.1.

• USB 2.0: 480Mbps (60MB/s)
• USB 3.0: 5Gbps (625MB/s)
• USB 3.1 Gen 1: 5Gbps (625MB/s)
• USB 3.1 Gen 2: 10Gbps (1,250MB/s)

A real world example of this is Apple's Retina Macbook, the one with a single USB 3.1 Type-C connector and no other ports. Even though it sports a fancy Type-C connector and is labeled as USB 3.1, it's a Gen 1 spec, so really it's USB 3.0 in a pretty dress or spiffy suit.

 

Asus TUF Sabertooth Z97 Mark 2/USB 3.1
USB shenanigans aside, the Sabertooth Z97 is a "military grade" motherboard for socket LGA 1150 processors. It has four DIMM slots supporting up to 32GB of DDR3-1866 memory, two PCI-E 3.1 x16 slots, a single PCI-E 2.0 x16 slot, three PCI-E x1 slots, half a dozen SATA 6Gbps ports, two rear USB 3.1 Gen 2 ports, two rear USB 3.0 ports, four rear USB 2.0 ports, and a host of other features.
The board is available now for $175(Kshs.19,000)

Samsung mass-produces 256Gb 3D vertical NAND chips

Samsung Electronics has begun mass production of 256-gigabit 3D vertical NAND (V-NAND) flash memories, clearing the way for higher-capacity memory that doesn't take up more room.

The memory is built on 48 layers of three-bit multi-level-cell (MLC) arrays and offers 32 gigabytes of data storage on a single chip. It doubles the density of conventional 128Gb NAND flash chips and consumes at least 30% less power than the previous Samsung chip, the company said Tuesday.
Smaller than a fingertip, Samsung's third generation V-NAND flash memory follows the second-generation chip, which featured three bits per cell and 32 vertically stacked cell layers.
In contrast to planar NAND, Samsung's V-NAND incorporates a 3D Charge Trap Flash (CTF) structure of 48 stacked array layers, like a miniature skyscraper. The arrays are connected electrically through 1.8 billion holes punched through them. Each chip has more than 85.3 billion cells, each storing three bits of data.

Samsung said it plans to make the chips for enterprise and data storage users as well as to promote the adoption of terabyte-grade SSDs (solid state drives). In June, Samsung announced a 2TB SSD in a 2.5-inch form factor for laptops and desktops.
The 850 Pro and 850 EVO SSDs included 3D V-NAND technology and doubled the capacity of their predecessors, with 120GB, 250GB, 500GB, 1TB and 2TB models

Origin Millennium (2014) PC.Build One Like This!

 

Millennium desktop, which liquid-cools the processor and overclocks it to 4.5 GHz. That’s 1.5 GHz higher than the stock model’s base clock.

Origin didn’t stop there. They’ve paired the processor with three Nvidia GeForce GTX 980 video cards, 16GB of DDR4 RAM, and an ASUS X99-Deluxe motherboard.A large side panel window, and a complex, remote-controlled LED lighting system puts the tower’s enthusiast chops on full display. It boasts a stealthy, unassuming look. Turn the LEDs on, and it’s Superman. Turn them off, and it’s Clark Kent.

Metal and plastic are used in the Millennium’s chassis. The former feels premium and sturdy, but the latter is a let down. A few build quality issues can be found along the top panel, where plastic is used to form a series of scoops and dips that enhance the rig’s visual appeal.

They look like handles, but they’re actually flimsy. The system even ships with big red labels warning users that these are “NOT A LIFT POINT.” Design students should take note. If it’s not a handle, don’t make it look like one.

The top of the system also offers excellent connectivity. Here, we found four USB 3.0 ports, and individual headphone/microphone jacks. Fan control knobs, and the usual power and reset buttons are present too. Many competitors have just two forward-facing USB ports that may or may not support USB 3.0 speeds.

Around back, there’s an incredible array of ports, including 10 additional USB 3.0 connectors, two USB 2.0, two Ethernet jacks, 5.1 audio, and antenna connections for the motherboard’s built-in Wi-Fi, and Bluetooth. The three GTX 980s each provide three DisplayPort, one HDMI 2.0, and one DVI.

Note, though, that not every video output can be used simultaneously. Nvidia’s drivers currently support up to four monitors at 4K. That cap does not increase when extra cards are installed.

VADirect X99 Gaming System.Build One Like This!

custom case doesn’t come free, though, which is why many gaming PCs have an eye-watering price. AVADirect takes a different approach, building systems to order from off-the-shelf parts. They’re like a friend who builds computers charging you to put one together, and delivering it to you when they’re done. An incredibly talented friend, with access to the sweetest components.
In theory, this approach offers more bang for your buck, and the X99 Gaming System we received was configured to prove that point. It arrived with an octa-core Intel Core i7-5960X, which has a base clock of 3.0GHz, and 16GB of Corsair DDR4 RAM. It also had a pair of EVGA Superclocked GTX 980s, which is what you’ll need if you want to game 4K. For storage, AVADirect included a one terabyte Samsung 850 EVO SSD, and a two terabyte Western Digital Black drive.
That’s pretty close to the Origin Millennium from late 2014, and the Falcon Northwest Mach V, though those machines boasted a third video card. The AVADirect system is cheaper than both of them, too, but whether it can hold its own in a world of beautiful custom-built PCs remains to be seen.

Processor Model LGA 2011-v3 Core i7-5960X
Manufacturer: Intel
Speed: 3 GHz
Cores: 8 Cores
64-bit: Yes
16GB of Corsair DDR4 RAM
Graphics Model: GeForce GTX 980 SuperClocked ACX 2.0
Manufacturer: EVGA
VRAM: 8 GB
Storage type: SSD
Size: 2TB
Speed: 520 Mb/s
Power Supply: Corsair AX1200i PSU


The Corsair 760T case sits over 22 inches tall, and weighs just under 25 pounds before there are any parts in it. It’s much larger than your standard desktop tower, though it’s actually smaller and lighter than competing full-tower PCs from Origin and Falcon Northwest.

Top-shelf performance

At the core of the AVADirect X99 Gaming System is a top-of-the-line Intel Core i7-5960X, an octa-core processor with a 3GHz clock speed, and 3.5GHz Turbo Boost. Backing it up is 16GB of Corsair memory, accompanied by a pair of super-clocked EVGA GTX 980s in an SLI configuration for a total of 8GB of VRAM. For storage, AVADirect included a one terabyte Samsung 850 EVO SSD, and a two terabyte Western Digital Black drive.