Disclaimer: I have nothing to do with these guys nor do I get anything from posting this. Pyroterra just announced V3 of their Visual Poi: 80 pixels, 12-bit, 49cm long. Also saying “thousands of images possible” but are being tight lipped about how … $1,490 for two sticks. FB album: https://www.facebook.com/media/set/?set=a.744261892292081.1073741834.337143916337216
I just ordered 3 pair. @ $ 1, 490 I was silly not to. dies
With a large enough storage device millions of images could be possible.
They’re not saying how though. But they did implement some kind of playlist which they are calling “sequences”. You can program up to four sequences of images and select which one to use during a show. This is on my todo list as well, but why limit it?
couldn’t it just be read from an SD card? suppose all you have to do is encode it one line at a time and then just read it out one line at a time
I was thinking of using folders and use the leds to in some way help move through the file system. But I fear it would be over complicated.
I’m just guessing here so feel free to shoot me down, if the poi have a row of say 20 pixels, its literally just a matter of encoding 20*3 RGB values per row in a text file, then read this file out sequentially
read line 1
write to led strip
diplay
read line 2
write to strip
display
…
…
…
and so on,
then just pick another filename to open with a case command. In theory a 8GB card could hold thousands of patterns, in fact go a step further call each pattern a numeric value and have the arduino generate a random number to call it’s data from the SD card, this way the arduino doesn’t need to be aware of file names.
Uh, that’s exactly what I’m doing. However, I don’t work with text files for the LED data - too much overhead. I pre-process the images into binary data files, one byte for each of R, G, and B. Then I read in the exact amount of pixel data that I need straight into the leds[] array.
My point about Pyroterra is that with V1 and V2, everything was stored in the microcontroller’s memory, so they were limited in how many they can store. Now they’re saying V3 can store so much more, but they don’t say how. They could’ve moved away from using the controller and using an SD card, I don’t know that. And until they release details about that, all anyone can do is speculate.
I R Mind Reader 
lolol
Not enough ui on prop = advanced features suck. And who wants to HAVE to carry their smart phone with them all the time…
Could they be using wireless streaming?
Highly doubt it.
I’m streaming out to six different light bars with different animations on each, 52 leds per pole and it’s silky smooth, using artnet to an uno with wifi shield on each pole, granted they are stationary so you don’t get a stroboscope effect
Stationary objects is one thing. Something that’s spinning at a high rate and in all different directions, is a whole different ball of wax.
If their previous versions are any indication, they stuck with reading in bitmaps. The possible difference here is that they moved away from using the controller’s available memory and use an SD instead. OR, they added additional memory to the unit. That’s also a possibility.
My money’s on an sd card
That’s my guess too. It’s the same thing I’m doing.