Consistent with what they said at the developer’s conference about wanting to extend the reach of their A.I. Samsung has announced a new System on Chip (SoC) with some A.I. related features. The Exynos 9 Series 9820 processor. The processor contains an NPU, a unit for processing neural networks at speeds faster than what could be done with a general purpose processor alone. The presence of this unit on the device hardware makes possible device side experiences that would have previously required that data be sent to a server for processing. This may also translate into improvements in AR and VR experiences.
The NPU isn’t the only upgrade that comes with the processor. Samsung says the 9820’s new fourth generation custom core delivers a 20% improvement in single core performance or 40% in power efficiency compared to is predecessor. Multicore performance is said to be increased around 15%. The Exynos 9820 also has a video encoder capable of decoding 4K video at up to 150 frames per second in 10-bit color. The processor goes into mass production at the end of this year.
Samsung says they would like to have AI implemented in all of their products by 2020. From the visual display shown during the SDC 2018 conference it appears their usage of “all” is intended to be widely encompassing. Phones, car audio systems, refrigerators, air conditioners…
Samsung is inviting developers to start engaging in development for their conversational AI. Now they have made the same tools that they use for Bixby development internally available publically. The development portal and the development tools for Windows and OS X are available now at: https://bixbydevelopers.com
Galaxy Home, a Bixby enabled smart speaker, was showcased as a target implementation for the SDK.
The “Media Control API” will be available to content partners this December for adding deeper control into applications. Samsung says Netflix and Hulu are on board and will begin development with it next year.
The Samsung Frame TVs are also being opened to developers by way of the Ambient Mode SDK. This will allow developer content to show when the TV is in it’s standby mode.
One of the technology implementations displayed at SDC 2018 was their HOLOLAB. Curious as to what it is I walked over. It was a rig for performing 3D scanning of subjects. A few of the tangible outcomes included some pretty elastic 3D printed statues of people, a volume metric display showing a subject, and a display showing one of the scanned subjects dancing.
I was more curious about the implementation. I got a picture peering into the space to see how the physical structure was built. Inside the space the ambient lighting was ramped up by way of LEDs distributed around the space and 52 Point Grey Research cameras, all mounted to various places on an 8020 rig.
The software is custom built. I asked of its commercial availability and the Samsung rep I spoke to told me, “it was a work in progress.” The rep indicated that they are planning to have some popup locations available in the future for more people to try it and receive a copy of their models.
It is still early in the conference, but I might take a moment to try it out myself.
Tizen 5.0 was released a few days ago. This is a week in advance of the Samsung Developer’s Conference 2018. For those keeping count until now the most recent version of Tizen prior to now (4.0) could be found on the Galaxy Watch and Samsung TVs. There are Mobile devices that run Tizen. But I am in the USA and those devices are not sold here (so I won’t speak on them much).
What’s new in Tizen 5.0? There is improved IoT support, support for Bixby, and support for glTF. glTF is a format that aims to provide efficient loading of 3D scenes, but it is being added to Tizen with intent for it to be applied to watch faces. With the upgrade we are also getting improved debugging support and a new version of Tizen.Net. It also looks that they are deprecating the UI Builder tool in Tizen.
I get the feeling that I’ll hear more about this update when I go to the conference next week. I’ll post more from the conference as I find out 🙂
I previously mentioned that the dates for the Samsung Developer’s conference were announced. Registration is now open. If you register before September 12 you can register for the lowest available price. The registration form is available over at https://www.samsungdeveloperconference.com/ . Registration is also possible on site. Currently Registration is 299 USD (+tax). After the 12th it will go up to 399 USD. On site registration is 499 USD.
Playing with the code that I was using to get data from my car and stream it to the cloud I did something that I knew was a no-no; I hard coded the connection string in the code. There’s a number of reasons to not do this*; it’s less secure as someone can potentially extract the connection string and use it for unauthorized access and if the connection string ever needs to change then code needs to be recompiled and redeployed.
When a Windows IOT device is provisioned there is a connection string that is managed by the device; your application can take advantage of this and need not worry about the details of how it is stored. To make use of this there are a few libraries that you need to add to your UWP project. These include the followings.
With the classes in these libraries you can obtain the ID of the device and then use that ID to request an Azure DeviceClient class that is initialized with the connection string that the device is managing.