Subject:
|
Lnp collision under Linux
|
Newsgroups:
|
lugnet.robotics.rcx.legos
|
Date:
|
Mon, 11 Jun 2001 19:05:40 GMT
|
Viewed:
|
1417 times
|
| |
|
|
Hi !
I'm having trouble trying to make a Linux box (Mandrake 7.2 kernel
2.2.17-21mdksmp) communicate with my RCX. My ultimate goal is to try and
make the PC control the RCX via infrared (so as not to suffer from the RCX
memory limitations). To do so, I must have a fast two way communication. As
stated in the subject, the problems arrise from collisions.
1) I'm wondering: are the tower and RCX supposed to be full-duplex-able
anyway ?
2) Since I was unable to send more than a couple of chars without getting an
"inter-byte timeout" message from lnpd, I changed the
LNP_BYTE_TIMEOUT_FACTOR in lnpd+liblnp/lnpd/lnp.h from 3 to 10 (after a few
tries with smaller values) and it seems to work fine.
3) Then, I had hangups from the PC side wich would never get out of a
lnp_integrity_write called from within a handler. I solved that by moving
the call to some less-appropriate-but-not-handler function, but it still
occured from time to time. I then tried what Martin Cornelius had suggested
in post #552 and, once again, it seemed to work.
4) Now the trouble comes from the RCX side : if I do not wait long enough
between messages, the RCX hangs up very badly (battery removal needed). The
"fun" thing is : this "long enough" depends on the timeout_factor I had to
increase (cf point 2) and is around 0.2 secs wich is *way* too long for
remote robot control.
I tried to find posts that could be of help, but found few answered ones
save for #552 (see above) and #827's patch which didn't make any difference.
Did I miss something ?
In a nutshell : I get RCX hang-ups, probably from collisions, if both RCX
and PC are emitting at too small intervals. The exact duration depends on a
timeout factor wich I had to increase or no messages could get through. If
I'm above this interval, everything works fine.
NB: These problems occur both in close and far mode, with lights on or off,
windows open or closed, on several PCs & RCXs and in different rooms.
Any help would be very appreciated.
|
|
Message has 1 Reply:
 | | Liblnp bug found and squashed ?
|
| This is no good: I'm beginning to talk to myself ;] This is a follow-up on my previous post, and (hopefully) an answer to problem number 3. 'Bit technical I'm afraid. I'M STILL STUCK ON THE OTHER PROBLEMS (as of tonight anyway it's 23:48 and I'm at (...) (23 years ago, 14-Jun-01, to lugnet.robotics.rcx.legos)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
