Subject: Mobile | October 17, 2017 - 03:50 PM | Sebastian Peak
Tagged: SoC, Snapdragon 636, snapdragon, qualcomm, octa-core, mobile platform, Kryo 260, Kryo, cpu, adreno, 8-core
Qualcomm's latest mobile platform is the Snapdragon 636, positioned (at least numerically) between the Snapdragon 630 and 660 introduced earlier this year, and offering a very impressive set of features for mid-range devices - even reaching parity with the Snapdragon 800-series in some respects.
Qualcomm claims CPU performance gains of up to 40% from the Kryo 260 cores in the Snapdragon 636 compared to the ARM Cortex-A53 cores found in the Snapdragon 630, and the switch to Kryo brings the new Snapdragon 636 closer to the specs of the Snapdragon 660 - also an 8-core Kryo 260 design (though the higher-numbered platform does boast slightly higher clocks from its eight CPU cores at 2.2 GHz vs. 1.8 GHz from the 636).
The Snapdragon 636 also features the same X12 LTE modem found in the existing Snapdragon 630/660, which is capable of up to 600 Mbps download speeds (3 x 20 Hz carrier aggregation, 256-QAM) and 150 Mbps peak upload (2 x 20 Hz aggregation, 64-QAM).
Graphics duties are performed by the Adreno 509, and 18:9 FHD+ displays are supported. The Snapdragon 636 also includes the Hexagon 680 DSP (which we first saw in the Snapdragon 820) with Spectra 160 ISP for supported image capture "of up to 24 megapixels with zero shutter lag while supporting smooth zoom, fast autofocus and true-to-life colors for outstanding image quality", according to Qualcomm.
This new Snapdragon 636 also offers Qualcomm's Aqstic codec (another feature inherited from the 800-series) for high-resolution audio up to 24-bit/192 kHz PCM, along with dual-oscillator support (separate clock generators for 44.1 kHz and 48 kHz based sample rates!) and a 130dB dynamic range with a very low THD+N of -109dB.
To expand on what the Aqstic codec in the SD636 provides, the separate clock generators are a fascinating addition in a world where many codecs resample the common 44.1 kHz - pretty much all digital music at or below CD quality - to 48 kHz during playback. Having a proper 44.1 kHz clock means native playback without the interpolation and subsequent filtering required when altering the original signal to an incompatible sample rate.
The Snapdragon 636 - which is both "pin and software-compatible" with existing Snapdragon 660 and 630 mobile platforms, according to Qualcomm - is expected to ship to customers beginning in November.
Subject: Mobile | October 16, 2017 - 10:23 AM | Sebastian Peak
Tagged: SoC, smartphone, phone, Oreo, mobile, Mate 10 Pro, Mate 10, Kirin 970, Huawei, Android 8, Android
Huawei has announced the successor(s) to the Mate 9 smartphone with the new Mate 10 and Mate 10 Pro, which feature a new "3D Glass Body" industrial design along with the new Kirin 970 processor and other improvements.
The key features from Huawei include:
- Kirin 970, the world’s first AI processor for smartphones with a dedicated Neural Network Processing Unit (NPU)
- A 3D Glass Body featuring a barely-there-bezel, HUAWEI FullView Display and HDR10 supported technology for intensely vivid and brighter colors
- TÜV Fast-Charge Safety Certified HUAWEI SuperCharge and 4000 mAh battery with AI-powered Battery Management
- New Leica Dual Camera with SUMMILUX-H lenses, with both featuring an aperture of f/1.6, and intelligent photography including AI-powered Real-Time Scene and Object Recognition and AI-powered Bokeh Effect;
- An all-new, simplified EMUI 8.0 based on Android 8.0
The Mate 10 Pro features an 18:9 OLED display
The Mate 10 is a 5.9-inch device with a 16:9 IPS display supporting HDR10, while the Mate 10 Pro offers an 18:9 OLED display (also with HDR10 support).
The new dual-camera system is again a joint effort with Leica, and combines a 12 MP color sensor with a 20 MP monochrome sensor, using lenses with a aperture of f/1.6 - and Huawei says this aperture is the "world's largest" for a smartphone. The digital zoom and bokeh effects are AI-powered, along with real-time scene and object recognition.
The new Kirin 970 combines an 8-core CPU with a 12-core Mali-G72 GPU, and includes an NPU (neural processing unit) for AI-related tasks as well as a new dual ISP for the AI-powered camera features mentioned above.
Both phones include a 4000 mAh battery which offers "smart battery management" which Huawei states "understands user behavior and intelligently allocates resources to maximize battery life". The new TÜV-certified fast charging feature supports low-voltage charging of 4.5V / 5A, and Huawei states this will charge the phones from 1% to 20% in 10 minutes, or 1% to 58% in 30 minutes.
The Mate 10 lineup
The Mate 10 and Mate 10 Pro ship with Android 8.0 and a new "simplified" version of Huawei's EMUI interface. Pricing and availablity for the U.S. was not revealed, but the phones will go on sale internationally starting this month for the Mate 10, and mid-November for the Mate 10 Pro.
The Mate 10 Pro lineup
While we don't have U.S. pricing yet, European pricing for the Mate 10 with 64GB of storage and 4GB memory is set at €699, and the Mate 10 Pro with 128GB/6GB will be €799.
A New Standard
With a physical design that is largely unchanged other than the addition of a glass back for wireless charging support, and featuring incremental improvements to the camera system most notably with the Plus version, the iPhone 8 and 8 Plus are interesting largely due to the presence of a new Apple SoC. The upcoming iPhone X (pronounced "ten") stole the show at Apple's keynote annoucement earlier this month, but the new A11 Bionic chip powers all 2017 iPhone models, and for the first time Apple has a fully custom GPU after their highly publicized split with Imagination Technologies, makers of the PowerVR graphics found in previous Apple SoCs.
The A11 Bionic powering the 2017 iPhones contains Apple’s first 6-core processor, which is comprised of two high performance cores (code-named ‘Monsoon’) and four high efficiency cores (code-named ‘Mistral’). Hugely important to its performance is the fact that all six cores are addressable with this new design, as Apple mentions in their description of the SoC:
"With six cores and 4.3 billion transistors, A11 Bionic has four efficiency cores that are up to 70 percent faster than the A10 Fusion chip, and two performance cores that are up to 25 percent faster. The CPU can even harness all six cores simultaneously when you need a turbo boost."
It was left to improvments to IPC and clock speed to boost the per-core performance of previous Apple SoCs, such as the previous A10 Fusion part, which contained a quad-core CPU split in an even arrangement of 2x performance + 2x efficiency cores. Apple's quad-core effort did not affect app performance beyond the two performance cores, with additional cores limited to background tasks in real-world use (though the A10 Fusion did not provide any improvement to battery life over previous efforts, as we saw).
The A11 Bionic on the iPhone 8 system board (image credit: iFixit)
Just how big an impact this new six-core CPU design will have can be instantly observed with the CPU benchmarks to follow, and on the next page we will find out how Apple's in-house GPU solution compare to both the previous A10 Fusion PowerVR graphics, and market-leading Qualcomm Adreno 540 found in the Snapdragon 835. We will begin with the CPU benchmarks.
Subject: Systems | September 25, 2017 - 04:15 PM | Jeremy Hellstrom
Tagged: asus, tinker, SoC, Rockchip, rk3288, Mali-T760, Cortex-A17
ASUS' take on single board computers is the new Tinker Board, powered by a 1.8 GHz Cortex-A17 based Rockchip RK3288 and a 600MHz Mali-T760 GPU which share 2 GB of LPDDR3. Storage is handled by a microSD slot, or the four USB 2.0 ports and the Tinker offers Gigabit wired connectivity as well as optional WiFi. You have a choice of operating systems, either Marshmallow flavoured Android or the Debian based Tinker OS, depending on which you prefer.
The Tech Report tested out the Tinker Board and found the hardware to outpace competitors such as Raspberry Pi, however the lack of software and documentation hamstrung the Tinker Board badly enough that they do not recommend this board. This may change in time but currently ASUS needs to do some work before the Tinker Board becomes an actual competitor in this crowded market.
"Asus' Tinker Board single-board computer wants to challenge the Raspberry Pi 3's popularity with a more powerful SoC and better networking, among other improvements. We put it to the test to see whether it's a worthy alternative to the status quo."
Here are some more Systems articles from around the web:
- PCSpecialist Apollo X01 (i7-7820X & 1080 Ti) System @ Kitguru
- HP Omen Desktop PC @ Techspot
- Upgrade My PC Please! Episode 1: The First 5 @ Techspot
- The Tech Report System Guide: September 2017 edition
Subject: Processors | March 14, 2017 - 03:17 PM | Jeremy Hellstrom
Tagged: nvidia, JetsonTX1, Denver, Cortex A57, pascal, SoC
Amongst the furor of the Ryzen launch the NVIDIA's new Jetson TX2 SoC was quietly sent out to reviewers and today the NDA expired so we can see how it performs. There are more Ryzen reviews below the fold, including Phoronix's Linux testing if you want to skip ahead. In addition to the specifications in the quote, you will find 8GB of 128-bit LPDDR4 offering memory bandwidth of 58.4 GB/s and 32GBs of eMMC for local storage. This Jetson is running JetPack 3.0 L4T based off of the Linux 4.4.15 kernel. Phoronix tested out its performance, see for yourself.
"Last week we got to tell you all about the new NVIDIA Jetson TX2 with its custom-designed 64-bit Denver 2 CPUs, four Cortex-A57 cores, and Pascal graphics with 256 CUDA cores. Today the Jetson TX2 is shipping and the embargo has expired for sharing performance metrics on the JTX2."
Here are some more Processor articles from around the web:
- Hands-On Nvidia Jetson TX2: Fast Processing for Embedded Devices @ Hack a Day
- AMD Ryzen 7 1700X Review; Testing SMT @ Hardware Canucks
- AMD Ryzen 7 1700 Linux Benchmarks: Great Multi-Core Performance For $329 @ Phoronix
Subject: Processors | March 7, 2017 - 09:02 AM | Tim Verry
Tagged: SoC, server, ryzen, opteron, Naples, HPC, amd
Over the summer, AMD introduced its Naples platform which is the server-focused implementation of the Zen microarchitecture in a SoC (System On a Chip) package. The company showed off a prototype dual socket Naples system and bits of information leaked onto the Internet, but for the most part news has been quiet on this front (whereas there were quite a few leaks of Ryzen which is AMD's desktop implementation of Zen).
The wait seems to be finally over, and AMD appears ready to talk more about Naples which will reportedly launch in the second quarter of this year (Q2'17) with full availability of processors and motherboards from OEMs and channel partners (e.g. system integrators) happening in the second half of 2017. Per AMD, "Naples" processors are SoCs with 32 cores and 64 threads that support 8 memory channels and a (theoretical) maximum of 2TB DDR4-2667. (Using the 16GB DIMMs available today, Naples support 256GB of DDR4 per socket.) Further, the Naples SoC features 64 PCI-E 3.0 lanes. Rumors also indicated that the SoC included support for sixteen 10GbE interfaces, but AMD has yet to confirm this or the number of SATA/SAS ports offered. AMD did say that Naples has an optimized cache structure for HPC compute and "dedicated security hardware" though it did not go into specifics. (The security hardware may be similar to the ARM TrustZone technology it has used in the past.)
Naples will be offered in single and dual socket designs with dual socket systems offering up 64 cores, 128 threads, 32 DDR4 DIMMs (512 GB using 16 GB modules) on 16 total memory channels with 21.3 GB/s per channel bandwidth (170.7 GB/s per SoC), 128 PCI-E 3.0 lanes, and an AMD Infinity Fabric interconnect between the two processor sockets.
AMD claims that its Naples platform offers up to 45% more cores, 122% more memory bandwidth, and 60% more I/O than its competition. For its internal comparison, AMD chose the Intel Xeon E5-2699A V4 which is the processor with highest core count that is intended for dual socket systems (there are E7s with more cores but those are in 4P systems). The Intel Xeon E5-2699A V4 system is a 14nm 22 core (44 thread) processor clocked at 2.4 GHz base to 3.6 GHz turbo with 55MB cache. It supports four channels of DDR4-2400 for a maximum bandwidth of 76.8 GB/s (19.2 GB/s per channel) as well as 40 PCI-E 3.0 lanes. A dual socket system with two of those Xeons features 44 cores, 88 threads, and a theoretical maximum of 1.54 TB of ECC RAM.
AMD's reference platform with two 32 core Naples SoCs and 512 GB DDR4 2400 MHz was purportedly 2.5x faster at the seismic analysis workload than the dual Xeon E5-2699A V4 OEM system with 1866 MHz DDR4. Curiously, when AMD compared a Naples reference platform with 44 cores enabled and running 1866 MHz memory to a similarly configured Intel system the Naples platform was twice as fast. It seems that the increased number of memory channels and memory bandwidth are really helping the Naples platform pull ahead in this workload.
AMD further claims that its Naples platform is more balanced and suited to cloud computing and scientific and HPC workloads than the competition. Specifically, Forrest Norrod the Senior Vice president and General Manager of AMD's Enterprise, Embedded, and Semi-Custom Business Unit stated:
“’Naples’ represents a completely new approach to supporting the massive processing requirements of the modern datacenter. This groundbreaking system-on-chip delivers the unique high-performance features required to address highly virtualized environments, massive data sets and new, emerging workloads.”
There is no word on pricing yet, but it should be competitive with Intel's offerings (the E5-2699A V4 is $4,938). AMD will reportedly be talking data center strategy and its upcoming products during the Open Compute Summit later this week, so hopefully there will be more information released at those presentations.
(My opinions follow)
This is one area where AMD needs to come out strong with support from motherboard manufacturers, system integrators, OEM partners, and OS and software validation to succeed. Intel is not likely to take AMD encroaching on its lucrative server market share lightly, and AMD is going to have a long road ahead of it to regain the market share it once had in this area, but it does have a decent architecture on its hands to build off of with Zen and if it can secure partner support Intel is certainly going to have competition here that it has not had to face in a long time. Intel and AMD competing over the data center market is a good thing, and as both companies bring new technology to market it will trickle down into the consumer level hardware. Naples' success in the data center could mean a profitable AMD with R&D money to push Zen as far as it can – so hopefully they can pull it off.
What are your thoughts on the Naples SoC and AMD's push into the server market?
- Zen and the Art of CPU Design
- AMD Zen Architecture Overview: Focus on Ryzen
- Dissecting AMD Zen Architecture - Interview with David Kanter
Introduction and Specifications
The Mate 9 is the current version of Huawei’s signature 6-inch smartphone, building on last year’s iteration with the company’s new Kirin 960 SoC (featuring ARM's next-generation Bifrost GPU architecture), improved industrial design, and exclusive Leica-branded dual camera system.
In the ultra-competitive smartphone world there is little room at the top, and most companies are simply looking for a share of the market. Apple and Samsung have occupied the top two spots for some time, with HTC, LG, Motorola, and others, far behind. But the new #3 emerged not from the usual suspects, but from a name many of us in the USA had not heard of until recently; and it is the manufacturer of the Mate 9. And comparing this new handset to the preceding Mate 8 (which we looked at this past August), it is a significant improvement in most respects.
With this phone Huawei has really come into their own with their signature phone design, and 2016 was a very good product year with the company’s smartphone offerings. The P9 handset launched early in 2016, offering not only solid specs and impressive industrial design, but a unique camera that was far more than a gimmick. Huawei’s partnership with Leica has resulted in a dual-camera system that operates differently than systems found on phones such as the iPhone 7 Plus, and the results are very impressive. The Mate 9 is an extension of that P9 design, adapted for their larger Mate smartphone series.
Subject: General Tech | January 3, 2017 - 09:10 PM | Sebastian Peak
Tagged: VR, SoC, snapdragon 835, qualcomm, processor, mobile, CES 2017, CES, AR
Qualcomm Technologies, Inc and ODG (Osterhout Design Group) have announced that the R-8 and R-9 smartglasses will be the first devices powered by the new Qualcomm Snapdragon 835 SoC. ODG is a developer of "mobile headworn computing and augmented reality technologies and products", and these new models leverage the reduced size and thermal requirements of the new Snapdragon 835 processor.
The R-8 smartglasses, seated next to a glass mug for scale
"The premium Snapdragon 835 processor was designed from the ground-up to support new and innovative products and experiences beyond mobile phones, and it’s great to see that the first announced Snapdragon 835 devices will be ODG’s smartglasses," said Raj Talluri, senior vice president, product management, Qualcomm Technologies, Inc. "Thermal dissipation on a heavy compute but small device is very difficult so higher power efficiency is a must. The Snapdragon 835 processor, with our unique SoC design expertise on a 10nm process node, enables ODG to meet their design goals and develop lighter, smaller and sleeker smartglasses that take advantage of the new processor’s superior performance and power efficiency."
The R-9 smartglasses
The Snapdragon-powered R-8 smartglasses are "lighter, smaller and sleeker than any other device in ODG’s portfolio", which should make their use a more attractive option for those interested in AR, VR, and Mixed Reality overlay capabilities. For their part the larger R-9 smartglasses are "based on ODG’s award-winning 50° FOV and 1080p Project Horizon platform". The company's previous smartglasses, the R-7, were powered by a Snapdragon 801 SoC.
Follow all of our coverage of the show at https://pcper.com/ces!
Subject: Processors, Mobile | October 18, 2016 - 11:32 AM | Sebastian Peak
Tagged: SoC, Snapdragon 653, Snapdragon 626, Snapdragon 427, snapdragon, smartphone, qualcomm, mobile
Qualcomm has announced new 400 and 600-series Snapdragon parts, and these new SoCs (Snapdragon 653, 626, and 427) inherit technology found previously on the 800-series parts, including fast LTE connectivity and dual-camera support.
The integrated LTE modem has been significantly for each of these SoCs, and Qualcomm lists these features for each of the new products:
- X9 LTE with CAT 7 modem (300Mbps DL; 150Mbps UL) designed to provide users with a 50 percent increase in maximum uplink speeds over the X8 LTE modem.
- LTE Advanced Carrier Aggregation with up to 2x20 MHz in the downlink and uplink
- Support for 64-QAM in the uplink
- Superior call clarity and higher call reliability with the Enhanced Voice Services (EVS) codec on VoLTE calls.
In addition to the new X9 modem, all three SoCs offer faster CPU and GPU performance, with the Snapdragon 653 (which replaces the 652) now supporting up to 8GB of memory - up from a max of 4GB previously. Each of the new SoCs also feature Qualcomm's Quick Charge 3.0 for fast charging.
Full specifications for these new products can be found on the updated Snapdragon product page.
Availability of the new 600-series Snapdragon processors is set for the end of this year, so we could start seeing handsets with the faster parts soon; while the Snapdragon 427 is expected to ship in devices early in 2017.
Subject: Processors | October 10, 2016 - 02:25 AM | Tim Verry
Tagged: SoC, Intel, FPGA, Cortex A53, arm, Altera
Intel and recently acquired Altera have launched a new FPGA product based on Intel’s 14nm Tri-Gate process featuring an ARM CPU, 5.5 million logic element FPGA, and HBM2 memory in a single package. The Stratix 10 is aimed at data center, networking, and radar/imaging customers.
The Stratix 10 is an Altera-designed FPGA (field programmable gate array) with 5.5 million logic elements and a new HyperFlex architecture that optimizes registers, pipeline, and critical pathing (feed-forward designs) to increase core performance and increase the logic density by five times that of previous products. Further, the upcoming FPGA SoC reportedly can run at twice the core performance of Stratix V or use up to 70% less power than its predecessor at the same performance level.
The increases in logic density, clockspeed, and power efficiency are a combination of the improved architecture and Intel’s 14nm FinFET (Tri-Gate) manufacturing process.
Intel rates the FPGA at 10 TFLOPS of single precision floating point DSP performance and 80 GFLOPS/watt.
Interestingly, Intel is using an ARM processor to feed data to the FPGA chip rather than its own Quark or Atom processors. Specifically, the Stratix 10 uses an ARM CPU with four Cortex A53 cores as well as four stacks of on package HBM2 memory with 1TB/s of bandwidth to feed data to the FPGA. There is also a “secure device manager” to ensure data integrity and security.
The Stratix 10 is aimed at data centers and will be used with in specialized tasks that demand high throughput and low latency. According to Intel, the processor is a good candidate for co-processors to offload and accelerate encryption/decryption, compression/de-compression, or Hadoop tasks. It can also be used to power specialized storage controllers and networking equipment.
Intel has started sampling the new chip to potential customers.
In general, FPGAs are great at highly parallelized workloads and are able to efficiently take huge amounts of inputs and process the data in parallel through custom programmed logic gates. An FPGA is essentially a program in hardware that can be rewired in the field (though depending on the chip it is not necessarily a “fast” process and it can take hours or longer to switch things up heh). These processors are used in medical and imaging devices, high frequency trading hardware, networking equipment, signal intelligence (cell towers, radar, guidance, ect), bitcoin mining (though ASICs stole the show a few years ago), and even password cracking. They can be almost anything you want which gives them an advantage over traditional CPUs and graphics cards though cost and increased coding complexity are prohibitive.
The Stratix 10 stood out as interesting to me because of its claimed 10 TFLOPS of single precision performance which is reportedly the important metric when it comes to training neural networks. In fact, Microsoft recently began deploying FPGAs across its Azure cloud computing platform and plans to build the “world’s fastest AI supercomputer. The Redmond-based company’s Project Catapult saw the company deploy Stratix V FPGAs to nearly all of its Azure datacenters and is using the programmable silicon as part of an “acceleration fabric” in its “configurable cloud” architecture that will be used initially to accelerate the company’s Bing search and AI research efforts and later by independent customers for their own applications.
It is interesting to see Microsoft going with FPGAs especially as efforts to use GPUs for GPGPU and neural network training and inferencing duties have increased so dramatically over the years (with NVIDIA being the one pushing the latter). It may well be a good call on Microsoft’s part as it could enable better performance and researchers would be able to code their AI accelerator platforms down to the gate level to really optimize things. Using higher level languages and cheaper hardware with GPUs does have a lower barrier to entry though. I suppose ti will depend on just how much Microsoft is going to charge customers to use the FPGA-powered instances.
FPGAs are in kind of a weird middle ground and while they are definitely not a new technology, they do continue to get more complex and powerful!
What are your thoughts on Intel's new FPGA SoC?
- Microsoft Goes All in for FPGAs to Build Out AI Cloud
- Microsoft Focusing Efforts, Forming AI and Research Group
- Stratix 10 Architecture Video
- Are FPGAs the future of password cracking and supercomputing?