It would be an understatement that we here at Maniacal Labs loves all things LED. With our first major product we brought you the AllPixel, allowing super simple LED control from anything with a USB port and Python. At the time one of the major reasons we went that direction was that the things we wanted to do were too big (like Colossus) for something like the original Raspberry Pi to handle. But a lot has changed in the intervening years and we found ourselves building more and more with a Raspberry Pi at the heart, running the show.

But connecting LEDs to the Pi and providing power was never as simple as it was with the AllPixel. We’d have some jumper wires hanging off the GPIO header manually wiring power into the LEDs. So, in a fit of “spend a bunch of time to save a bunch of time” we designed the PiPixel kit.

It provides all you need to drive LEDs directly off your Raspberry Pi with as little hassle as possible: Power input, signal level shifting, and data/power output. Data output is selectable and available on SPI, GPIO 13, and GPIO 18. Via these data outputs and our BiblioPixel LED animation framework, it can handle the control of the following LED types:

  • APA102 (“DotStar”)
  • SK9822
  • WS2801
  • LPD8806
  • WS281x (“NeoPixel”)

As you may note, this is not as many protocols as the AllPixel, some just cannot be handled by the Raspberry Pi, but this does cover most of the popular chipsets.

The PiPixel will also work with any other library that utilizes the SPI port (/dev/spi0.0), GPIO 13, or GPIO 18. But, of course, we highly recommend that you check out our feature rich BiblioPixel library.

Best of all, we designed the PiPixel to be super cheap and you can pick one up on Tindie for a mere $8.50.

Of course, as with everything else we do, the PiPixel is open source and you can checkout all the design files over on GitHub.

Assembly and usage documentation is available at the above linked GitHub repo, but be sure to check out the video below for a full walkthrough.

For questions, comments, or support, head over to our Maniacal Labs Users Google Group.

Friend of Maniacal Labs and contributor to BiblioPixel, Maksim, has created an awesome full RGB 7-segment display module based on the WS2812B (NeoPixel) LED.

These awesome little modules simply snap together and provide full RGB control over each segment of each digit, allowing for some pretty awesome effects. Plus, they are certainly the easiest 7-segment modules to interface that we’ve even seen. He’s running a crowdfunding campaign right now over at CrowdSupply.

But we’re not just writing this to promote a campaign… Maksim was kind enough to send over some pre-production units so that we could get them working with BiblioPixel with as little fuss as possible. We’ll just let this demo video do the talking for how that turned out:

The new NeoSegment module for BiblioPixel automatically handles dealing with writing numbers and text to each digit and provides a full built-in font covering a good facsimile of every character. For all the details on using it with the NeoSegment, check out the GitHub Repo.

We wish the best of luck to Maksim on this and future projects and hope to see awesome projects made with BiblioPixel and the NeoSegment!

While we’d love for everyone who uses our code or hardware to send us an email and show off their projects, not everyone does. So we keep an eye on the internet for cool projects that use our products. Last night we were happy to find the project shown in the video above. The video description confirms they use BiblioPixel and we’re pretty sure they mentioned the AllPixel as well. But even better, this thing looks so incredibly similar to Colossus, just a little bit smaller. Same resolution, same basic construction, similar power routing… We put so much time into that project that it’s just fantastic to see someone else build their own!

Great work guys! If you see this, please send more pictures and video! 🙂

BiblioPixel supports a wide array of hardware through its driver system but one of the biggest annoyances was always having to actually have that hardware in order to test your code. That’s why, very early on, we added the visualizer to the library which worked well for small displays, but was never great with high pixel counts or fast framerates. For over two years we’ve wanted to replace it with something better but could never settle on something that would have high performance and run on every operating system that BiblioPixel already does.

But at the 2015 SparkCon, here in Raleigh, we met Michael Clayton who I actually now work with at Red Hat. At SparkCon, Michael showed off his awesome Kimotion project which shows high performance, WebGL-based, animations in a browser using data captured from an XBox Kinect sensor via a Python server. This was clearly perfect for replacing the visualizer in BiblioPixel but the idea remained just an idea for another year while other things were worked on.

But after some further discussion with Michael we decided that, mainly because he is a WebGL wizard, we could probably knock out a new visualizer in a day. So we met up one Saturday and, in about 9 hours, threw together something that put the original visualizer to extreme shame. We call it SimPixel.

This baby will do buttery smooth 60 frames per second with thousands of pixels without skipping a beat. We actually use it to stress test BiblioPixel!

Being that it’s written in JavaScript and WebGL, all you need to run it is any relatively modern web browser. BiblioPixel communicates with it over super fast websockets which actually means you can even connect to another system on the same network for the pixel data.  Check out Michael’s blog post for more information about how SimPixel works. In general, we use it locally on the same machine. This is what enables the bibliopixel demo command in the new BiblioPixel v3.0 to show off the functionality of BiblioPixel without any required hardware. It’s so easy even, that you can go from nothing installed to seeing awesome pixels in 16 seconds!

And because SimPixel is web-based there’s not even anything to install. In the video above, BibliPixel simply opens a browser to SimPixel.io which is all you need. If you do, however, want to see the code (it’s open source, of course!) or run it locally you can find all the details at the SimPixel GitHub repo.

Because of SimPixel, you can easily see BiblioPixel in action with absolutely zero code.

pip install bibliopixel
bibliopixel demo

That’s it! Install BiblioPixel and then run the demo! http://SimPixel.io will automatically launch and connect to the BiblioPixel animations that are now running. SimPixel is also fully supported in BiblioPixel’s new “Projects” feature, meaning you can create your own setup with nothing but some basic configuration.

Stay tuned as we’ll be highlighting more of BiblioPixel’s great new features soon 🙂

It’s amazing to think that nearly 5 years ago, before Maniacal Labs even, I released RPi-LPD8806, which would eventually become BiblioPixel, our pure Python pixel animation framework. Back then it was only for the Raspberry Pi and only for LPD8806 LEDs… and I barely knew Python at the time!

Over the years, it has morphed and grown into a massively capable framework with support for so much more than that original library. But no year in BiblioPixel’s history has been more significant than this last year. I’d always had future plans for it but mainly updates happened as necessary to support other projects I was working on. The most significant, of course, being the AllPixel. But for more than a year now, Tom Ritchford (who actually first contacted me because of RPi-LPD8806 way back in 2013) has taken the vision of BiblioPixel to a whole new level. And if you’ll believe it, this is just the start.

So, who’s Tom? In his own words:

Tom Ritchford has been writing computer programs for a living since 1979. His alter ego Tom Swirly does music, lights and sound with computers for fun. He currently lives in Amsterdam where he is writing a book.

After thousands of emails and hundreds of commits, BiblioPixel 3.0 is faster, cleaner, stronger, happier and in every conceivable way better than it’s predecessor. You will go from zero to blinking lights in under a minute!

What’s new? Just to name a few highlights…

  • Full Python 3.4+ support! (In fact we dropped Python 2 support)
  • Near complete core rewrite that’s faster than ever!
  • Run BiblioPixel with nothing more than a config file, zero code!
  • SimPixel, our all-new WebGL based visualizer. Get started with no hardware at all!
  • Tight integration with BiblioPixelAnimations, our repository of pre-made animations.

Even with all this, it’s still about 90% compatible with all your old code and only requires minimal modifications.

Get started and install now!

Footnotes

  • Talk about blast from the past, here’s a demo video I posted of RPi-LPD8806 from October, 2012:
  • Original post (on a defunct website) I made about the RPi-LPD8806 library which is how Tom got involved.
  • So, what does nearly 5 years of work on BiblioPixel look like? Thanks to gource we get this pretty video: