NullColaShip

I guess it really depends on how quickly you can take and process your samples. I've currently got a frame delay of about 40ms, which means it'd be 25fps if it was able to calculate each frame and push it out instantaneously - so it could easily go quite a bit faster. The video in the first post in this thread should give some idea of how fast it is - particularly the rainbow fade sequences, which run very smoothly (at least in person - the video might not look quite as smooth given its limited framerate).

Funny you should say that - I started in January, and one of the reasons I started the project was to improve on last year's attempt, which used dirt cheap non-addressable strips to make a fixed message change colour. When I'm back up and running on a new laptop, I'll refactor the code (quite possibly changing the text display code to use the newly-written image display code, since the data is entirely compatible) and make it available in my Git repo. At the moment, I have a new function that simply draws an 8x8 monochrome image, with optional background transparency. I then draw a base

I finally found the time and effort to finish sewing the display together (the LED strips are now sewn onto a black felt backing, which can be rolled or folded up for storage), and I've coded up a quick Christmas sequence. I'll tidy up the code (I wrote some new functions to output 8x8 images, which I can make a bit more generic) and push it up to Github some time in the not too distant future (I need to get a new laptop), but in the mean time, here's a video of it in action (as part of a window display with a few other cheap lights):

I'm keen for a few - no idea what I'll do with 'em, but can't hurt to have 'em! Kits would be fantastic; otherwise it might take me a while to scrape together enough components to make shipping certain components to me in New Zealand cheap enough... Bluehash ------------ 2 Rob ------------------ 2 Cubeberg------------ 5 SugarAddict--------- 5 GeekDoc------------- 2 NullColaShip---------3 [international]

Hi Rob, That looks great. When/how can we get hold of them? I imagine there'll be strict per-person limits since your initial run is quite small, but I'd be happy with one to start with... here's hoping you can work out a way to produce these in moderate quantities (and at reasonable cost)!

Looks like this is coming along nicely - Rob, do you have an estimated price (for an assembled booster) yet? I have at least one potential use lined up already, though it's contingent on a couple of other things falling into place - but I'd certainly be keen to grab a couple of these when they're ready! Also, one quick question: how easy/difficult would it be to make the Launchpad be the device which sends out HTTP requests rather than being the server listening for them? My main idea would require it to be making requests (even if it can't receive the response, though it'd be good if it

The framerate depends on how many LEDs you're driving - they're not addressable in the sense that you can say "change LED X to colour Y" but rather in that you send it a sequential list of colour values, with the Nth LED along the strip taking the Nth colour value you spit out in a given update cycle. Therefore, the fewer LEDs you're driving, the faster you can pump out a full frame - i.e. the faster the maximum framerate. I'm achieving pretty smooth animation on my 32x8 display, but I can't recall what kind of framerate it's actually pulling - check the video in the first post if you want t

I had some spare screw terminals kicking about, but they weren't quite the right pitch - so I tore the plastic off and used the individual terminals within. They were just slightly too wide to comfortably fit side-by-side without touching, so I left a single gap between each one. I've contemplated using some Sugru to turn them into a solid block again, but they seem fine as they are so I'll probably just leave them bare.

I've been too busy to do much on this project lately, but since I just received the stuff I bought from the 43oh store, I couldn't resist transferring everything to a ProtoPad BoosterPack. Here's what it looks like now: 1. Microchip 23K256 32KB SRAM chip (for storing pixel data) 2. 5V input for LED array 3. To LED array (from left to right: GND, data, clock, +5V) 4. Socket for Bluetooth module (not yet wired in), made from stackable headers with the pins bent 90 degrees I have the Bluetooth module, but haven't yet wired it in or written any code to support it. The module I have is

Thanks - I have much more planned for it (though I'm not sure how much I'll be able to do). I'll certainly add it to the next POTM if I notice one's happening

Bump (initial post updated - including video!)

I finally managed to get the 23K256 up and running, and now have my LED array code storing pixel data in the 23K256. The pixel data is streamed directly from SRAM to the LED strip - primarily for performance reasons (and because there's no reason to pull chunks of the data back into the MSP430), though I also had trouble trying to read the pixel data out in chunks for some reason. I'll probably look into that again at a later date, but for now I'll continue to focus on adding functionality to the display. I also boosted the CPU clock to 16MHz, which is giving a pretty fast refresh rate.

I've just ordered a few of these, and I intend to use one to store pixel data for a display built from addressable RGB LED strips - I'll keep an eye on this thread and report back on progress once I have the chips, but I'll also be posting updates in my project thread.

Thanks for the suggestion - I've ordered a few 23K256 chips from Microchip Direct (Mouser wanted to charge nearly three times as much for shipping, in addition to charging twice as much for the chips). They reckon they'll be here by Monday (maybe Tuesday, depending on which timezone their estimate is in), so hopefully I can crack on with this project next week!

Constants aren't really a feasible option due to dynamic nature of what I want to do with it, though I might be able to do some of it by calculating output on-the-fly from a set of constants rather than generating an array of pixel data in RAM and then pushing it out... Rusty, do you have any suggestions for cheap, reasonably easy to use SPI or I2C RAM chips? Preferably something for which sample code is available, though I could probably cobble something together as long as there's decent documentation available.