Tag: GPU

Apple gets Thor about Thunderbolt display and smites it

Thor.TDW.battle.cropApple is giving up on its Thunderbolt Display which it first introduced in the summer of 2011.

A spokesApple said that the the display will be available through Apple.com, Apple’s retail stores and Apple Authorised Resellers while supplies last, but it will not make any more of them.

The move has pundets puzzled and the Tame Apple Press thinks it means a new 4K or 5K display is on the horizon.

Stock shortages ahead of WWDC sparked rumors that Apple might be planning to introduce a new display at the event. Nothing happened and and Apple instead focused on dull software for iOS devices, Macs, Apple TVs, and Apple Watch devices.

But rumors that Apple was working on a 5K display have been around for a while. If they are true then it would have  resolution of 5120 x 2880 pixels, USB-C/Thunderbolt 3 peripherals.

It might have an AMD graphics chip inside so that anyone with the right connection could get a better looking screen.  Another suggestion is that it might have a DisplayPort 1.2 Multi-Stream Transport setup to stitch two halves of a display together to make one display.

However this is normal PR spinning after Apple kills off one product. The assumption is that it must have “something new” when it might just not be doing much at all.


AMD shows off Polaris-based Radeon RX 470 and RX 460

4528082378_4d5b9fb99e_zAMD has been showing off its latest Polaris based GPUs at E3 2016.

For those who came in late, Polaris is AMD’s bright new hope in the GPU world – a bit like Zen is for the CPU, only it appears to exist whereas Zen doesn’t.

THe Radeon RX 470 and RX 460  join the recently announced RX 480 as part of the company’s new Polaris family. Polaris is AMD’s newest GPU micro-architecture, which is based on the 14nm FinFet production process.

AMD is not telling us the prices of its new GPU, but it is possible to have stab at it. The  RX 480 is made for 1440p gaming, and the RX 470 will focus on delivering a “refined, power-efficient HD gaming” experience. The RX 460 will offer a “cool and efficient solution for the ultimate e-sports gaming experience.”

The 4GB version of the RX 480 will start out at $200, it’s safe to assume that these two other cards will launch at lower price points.

AMD says the chips are extremely thin, offering a very low Z-height, and will fit into thin and light gaming notebooks.

The entire new RX line will also support a wide variety of features that include DX12, Vulkan, HDR, HDMI 2.0b, DisplayPort 1.3/1.4, and H.265 encoding/decoding.

There is no release day  but since the RX 480 is scheduled to launch on June 29 the other two should be soon after. AMD is claiming that card outperforms $500 graphics cards in VR.

Google claims its TPU improves machine learning

victorian-education-2Google claims that its Tensor Processing Unit (TPU), advances machine learning capability by a factor of three generations.

Google CEO Sundar Pichai told the Google’s I/O developer conference that TPUs deliver an order of magnitude higher performance per watt than all commercially available GPUs and FPGA.

Pichai said the chips powered the AlphaGo computer that beat Lee Sedol, the world champion in the incredibly complicated game called Go. Google still is not going into details of the Tensor Processing Unit but the company did disclose a little more information in its blog.

“We’ve been running TPUs inside our data centres for more than a year, and have found them to deliver an order of magnitude better-optimised performance per watt for machine learning. This is roughly equivalent to fast-forwarding technology about seven years into the future (three generations of Moore’s Law),” the blog said. “TPU is tailored to machine learning applications, allowing the chip to be more tolerant of reduced computational precision, which means it requires fewer transistors per operation. Because of this, we can squeeze more operations per second into the silicon, use more sophisticated and powerful machine learning models, and apply these models more quickly, so users get more intelligent results more rapidly.”

The tiny TPU can fit into a hard drive slot within the data centre rack and has already been powering RankBrain and Street View, the blog said.

What Google is not saying is what a TPU actually is and if it will be a replacement for a CPU or a GPU. Word on the street is that the TPU could be a form of chip that implements the machine learning algorithms that are crafted using more power hungry GPUs and CPUs.

AMD users see Crimson after new driver burns their cards

AMD logoAMD’s new Crimson drivers, which killed off the Catalyst name and software, have a serious problem – they make cards overheat.

There are widespread reports of cards overheating and dying. The new driver is setting the video card fans to 20 percent and then leaving them there.

Fan speed should increase as the GPU temperature goes up, but Crimson sticks them there at 20 percent, even during games and intensive workloads.

This means that GPU temperatures  climb to more than 90° C which causes high temperatures, thermal throttling, graphical glitches and crashes.

Some users are reporting permanent hardware damage. Although the GPU itself throttles when it overheats, there’s speculation that other components on the cards, such as the VRMs, can still be damaged.

AMD has acknowledged the fan speed problem and says that a “hot fix” should be rolled out from today.

It seems AMD did not learn from Nvidia which had a similar problem in 2010 and 2013.

AMD finally releases Fiji

fiji_resort-t2AMD announced its new graphics chips rangie including the much awaited Fiji Graphics cards with High Bandwidth Memory (HBM).

There were a lot of announcements yesterday. AMD released the new range of R7 300 Series cards that is aimed squarely at gamers. For gamers  there’s the new R9 300 Series (which are basically R9 280s with higher clocks and 8GB of memory).

This family included:

• Radeon R9 380: $199, 4GB GDDR5 memory, 1440p gaming resolution
• Radeon R9 390: $329, 8GB GDDR5 memory, 4K gaming resolution
• Radeon R9 390X: $429, 8GB GDDR5 memory, 4K gaming resolution

Most of us felt that there was “little to see here move on please” and rushed to the Fiji Graphics cards which AMD CEO Lisa Su described.  Fiji is the “Most complex and highest performance GPU we have ever built.”

These offer three times the performance-per-watt of GDDR5. Fiji has 1.5x the performance-per-watt of the R9 290X, and was built with a focus on 4K gaming.

The chip has 4096 stream processors and is comprised of 8.9 billion transistors.

It has a graphics core clock of 1050MHz and is rated at 8.6 TFLOPs. AMD says there will also be plenty of overhead for overclocking.

Fiji will initially be available in two variants: the first is the Radeon R9 Fury (air cooled), while the second is the Radeon R9 Fury X (water-cooled). The Radeon R9 Fury will go on sale July 14 for $549 while the Radeon Fury X (1.5x performance-per-watt of a Radeon R9 290X) will be available June 24 for $649.

AMD also took the opportunity to show off its “Project Quantum,” which is a small form-factor PC that manages to cram two Fiji GPUs inside. The processor, GPUs, and all other hardware are incorporated into the bottom of the chassis, while the cooling solution is built into the top of the case. AMD says that it’s working with its PC partners to bring this solution to market.

Nvidia thinks it can make a billion from the cloud

nvidiaMaker of chips that help you see things, Nvidia, expects its cloud computing revenue to hit $1 billion in the next two to three years.

It says that demand for big data analysis drives growth in graphics chips.

CEO Jen-Hsun Huang told reporters a day before the Taipai Computex show that cloud computing is the company’s fastest-growing segment, with revenue increasing at about 60-70 percent a year.

Cloud computing allows people to play graphics-heavy games over the Internet, Huang said. He also noted that the company’s GPUs can now be used for a wide variety of applications, such as voice commands like those used by Microsoft’s search engine.

However he warned that it is going to be a while before people can start playing streamed games at the 4K resolution.

While it is possible to stream 4K movies from online services like Netflix to PCs, TVs and set-top boxes, streaming games from the cloud requires many infrastructure changes, said Jen-Hsun Huang said.

Nvidia can currently stream 1080p games at 60 frames per second from its Grid online gaming service, but the technology needs to be developed for 4K streaming and a lot of fine-tuning is needed at the server level, Huang said.

“It’s going to be a while,” Huang said.

The cloud is not the only area that Nvidia has also been moving into. Lately it signed up for an automotive chip programme with automaker Tesla Motors.

Weak PC sales hurts Nvidia

nvidiaNvidia predicted lower-than-expected revenue for the second quarter because falling PC sales have weakened the need for its graphic processor units.

Nvidia also reported first-quarter revenue and profit below what the cocaine nose jobs of Wall Street estimated.

Chief Financial Officer Colette Kress attributed the tepid forecast to a fall in demand from OEMs and PC market which is softer than a baby’s bottom.

Worldwide PC shipments fell about 6.7 percent to 68.5 million units in the first quarter, and are expected drop 4.9 percent during the year.

Rival chipmaker AMD reported a steep fall in first-quarter sales last month and said it expected weak demand for PCs to continue for some time.

Nvidia was also hurt by the strong dollar, which has risen about 9 percent. The outfit does a lot of its business in US dollars which has made its GPU gaming more expensive.

The outfit forecast second-quarter revenue of $1.01 billion, plus or minus two percent, below the average analyst estimate of $1.18 billion.

The company’s net income fell to $134 million in the first quarter ended April 26.

Revenue rose 4.4 percent to $1.15 billion, but missed the average estimate of $1.16 billion.

AMD opens Nasdaq at Financial Day ‘15

AMD Analyst Day '15_2
NEW YORK, NY – AMD’s CEO, Dr. Lisa Su, outlined a multi-year strategy designed to return the company to profitability and growth at today’s AMD 2015 Financial Analyst Day at the Nasdaq MarketSite studio in New York City.

The embattled company is a little over seven months into Su’s leadership role and is attempting what can only be called a “breakout” from the constraints of being locked into the number two position in the deflating PC market segment.

IP and Core Technology Updates
AMD showcased a number of new items at the event, including details on its next-generation 64-bit x86 and ARM processor cores, future graphics cores expected to deliver a 2x performance-per-watt improvement compared to current generation offerings, and modular design methodology that reduces system-on-chip (SoC) development costs and accelerates time to market.

Technology-related announcements include:
• Development of a brand new x86 processor core codenamed “Zen,” expected to drive AMD’s re-entry into high-performance desktop and server markets through improved instructions per clock of up to 40 percent, compared to AMD’s current x86 processor core. “Zen” will also feature simultaneous multi-threading (SMT) for higher throughput and a new cache subsystem.
• Updates on the company’s first custom 64-bit ARM core the “K12”. These enterprise-class 64-bit ARM cores are designed for efficiency and are ideally suited for server and embedded workloads.
• AMD’s plans to extend its graphics technology to the first high-performance graphics processing unit (GPU) featuring die stacked High Bandwidth Memory (HBM) using a 2.5D silicon interposer design. AMD plans to introduce this packaging in the second half of the year with its latest GPU.

Computing and Graphics Segment Updates
Additionally, AMD announced updates to its Computing and Graphics (CG) product roadmaps for accelerated processing unit (APU), central processing unit (CPU), and GPU products planned for introduction in 2016 and beyond. The upcoming products address key customer priorities, including increased performance, longer battery life, and improved energy efficiency. AMD also provided further details and publicly demonstrated its 6th Generation A-Series APU, formerly codenamed “Carrizo,” as well as its next-generation GPU offerings launching in the coming months.

AMD’s updated CG product roadmap includes:
• New AMD FX CPUs based on the “Zen” core and built using FinFET process technology. Featuring high core counts with SMT for high throughput and DDR4 compatibility, these CPUs will share the AM4 socket infrastructure with AMD’s 2016 Desktop APUs.
• 7th Generation AMD APUs will enable a discrete-level GPU gaming experience and full HSA performance in the FP4 Ultrathin Mobile Infrastructure.
• Future generations of high-performance GPUs will be based on FinFET process technology, which will contribute to a doubling of performance-per-watt. These cutting-edge discrete graphics will include second generation HBM technology.

Enterprise, Embedded, and Semi-Custom Segment Updates
AMD laid out a long-term strategy for its Enterprise, Embedded and Semi-Custom Business Group (EESC) to grow across a number of high-priority markets based on leveraging high-performance CPU and GPU cores that allow customers to build differentiated solutions. The near-term will bring continued focus on enabling scalable, semi-custom solutions and growth in the embedded pipeline. Looking ahead, next-generation “Zen” and “K12” cores will bring high performance to the datacenter, a space where AMD plans to regain share with a portfolio that includes x86 and ARM processors, increased power efficiency, and a renewed presence in the high-performance x86 server market.

“AMD’s high-performance IP, efficient modular design methodology, and evolved semi-custom business model will fuel strong growth opportunities across multiple markets,” said Forrest Norrod (a recent AMD hire formerly general manager of Dell’s Data Center Solutions), senior vice president and general manager, EESC. “In addition to driving sustained growth in our semi-custom and embedded businesses, we’re reaffirming our commitment to high-performance server computing based on our strong set of new product offerings.”

AMD’s EESC roadmap details include:
• Next-generation AMD Opteron™ processors, based on the “Zen” core targeting mainstream servers that will enable a broad spectrum of workloads with significant increases in I/O and memory capacity.
• Building off of the expected availability of “Seattle”-based systems later this year, AMD detailed plans for its next-generation ARM processors featuring the upcoming “K12” core.
• AMD also provided a glimpse into its new high-performance APU targeting HPC and workstation markets that is intended to deliver massive improvements to vector applications with scale-up graphics performance, HSA enablement, and optimized memory architecture.

Techeye Takeaway
AMD under Lisa Su is just now beginning to find its footing. The company is in dire need of an all-around “morale lift” to gain traction – this includes customers, employees and investors alike. In many respects the company committed the same mistake as did Intel in refusing to acknowledge the full impact their absence from mobile market would have on their future earnings. They are still in recovery mode from that mistake.

Another troubling factor is their misjudgment of the amount of time required to successfully penetrate the x86 dominated server segment with their 64-bit ARM based “Seattle” processors and ambidextrous plan. In fact, a fair number of those in the analyst community completely misjudged this call. Time has now nearly corrected this error – with the introduction of the “Zen” series AMD is once again preparing to reenter the x 86 server market as a player.

The company announced the use of High Bandwidth Memory (HBM) with their GPUs and plans product shipments in the second half of this year placing AMD 6 months ahead of Nvidia.

AMD was a lead proponent of the HBM JEDEC standard. Sources indicate AMD will be using HBM memories in a wide range of future CPU, APU and GPU product designs – dramatically increasing performance while keeping a lid on power.

AMD appears to be betting on the PC market from an advanced performance viewpoint, addressing the ever increasing demands of virtual reality, 4K screens and real-time gaming – something the power user community can appreciate. The company’s segue back into the low-end and mid-range server market with the aptly named “Zen” cores – something that has been obvious to their Asian customer base for an eternity.

Will AMD execute this time around?…,

Nvidia gives the world a new road map

Nvidia has announced a new roadmap for its GPU families at the GTC conference and it appears to have carried out some major surgery.

Pascal, Nvidia’s latest GPU architecture, is being introduced in between Maxwell and Volta. It has absorbed old Maxwell’s unified virtual memory support and old Volta’s on-package DRAM, integrating those feature additions into a single new product. We always thought that Pascal was a rubbish name for a GPU, after all, who wants a chip which breaks because you forgot to put a semi-colon in the middle of a nest.

Volta was supposed to follow Maxwell in 2014. Volta’s marquee feature would be on-package DRAM, using Through Silicon Vias (TSVs) to die stack memory and place it on the same package as the GPU.

What appears to have changed is Nvidia’s definition of Maxwell and Volta. Maxwell has lost its promised unified virtual memory feature which will now be under the bonnet of the chips after Maxwell. All users can hope for from Maxwell will get is the software based unified memory feature being rolled out in CUDA 6.

Nvidia is not telling anyone about its second generation Maxwell GPUs and how those might be integrated into professional products.

Maxwell’s best feature will be DirectX 12 support and will ship in 2014 as scheduled.

Meanwhile Volta has been pushed back and stripped of most of what people will find interesting, Its on-package DRAM will be promoted to the GPU before Volta, and while the name Volta still exists, all anything knows about the chip is that will come after the 2016 GPU.

Nvidia has not said anything else directly about the unified memory plans that Pascal has inherited from old Maxwell. Pascal will get NVLink which is an attempt to supplant PCI-Express with a faster interconnect bus.

Nvidia thinks that the 16GB/sec made available by PCI-Express 3.0 is not enough when compared to the 250GB/sec+ of memory bandwidth available within a single card. PCIe 4.0 will bring higher bandwidth yet but Nvidia wants to push its own bus to achieve the kind of bandwidth it wants.

According to the roadmap, the result is a bus that looks a whole heck of a lot like PCIe but uses tighter requirements and a true point-to-point design. NVLink uses differential signalling, with the smallest unit of connectivity being a “block.”

A block contains eight lanes, each rated for 20Gbps, for a combined bandwidth of 20GB/sec. In terms of transfers per second this puts NVLink at roughly 20 gigatransfers/second, as compared to an already staggering 8GT/sec for PCIe 3.0.

Nvidia has knocked up a Pascal prototype and it will be put on a motherboard parallel to the board with each Pascal card connected to the board through the NVLink mezzanine connector. This allows GPUs to be cooled with CPU-style cooling methods in a server rack. 

AMD to build gaming tablet

Waiting for AMD to adjust itself to the consumer mobile market was a little like waiting for Godot, but now it seems that the chipmaker really wants to contribute something.

Word on the street is that AMD is planning to build a tablet of its own to show its suppliers how it should be done.

According to TechRadar, AMD wants to build a game tablet, code-named as “Project Discovery” and the beast is so cool that it has already won an award for innovation at CES 2014, which will take place in January, 2014.

Images show that AMD’s upcoming gaming tablet will be similar to the Windows 8 Razer Edge gaming tablet. Thus, AMD’s gaming tablet would come equipped with a game controller and a docking station. If it will really see the sunlight, then it will obviously come equipped with Windows 8.1.

It looks like AMD’s “award winning” tablet will only be a prototype for now, with a commercial product being launched on the market later.

AMD said that it did not plan to enter the market with a branded tablet and peripherals at this time, which might suggest the new tablet might some sort of deal with another supplier, or it might be lying.

Either way, AMD entering the market will make it as popular as Microsoft was when it came out with its Surface tablet. The feeling out there among OEMs is that chipmakers should stick to making chips and leave the plastic and the tin to the experts.

AMD’s Mullins chip is a 64-bit, x86-based chip and the low-power Mullins APU is supposed to be AMD’s answer to Intel, Nvidia and Qualcomm. Mullins has a claimed double per-watt performance over its predecessor, Temash.