AMD to release Dual 'Fiji' GPU Graphics Card sooner than expected

Dual 'Fiji' GPU R9 Fury X2 Graphics Card

A few days back shipping information listing 'Fiji Gemini' being shipped from AMD's headquarters in Canada. 'Gemini' is the code name that has been used in AMD's previous Dual-GPU Radeon R9 295X graphics card. Although only speculation, but the new graphics card is likely to be called the Radeon R9 Fury X2.
It is certain that AMD will include 2 x 'Fiji' GPU's that are currently used in the R9 Fury, R9 Fury X and R9 Fury Nano graphics card. The shipping information also states that the dual 'Fiji' card will include a Cooler Master heat sink, and most certainly come with an all-in-one water cooling solution to keep temperatures down.

Shipping information for the Radeon R9 Fury X2

The dual 'Fiji' graphics card will feature a whopping 8192 Stream Processors with 128 GCN Compute units, 128 Render Output Units, 512 Texture Mapping Units, and 8GB of HBM (High-Bandwidth-Memory) with 4GB per GPU chip. The memory bandwidth and frequency are expected to remain the same speed as the R9 Fury X.

The sudden move by AMD to launch the R9 Fury X2 sooner could be a response to Nvidia's plans to release it's dual 'Maxwell' graphics card.

Source: WCCFtech

Nvidia's 'Pascal' set to feature HBM 2.0

Leaked slides from Nvidia's GTC Taiwan Presentation show HBM Memory in their Next-Gen GPU Range

 Nvidia's upcoming next-generation 'Pascal' is expected to feature HBM 2.0 (High-Bandwidth-Memory) in it's product line due to be released sometime in the first quarter of 2016. A leaked slide from Nvidia's GTC Taiwan 2015 presentation shows clearly an illustration of it's next generation GPU with HBM memory on the GPU die. According to the leaked slides, Nvidia will name it's HBM stacked memory as '3D Memory'.

The move to HBM by Nvidia makes sense as their main competitor AMD has recently released it's 'Fiji' range with HBM. If Nvidia were to include HBM 2.0 in it's upcoming GPU lineup, it will have a memory bandwidth of around 1 TB/s, which is double the speed of AMD's current generation 'Fiji' based GPUs.

A leaked slide showing Pascal's key features including '3D Memory'

Current rumors also indicate that Nvidia will include up to 16 GB of HBM 2.0 memory, however looking at the illustration on the slide, it's unlikely that this will be true as the picture shows the GPU with 4 stacked memory modules, AMD have the same setup with 1 GB of HBM on each stacked module, this would mean that each module would have to be 4 GB each. As HBM technology is at it's early stages, it's unlikely that 16 GB can be achieved.

What we do know is that Nvidia will most likely manufacture it's 'Pascal' GPU in 16nm FinFET process by TSMC.

Intel's Cannonlake to offer CPU's with more Cores

Intel's Roadmap to 'Cannonlake'

 Although Intel are no stranger to offering high-performance CPU's with a high core count such as the Intel Core i7-5960X processor with 8 cores and 16 threads, these processors are top of the line and carry a heavy price tag. For years Intel have only offered it's mainstream processors with 2 and 4 core count and a maximum of 8 threads on models with Hyperthreading.

This is all about to change according to a LinkedIn profile spotted by Motley Fool, of an Intel engineer which the profile describes he or she is working on the upcoming 'Cannonlake' SoC that "integrates 4/6/8 cores and Converged Coherent Fabric (CCF) which acts like a NorthBridge".

The reason of Intel's move to offer more cores could be a response to AMD's upcoming 'Zen' microarchitecture which will offer high core count CPU's. As Intel will move to a smaller 10nm manufacturing process in 'Cannonlake' it's plausible that Intel will increase the amount of CPU cores due to the smaller die.

AMD Releases New Embedded GPU to it's Lineup

AMD's New Embedded Discrete Graphics

AMD have announced that they will be expanding their discrete graphics line up with 3 new products from power efficient (192 GFLOPS and 20w) to high performance (3TFLOPS and 95w) offering more choices and reinforcing their commitment to the embedded market.

The new embedded GPU line up offers 3 levels of performance:

* AMD Radeon E8950MXM Embedded Module, this it the Ultra-High-Performance Model. The high-end model is features a powerful discrete GPU that is supports 4K decode & encode, with up to 6 x 4K displays. Its especially designed for high-end machines such as: casino & arcade gaming machines, medical imaging, military applications and flight simulators.

* AMD Radeon E8870 Embedded Series (MXM and PCIe), a mid-range High-Performance Model. This once is designed to provide exceptional performance at a lower power consumption, this is ideal for smaller form factor machines that can also include: arcade and casino gaming machines, many medical applications, and digital signage.

* AMD Radeon E6465 Embedded Series (MCM, MXM and PCIe), a Power-Efficient Model. Designed for providing graphics solutions with very low power consumption. This is suitable for retail, digital signage, thin client computers, and head-up displays for aviation.

Specifications:

E8950MXM

•32 Compute Units; 3 TFLOPS single precision (Peak)
•8GB GDDR5 Memory; 256-bit wide
•95W Thermal Design Power
•Support for 4K hardware-accelerated decode and encode
•AMD Eyefinity technology for up to 6 display outputs
•Support for DirectX 12, OpenGL 4.5, and OpenCL 2.0

E8870MXM

•12 Compute Units; 1.5 TFLOPS single precision (Peak)
•4GB GDDR5 Memory; 128-bit wide
•75W Thermal Design Power
•Dual HD decode of H.264, VC-1, MPEG-4 and MPEG-2
•AMD Eyefinity technology for up to 6 display outputs
•Support for DirectX 12, OpenGL 4.5, and OpenCL 2.0

E6465MCM and E6465MXM

•2 Compute Units; 192 GFLOPS single precision (Peak)
•2GB GDDR5 Memory; 64-bit wide
• 20W Thermal Design Power
•Dual HD decode of H.264, VC-1, MPEG-4 and MPEG-2
•AMD Eyefinity technology for up to 4 display outputs
•Support for DirectX 11.1, OpenGL 4.5, and OpenCL 1.2

Source: techpowerup.com

GlobalFoundries Confirm that Manufacturing AMD's 'Zen' Will be on Schedule

Roadmap of GlobalFoundries Chip Manufacturing

Responding to rumors that AMD will contract TSMC instead of GlobalFoundries to manufacture AMD's next generation 'Zen' processor. Jason Gorss, Senior Manager, Corporate and Technology Communications at Globalfoundries has confirmed that they will manufacture AMD's 'Zen' on time and that their 14nm FinFET manufacturing process is actually ahead of schedule stating:

"Our 14nm FinFET ramp is exceeding plan with best-in-class yield and defect density. The early-access version of the technology (14LPE) was qualified in January and is well on its way to volume production, meeting yield targets on lead customer products. The performance-enhanced version of the technology (14LPP) is set for qualification in the second half of 2015, with the volume ramp beginning in early 2016. Prototyping on test vehicles has demonstrated excellent logic and SRAM yields and performance at near 100% of target."

In the past GlobalFoundries haven't had a good track record in manufacturing processors for AMD so it's great news that GloFo have 'Zen' on track.

Source: WCCFtech

Nvidia to Launch a Dual-GPU Graphics Card with 2 x GM200 'Maxwell' Chips

Nvidia's new graphics card will feature 2 x 'Maxwell' chips

Nvidia is set to launch it's first ever dual-GPU graphics card that will probably be the fastest graphics card on the market. The card will feature 2 of Nvidia's top end 'Maxwell' chips named the GM200. These chips currently power the GTX980 Ti and Titan X graphics cards. This is definitely a response to AMD's upcoming R9 Fury X2, a dual graphics card featuring the new 'Fiji' chip with HBM (High Bandwidth Memory).

It is known that Nvidia has had a secret press conference in New York City a week ago and has invited only select members of the press to showcase it's upcoming product.

What we do know is that the dual GM200 graphics card will feature double the amount of CUDA cores which can be up to a staggering 6144 CUDA cores and the memory should be at least on par with the Titan X with 12 GB of GDDR5 memory. Expect the power consumption to be high on the dual-GPU card, which should be around 400-500w TDP.

Source: wccftech.com

Intel Releases its Ultra Fast PCI-Express Solid State Drives - The Intel SSD DC P3608

The Intel DC P3608 Solid State Drive

 Intel have released a whole new range of high capacity ultra fast PCI-Express solid state drives. The SSD DC P3608 is Intel's new 'Data Center' range which are primarily designed for enterprise level. The new drives come with a capacity of up to 4 TB with Read & Write speeds of up to 5GB/s & 3GB/s. What makes the DC P3608 drives achieve such high speeds is because the drive features 8 x PCI-Express lanes as opposed to 4 x lanes in previous models.

There are 3 capacities with the Intel SSD DC P3608 being 1.6TB, 3.2TB, and 4TB. The 1.6TB has a read speed of 5GB/s and write speed of 2GB/s, the 3.2TB has a read speed of 4.5GB/s and write speed of 2.6GB/s, and the 4TB has a read speed of 5GB/s and write speed of 3GB/s.

Some of the DC P3608's features include:

Enhanced Power Loss Management - Protection from unplanned power loss called PLI (Power Loss Imminent).
20mm HET MLC NAND (High-Endurance Technology) - The drive has extremely high durability with the 4TB drive being able to write up to 21.9PB over the life of the drive before wearing out.
End-to-End Data Protection - End-to-end data protection and ECC (Error Correction Code) Dual NVMe Controller - More effecient than SATA or SAS, the dual controller architecture allows the SSD to simultaneously transfer data with this improved efficiency.

Pricing for the DC P3608 starts from $3509 for the 1.6TB model and $8759 for the 4TB model.

AMD's Next Generation 'Zen' - What we know so far

AMD's Next Generation 'Zen' x86 Processor

 Back in September 2014 AMD have confirmed that they will be building a whole new x86 processor from scratch and putting its current Bulldozer based processor architecture finally to rest. AMD promises that the new x86 architecture codenamed 'Zen' will finally be able to complete head on with Intel's high performance processors. AMD claim there will be significant improvements in performance, efficiency and power consumption which AMD currently lags behind its competitor Intel.

Work on the new processor started in early 2015 shortly after AMD hired Jim Keller, the engineer responsible for designing the highly successful Athlon64 codenamed 'ClawHammer' back in 2003. Most recently Jim Keller has left AMD, indicating that the processor design has been completed. It is also known that for the very first time in several years that the engineers working on 'Zen' have been given total freedom in designing the new processor architecture. As stated by Zen team leader Suzanne Plummer:

"It is the first time in a very long time that we engineers have been given the total freedom to build a processor from scratch and do the best we can do. It is a multi-year project with a really large team. It’s like a marathon effort with some sprints in the middle. The team is working very hard, but they can see the finish line. I guarantee that it will deliver a huge improvement in performance and (low) power consumption over the previous generation."



A comparison of the current 'Excavator' & the new 'Zen' core.

Here are the key features that will be delivered to AMD's 'Zen' processor.

 SMT (Simultaneous Multithreading)

AMD will abandon it's hugely unpopular modular CMT (Clustered Multithreading) that is currently used in AMD's microarchitecture since Bulldozer to SMT (Simultaneous Multithreading) which is a similar model to Intel's current Hyperthreading model, a technique that allows multiple tasks to be executed on one CPU core, improving the CPU's overall efficiency.

DDR4 Support

It is confirmed that AMD's 'Zen' will fully support DDR4 memory. DDR4 memory is currently supported in Intel's Haswell-E, Broadwell and Skylake processor family.

Improvement in IPC (Instruction Per Cycle)

One of AMD's biggest focuses on Zen will be improving IPC, by increasing the number of instructions executed per clock cycle and allowing the processor to perform more efficiently at lower clock speeds. Currently AMD must rely on significantly higher CPU clock speeds to achieve a similar performance of its counterpart Intel.

Another way AMD will achieve a better IPC is to have an individual FPU (Floating Point Unit) per CPU core. Since 'Bulldozer' AMD have been using a shared FPU per 2 cores, although this setup allowed AMD to pack more cores in their CPU's, their single-threaded performance was hampered and fell behind Intel.

14nm or 16nm FinFET

AMD currently manufactures its processors in 28nm SHP and 32nm SOI, which is far behind Intel who started manufacturing their entire CPU lineup on 14nm process since 'Broadwell'. AMD will produce 'Zen' on 14nm FinFET with GlobalFoundries. There is a possibility that AMD may choose TSMC over GlobalFoundries to manufacture 'Zen' due to GlobalFoundries' poor track record and delays with manufacturing previous generation AMD processors. If AMD is to contract TSMC over GlobalFoundries, 'Zen' will be manufactured on a slightly larger 16nm FinFET fabrication process.

Switching to a smaller 14nm or 16nm fabrication process will allow AMD's 'Zen' to use larger CPU cores with a single Floating Point Unit per core.

A diagram of a Quad Core Zen processor.

AMD has generated a lot of hype since it's announcement in 2014. Many people are being cautious and waiting to see the final result. Since the launch of 'Bulldozer' AMD have hyped up the market but ended in disappointment with 'Bulldozer' performing worse in some cases compared to it's predecessor.

AMD to Manufacture 'Zen' CPU with TSMC at 16nm instead of Globalfoundries

An overview of AMD's new x86 high performance processor

Allegedly AMD have decided to contract TSMC to produce it's make-or-break 'Zen' CPU on its 16nm FinFET process over fears that Globalfoundries will be unable to manufacture the Zen on the 14nm FinFET process on time and further delay it's crucial launch during 2016.

If the rumours are true then it comes as no surprise. AMD have had issues with Globalfoundries in the past, most notably in 2011 where AMD's Bulldozer based FX chips were delayed for 9 months and eventually launched in Q4 2011. At that time Globalfoundries had several challenges getting the Bulldozer manufactures in the 32nm SOI process.

AMD's new Zen is very important for the company. It will be a whole new x86 architecture built from the ground up, focusing on IPC (Instructions Per Clock), greater performance per watt, high bandwidth low-latency cache system, DDR4 support, and most notably Simultaneous Multithreading (SMT) which will be similar to Intel's Hyperthreading.

If successful, AMD's Zen will finally be able to compete with Intel's high performance CPU's, something AMD was struggling with in nearly half a decade. That is why AMD can't afford anymore delays and need to ship Zen next year, even if it means contracting with TSMC using a larger 16nm manufacturing process.

Source wccftech.com

Intel announces it's Skylake embedded processor lineup

Intel's Skylake Architecture using 14nm technology

Earlier in September Intel has announced its new Skylake desktop processors. Starting with the enthusiast level 'K' Series Core i5 6600k & i7 6700k processors followed by it's powerful 'S' Series  and more recently it's mobile Skylake-H and Skylake-U processors.

Now Intel announces it's consumer level Skylake-T and embedded Skylake-TE processor series to it's lineup. These processors will all be on the new LGA1151 socket. The 'T' series lineup will start off with the entry level Pentium G4400T, and on to the Core i3-6100T, i5-6500T and i7-6700T. The embedded 'TE' series share the same processors as the Skylake T but with slightly different CPU clock speeds. Here is the a table of the lineup below.

Model	Cores / Threads	Frequency / Turbo	L3 cache	Graphics	GPU Frequency	TDP
Pentium G4400T	2 / 2	2.9 GHz	3 MB	HD 510	350 / 950 MHz	35W
Pentium G4400TE	2 / 2	2.4 GHz	3 MB	HD 510	350 / 950 MHz	35W
Core i3-6100T	2 / 4	3.2 GHz	3 MB	HD 530	350 / 950 MHz	35W
Core i3-6100TE	2 / 4	2.7 GHz	4 MB	HD 530	350 / 1000 MHz	35W
Core i5-6500T	4 / 4	2.5 / 3.1 GHz	6 MB	HD 530	350 / 1100 MHz	35W
Core i5-6500TE	4 / 4	2.3 / 3.3 GHz	6 MB	HD 530	350 / 1000 MHz	35W
Core i7-6700T	4 / 8	2.8 / 3.6 GHz	8 MB	HD 530	350 / 1100 MHz	35W
Core i7-6700TE	4 / 8	2.4 / 3.4 GHz	8 MB	HD 530	350 / 1000 MHz	35W

The new embedded desktop series replace the current 'Haswell' Pentium G3320TE, Core i3-4340TE, i5-4570TE and i7-4770TE and have a slight performance increase with about 100MHz increase in clock speed, a lower TDP thanks to Skylake's 14nm manufacturing process and support for DDR4 memory.