This is my test to check if serial communication work:
I use 3 arduinos nano with MrlComm inside. I connect each of them on serial1, serial2 and serial3 on the mega left or right. I Connect one servo on all nano.
With MyRobotLab, I move 3 servo.
With MrlComm 55, After few minutes, MrlComm 55 break the communication, doesn't work.
With MrlComm 41, i have no problem, the test work all the time. This is called reliability.
So, moz4r, do this test and give your result !!
Dom.
You are always using the
You are always using the MrlComm version which came from the corresponding myrobotlab.jar correct ?
I want to make sure, because people have been confused about this.
The only one "expected" to work is the one which the myrobotlab.jar came with - therefore your MrlComm 55 came from a specific myrobotlab.jar .. and those two things are only used at the same time.
If you switch to another MrlComm - you better be switching the jar too ..
This is one reason why we say to keep different versions in different directories ..
Sorry if this is all old news to you .. just want to be sure
Of course Grog i use MrlComm
Of course Grog i use MrlComm which came with MRL. I'm not a noob...
i just inititiate a Serial
i just inititiate a Serial com to Serial1+2 and was worky without crash or stuck.
Then I started neopixel, and light was beautiful.
Ok can you try to launch minimal things to test
: https://github.com/MyRobotLab/pyrobotlab/blob/develop/home/Calamity/mul…
+ 1 servo
and send a noworky if crash
No offense :) although the
No offense :)
although the noobie avatar throws me off :D
@Grog, no problems. @moz4r,
@Grog, no problems.
@moz4r, your test is not anough because it's only one direction. You test mega -> other arduino
Do MY test and you see. Minimal script is not enought too. Do the test MrlComm -> MrlComm !!
Dom.
MY test
Hi Dom
Could you please provide the script that you are testing with, so that the problem can be reproduced. The more information you can provide about your setup, the easier it is for us reproduce and solve the problem.
There has been a major rework of the Arduino / MRLComm code between version 41 and 55 and I don't think all parts of the code has been tested with real hardware yet.
/Mats
I do an other reliability
I do an other reliability test about 10 minutes
Poll 10000 time a pin / second on the remote serial arduino
( usefull to find dark matter in the galaxy )
and moveto a remote servo based on those values
Result : I didnt find dark matter , but mrl didnt crash
10000 is really too much it was just a stress test.
https://github.com/MyRobotLab/pyrobotlab/blob/master/home/moz4r/Test/da…
And with that you do not know
And with that you do not know how to make a test MrlComm -> MrlComm ... pffff
For me, serial port aren't usable. That it.
Because of this I compile my own MRL based on LAMIAK, without service InMoov because too restrictive.
I prefer encoding services in python in order to remain master in the behavior of my robot.
Dom.