Subscribe to:

Subscribe to :: ::

Are Both AMD And nVidia Readying To Release A 14nm GPU?

August 25, 2015 by Michael  
Filed under Computing

AMD and Nvidia both appear to be certain to get their “14 nm” out next year.

According to TweakTown Nvidia is apparently dotting the “I” and working out where to put in the semi-colons for its Pascal GPU using TSMC’s 16nm FinFet node. AMD rumored has been wining and dining its old chums at GlobalFoundries to use its 14nm process for its Greenland GPU.

Although these sound like different technologies the “14nm and 16nm”  is difference how you measure a transistor. The outcome of both 14 and 16 should be a fairly same sized transistor with similar power features. TSMC calls its process 16nm FinFet, while Samsung and GloFo insist on calling it 14nm FinFet.

The dark satanic rumor mill suggests that the Greenland GPU, which has new Arctic Islands family micro-architecture, will have HBM2 memory. There will be up to 32GB of memory available for enthusiast and professional users. Consumer-oriented cards will have eight to 16GB of HBM2 memory. It will also have a new ISA (instruction set architecture).

It makes sense, AMD moved to HBM with its Fury line this year. Nvidia is expected to follow suit in 2016 with cards offering up to 32GB HBM2 as well.

Both Nvidia and AMD are drawn to FinFET which offers 90 percent more density than 28nm. Both will boost the transistors on offer with their next-generation GPUs, with 17 to 18 billion transistors currently being rumored.



Is nVidia King Of The GPU Arena?

August 24, 2015 by Michael  
Filed under Computing

Nvidia’s desktop GPUs accounted for nearly 80 percent of all sales in the segment in Q2 2015, its highest market share ever.

According to beancounters at the market research firm Mercury Research the GPU market is slowly dying.

The latest quarter was a decrease of 11 percent from Q1 2015 and a year-on-year decline of 21.7 percent so Nvidia is the undisputed king of a much smaller kingdom.

Mercury Research notes that the notebook GPU segment also witnessed a decrease to the tune of 34.1 percent year-on-year, mainly due to the continued improvements in the iGPU segment.

However when comparing both number of GPUs sold to partners and a four-quarter average of sales, Nvidia is the Windows and AMD is the FreeBSD.

AMD is dependent on its latest Radeon 300 series of cards to claw back something but at the moment it is looking like Nvidia is unstoppable.

Nvidia has continued to amass more sales over the course of the last year, and with its Maxwell-generation cards now available across all price tiers, it is unlikely has much to worry about from AMD.


AMD Still Losing Ground

August 21, 2015 by Michael  
Filed under Computing

AMD is continuing to lose market share to Nvidia, despite the fact that its new best video card, the Fury is out.

AMD always had a get out of jail card when the last GPU market share numbers were out on the basis of it not having released anything. At the time NVidia had 76% of the discrete GPU market. This was when Nvidia’s best card was the GeForce GTX 980.

A lot happened in that time. There was the release of the Titan X in March, and before the GTX 980 Ti in June. AMD had its Hawaii architecture inside of the R9 290X, and the dual-GPU in the form of the R9 295X2. It was expected that the R9 390X might turn AMD’s luck around but that turned out to be another rebrand. Then there was the arrival of the R9 Fury X.

AMD has new products on the market: the R9 Fury X, R9 Fury, R9 390X and a bunch of rebranded 300 series video cards. But according to Mercury Research’s latest data, NVIDIA has jumped from 76% of the discrete GPU market in Q4 2014 to 82 per cent in Q2 2015.

AMD has 18 per cent of the dGPU market share, even after the release of multiple new products.

It is not that the Fury X isn’t selling well, but because of yield problems there will only 30,000 units made over the entire of the year.

AMD also rebranded nearly its entire product stack thus making no reason to buy a R9 390X if you own an R9 290X.

Sure there is 8GB of GDDR5 on board compared to the 4GB offered on most R9 290X cards, but that’s not enough to push someone to upgrade their card.

Tweaktown noted that  there was a big issue of the HBM-powered R9 Fury X not really offering any form of performance benefits over the GDDR5-powered GeForce GTX 980 Ti from NVIDIA. The 980 Ti beating the Fury X in some tests which it should not have.

Nvidia has plenty of GM200 GPUs to go around, with countless GTX 980 Ti models from a bunch of AIB partners. There is absolutely no shortage of GTX 980 Ti cards. Even if you wanted to get your paws on a Fury X, AMD has made it difficult.

Now it seems that next year could be a lot worse for AMD. Nvidia will have its GP100 and GP104 out next year powered by Pascal. This will cane AMD’s Fiji architecture. Then Nvidia will swap to 16nm process when its Maxwell architecture is already power efficient. Then there is the move HBM2, where be should see around 1TB/sec memory bandwidth.

All up the future does not look that great for AMD.


Is Intel’s Skylake Processor Creating Issues?

August 20, 2015 by Michael  
Filed under Computing

There is no doubt that Skylake is rather impressive architecture with a huge focus on mobile. At the same time, the Skylake Core i7 6700K desktop part seems to be the most disconnected product launch Intel did in the last few years.

The desktop part has been launched, but only one of the two CPU’s launched can be purchased. There is a serious shortage of desktop processors, and when we asked around everyone we know thinks that the product launch came a big prematurely. The Skylake desktop CPU comes just months after Broadwell, which totally confused the channel.

Broadwell, now “previous generation” processor, has faster graphics performance then the top end desktop Skylake at the moment. Broadwell Core i7 5775C was launched on June 2nd for $366 on LGA 1150 socket and with a 65 W TDP. We are still scratching our heads to see Broadwell launching on June 2nd only to be replaced by Skylake in early August.

Skylake was launched with 91W TDP, with the highest CPU clock of 4GHz but slower GPU performance compared to Iris graphics of Broadwell. This is not all, as Skylake requires more expensive DDR4 memory, a new Z170 motherboard and yet again new 1151 socket. Two sockets in two months, this is a new high for Intel. The GPU is called Intel HD Graphics 530 this time around. Intel gets away from four digit to three digit graphics differentiator, and this is not confusing at all. I wonder what might be faster, Iris Pro 6200, Haswell’s HD 4600 or Intel HD Graphics 530.

All of the above creates quite a mess in the channel, as end users don’t know what is compatible with or comparable to what. Well, your new motherboard that you got a few months ago won’t work with your brand new Skylake processor, that’s for sure.

It just looks that Intel gets away pretty easy with all the mess that it creates occasionally. The press is usually faster to point out what AMD did wrong then to point out obvious issues with Intel desktop strategy. It might be that many looked trough Intel’s fingers as people see Intel fighting with Qualcomm / Apple for its place under the hood. Most of the press have been busy enough with Qualcomm for Snapdragon 810 and AMD for its market value.

Intel Skylake looks like it is going to turn into quite a good processor, but it is questionable whether it will bring enough of a performance upgrade versus Haswell and Haswell refresh to make the purchase viable.

It is interesting that none of the deep dive articles about Skylake have found their place on the internet. We expect this to happen after IDF 2015 as it started on Tuesday August 18th.


IBM Shows Off Linux Only Mainframe

August 19, 2015 by Michael  
Filed under Computing

IBM has announced a new Linux-only mainframe aimed at both ends of the enterprise market.

The new mainframe is led by two servers called LinuxONE that Biggish Blue claims is the “world’s most advanced Linux system.” They both possess the fastest processor in the industry in and are designed for the “new application economy” and hybrid cloud era.

At the top of the range sits the LinuxONE Emperor which is based on IBM z13. It has a new processor design, faster I/O and the ability to address up to 10TB of memory — three times as much as its predecessor. It can house up to 141 processor units in a single system and run as many as 8,000 virtual servers, the company says.

At a maximum 5GHz, the z13′s processor is slower in terms of clockspeed than the chip in the z12, but IBM says it more than compensates for that with other improvements. The chip has eight cores compared with six for its predecessor, and it’s manufactured on a newer, 22 nanometer process, which should mean smaller, faster transistors.

It also has the security and advanced encryption features needed by enterprises. The LinuxONE Rock hopper is an entry level product geared towards speed, security and availability benefits of the mainframe.

IBM is giving shedloads of access to developers. As part of the Linux Foundation’s ‘Open Mainframe Project’ it has contributed some 500,000 lines of code including code related to IT predictive analytics that are on the lookout for unusual system behaviour to stop issues becoming failures.

The LinuxONE Developer Cloud acts as a virtual R&D engine for creating, testing and piloting of emerging applications such as links to engagement systems, mobile apps and hybrid cloud apps.

LinuxONE is provisioned as a virtual machine using the open standards-based KVM hyper-visors.

It is all in the shops now IBM doesn’t give pricing for its new mainframes, but will be more than $100,000.



TSMC Will Produce 16nm FinFET+ Processors Very Soon

August 18, 2015 by Michael  
Filed under Computing

TSMC has mentioned that it will be looking to commence production of its 16nm process in the third quarter now it looks like it is happening.

The outfit told the Taiwan Stock Exchange that its 16nm chips “smoothly entered” volume production as expected.  It sounds like someone in TSMC’s PR department has been reading 40 Shades of Grey again.

Production of a 16nm  node was underway, with higher yields expected later this quarter. Nvidia wants to get its GP100 Pascal GPU manufactured on this node as it switches to 16nm next year.

TSMC’s 16nm FinFET+ node will deliver twice the density and 65 percent higher speed at 70 percent less power than the current 28HPM process.

So far there has been a shortage of dates mentioned. We think that it will be the first quarter of 2016, with the hardware becoming available in the second quarter.

Both will still be behind Samsung which is going be 14nm fairly soon.



More Details Uncovered On AMD’s ZEN Cores

August 17, 2015 by Michael  
Filed under Computing

Our well informed industry sources have shared a few more details about the AMD’s 2016 Zen cores and now it appears that the architecture won’t use the shared FPU like Bulldozer.

The new Zen uses a SMT Hyperthreading just like Intel. They can process two threads at once with a Hyperthreaded core. AMD has told a special few that they are dropping the “core pair” approach that was a foundation of Bulldozer. This means that there will not be a shared FPU anymore.

Zen will use a scheduling model that is similar to Intel’s and it will use competitive hardware and simulation to define any needed scheduling or NUMA changes.

Two cores will still share the L3 cache but not the FPU. This because in 14nm there is enough space for the FPU inside of the Zen core and this approach might be faster.

We mentioned this in late April where we released a few details about the 16 core, 32 thread Zen based processor with Greenland based graphics stream processor.

Zen will apparently be ISA compatible with Haswell/Broadwell style of compute and the existing software will be compatible without requiring any programming changes.

Zen also focuses on a various compiler optimisation including GCC with target of SPECint v6 based score at common compiler settings and Microsoft Visual studio with target of parity of supported ISA features with Intel.

Benchmarking and performance compiler LLVM targets SPECint v6 rate score at performance compiler settings.

We cannot predict any instruction per clock (IPC improvement) over Intel Skylake, but it helps that Intel replaced Skylake with another 14nm processor in later part of 2016. If Zen makes to the market in 2016 AMD might have a fighting chance to narrow the performance gap between Intel greatest offerings.


Intel Plans To Bring Xeon Processors To Laptops

August 11, 2015 by Michael  
Filed under Computing

Intel is bringing its Xeon server processor technology to notebook computers.

The Xeon notebook chip is designed for “professionals who needs workstation-class capabilities in a portable device”, Intel said, and will launch in the form of the E3-1500M v5 processor family based on the recently-launched Skylake architecture.

The chipmaker’s decision to bring server chips to notebooks is down to the increasing number of creative professionals and engineers needing a more portable machine for power-hungry applications.

“This family of processors … will deliver high precision computing horsepower in notebook form factors, delivering the right balance of power and mobility,” said the firm.

“[The] Xeon-based mobile workstations will have key features such as error-correcting code memory that automatically detects and repairs errors on-the-fly that cause data corruption and system crashes for peace-of-mind reliability.”

The new systems, dubbed “mobile workstations”, will also have the benefits of the unique hardware-assisted security, manageability and productivity capabilities of Intel vPro Technology, Intel said. They will feature Thunderbolt 3 USB Type-C connectivity for faster and simpler I/O.

Intel said that it is “not quite ready to unveil all the details” regarding the new processor, but promised that the announcement is something that a large number of content creators, designers and engineers “can get excited about”.

Intel announced the first wave of processors based on the 14nm Skylake architecture last week, naming them the ’6th-generation Core’ family.

The chipset is the first mainstream Intel desktop platform to support DDR4 memory, and is claimed to deliver 30 percent better performance than a three-year-old PC based on Ivy Bridge architecture, 20 percent better performance than a two-year-old PC (Haswell), and 10 percent better performance than a one-year-old PC (Broadwell).

Skylake is the successor to the chipmaker’s Broadwell architecture, and was first put on the radar at Intel’s Developer Forum last year when the firm previewed the chip, touted to deliver significant increases in performance, battery life and power efficiency.

Processors based on the Skylake architecture have a new chip design, despite being fabbed on the same 14nm process as Broadwell, making Skylake a ‘tock’ iteration in Intel’s ‘tick-tock’ chip architecture cadence.


AMD Coherent Data Reaches 100 GBs

August 10, 2015 by Michael  
Filed under Computing

After a lot of asking around, we can give you some actual numbers about the AMD’s coherent fabric.

The inter-connecting technology already sounded very promising, but now we have the actual number. The HSA, Heterogeneous System Architecture MCM (Multi Chip Module) that AMD is working on can give you almost seven times faster score than the traditional PCIe interface.

Our industry sources have confirmed that with 4 GMI (Global Memory Interconnect) links AMD’s CPU and GPU can talk at 100GB/s. the traditional PCIe 16X provides 15GB/s at about 500 ns latency. Data Fabric eliminates PCIe latency too.

AMD will be using this technology with the next gen Multi Chip module that packs a Zeppelin CPU (most likely packed with a bunch of ZEN cores) and a Greenland GPU that of course comes with super fast HBM (High Bandwidth Memory). The Greenland and HBM can communicate at 500 GB/s and can provide highest performance GPU with 4+ teraflops.

This new MCM package based chip will also talk with DDR4 3200 memory at 100GB/s speed making it quite attractive for the HSA computation oriented customers.



Intel Shows Off 6th Generation Processor Based On 14nm Skylake

August 6, 2015 by Michael  
Filed under Computing

Intel has announced the first wave of processors based on the 14nm Skylake architecture, naming them the ’6th-generation Core’ family.

Intel’s latest chipset is the first mainstream Intel desktop platform to support DDR4 memory, and is claimed to deliver 30 percent better performance than a three-year-old PC based on Ivy Bridge architecture, 20 percent better performance than a two-year-old PC (Haswell), and 10 percent better performance than a one-year-old PC (Broadwell).

Skylake is the successor to the chipmaker’s Broadwell architecture, and was first put on the radar at Intel’s Developer Forum last year when the firm previewed the chip, touted to deliver significant increases in performance, battery life and power efficiency.

Processors based on the Skylake architecture have a new chip design, despite being fabbed on the same 14nm process as Broadwell, making Skylake a ‘tock’ iteration in Intel’s ‘tick-tock’ chip architecture cadence.

Arriving on the market today, the new quad-core chip designs are the Skylake-K variants of Intel’s latest generation of processors, comprising the Core i7-6700K and the Core i5-6600K.

These SKUs are aimed at gamers and computing enthusiasts, and Intel said that the rest of the product family, which will make up the bulk of Skylake processors, will arrive later this year.

The Core i7-6700K has a base clock speed of 4GHz and a max Turbo speed of 4.2GHz, supporting eight threads. The Core i5-6600K has a base clock speed of 3.5GHz and a max Turbo speed of 3.9GHz, supporting four threads.

The chips also support DDR4 memory at up to 2133MHz, or DDR3 at 1600MHz in two memory channels, with two DIMMS per channel. These can be fitted with up to 64GB of DDR4 memory.

The driving force behind releasing these two chips first is the gaming market, Intel said, as customers will see a noticeable performance improvement.

The chips cost $350 (£224) for the Core i7-6700K and $243 (£156) for the Core i5-6600K.

Intel also introduced a new chipset, the Z170, and a new LGA 1151 motherboard socket. The firm said that this will offer up to 40 percent more I/O interfaces, including storage attached to the PCI Express bus that is now supported by Intel’s Rapid Storage Technology.

Intel CEO Brian Krzanich announced last month that the firm’s shift from one transistor size to another is stretching from two to 2.5 years, putting Moore’s Law into question.

Moore’s Law is the prediction made by Intel co-founder Gordon Moore that computing power would double every two years. The law turned 50 earlier this year, and Intel said that the best is yet to come, and that the law will become more relevant in the next two decades as everyday objects become smaller, smarter and connected.

Intel has adhered to this in the past, using a ‘tick-tock’ strategy when launching new processors. The ‘tick’ refers to a shrinking of the manufacturing process, while the ‘tock’ is an improvement of the design and architecture at the same size.

However, Krzanich casted doubt over this during Intel’s earnings call in June, saying that manufacturing processes haven’t advanced at the same rate as in the past.

“The [tick-tock] strategy created better products for our customers and a competitive advantage for Intel,” said Krazanich.

“It also disproved the death of Moore’s Law predictions many times over. The last two technology transitions have signalled that our cadence today is closer to 2.5 years than two.”

He added that to address this, Intel plans to introduce a third 14nm product, codenamed Kabylake, in the second half of 2016 built on the foundations of the Skylake micro-architecture but with performance enhancements.

The firm will then launch its first 10nm product, codenamed Cannonlake, in the second half of 2017.


AMD’s x86 16-core Heterogenous EHP Processor Spotted

August 6, 2015 by Michael  
Filed under Computing

It was rumored back in April about AMD’s upcoming Exascale Heterogeneous Processor (EHP) with 16 cores and a Greenland APU, and now it seems that the rest of the world has caught up to the news.

A paper was submitted at IEEE and it was the first time AMD mentioned sixteen Zen cores wrapped around the GPU and powered by HBM 2 memory. We believe that this is a 16-core processor with 32 thread support and not 32 core as many reported. We will know soon enough and then can have another “we told you so” headline.

We would not be surprised if we hear more about this AMD processor at the Hot Chips conference on August 23rd. The EHP computing solution uses a silicon interposer and an APU chip that, almost as a raison d’être for AMD over the past several years, packs a GPU and CPU into a well-tuned band.  All this will be surrounded by die-stacked DRAM.The Italian website that brought this news back to life claims that AMD expects to ship the product between 2016 and 2017. That is the sort of timing you can expect with the rest of the ZEN based cores on the market. One can only hope that it will happen sooner rather than later. AMD needs to get more of the high performance compute market and earn some profits.

The IEEE article gives a bit of light on AMD exascale computer strategy:

Exascale computing requires very high levels of performance capabilities while staying within very stringent power budgets. Hardware optimized for specific functions is much more energy efficient than implementing those functions with general purpose cores. However, there is a strong desire for supercomputer customers to not have to pay for custom components designed only for high-end HPC systems, and therefore high-volume GPU technology becomes a natural choice for energy-efficient data-parallel computing. To fully realize the capabilities of the GPU, we envision exascale compute nodes comprised of integrated CPUs and GPUs (i.e., accelerated processing units or APUs) along with the hardware and software support to enable scientists to effectively run their scientific experiments on an exascale system.  [In the paper submitted to IEEE...] We discuss the hardware and software challenges in building a heterogeneous exascale system, and we describe on-going research efforts at AMD to realize our exascale vision.


AMD To Use Superfast Fabric

August 5, 2015 by Michael  
Filed under Computing

AMD has come up with a new interface, interconnect technology it calls coherent fabric.

It is a new inter-chip connection that will be used in the Exascale Heterogeneous Processor (EHP). Coherent fabric is primary interconnect for IP on AMD SOCs, APUs, CPUs and GPUs beginning with the next generation CPU SOC and followed by the Greenland GPU ASIC. Incoherant fabric, on the other hand, is something worn by loud American tourists and is sometimes called a Hawaiian shirt.

AMD’s GMI – Global Memory Interconnect, extends the coherent fabric between chips on the same multi-chip module (MCM) package. The coherent fabric extends between packages using combo phys which also support PCIe interconnection.

All this sounds rather complicated, but is basically a faster way for Zen SoC to talk to the Greenland Die on future processors.

Coherent fabric will speed up the communication in Zen CPUs between the cores and cache, host controllers, eg. USB, SATA or GbE, memory controllers, PSP, timer, counter, ACPI or legacy interface, combo phys and GMI layers. GMI phys of Zen die will talk directly with GMI phys of Greenland HBM 2.0 graphics die and uses coherent fabric too.

Our industry sources did not want to go into the specifics of achievable speeds of this interconnect, but the promise of a significant improvement is there. And of course the upcoming AMD’s 16 core Heterogenous EHP Processor will benefit from the fast coherent fabric interconnect.

AMD’s coherent fabric is the direct competitor to Nvidia’s NVlink interconnection that is expected in 2016 with the Pascal GPUs.


Is Intel Trying To Divide To Conquer?

July 30, 2015 by Michael  
Filed under Computing

Intel is trying to boost promotion of its desktop CPU platforms by dividing the market into six pieces.

According to Digitimes which has its paws on the cunning plan said that Intel is talking about something called an Enthusiast Tower.

An Enthusiast Tower, is not a ride at Disneyland, it is the gaming, video/audio content and high performance sector. What Intel defines as “mainstream” has high performance-price ratios.  “All-in-one (AIO) PCs”, “Mini PCs (NUC)”, “Portable AIO PCs” and “Compute Sticks” make up the remaining pieces of Intel’s marketing pie.

The Enthusiast Tower part of Intel’s business is doing well.  It is seeing growing sales, while demand for NUC products and Compute Sticks is also gradually picking up.

Intel said that its MiniPCs will support both Windows and Chrome OS, and the other five only Windows 8.1/10.

In early August, Intel will announce several K-series processors including Core i7-6700K, and Z170 chipsets and will unveil Skylake-S and Skylake-U series processors and H170/B150 chipsets in early September.

Intel will start mass shipping Skylake processors in October and November. Its top-end six-core and eight-core Broadwell-E processors will be in the shops in the first quarter of 2016. They will use LGA 2011-3 and supporting the X99 chipsets and DDR4 memory.


Will Intel’s Skylake Come To NUCs?

July 29, 2015 by Michael  
Filed under Computing

Intel is expected to upgrade its Compute Stick and NUC solutions to Skylake processors starting October.

ECS, Gigabyte, Asustek and ASRock are expected to launch related products.

Sales for the Compute Stick and NUC have been rising and it appears that Intel sees gold in the mini PC segment’s potential. NUC s are seeing stable demand in Japan, China, Southeast Asia, Latin America and Europe.

Intel is set to release solutions with its new Core m5/m3 processors codenamed Cedar City in the fourth quarter for the Compute Stick.

The new Compute Stick will feature 4GB of memory, 64GB of storage space and support Ultra HD. It will be based around the Core m3-6Y30 processor, which is set to release in October. It will also have Windows 10.

The version with the  Core m5-6Y57 vPro processor, will not come with a pre-installed operating system. In the first quarter of 2016, Intel is planning to launch inexpensive Atom x5 processors.

In November, Intel will launch two Skylake-based processors codenamed Swift Canyon, specifically for the NUC segment and will release high-end Core i7 processors at the end of the first quarter 2016 to improve the product line’s specifications and functions.


Samsung Targets 10nm FinFET

July 23, 2015 by Michael  
Filed under Computing

Samsung has put 10nm FinFET in its roadmap to stop its customers migrating to TSMC.
There were some rumours that Samsung may alter its schedule in order to prevent clients that might consider switching to 10nm chips from TSMC as that outfit is expected to skip the 14nm process and go straight to 10nm

Kelvin Low from Samsung Foundry confirmed in a video posted on YouTube that Samsung has formally added 10nm FinFET into the process roadmap, for chip designers working in mobile, consumer or networking market segment the new chips will provide significant performance and power consumption improvements.

Samsung LSI division has already shown off its first 10nm wafers which was a symbolic message to major clients that Samsung is more than capable of getting its 10nm production lines up and running without much hassle. Low expected 10nm products to appear at the end of 2016