So I originally saw a project on here to make an LED coffee table, and read into it's workings and decided to buy some indexable (mappable) LEDs based on WS2812b to achieve this. I quickly laid them out in a grid layout 10x5 and began programming from scratch.
After I got a few patterns going, I realised there was potential for it to display text, so set about creating character maps for A-Z and 0-9 to display, based on a standard 7 segment display, and was pretty impressed with the result. (You'll have to excuse the overexposed image, it looks better in real life.)
I also do a lot of theatre lighting in my area and at a technical meeting for the next production it was discussed that it would fit the scene/song well if they utilised their existing projector to show a stock exchange style ticker. Knowing that I was already on the conceptual route to this, I volunteered to make a large scale version for the production.
It is constructed from free hardboard from the IKEA recycling area (the front panel is actually half of the back of a wardrobe, it had zero damage, so not really sure why it had been discarded) which was then stapled onto a 2"x1" batten frame around the edges.
As the final product is 42x7 (294) pixels, the most boring part was drilling the 294 holes required to mount them! So I plodded through doing a few columns every night for a few nights and was finished soon enough!
Due to the way I incrementally built up the code, bolting functions on as it developed, I started to encounter major memory issues along the way, and had to trim a lot of non-essential parts of code out and radically change other functions in order for it to run correctly. This included removing lower case character maps (not used for the product in this setup) and changing the pixel layout from being stored in a 42x7 INT array into a mathematical equation which returned a pixel ID for any given x, y coordinate. I decided to stick with the memory constraints of the UNO as it improved my programming skills by analysing the memory usage and seeing where savings could be made, although as time became incredibly tight (it wasn't working in time for our first production, unfortunately) I did consider 'cheating' and buying a MEGA just to get it working, but I'm glad I stuck with it.The Final Product
As you may be able to see from the video, the last 44 LEDs (6 full columns and the top 2 LEDs of the 7th column on the left) have a slightly different colour temperature on the blue LED, which is likely due to the fact they were from different manufacturers (I bought 1 string of 50, then 2 more to expand the prototype to half size, then 3 more to finish it off). In the future, I'd recommend buying all required strings in one go, to increase the chance of them all being from the same batch.
There isn't really much more to add for now, but feel free to add any comments for improvements or queries as you wish :)