bms001:
You can easily increase this by modifying this line#define REFRESH_INTERVAL 20000 // minumim time to refresh servos in microsecondsin Servo.h
e.g. reduce to 10000 which equates to 100Hz.
How much you can increase this is limited by the need to fit 4 2000us pulses (plus some overhead) in this window.Obviously if you have some conflict between servo and another library then this may make that worse. But on its own I found that making this change gave a significant improvement in terms of getting a quick and smooth response from a quadcopter.
How was it? Better/worse/no difference? What made you choose to use the 'servo' approach instead?
bms001:
You can easily increase this by modifying this line#define REFRESH_INTERVAL 20000 // minumim time to refresh servos in microsecondsin Servo.h
e.g. reduce to 10000 which equates to 100Hz.
How much you can increase this is limited by the need to fit 4 2000us pulses (plus some overhead) in this window.Obviously if you have some conflict between servo and another library then this may make that worse. But on its own I found that making this change gave a significant improvement in terms of getting a quick and smooth response from a quadcopter.
How was it? Better/worse/no difference? What made you choose to use the 'servo' approach instead?
guys
i have tried servo at first but it made conflict with virtual wire so i use now servotimer2 instead
there was no deference what ever i changed pins or change data rate in virtual wire , using servo or servotimer2 alone gave good results but when i just add the virtual wire at the head of the code without even put any code related to it i got constant value intervals as i see in excel sheet which data is brought from serial monitor of D controller values ,when i remove the library it goes fine
i have now used Leonardo board with virtual wire and software servo and there is no deference
virtual wire slowers every thing
any opinion