Timeline for Arduino time keeping using millis() is not accurate or correct?
Current License: CC BY-SA 3.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 12, 2014 at 13:08 | comment | added | TheDoctor |
A more optimized getECG: float eHealthClass::getECG(void) { return analogRead(0) * 5 / 1023.0; }
|
|
| Mar 11, 2014 at 16:36 | comment | added | hawkar | For those still interested, see the comment I have posted on JRobert's answer, I didn't want to reply with my own answer since I don't have one, I have just rephrased the problem. | |
| Mar 11, 2014 at 3:27 | comment | added | EternityForest | Millis() still works on a timer that ticks every 1.024ms, but they added error compensation, in the form of incrementing whenever an error meter variable gets too high. I think it's Roman Black's algorithm actually. So timing should be much closer to 1ms exactly.github.com/arduino/Arduino/blob/master/hardware/arduino/cores/… | |
| Mar 10, 2014 at 15:33 | comment | added | Cybergibbons | Resonators have a poor initial tolerance (often 0.5-2%) and poor temperature stability, but if you calibrate timing loops when using them, they can be fine as long as temperature doesn't move. | |
| Mar 10, 2014 at 14:33 | comment | added | Chris O | @jfpoilpret Ah good to know. Looking at the schematic, this would be the CSTCE16M0V53-R0 Murata CERALOCK device. | |
| Mar 10, 2014 at 9:28 | history | edited | Cybergibbons | CC BY-SA 3.0 |
Add detail on easier way to get constant timing.
|
| Mar 10, 2014 at 9:25 | comment | added | jfpoilpret | Please note that, for the Uno, the clock is not crystal-driven, but just uses a ceramic resonator which is less accurate than a crystal. | |
| Mar 10, 2014 at 9:15 | history | answered | Cybergibbons | CC BY-SA 3.0 |