#technical machine

Pixel dailies: Technical Machine

http://project-nuzlockedex.wikia.com/wiki/Category:TM_Lists

Usually people arrange TM Lists for certain Pokemon games by numerical order, but we at Project Nuzlocke Dex have made another list to put the TMs in order of plot/story progression to help players who are looking for a specific TM and how long it will take to get it from their specific point in the main story. 

This is easier to do with the more modern Pokemon games, as they have become more linear compared to Gen I-IV. 

Each TM List is divided into certain important events, such as if the player has earned a specific badge or triggered a certain event. It accommodates up to the next roadblock that the player faces, whether it be a certain HM or actual roadblock that prevents the player from going further. 

3/10/15– Eric Kolker

Hey Tesselators, Eric here. Since we announced Tessel 2 last week, we have gotten a lot of questions about the new hardware.

Shiny new boards back from the manufacturer!

Tessel 2 at a glance

We packed a whole lot of hardware in there:

A 580MHz WiFi router system on chip (Mediatek MT7620n) running linux (OpenWRT)

64 MB of DDR2 RAM

32 MB of flash storage

2 High-speed USB 2.0 ports

a micro USB port

A 10/100 Ethernet port (RJ-45 jack)

A 48MHz ARM Cortex M0 microcontroller (Atmel SAMD21)

Two module ports that are much more capable than their predecessors

a button and a bunch of LEDs, because what’s a Tessel without blinky?

The board’s bill of materials and physical characteristics are only part of the picture. We spent a long time thinking about how we wanted to architect Tessel to push it beyond “another dev board” and clear into “this platform is exactly what I needed!” territory.

A few features under the hood (in addition to the ones current Tessel users know and love, including the expansive plug-and-play module ecosystem and high-level language support for low-level hardware features) include:

Router-grade 802.11b/g/n WiFi, including access point mode (Tessel can be a router)

16 GPIO broken out as a pair of multi-purpose module ports

Individual control over and protection for all outward-facing power buses (USB and module ports)

A form factor designed for abstraction and flexibility in the hardware, software, and mechanical worlds as you scale from prototype to production

Abstraction boundaries

One of the things which makes software so powerful is a heavy emphasis on frameworks and abstraction. Although there is no shortage of “standards” (official or otherwise) in the hardware space, one thing nobody has done particularly well yet is cleanly and clearly share abstraction boundaries between the hardware and software layers. We’re looking to change that.

The high-level system diagram for Tessel 2...and most other single-board computers, too.

The diagram above is a high-level system diagram for Tessel 2. Let’s dive into where we drew the lines internally.

The board employs a processor/coprocessor architecture. The Mediatek runs your user code, hosts USB devices, handles the network connections (be they wired, wireless, or cellular over USB), and communicates with the SAMD21.

The SAM acts as a coprocessor and handles real-time, low-level IO through the module ports, USB comms through the Micro USB port, and programming the device as a whole.

The two chips are connected by a SPI bridge that also includes the onboard flash (the readme for Tessel 2’s firmware repo goes into more detail here).

The whole system is powered from the single Micro USB (device) port, and its specific functional blocks look more like this:

Functional blocks of Tessel 2

This arrangement, which also very closely mirrors where the related parts are located on the hardware itself, allows us to draw the boundaries at the both the mechanical and conceptual level as follows:

Functional groups in Tessel 2’s architecture

Or, on the board itself:

Functional groups on Tessel 2’s prototype hardware

Consciously drawing these same boundaries when creating both the software and the hardware lets us make developing on the Tessel platform simple and consistent throughout an entire product cycle, which is a huge win. I’ll talk more about this in another post, but suffice it to say that most of the optimization and integration story relies on the fact that we kept these boundaries at the top of the list when making design decisions about how to build Tessel 2.

New and improved module ports

The two module ports on the new Tessel look and behave the same as the ones on the original board, but they’re actually a lot more versatile. In fact, don’t lock yourself into thinking that they’re “just module ports”; think of them as mini GPIO banks. Each pin on the two 10-pin headers is unique and can be reconfigured to do almost anything from speaking alternate comms protocols to clock generation. For example, if you decide you don’t want SPI, feel free to give yourself another I2C or UART with minimal changes to the SAMD21’s firmware. Touching only JS, you can forgo the fancy comms in favor of just plain GPIO, which gives you access to as many as 16 of them.

Plus, a nifty new power architecture gives you individual control over the 3.3V rails on each port, so you can turn modules off when they’re not in use to save power. This essentially converts the 3.3V rails on the ports into two high-current (at least 250 mA) output pins that just so happen to power modules most of the time. ...Or, put another way, this is our pass at solving hot-plugging for low-level hardware.

Last but not least, all eight pins on Port B are also inputs to a 12-bit, 350ksps ADC, with adjustable gain that can operate in differential mode, if that floats your boat. Not too shabby.

Close

We’re pretty excited about the new hardware and what in enables, and hope you are too. It’s been in the works for a few months now, and at this point it’s time for us to start cranking out docs and answering questions, so don’t hesitate to reach out on the forums or over email.

On the subject of docs, here is what we’ve released so far, including a preliminary schematic (full KiCAD project files are on their way soon). For now, head to our website to see more.

Until next time, ~e

This post has been translated:

3/5/15– Kelsey Breseman

Technical Machine announced Tessel 2 today: an open source $35 development platform you can take into production.

Why Tessel 2?

Tessel 2 is an affordable, accessible, open, and robust platform to build connected hardware devices and funnel them into production.

The Tessel platform was created to abstract away the initial hurdles of hardware development. Tessel opens up connected device development to people who want to build embedded devices, but don’t have the time to start with Ohm’s law and work their way up.

Tessel 2 takes the promise of Tessel 1, adds features and a path to production, and cuts the cost in half.

The original Tessel board fulfilled the goal of electrical accessibility. High-level languages, plug-and-play extensibility, and the use of a great package manager for installation all made Tessel the fastest way to build a prototype out of an idea. But it was also frustrating: fast as it is to get started on a Tessel 1, it’s hard to go beyond a single prototype. In the year since we shipped Tessel, we’ve been collecting feedback on user experiences and investigating solutions. Tessel 2 is based on that feedback.

With Tessel 2, we are building for users who want to take an idea, prototype, and then produce in volumes of hundreds to thousands.

For Tessel 2, we sought out the most reliable WiFi chips on the market, and found a solution in wireless router systems-on-chips (SoC’s, combined processor/wifi solutions). These chips are built into millions of routers and have very robust hardware and firmware. As an added bonus, they are produced in such high quantity that they cost less than $3 in quantity, compared to Tessel 1's $13 for WiFi + microcontroller.

We’re now able to offer a device that runs 20x as fast, offers full Node and io.js, has router-grade Wifi reliability, USB ports, and ethernet, and costs less than the original Tessel. That offering is Tessel 2.

Putting Tessel in a Product

At this much lower cost, Tessel 2 is cheap enough to just insert into a product (as people do with BeagleBone and Raspberry Pi). And it's open source, so you're free to do so. Unlike BeagleBone and Raspberry Pi, however, we’re going out of our way to make that easy for you.

At the 100s-1000s scale, we are able to automate some simple optimization on the boards and components, thus reducing your end cost for embedded hardware.

We're also offering services to support you as you scale up to production. Manufacturing is opaque and intimidating. It doesn't have to be– we're aiming for a Heroku-like experience of physical product deployment.

Please email us if you’re interested in discussing this with us.

Will Tessel 2 act like Tessel 1?

In a word, yes. For all intents and purposes, using the Tessel 2 should feel the same as using the original Tessel (but more reliable and a little faster). However, there will be a few changes to the underlying technology: original Tessel runs the Colony compiler to turn JS into Lua and is thus “Node compatible”; Tessel 2 just runs Node on V8– so you won’t run into the same incompatibility bugs. You can read more about this in the blog post Moving Faster with io.js.

There will also be a couple of differences in modules: Tessel 2 supports USB modules, so a USB webcam, USB speakers, and a few other modules will move into this sphere. This will allow us to reduce the cost of these modules and get you a higher-quality experience. USB compatibility also lets us easily support new third-party plug-and-play hardware capabilities in a similar capacity to current Tessel modules.

Is it still JavaScript?

Tessel 2 will not only run JavaScript with io.js (a fork of Node), but also support Rust, Python, and potentially other languages.

At the time of writing, we have already run Rust and Python code on a Tessel 2.

Around the time of ship, we will release examples of running Rust and Python on Tessel 2, as well as documentation on library porting for interested members of the community. The same process of plugging in a hardware module and installing the driver from a package manager will apply for these languages. Both languages will eventually be fully supported on Tessel 2.

Multiple language support is made possible by the change in underlying technology on Tessel. We’re excited to see where you take it.

What about low-power applications?

Based on feedback from the original Tessel, we’ve learned that people building connected device systems often need either a relatively high-power system to control and process various inputs and outputs, or they want simple, low-power sensor nodes.

WiFi chips are now more reliable and also cheaper than they were a year ago, but Wifi as a protocol is not designed to be low power. Tessel 2’s power draw is slightly less than that of the original Tessel, but not much.

For people seeking low-power solutions, we’re working on adding that capability to the Tessel ecosystem as well. Expect another big announcement next month.

Open source

Tessel is open source, software and hardware, and we owe a lot to our community. We have open sourced the software work we’ve done so far on Tessel 2, along with the schematics to tessel.io/opensource, and will continue developing in the open. As the hardware is finalized, we will release layouts.

We will be adding more about the architecture of Tessel 2 to the open source page in the next few days.

What will you build?

Our vision at Technical Machine is to build an open and scalable platform for hardware development. We’re reimagining the hardware development process with a focus on creating the best possible developer experience. Tessel 2 is the next piece of that vision.

With Tessel 2, we’re giving you the experience of Tessel, at the price of a Pi, with scaling opportunities that no other development board offers.

For us, Tessel 2 lays out the next leg of the journey on your road to production. We’re waiting in anticipation for you to take your next steps.

Kelsey Breseman Director of Community, Technical Machine

2/23/15– Kelsey Breseman

Jay Henderson spills over with excitement, ideas, and passion. He’s building a code school and a hardware startup both from the ground up right now, trying to improve drought conditions in north Texas, working to bring together his community, and working to teach himself ever more about technology and programming.

Just home from his day job, Jay sat on the couch to Skype me, all smiles and enthusiasm. His girlfriend’s dog, Dylan, jumped up more than once to try to join the call.

Kelsey: What are the main things you’re working on right now?

Jay: The main thing that I’m working on right now is a water barrel app. It was part of the North Texas Apps Challenge. During the summer months, there’s water restrictions in Dallas County. We wanted to create a rain barrel app that tracked how much water you got from your roof, and then how much water you started to dispense on your yard.

The other thing I’m working on is, I’m trying to start a learning lab that teaches people how to do software, and then eventually, blend that into hardware.

Kelsey: Can you tell me more about your water barrel app?

Jay: We met with an environmentalist at the University of North Texas, and he told us about how he made these products for people who wanted them in their home. He said, if there was a way to calculate how much water they’re using, and have an app that displaces the water out as they use it, that would be cool. So we made Barrl.

I went on Adafruit and bought a liquid flow meter, a soil / moisture temperature sensor, and a brass solenoid valve, which we plan to use with to tell us how much water is going into the barrel.

We’re trying to work on that as our first product as a startup, as Intelliflux Labs. We’re hoping to bootstrap it. That’s why we love Tessel, because I have a frontend development background, so the JavaScript makes it quite easy for me.

Kelsey: You said you have a frontend development background?

Jay: My background is in front-end: HTML, CSS, JavaScript, user interface; user experience is my forte. In the last year, I have really immersed myself in full-stack development. Experimenting with that has been fun.

The reason why I want to learn as much as I can is so that I can teach other people who are unaware of what’s going on with JavaScript, and see what they can make it do with physical computing.

Kelsey: Can you tell me more about your learning lab plans?

Jay: About three years ago, I was going to school and working at the Apple store, but courses became really expensive. I couldn’t go to school anymore, and then I lost my job. But then I started to teach myself during my unemployment. I created a Personal Learning Environment, a PLE, started navigating content, learning things that I was personally interested in for programming, and started to get an understanding.

I’d started to see different code schools pop up, and I decided to have a trial run of a code school here.

Eight people paid for it, and then we got ten scholarships from a company. It was really exciting meeting the different personalities. I had a gentleman who was developing in C#, who wanted to learn more about HTML, CSS, and JavaScript. I had a school teacher who knew nothing about programming or anything like that.

I started the course with an introduction to PLEs; I asked each person, what do you want to build? What is it you’re passionate about every day? Your learning is going to grow based on your personal learning environment. I’m teaching you how to go out and find these tools, but in the end it’s going to be up to you. When you’re faced with a problem, you’ll need to reach out to others in the community to get answers.

One gentleman was doing graphic design, and he loved the job, but it was so demanding, and he wasn’t getting a lot of compensation for the work he was doing. During the course, I had a speaker series where I had professionals come up. They spent most of their time answering questions and engaging with the students. One speaker came in and talked about the Myo, different human interactions that are coming out, and that really engaged him [the graphic designer]. He started prototyping UI with some of his graphic UI knowledge, and he just landed a job at the Dallas Morning newspaper here just from interacting with the course.

I helped teach a man who was homeless, in the course. He took it on quite well, and now he’s working in a coworking space, and developing clients with a unique support system.

The testimonials and success stories are starting to come out, and we only did just a trial course.

Kelsey: That’s amazing. What are you planning to do differently with your next course?

Jay: I think I condensed too much into five weeks. We covered a lot of UI and frontend, but I didn’t go into detail with programming logic. I think that the concept of doing something physical for that will be easier to grasp, because you’re dealing with physical components. You can actually see physical objects come to life coding physical elements, such as, if this is true, then have a led blink.

I’m hoping to do more– from surveying students from the course, participants were more on the marketing side, more graphic artists and designers, and I’m starting to brainstorm about how I can help bridge the visual with the programming.

Kelsey: Are you planning to use the Tessel for that?

Jay: One of the things that I want to do with Tessel is to take two or three projects from the projects board and teach that to the class in workshops. Once they get a foundation with JavaScript, introduce them to the things that they can do outside of JavaScript. In which we’ll open up the possibility of creating products.

[Tessel] is actually making it a whole lot easier for software developers to create real products with only a little bit of knowledge of hardware. That’s one thing that I’m trying to do, is give frontend developers and software developers a chance to interface with hardware.