Jump to content
43oh

Adopt a noob?


Recommended Posts

I know I've "proposed" some projects in the forums, and I've come to realize that pretty much a ALL of that is over my head. Blinking lights at a Dragon's breath, using someone else's code isn't exactly something to rejoice over. Granted, it looks cool and I may be even able to claim a bit of pride since I did tweak the code a little for some added drama, but that is just added drama, and it isn't getting me closer to the RFID car security system, simulating an engine... or even fuel injection. I find myself playing more with the variables than trying something new.

 

I'm having issues making the jump from "this" to "that", so, I'm hoping someone with a good grasp on things may be willing to waste a bit of thought with me. Mind you, nearly everyone here seems to have few issues moving back and forth between architecture or capability, primarily because "you" have a grasp on it. Back in the day, I may have animated the MCP (from Tron, and it sucked and was a hobby) and such, but an Apple computer is a bit different than what I'm looking at.

 

I see the benefits of doing things like LED cubes and such- they are "cubes' and all the math and such fits in to shift registers and like. I've 50 74HC595 shift registers on order (sucker changed my bid from $1 for 5 to $10 for 50, but it's okay: I can afford it, and I'll dispense a few if anyone needs). I bought them because my nephew and a few others like the cube displays, and those are easy-enough to build... though it also means I need to buy more MSP's. >sigh< However, a cube is a nice ^2 thing, no matter which way you slice it, so it makes sense using shift registers: shift bits and instantly things change to the next line, row, column....

 

But, what caught my eye, was why there aren't any pyramids or any other complex shapes out there. I mean, other than the spinning LED's for POV and such, it's 3x3x3, 4x4x4, 8x8x8 et cerebral. I immediately realized it was because the "math" doesn't work well with shapes whose sides are not a derivative from ^2. It's not fluid, per se. Though one could map a 5x5 array to a solid pyramid with 25 LED's: 12 on the base, 8 at the next level, 4 at the next and one at the top, 5^2=25, being a "square", doesn't permit simple bit shifting to work since the geometry isn't "square" or "cubed". That immediately presents issues for animation.

 

Another thing I thought of about was pulling and squishing a proper cube in to a "diamond"-like shape, but then it would only be regular cube, pulled and squished.

So, maybe someone pushing me would be helpful. And given the thoughts I presented above, that may push someone else to help me? :)

 

 

So, if interested, PM me, and we'll start the discussions. Otherwise, just reply here and we'll see where it goes.

Link to post
Share on other sites

i usually order jelly bean stuff from tayda, they arent as cheap as some others on ebay but i can get stuff from them to new zealand within 3 weeks. dealextreme has some nice stuff now too but they take like 2 months. shocking.

 

on the sparkfun free day i got one of their id-20 rfid readers, used it with an arduino and some really cheap tags from ebay, doubtful it would work on 3.3v.

 

if you need a couple more msp430s to play with you could order some samples, like msp430g2253 and msp430g2231.

 

perhaps try a led matrix display that you can write afew characters on, or write up conways game of life

Link to post
Share on other sites

The thing that I've found works really well - if you've got a big project - bite off one tiny bit at a time. My first project was an RC car with ultrasonic sensors. I worked out Bluetooth communication first, then the sensors, then controlling the car, and then I combined all the pieces into one project. That being said - an RC car conversion isn't the hardest project - so something like an LED cube (maybe add something else in for complexity - MSGEQ7 maybe?) would be a good start.

Link to post
Share on other sites

This will come out sounding wrong....

i prefer stripboard compared to that pad per hole stuff, but i dont know if it would be any good for a matrix, perhaps you could learn to use eagle. seeed does 10x 50x50mm boards for 10$.

So, I'm supposed to order boards and wait for how long, before I can play? ;)

(like I said, it'll come out sounding wrong)

 

Radio Shack is 1/4 mile away and I'll be back in 15 minutes.

Link to post
Share on other sites

I always like to build a prototype first - I only order boards if everything works and I'm sharing or need several. I think that people who have been doing this longer are more patient. I hate waiting for everything. I stock up on parts from ebay, tayda, etc. To prevent expensive RS trips. Takes planning though.

Link to post
Share on other sites

My experience with attempting to build a 5x5x5 cube.........

 

Plan, Plan, Plan ahead.

Building it from scrap parts does not turn out well, I tried , and I failed.

 

I got it started, but the wire I had on hand was not stiff enough for the build, My soldering job was horrible.....enameled wire sucks to solder. and the cheap leds that I had on hand, 1/4 were burned out/DOA/New?!

 

What I did learn and may help:

build a jig, it will help when trying to solder each level

use premium LEDs, you don't want to build the entire thing to find a dead led in the middle

One step at a time, and take your time! you dont want to spend a bunch of money and turn it into a turd pile.

Think outside the box.

 

 

I plan on coming back to my failure next year, but I am making a geometric shape out of ping pong balls as the LED diffuser, connected together with stiff plastic tubes as the frame.

 

 

 

To get a good start, I would start with LED control, either SPI/shift register, or TLCxxxx led driver

 

Just remember, one step at a time, and if that step is to big break it up to smaller steps.

 

Just My 2cents.

Link to post
Share on other sites

 

But, what caught my eye, was why there aren't any pyramids or any other complex shapes out there. I mean, other than the spinning LED's for POV and such, it's 3x3x3, 4x4x4, 8x8x8 et cerebral. I immediately realized it was because the "math" doesn't work well with shapes whose sides are not a derivative from ^2. It's not fluid, per se. Though one could map a 5x5 array to a solid pyramid with 25 LED's: 12 on the base, 8 at the next level, 4 at the next and one at the top, 5^2=25, being a "square", doesn't permit simple bit shifting to work since the geometry isn't "square" or "cubed". That immediately presents issues for animation.

 

 

This got me musing - why not make non-cubic LED structure.  

A LED tetrahedron should be quite manageable. 

Electrical wiring could be just the same as a LED cube - (columns of LEDs and levels - select a level and turn on all the LEDs

on that level by driving appropriate columns)  Only difference would be only the first level has all the LEDs on it.

(Possible optimization to use fewer levels is obvious, but wiring might be a bother).

 

Physical construction - not sure if much different in difficulty than cubic lattice.

Presumably would use a jig, as with cube, not sure if triangular support structure would be any harder to build between planes. Once constructed it might be stiffer than a cubic lattice (no trapezoid deformation).

Repair (e.g. replacing a bad LED) might be more difficult (less space in structure).

 

 As far as programming - the coordinate system seems pretty easy

(use triples of (p, q, r) - except unlike a cube, for any given level p, q can only run from 0 to p, and for any column q, r can only run from 0 to q).

 

Usual primitives - set a led, read LED, set a line (along one of the 3 axes), set a plane (perpendicular to one of axes), 

shift along an axes

 

Frame buffer - could just use regular 3 D matrix (waste space), or store in linear array and use table or/explicit calculation to translate coordinates into offsets.  (Either way - hidden behind primitive access functions).

 

 

Mapping something like a font onto such a grid might present some challenges,

but just coming up with abstract animations/blinkenlights shouldn't be that hard.

(cascades, random walk, etc.)

 

 

Caveat - while I have read about LED cubes, I haven't built or programmed one, so these comments are just from theory.

 

"In theory there is no difference between theory and practice, but in practice there is."

- Jan L. A. van de Snepscheut

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...