Need a different way of thinking here:

Archive FastLED Archive
1

Need a different way of thinking here:

void snowFlake() {
if (flakePause > 0) {
if (millis() - lastRun > ((flakeCnt > 0) ? flakeSpeed : flakePause)) {
pushString(0); // push pixels down the string
leds1[NUM_LEDS - 1].r = flakeColor;
leds1[NUM_LEDS - 1].g = flakeColor;
leds1[NUM_LEDS - 1].b = flakeColor;
LED1.showRGB((byte*)leds1, NUM_LEDS);
lastRun = millis();
flakeColor >>= 1;
flakeCnt++;
if (flakeCnt > ((NUM_LEDS - 1) + 8)) {
// flake reached bottom
flakeCnt = 0;
flakeColor = random(20, 256);
flakePause = random(250, 1000);
flakeSpeed = random(15, 51);
}
}
} else {
flakePause = random(500, 3000);
flakeSpeed = random(15, 51);
flakeColor = random(20, 256);
}
}

It does what it’s called, a snowflake … falling, at different speed, random interval, and random brightness (which also affects the trail length.) Problem is, this gets rather convoluted when you have multiple strings on the same AVR as I now have to keep track of multiple variables for each string (otherwise they all display the same exact thing and that just isn’t nice.)

So, suggestions anyone?

2

You want to make a class to represent a line of snowflakes - something like this:

class Snowflake {
long lastRun;
int flakeCnt;
int flakeSpeed;
int flakePause;

CRGB *myLeds;
int numLeds;

public:
Snowflake(CRGB *baseLeds, int firstLeds, int ledCount) {
myLeds = baseLeds + firstLeds;
numLeds = ledCount;
reset();
}

void reset() { 
	flakePause = random(500, 3000);
	flakeSpeed = random(15, 51);
	flakeColor = random(20, 256);
	flakeCnt = 0;
}

void pushString(int led) { 
	while(led < (numLeds-1)) {
		myLeds[led] = myLeds[led+1];
	}
}

// run a snowflake cycle, returns whether or not it changed anything
bool runSnowflake() {
	if( (millis() - lastRun) > ((flakeCnt > 0) ? flakeSpeed : flakePause)) {
		leds[numLeds-1].r = leds[numLeds-1].g = leds[numLeds-1].b = flakeColor;
		lastRun = millis();
		// compiler will turn this into a shift automatically, this makes it
		// clearer why you're doing this
		flakeColor /= 2;
		flakeCnt++;

		if(flakeCnt > ( (numLeds-1) + 8 )) { 
			reset();
		}

		return true;
	} else {
		return false; 
	}
}

};

#define NUM_LINES 5
Snowflake *flakes[NUM_LINES];
CRGB leds[NUM_LINES][NUM_LEDS];

CLedController *controllers[NUM_LINES] = { &LED1, &LED2, &LED3, &LED4, &LED5 };

void setup() {
// do the led setup
// …

// now setup the snowflakes
int offset = 0;
for(int i = 0; i < NUM_LINES; i++) { 
	flakes[i] = new Snowflake(leds[i], 0, NUM_LEDS);
}

}

void loop() {
for(int i = 0; i < NUM_LINES; i++) {
// if any of these calls change led data, changes will become true
if(flakes[i]->runSnowflake()) {
controllers[i]->showRgb(leds[i], NUM_LEDS);
}
}
}

3

damnit, g+ destroyed my indentation :frowning:

4

How does that keep track of the individual strings though? It’s not meant to be one-string-at-a-time. All strings can, at any given point, have a flake going across them, at a random speed, and brightness.

5

So that’s the great thing about classes, each object had its own set of variables in the class, so there’s five sets of counts, speeds , etc… And because each object sets those values independently, you can have different strings at different speeds :slight_smile:

6

Ill explain more when not on my phone in a bit.

7

So basically - what that code does is create 5 snowflake objects, each with its own string of led data to play with, and its own set of values for things like speed and color and such.

This is what I do with my own projects. I create objects for the different types of things that I want to do - and those objects only know about the portion of CRGB array that they’re told about. Then, in my loops I have each pattern object do whatever it wants to do with the leds (if anything, on that round), and then I show all the led data.

8

Ok, that all makes sense. I’ll try it out and work some magic. This will give me a good reason to start working with classes for some of the other stuff. Normally I end up with either a single string (SPI) or two strings (bitbang, and that too gets funky.) Now that I can more easily run multiple strings on the same AVR, I need to start rewriting the code …

9

Yeah, that didn’t work. However, I blame myself for not understanding classes and what not. So this’ll call for some more screwing around on my part …

10

Email me your code - what I threw out was off the cuff before running back out the door - it may have some flaws in it.

11

I will. But, let me try to figure things out first. I’m not ready to give up yet. If I can’t figure it out, expect an e-mail. :slight_smile: