TweakTown NewsRefine News by Category:
It looks like AMD has some interesting things coming in 2016, with the current rumor is that its next-gen APU would arrive as a 16-core processor based on its Zen architecture, as well as a "Greenland" GPU with HBM memory.
The new APU will replace the Godaveri platform which we were introduced to with the Carrizo APU, with the Carizzo-based notebooks arriving in the next couple of months, but most likely at Computex 2015 in June. The new Zen APU would feature quad-channel DDR4 support, up to 16 processor cores based on the Zen architecture, and the HBM-powered Greenland GPU.
We should expect Greenland to be based on the Fiji architecture, which will be powering AMD's upcoming flagship video card: the Radeon R9 390X. We don't know if AMD will be using HBM1 or HBM2 on the Zen APU, but with the R9 390X using HBM1, we should expect AMD to quickly shift to HBM2 sometime in 2016.
We just got back from a briefing of the Knights Landing platform at Intel Jones Farm. This new platform compacts a large number of cores into a small package that consisted of 1U blades.
Here is the basic Knights Landing (KNL) information screen. Knights Landing will be based on 14nm process node and have a TDP of ~300watts. The number of cores shown on the slide shows over 60 cores, we assume there will be several different SKU's with different core counts and other spec's.
The KNL core itself is based on Silvermount, these cores have full Xeon capability and features which have been modified to meet the new design platform. The system itself will have both Windows and Lunix capability with very little code modification to applications if any. While running a Windows OS on this platform you can see all the logical processors in the task manager which shows the OS actually sees all the cores unlike Knights Corner which was a PCIe coprocessor.
As seen on the Watercooled PC Facebook page, one users' comment to the following photo is that it "looks like the pins were at a football game doing the wave." All we know is that someone messed up - badly.
We've seen plenty of incompetent PC builders throughout the years, however this one almost takes the cake. Not only have they put their thermal paste on the pins, they've obviously pushed either the wrong socket processor in or not lined it up properly, bending many pins involved. But maybe there's something we're missing here - more thermal paste might be the solution we've all been looking for.
Don't try this at home people, unless you've got a lot of money to throw around.
With tongues wagging over the announcement of the Galaxy S6, we haven't seen any real-world performance numbers - with the biggest question being the battery life of Samsung's super-powered handset.
Analysts with Moor Insights & Strategy took to their blog over Mobile World Congress, which we're only getting around to now, with an article titled "Not All 4G LTE Modems Are Created Equal According To Tests With Qualcomm And Samsung". Where most are talking up the internal power and benchmarking abilities of the Galaxy S6, the hit on battery life thanks to Samsung baking in its own modem could be a bigger problem than previously thought. It looks like Qualcomm has the better modem chops, but Samsung has decided to use more of its own components than ever in its latest flagship.
The blog post covers this, where they said "Some companies, like Qualcomm and Samsung Electronics, have developed their own front end solutions in order to improve their modem solutions and make vast improvements in all aspects. They have accomplished these improvements in multiple ways, including increased throughput going from Category 4 150 Mbps download speeds to Category 6 300Mbps and Category 9 450 Mbps download speeds as well as reducing power consumption at the same time". While that might not sound like much, what they say next should have you opening your eyes a little wider.
MWC 2015 - Qualcomm had some interesting things to talk about during Mobile World Congress this year, apart from having its chips in various new flagship handsets like the HTC One M9 and Samsung Galaxy S6 and Galaxy S6 edge.
Qualcomm confirmed the existence of its new 'Kryo' processor, a 64-bit chip that would be manufactured on the FinFET process, and sampled sometime in the next 6-12 months. The new Snapdragon 820 processor will be the first SoC to use Qualcomm's new "cognitive computing platform", something known as 'Zeroth', writes Patrick Moorhead for Forbes.
Moorhead expects to see Snapdragon 820 samples in the next 6-9 months, with a Snapdragon 820-powered smartphone to be unveiled in "December 2015, worse case October 2016". It could happen even sooner, with Snapdragon 820-powered devices shown off at MWC in March 2016.
Intel will soon be launching its new Atom processors, the "Braswell" based chips later this year, but before that happens we're going to see a new classification of its current mobile chips. This will help consumers have a better understanding of the power inside of the Atom processor.
Later this year Intel will be offering three different types of Atom processors, naming them like their Core processors with Atom x3, Atom x5 and Atom x7. Currently, Intel offers the Core i3, Core i5 and Core i7 processors. The Atom x7 for example, will feature more cores, better graphics engines and higher clock speeds than its Atom x3 processor.
At the bottom end of the scale we'll soon have the Atom range of processors, in the middle we'll have the Core M processors for "PC-level performance in tablet-thin designs" and the usual Core range of processors for everything else.
The International Solid-State Circuits Conference (ISSCC) has just kicked off in San Francisco, where Samsung has just unveiled the world's first 10nm FinFET technology. Samsung Electronics Semiconductor Business chief, Kim Ki-nam teased the future for Samsung when he took the stage.
Ki-nam teased the company's 10nm DRAM technology, as well as a sneak peak at its 3D V NAND technology. Samsung expects that its technology will be used in countless devices going into the future, where Kim said Samsung is "expected to come in the future advent of IoT (Internet of Things) spread into a wide range of IT equipment, including the age of the data center (Data Driven World) 'and the silicon semiconductor technology it is possible to ensure the performance and low-power solution for the semiconductor chip that can process these data through the innovation".
We should hear more about Samsung's 10nm FinFET process as we move deeper into the year, as well as what devices and products will be powered by the 10nm technology. Exciting times!
Intel will be providing more details on its upcoming 10nm manufacturing process this week at the 2015 International Solid-State Circuits Conference (ISSCC), and how its new research will continue pounding on the door of Moore's law when it hits 7nm, and beyond.
The chipmaker expects to provide the first 10nm-based processors late 2016 or early 2017, as the company is hoping to dodge the delay train it hit with Broadwell at 14nm. Before 10nm is even here, Intel is teasing 7nm, saying that it will need to use new materials in order to build it. This means that 10nm will be the last product Intel builds using silicon, with Intel eyeing down a replacement for silicon, such as III-V semiconductor, such as indium gallium arsenide (InGaAs).
Then we have even more interesting points of Intel's shift to 7nm, which could see the company using new types of packaging. This includes 2.5D, which is something AMD is using on its upcoming Radeon R9 390X which uses HBM memory. 2.5D has separate dies which are placed side by side on an interposer. Intel would also be looking at 3D, where each die is stacked directly on top of one another. When it comes to 10nm, Intel is hoping to continue pushing Moore's law against the wall, all while reducing the price per transistor. 7nm is going to be a very exciting milestone, as it will shift away from silicon that has been used for decades now. Imagine the possibilities of a 3D stack of 7nm dies... that should have any enthusiast begging for more.
The time has come for Samsung to tease to the world that it has started the mass production of its 14nm FinFET process technology, moving from its current node at 20nm.
Executive Vice President of Sales & Marketing, System LSI Business, Samsung Electronics, Gabsoo Han, said: "Samsung's advanced 14nm FinFET process technology is undoubtedly the most advanced logic process technology in the industry. We expect the production of our 14nm mobile application processor to positively impact the growth of the mobile industry by enabling further performance improvements for cutting-edge smartphones".
This is quite the achievement, as the new 14nm FinFET processors offer up to 20% more speed, 35% less power consumption, and 30% productivity gain when compared to Samsung's current 20nm process technology. Samsung's new 14nm FinFET process will see its way into the company's upcoming Exynos 7 Octa later this year, as well as many other products in the near future.
With over 10 years of cooperation, Intel have been a mainstay in Apple hardware - taking over from Apple's co-designed chips with PowerPC due to them reportedly having issues with slimming down their laptop range, turning to the processor giant for some help.
In a recent interview with Business Insider, Intel's CFO Stacy Smith stated "for a customer like Apple you'd have to take a big step off performance to step off our architecture. That is what in essence enables us to win across different customers."
This is a rather big statement from Intel, but lets be honest here - Intel is at the top and it will stay that way for a long period of time no matter what happens.
There has been some rumors that Apple are looking to move in from Intel in order to utilize ARM architecture in their products, however nothing tangible has come to light as of yet.