On Wednesday at the Intel Developers’ Forum, Intel (NASDAQ: INTC) unveiled its plan to remove wires from the PC workplace.

During a keynote on Wednesday, Intel executives announced that Skylake — the Desktop and notebook-focused successor to Broadwell — would be “fully wireless”. That means that the future, according to Intel, is one where the daily activities for devices of docking, connecting one’s computer to a display, as well as charging will all be done without the use of wires for devices that use the company’s next-generation CPU platform.

This wireless experience will be done via a suite of technologies. For charging, Intel will jump onboard the Rezence standard — an alliance with members from most major manufacturers including Asus, Lenovo and Dell — while for wireless display Intel will use its own standard called “WiDi”.

The slide below explains specifically how Rezence will work on future Intel platforms. The coils on both ends will be tuned to 6.78 MHz, and Intel says that measures will be in place to prevent overheating of the coils or interference from other metallic objects that may be near the two devices. For example, the resonant coupling frequency will be set to 6.78 Mhz which apparently is a safe zone to avoid interference.

For wireless displays, Intel is pushing a self-developed standard called WiDi Pro. WiDi isn’t an inherently new standard, but Intel is looking for a mea culpa and redo for this new effort, that’s now GPU accelerated, starting next year with the fifth generation of Intel Core chips.  Intel is targeting this platform for enterprise environments, removing the hassle of plugging in displays, projectors and other such devices.

Lastly, Intel’s solution for wireless docking is called WiGig. WiGig is Intel’s solution for a wireless dongle free world where devices such as mice or keyboards connect to the user’s computer via a 60 GHz connection. Intel says the range for WiGig connected devices will be between 2 and 3 feet.

While these are all promising technologies, the real question is if this is the best direction for connectivity.

Wireless soup

Cluttering of the electromagnetic spectrum is a big problem for all users, especially those in an EM-crowded environment that are trying to connect to a wireless network.

In environments with hundreds of users attempting to connect to the internet via WiFi, WiFi quickly becomes overcrowded and unreliable. Technology journalists that attend big conferences such as CES of Computex will gripe non-stop about the unreliability of WiFi since the rush of people attempting to connect to the network overwhelms routers and leaves bandwidth something something to be desired.

An increase in EM traffic from a move away from cables means an increase in interference. In environments where the network is densely packed — such as a tradeshow — the potential for things stopping in their tracks due to EM crowding is immense. Any mission critical applications, such as an executive’s keynote, would always be done via traditional wired connections.

In the end this wireless suite of products may have its applications and usages. But institutional mistrust — particularly because of possibilities of technical failures or security issues with the wireless projectors — will stop these devices from going mainstream.

The post Wireless Connectivity Means We’re In For a Buggy, Laggy And Insecure Future appeared first on VR World.

With the introduction of the next-generation of Intel (NASDAQ: INTC) platforms, DDR 4 is finally getting its time to shine.

At IDF 2014, memory maker G.Skill turned up to show off its high-end DDR 4 RAM that came complete with a clockspeed of 3333 MHz in a 4GB x 8 modules configuration. G.Skill also brought with it to the show 8GB modules running at 3200 MHz in a 32 GB configuration. The RAM was running on a Intel i7-5960X CPU along with Rampage V Extreme X99 and X99-Deluxe motherboards from Asus (TPE: 2357).

The 3333 MHz chips have a 1.35V and latency of 16-16-16-36, which are the same as the 3200 MHz DIMMs.

“It’s very exciting to show the world what we can do with the new DDR4 memory standard for the latest Intel X99 platform. There is no doubt that breaking the 3000MHz memory barrier with ease is the next big thing in desktop performance, effectively doubling bandwidth throughput compared to the previous DDR3 memory standard,” Frank Hung, Product Marketing at G.SKILL, said in a press release.

This RAM comes with a price tag of $700 for the 16GB 3333 MHz kit. The kits are available now for pre-order with availability expected later this year.

The post G. Skill Demos High-End DDR 4 RAM at IDF 2014 appeared first on VR World.

For all Intel (NASDAQ: INTC) wants to hype new unproven markets such as wearables and mobile, desktop along with servers, are still the company’s cash cow.

Intel didn’t share much in the line of desktops at IDF, as the company already made its big desktop related announcements earlier this year at the Game Developer Conference in San Fransisco as well as in June at Computex. The company did hint at Skylake, the successor to Broadwell, but didn’t reveal any important technical information (Skylake, however, was old news as its existence had been long leaked already).

But as desktop is one of the two main pillars of the company, getting a gauge on the desktop “state of the union” is important. Bright Side of News*’ Sam Reynolds sat down with Lisa Graff, Intel’s GM and VP of the Desktop Client Platform division, to see where the company has gone and where it is going in the desktop space.

Bright Side of News*: Intel is clearly pushing hard to expand its presence in Shenzhen, as was clear at IDF Shenzhen earlier this year. But so far this has been mostly for Intel’s mobile efforts. What about the Desktop Group?

Lisa Graff: Shenzhen has come from doing lots of different devices — CD players in the past, tablets, etc. So there’s a number of things going on there with desktops, primarily with all-in-ones. You’ve got a whole lot of the larger ODMs as well as OEMs there [doing all-in-ones]. Primarily, that’s where their interest lies.

Overall the PC group is there, enabling them, and helping them build what they want to build.

BSN*: Right now Intel has a commanding lead in the enthusiast CPU space. Technically speaking, why, in your opinion, is Intel leading by such a large margin against the competition?

LG: We’ve invested in it. We’ve invested in multiple aspects of our processors to make them very, very high performance. We put a lot of focus in the design of the processor itself to get each core high performance. Some software scales well with multiple cores and threading, while some software does not.

So, you need a core to be high-performance even if that software is single-threaded. The per-core performance needs to be very good. We can always scale and add more cores to the die. Those are the two aspects from the design standpoint. We’re always looking at how we could increase that performance. There’s a lot of different ways you could do that, and many of them are hard. Our per-core performance compared to the competition is very good; four-cores does not equal four-cores as I’m sure you know.

Those are the design aspects, but our other advantage is in process technology. Intel, as you know, is years ahead in process technology and we’ve put a huge amount of investment building our own fabs with a tight coupling between designing the processor and designing the manufacturing process. This is all done years ahead of when the product comes out so you can fine-tune everything to get the most performance out of the chip. All of that pays off in the end if you can have faster transistors that are less expensive, which you can put more of them on a die.

We know we have a customer base that will pay for performance so we’ve invested in it.

BSN*: Let’s shift to Broadwell on the desktop. Can you clarify the situation regarding PCI-Express? Some say that it may be intentionally slowed to give Intel’s Iris graphics a leg up.

LG: We’re only saying one thing about Broadwell on desktop: which is we are going to bring Iris graphics. We’ll come out with more details later, but we have confirmed that we will bring Iris Pro to desktop — meaning LGA. We think that’s going to enable us to have nice little mini-desktops — you can save the power, heat, thermals and cost of a discrete graphics card in these smaller form factors.

BSN*: As a follow-up, how do these small form factors play into Intel’s  broader vision for the desktop?

LG: Look, regardless of these small form factors, a lot of people still buy regular towers. So I don’t think there will be one size fits all, where this new “nirvanna” thing replaces everything. People have different needs.

For minis I think it’s [a few] things: one is there is absolutely a desktop replacement where the usage is such that the space saving is one of the big things we’re going after. This might be people who are already using small form factor. Second, there are new usages for minis. That’s the easy-embedded, standards-embedded use case. Point of sale is an example.

BSN*: Thanks for your time.

The post Lisa Graff on 'The State of The Desktop Union' appeared first on VR World.

Intel (NASDAQ: INTC) is betting big on Broadwell-based processors that will be used in low-power devices like tablets, and more mainstream devices like notebooks. At the IDF, Intel showed off a 12.5-inch reference design tablet running a Core M 5Y70, which features a 2 GHz dual-core 64-bit CPU, 4 MB L3 cache, dual-channel LPDDR3 memory controller and a GPU with 24 execution units and 192 stream engines.

The tablet was put through Cinebench’s R11.5 multi-threaded benchmark and managed a score of 2.7, which is around 70% of that of a Core i3 4330, and significantly more than anything in the Atom line, or AMD’s Beema and Mullins series. In the OpenGL test, the Core M scored 16.96 FPS.

In Futuremark’s 3DMark Ice Storm benchmark, the reference tablet scored 46,097 which is twice as fast as what the Qualcomm Snapdragon 800 managed. Similarly, in the SunSpider JavaScript benchmark, the tablet’s score of 142.8 is far better than anything in this segment.

The benchmark scores are not the only metric when judging a CPU, but it is encouraging to see Broadwell doing so well. With a TDP of 4.5W, Intel is claiming a battery life of 9 hours, which is also a positive change from previous generations. One of the main reasons for Intel’s failure to lock in hardware wins has been due to overtly high energy consumption with the Atom series, and with Broadwell, it looks like Intel managed to fix those issues.

It should be interesting to see the price points at which Broadwell-based will launch, as well as the real world benchmarking figures they can achieve. October cannot come soon enough.

The post Broadwell Found To Be More Than Twice As Fast As Snapdragon 800 appeared first on VR World.

After the close of the first day of Intel’s Developer Forum in San Francisco, long term shareholders of Intel (NASDAQ:INTC) may be left with mixed feelings.

After all, the company put equal, but not equally prominent time, into its traditional high-margin market — servers — and also to the unproven, currently unprofitable, market of wearables and mobile. Granted, the much awaited and much needed refresh of Intel’s Xeon CPU was Intel’s first announcement of the show — but that occurred at a day zero event before IDF kicked off which lacked the prominence or importance of the forum’s opening keynote.

The reason for this is simple: Xeons don’t attract the attention from the usual mainstream consumer press that fashion-buzzword topics such as wearables and mobile do. Even though Xeons are essentially the brains that power the hundreds of millions of web-connected smart devices and their respective app stores, they don’t get the attention they deserve because the public simply is not interested in them. Instead, a market where Intel has yet to make a dollar — in fact is spending billions just to acquire its first customers — steals the show.

While it may be that the unproven markets of wearables, mobile, and the amalgam,  Internet of Things, one day proves to be a profitable division that day is not yet here. Intel does have some promising advancements with x86 in low-power (traditionally ARM dominated) fields, but these advancements really won’t be shown until Broadwell makes its way to these devices after the Atom line. Which, of course, relies on customers signing up to make devices that incorporates these chips. And that, of course, relies on Intel’s chips being the more competitive option than ARM.

All this is certainly possible. It could be that the next-generation of x86 — in the form of Broadwell and later Skylake — powers award winning devices of all form factors. There always could be high-end or extreme versions of ultra-mobile devices running on Core-class processors. But it’s also possible that this does not happen leaving Intel with an investment that would have been better spent elsewhere.

The post Intel’s Seeming New Focus: Low Margins But High Growth appeared first on VR World.

On Monday VR World reported on some irregularities with the “extra pin” found on Asus’ (TPE: 2357) X99 motherboards.

Asus claims that the additional undocumented pins found on Haswell-E and the sockets of Asus’ compatible boards are used to give users extra stability during overclocking. As documented by VR World’s Nebojsa Novakovic, in its own documentation Asus claims that these pins, better known as “socket 2084,” have numerous advantages when used including better monitoring of voltage, more strapping ability, higher frequency, and maximum Vcore.

However, sources that have spoken to VR World say these pins aren’t as special as Asus says they are and are merely used for CPU debugging  — and present a crash risk or even damage risk if used for other purposes. Further, sources say that Asus is attempting to patent the particular usage of the pins to keep them away from Foxconn (TPE: 2354).

But Intel (NASDAQ: INTC), for its part, has come out fairly neutral on the issue.

Lisa Graff, Intel’s boss of its desktop group, said in an interview with VR World that Intel was investigating the issue and won’t have an answer for some time.

“We understand that Asus has done some different things — and we understand that some of our customers will do different things with their boards — and beyond that at this point there’s nothing much we can say except talk to Asus about it,” she said.

Considering the number of reports of burned or otherwise dead Haswell-E chips that have appeared in the press shortly after the launch of X99, one might conclude that this was the cause. Graff doesn’t think this is the case and says that Intel won’t have an answer until it completes its own investigation.

“We have heard there have been some issues with Haswell-E based systems, but we do not know what the cause is at this point. We investigate all these kind of things. We don’t know, so we’re going to investigate,” she said.

“There’s no way to make a call on what exactly it is until we do an engineering analysis on it.”

The post Intel: No Answer Yet on Asus’ Extra Pin Situation appeared first on VR World.

Intel’s (NASDAQ: INTC) second developer forum of the year — the first taking place during April in Shenzhen — officially kicked off in San Francisco Tuesday with a mobile theme as the event’s keynote was anchored around Intel CEO Brian Krzanich and other Intel executives taking the stage to announce the availability of Intel’s XMM 7260 LTE baseband in its first hardware win, as well as a series of developer tools for Android developers, as well as other mobile themed announcements.

On stage Intel’s Kirk Skaugen, Vice President of the PC Client Group,  said that the first hardware win for Intel’s XMM 7260 baseband will be in Samsung’s Galaxy Alpha smartphone, which is now shipping in Europe, Asia (excluding China/Korea/Japan), Australia and Latin America. The XMM 7260 and the 7262 modems are Intel’s first LTE Category 6 compatible basebands.

Sticking with the mobile theme, Intel executives also said on stage that Intel would be providing Android developers with an Intel reference design program to assist developers in developing hardware and apps for Intel-powered Android platforms. Intel’s Doug Fisher, the company’s general manager of software, said that Intel is committed to providing updates to Android two weeks after Google pushes out updates to AOSP.

Another tablet win

Later during the keynote Krzanich was joined on stage by Michael Dell to show off the Dell Venue 8 7000 tablet which contains Intel’s new RealSense snapshot technology. This creates a depthmap and allows users to fine tune focus adjustments to create a more accurate and true-to-life picture. Exact specs, pricing on the tablet have yet to be released but Dell did mention on stage that it would have a 2K resolution (2560 × 1440 pixels) and an 8.4-inch OLED screen.

A rendering of the Dell 8 7000 series.

At the Shenzhen IDF conference earlier this year, Intel’s Krzanich said that Intel was targeting to have 40 million Intel-powered tablets shipped by year’s end. Krzanich did not provide much in the line of in-depth information detailing exactly how many tablets had been shipped to date, or how close the company is to meeting its goal,

Always a place for the PC

While the bulk of the keynote at IDF was about mobile, and emerging markets from mobile such as wearables, Intel’s Skaugen also gave attendees the first official preview of Skylake — Intel’s next-generation architecture that’s due out next year. While there were no technical specifications disclosed, Skaugen did demonstrate a reference design of a 2-in-1 notebook powered by Skylake playing 4K video.

In contrast to Broadwell — which is largely targeted at low-power mobile platforms — Skylake will broadly target Intel’s customers and will find a place in a variety of form factors. Skylake will be built on Intel’s 14nm process node, and Skaugen said that development of the platform is progressing and its “health is great.”

Edison finally available

Intel kicked off the year at CES by announcing Edison, a tiny computer the size of an SD card that comes packed with WiFi, Bluetooth, and an Intel SoC.

Targeted at a variety of use cases from wearables to makers, Intel hopes that the system would revolutionize hyper small form factor computing. At IDF, Krzanich said that Edison is now available with pricing starting at $50.

Check back later for more from San Francisco as IDF continues. 

The post Mobile Takes Center Stage at IDF 2014 appeared first on VR World.

IDF is busy, even before it starts – so it was this time in San Francisco with Gigabyte Overclocking competition just a day before the keynote.

The OC results from Cookie, Charles Wirth and others were good, with 5.8 to 6 GHz achieved LN2 cooling results on the Core i7-5960X on Gigabyte boards seen here. The RAM on trial also performed well, hovering above 3 GHz for the G.Skill and Kingston part, with Crucial reference DIMMs just below that.

However, something far more interesting was found on the overclocking floor (together with our friend Koen from Hardware.info). Remember Asus’ claims about additional pins on its LGA 2011-3 socket, aka “socket 2084,” which Asus claims brings extra performance and OC reliability by using undocumented pin holes on the Intel’s new CPUs? You can see Asus claims right here:

According to industry sources Asus was considering patenting the socket and preventing Foxconn, its manufacturer, from selling it to other vendors. On the other hand, another major industry source claimed that these pins are not what Asus claims, but just CPU debug and test pins brought back to the socket in this generation. Therefore, not only would they be useless for overclocking, but also a potential crash risk if connected in a production system.

Then, remember another point – the Core i7-5960X is just a reduced version of the same die used for the Xeon E5 v3 8-core Haswell-EP, with some features like dual QPI channels, memory ECC and few others disabled, including some approximately 150 signal pins related to that. So, the Enterprise Group might know even more than the Client group about the undocumented pins on these new processors?

Note: The Foxconn model marking for the normal socket 47191 and for the OC socket the model is 46391.

Next step – see the Gigabyte brand new LN2 cooling optimised mainboard for these CPUs, with this exact same “special” socket:

And compare with the old socket:

So, Gigabyte can get hold of this same socket, the question is what do those additional pins connect to. If the first possibility stated above is correct, i.e. that these are real additional power, ground etc. pins that allow better and more reliable OC, then Asus has no way to patent these as it directly links to Intel IP including the socket and its validation.

If, however, it is the second possibility, of test, debug or no-connect pins being exposed, then it is a serious marketing rubbish which could be used to deceive the high-end buyers, whom both Intel and its key OEMs, which Asus and Gigabyte are, surely cherish and treasure. And mind you, this is already in a product on the market, the Asus top end Rampage V Extreme.

Since Asus, Gigabyte or even Foxconn are not likely to be able to respond to this, it seems that Intel will be the one to resolve this mystery, especially as more of its top end CPUs get the performance enhancement and managed unlocking capabilities over time.

The big question is: what are the warranty implications if things fail because of CPUs being inserted into sockets with undocumented pins.

This post originally appeared on VR World, Bright Side of News*’ Asia Pacific sister site.

The post Haswell-E Controversy: What Should Intel Do About Asus And Socket 2084? appeared first on VR World.

IDF is busy, even before it starts – so it was this time in San Francisco with Gigabyte (TPE: 2376)  Overclocking competition just a day before the keynote.

The OC results from Cookie, Charles Wirth and others were good, with 5.8 to 6 GHz achieved LN2 cooling results on the Core i7-5960X on Gigabyte boards seen here. The RAM on trial also performed well, hovering above 3 GHz for the G.Skill and Kingston part, with Crucial reference DIMMs just below that.

However, something far more interesting was found on the overclocking floor (together with our friend Koen from Hardware.info). Remember Asus’ (TPE: 2357) claims about additional pins on its LGA 2011-3 socket, aka “socket 2084,” which Asus claims brings extra performance and OC reliability by using undocumented pin holes on the Intel’s (NASDAQ: INTC) new CPUs? You can see Asus claims right here:

According to industry sources Asus was considering patenting the socket and preventing Foxconn (TPE: 2354), its manufacturer, from selling it to other vendors. On the other hand, another major industry source claimed that these pins are not what Asus claims, but just CPU debug and test pins brought back to the socket in this generation. Therefore, not only would they be useless for overclocking, but also a potential crash risk if connected in a production system.

Then, remember another point – the Core i7-5960X is just a reduced version of the same die used for the Xeon E5 v3 8-core Haswell-EP, with some features like dual QPI channels, memory ECC and few others disabled, including some approximately 150 signal pins related to that. The Enterprise Group might know even more than the Client Group about the undocumented pins on these new processors?

Note: The Foxconn model marking for the normal socket 47191 and for the OC socket the model is 46391.

Next step – see the Gigabyte brand new LN2 cooling optimised mainboard for these CPUs, with this exact same “special” socket:

And compare with the old socket:

So, Gigabyte can get hold of this same socket, the question is what do those additional pins connect to. If the first possibility stated above is correct, i.e. that these are real additional power, ground etc. pins that allow better and more reliable OC, then Asus has no way to patent these as it directly links to Intel IP including the socket and its validation.

If, however, it is the second possibility, of test, debug or no-connect pins being exposed, then it is a serious marketing rubbish which could be used to deceive the high-end buyers, whom both Intel and its key OEMs, which Asus and Gigabyte  are, surely cherish and treasure. And mind you, this is already in a product on the market, the Asus top end Rampage V Extreme.

Since Asus, Gigabyte or even Foxconn are not likely to be able to respond to this, it seems that Intel will be the one to resolve this mystery, especially as more of its top end CPUs get the performance enhancement and managed unlocking capabilities over time.

The big question is: what are the warranty implications if things fail because of CPUs being inserted into sockets with undocumented pins.

The post Haswell-E Controversy: What Should Intel Do About Asus And Socket 2084? appeared first on VR World.

Today, as Intel (NASDAQ: INTC) launches the third generation of its Xeon E5 dual-CPU platform, many eyes are on the improvements it brings to the servers in the datacenter. However, the benefits are just as high – if not higher – on the high-end workstation front.

First of all, Haswell core means sped-up AVX floating point, by inclusion of fused multiply-add (FMA) ops for theoretical FP rate doubling in benchmarks like Linpack, for instance. Haswell’s AVX2 also, just as importantly, moves integer processing to the wide parallel AVX engines, essentially offloading anything aside the address calculations to the RISC-like, three-address AVX instruction format and wide register sets. For workstation apps, once re-compiled to take advantage of it, the benefits could be enormous, and also be another gradual move away from the antiquated X86 code base.

Then, the wide choice of a number of cores per SKU – from 8 all the way to 18 – enables you to pick the right balance of per-core speed (i.e. per-thread performance) and core number, depending on the parallelism of your application. Some apps scale less well across many cores, thus preferring high per-core speed, while others like ray tracing make the most out of many cores.

The initial workstation SKU in the Xeon-E5 v3 range, the E5 2687W v3 flavor, is a 3.1 GHz 10-core part that actually uses the 12-core die where 2 cores (and their associated caches) were turned off. Now, its predecessor, the 2687Wv2 on the Ivy Bridge platform, had full L3 caches even if some cores of the die were disabled, a benefit that, I guess, we will only see back in Broadwell-EP (E5 v4) SKUs next year.

Then we come to DDR4 – yes the initial DIMMs aren’t exactly speedy, especially latency-wise, but the lower voltage and other reliability features of DDR4, together with quick improvements in speed and latency expected over the next few quarters, should provide the users the never-before seen capacity on a dual-socket workstation, beyond 1.5 TB RAM, without sacrificing the bandwidth on high load situations like DDR3.

The improvements in the PCIe bandwidth, integrated voltage regulation, and sped-up QPI to 9.6 GT/s also round out the key extra benefits.

Putting it through its paces

Here we look at the initial reference workstation based on this SKU from Intel, packaged by BOXX. The machine itself is compact, using liquid cooling on a SuperMicro X10DAi workstation mainboard with three PCIe x16 v3 slots. This doesn’t max out the platforms theoretical quad-GPU full bandwidth capability, but should be enough for most users. In return, the board has space for 16 DDR4 DIMMs, i.e. a full terabyte of RAM if using 64 GB modules available early next year. The installed RAM was 128 GB, in 8 pcs of Samsung 16 GB ECC DDR4-2133 RDIMMs.

The system came with a Nvidia Quadro K2000, which I changed to AMD FirePro W9100, arguably the most powerful professional OpenGL card available as of today. With 16 GB VRAM and six DisplayPort outputs, the card is able to drive even 8K displays like the one from BOE Technology that we mentioned last week. Intel’s 240 GB + 400 GB (SATA + PCIe) SSD combo completed the picture.

The first benchmark was the brand new SPECwpc all-encompassing workstation productivity benchmark by SPEC, on this system. The suite, which takes a couple hours to run, covers everything from processor to graphics (a.k.a ViewPerf) to overall system performance, and seems to do the job with much less trouble than, for instance, BAPcO SysMark did years ago on the PCs.

Here are the first SPECwpc results, on the dual 3.1 GHz E5-2687W v3 system:

Next, we ran CineBench 15 – note that the system is about twice as fast as an overclocked 4+ GHz Core i7-5960X, the desktop Haswell-E brethren to these Xeons.

In CPU-Z, you can see the data about the CPU.

Then we come to the newest version of SiSoft Sandra. Here is the report on the key performance data.

In our next round, we will be focusing on the changes in performance obtained when changing – and tuning – the main memory, as well as looking at the opportunity for even higher CPU speed. In my own opinion, the workstation market can easily justify higher TDP – and maybe even unlocked – Xeons, especially in both 8 core and 18 core per socket configurations.

The post Haswell-EP Workstation Preview: Xeon E5 v3 Rocks, But Still More To Go appeared first on VR World.

Today, as Intel (NASDAQ: INTC) launches the third generation of its Xeon E5 dual-CPU platform, many eyes are on the improvements it brings to the servers in the datacenter. However, the benefits are just as high – if not higher – on the high-end workstation front.

First of all, Haswell core means sped-up AVX floating point, by inclusion of fused multiply-add (FMA) ops for theoretical FP rate doubling in benchmarks like Linpack, for instance. Haswell’s AVX2 also, just as importantly, moves integer processing to the wide parallel AVX engines, essentially offloading anything aside the address calculations to the RISC-like, three-address AVX instruction format and wide register sets. For workstation apps, once re-compiled to take advantage of it, the benefits could be enormous, and also be another gradual move away from the antiquated X86 code base.

Then, the wide choice of a number of cores per SKU – from 8 all the way to 18 – enables you to pick the right balance of per-core speed (i.e. per-thread performance) and core number, depending on the parallelism of your application. Some apps scale less well across many cores, thus preferring high per-core speed, while others like ray tracing make the most out of many cores.

The initial workstation SKU in the Xeon-E5 v3 range, the E5 2687W v3 flavor, is a 3.1 GHz 10-core part that actually uses the 12-core die where 2 cores (and their associated caches) were turned off. Now, its predecessor, the 2687Wv2 on the Ivy Bridge platform, had full L3 caches even if some cores of the die were disabled, a benefit that, I guess, we will only see back in Broadwell-EP (E5 v4) SKUs next year.

Then we come to DDR4 – yes the initial DIMMs aren’t exactly speedy, especially latency-wise, but the lower voltage and other reliability features of DDR4, together with quick improvements in speed and latency expected over the next few quarters, should provide the users the never-before seen capacity on a dual-socket workstation, beyond 1.5 TB RAM, without sacrificing the bandwidth on high load situations like DDR3.

The improvements in the PCIe bandwidth, integrated voltage regulation, and sped-up QPI to 9.6 GT/s also round out the key extra benefits.

Putting it through its paces

Here we look at the initial reference workstation based on this SKU from Intel, packaged by BOXX. The machine itself is compact, using liquid cooling on a SuperMicro X10DAi workstation mainboard with three PCIe x16 v3 slots. This doesn’t max out the platforms theoretical quad-GPU full bandwidth capability, but should be enough for most users. In return, the board has space for 16 DDR4 DIMMs, i.e. a full terabyte of RAM if using 64 GB modules available early next year. The installed RAM was 128 GB, in 8 pcs of Samsung 16 GB ECC DDR4-2133 RDIMMs.

The system came with a Nvidia Quadro K2000, which I changed to AMD FirePro W9100, arguably the most powerful professional OpenGL card available as of today. With 16 GB VRAM and six DisplayPort outputs, the card is able to drive even 8K displays like the one from BOE Technology that we mentioned last week. Intel’s 240 GB + 400 GB (SATA + PCIe) SSD combo completed the picture.

The first benchmark was the brand new SPECwpc all-encompassing workstation productivity benchmark by SPEC, on this system. The suite, which takes a couple hours to run, covers everything from processor to graphics (a.k.a ViewPerf) to overall system performance, and seems to do the job with much less trouble than, for instance, BAPcO SysMark did years ago on the PCs.

Here are the first SPECwpc results, on the dual 3.1 GHz E5-2687W v3 system:

Next, we ran CineBench 15 – note that the system is about twice as fast as an overclocked 4+ GHz Core i7-5960X, the desktop Haswell-E brethren to these Xeons.

In CPU-Z, you can see the data about the CPU.

Then we come to the newest version of SiSoft Sandra. Here is the report on the key performance data.

In our next round, we will be focusing on the changes in performance obtained when changing – and tuning – the main memory, as well as looking at the opportunity for even higher CPU speed. In my own opinion, the workstation market can easily justify higher TDP – and maybe even unlocked – Xeons, especially in both 8 core and 18 core per socket configurations.

This post originally appeared on VR World.

The post Haswell-EP Workstation Preview: Xeon E5 v3 Rocks, But Still More To Go appeared first on VR World.

While flying from Taipei to San Francisco for the usual September round of tech events, one story in this month’s MacWorld caught my attention: one of the editors was speculating how nice it would be to have a true full Mac in an iPhone size. Of course, Apple may be a little farther from merging the OS X and iOS than we thought earlier, but, what about looking at the same thing on the Wintel platform?

It could be argued that one big mistake Microsoft (NASDAQ: MSFT)  did with Windows Phone is to compete with Android and Apple (NASDAQ: APPL). Windows OS, as imperfect as it is, has one big advantage over both Android and iOS: it is a content creation-oriented environment, compared to content-playback environments on the other two. After all, Apple has OS X for content creation, while Google, well, searches for already made content, isn’t it?

Come back to the increasingly popular phablets, seemingly the smallest true computer format one can accept for everyday use. A 6-inch 2015 phablet, with a Bluetooth keyboard embedded inside its protective cover instead of wasting the screen space, can also have a say quad core 14-nm Atom processor, 4 GB RAM and 64 GB flash storage, not to mention Full HD or better screen resolution. Hmm, looks as good as today’s UltraBooks, doesn’t it? So technically it should be able to run a full version of Windows 8.1 or later, including Office and many other apps, without a hitch – something utterly impossible for any ARM-based phone, phablet or tablet, no matter how powerful it is.

Now, if I want to modify my Word document, Excel spreadsheet or PowerPoint presentation on the go, and I don’t want to open my PC for that, trying to do the job on those small Android ‘Office’ apps, including Microsoft’s own one, is either pain in the neck or just not doable. But, with true Office on true Windows, on an Intel (NASDAQ: INTC)-inside phablet the ultra-compact content creation and editing becomes possible.

Now, will Intel and Microsoft jump on this advantage while they have it, before the Android world catches up?

The post IDF 2014 Murmurings: A True Windows Phablet for Content Creationists? appeared first on VR World.