Hi,
I want to control some LED by a sinus signal and was trying around a bit.
Checking the sin() function over the serial plotter I was wondering why the signal has just 5 or 6 steps and the resolution was that poor.
Code as followin:
void setup()
{
Serial.begin(9600);
}
void loop()
{
Serial.println(sin(millis()/150));
}
norpen:
I want to control some LED by a sinus signal
Why? That only makes our (eyes) non-linear perception of brightness worst.
What do you mean by " signal has just 5 or 6 steps"?
void setup()
{
Serial.begin(9600);
}
void loop()
{
Serial.println(sin(millis()/150));
}
Let's see.... First read millis(), which is a monotonically increasing integer value (well, more often than not) - 1, 2, 3, 4, 5, 6....
Divide that by 150, producing another integer value
Pass that integer value to sin(), which takes a single argument, which is an angle expressed in radians - 2*PI radians == 360 degrees.
So...
For millis() anywhere from 0 to 149, this reduces to sin(0), which == 0
For millis() anywhere from 150 to 299, this reduces to sin(1), which == ~0.84
For millis() anywhere from 300 to 449, this reduces to sin(2), which == ~0.91
For millis() anywhere from 300 to 449, this reduces to sin(3), which == ~0.14
For millis() anywhere from 300 to 449, this reduces to sin(4), which == ~-0.76
....
Look familiar?
It really helps a lot to read the documentation for a function, and look at some examples, so you understand HOW it works, before using it.
Regards,
Ray L.
thanks a lot for your help
If you do the division in floating point you get a much smoother curve:
Serial.println(sin(millis() / 150.0));
If you do the division in floating point you get a much smoother curve:
True, but the curve depends in a nonobvious way on the Baud (sampling) rate. This program
void setup() {
Serial.begin(115200);
for (int i = 1; i < 300; i++) {
Serial.println(sin(millis() / 150.0));
}
}
void loop() {}
Produces this curve:
