LIRC Stops receiving command when playing video

[<DBakker@ar...>]

For some reason lirc stops receiving commands when I play video from any 
lirc enabled video app (xine, mythtv....). Im using the latest cvs of 
everything. I have debug on everything and dont see any errors. LIRC will 
send commands no prob and I can connect into it. If I restart the video 
app, LIRC will receive commands for about 3-5 minutes and then it stops 
receiving. Im using a home brew receiver with transmit diode on serial 
port 1. I dont see any IRQ conflicts. Any ideas?

Re: ati remote UHF controller (lirc_atiusb) don't work

[<lirc@ba...>]

Hi!

"ygosset@..." wrote:
[...]
> But for me, it's the same thing, that doesn't work.( in lirc-0.7.0pre4 ).

The error message that you have quoted does not exist in 0.7.0pre4.

[...]
> Can I try kernel 2.6 ? I can make all test with different kernel at
> home, I'm use to do it.

Please use 2.4 and 0.7.0pre4 for testing and give us a detailed problem =20
description.

Christoph

=?iso-8859-1?Q?Re:_ati_remote_UHF_controller_(lirc=5Fatiusb)_don't_work?=

[<ygosset@ma...>]

Hello.=0D=0A=0D=0ABut for me, it's the same thing, that doesn't work.( in lirc-0.7.0pre4 ).=0D=0AI can't find the revision 1.22, in official/stable version of lirc, or=0D=0Ain the  beta version.=0D=0A=0D=0APersonaly, I don't want to use infrared and serial ports, because I've=0D=0Aalready a device that can be work (perharps ...:( )=0D=0A=0D=0APlease help, I can be an another-betatester, I've only notions in=0D=0AC/C++, I don't want to make reverse engineering (I haven't time :-( )=0D=0A=0D=0ACan I try kernel 2.6 ? I can make all test with different kernel at=0D=0Ahome, I'm use to do it.=0D=0A=0D=0AThanks.=0D=0AYvan.=0D=0A=0D=0A---------- Ent=EAte initiale -----------=0D=0A=0D=0ADe      : lirc-list-admin@...=0D=0AA          : lirc-list@...=0D=0ACc          : =0D=0ADate      : 02 Apr 2004 22:59:00 +0200=0D=0AObjet : Re: ati remote UHF controller (lirc_atiusb) don't work=0D=0A=0D=0Agmt 0Hi!=0D=0Agmt 0=0D=0Agmt 0Paul Miller "pmiller9@...
 ceforge.n
et" wrote:=0D=0Agmt 0=0D=0Agmt 0> I just committed a change to lirc_atiusb to add support for this=0D=0Agmt 0> "get_queue" function.  Please let me know if this change does=0D=0Aor does not=0D=0Agmt 0> solve the problem.  (cvs revision 1.22)=0D=0Agmt 0=0D=0Agmt 0I have just undone this change because it makes no sense.=0D=0Agmt 0The OP is obviously using lirc_atiusb with a lirc_dev from=0D=0Alirc-0.6.6  =0D=0Agmt 0which does not work. This is not very surprising.=0D=0Agmt 0=0D=0Agmt 0Christoph=0D=0Agmt 0=0D=0Agmt 0=0D=0A=0D=0A=0D=0A----------------------------------------------------------------=0D=0A<Publicit=E9>=0D=0AHAUT DEBIT : PRIX EN BAISSE ! =0D=0AInternet Haut D=E9bit Illimit=E9 =E0 partir de 14,90 E/mois ?=0D=0AUne offre n9uf telecom ! =0D=0AInscrivez-vous vite, en cliquant ici : http://adsl.neuf.fr =0D=0A

Re: Analysis of pulse-space sequences in Mode 2

[<lirc@ba...>]

Hi!

Julio C. Martinez "julio@..." wrote:
[...]
> got the remote working, except that I cannot get to differentiate repeated
> signals from new ones -- the rep count is always 0.
>
> I understand exactly how the Pioneer SR-90 encodes its signals, but am not
> able to express it with the config parameters that are available.

Have a look at the VXX2801 config file. That's the way how to encode it  
with LIRC. This was a very recent change to LIRC, so I don't have any  
idea if WinLIRC already has merged support for it.

If not you will have to use your raw mode config file.

The reason why you sometimes see signal that are more than 180k us  
apart, simply is that WinLIRC could not match the input to any button in  
the config file. The reason why the repeat value always stays 0, is  
somewhere in your config file, which you havn't provided.

Christoph

Re: Analysis of pulse-space sequences in Mode 2

[<lirc@ba...>]

Hi!

Julio C. Martinez "julio@..." wrote:

> Forgot to ask how to access "remotes/generic/NEC.conf". I guess this is not
> in the database and needs to be obtained from "the distribution", which I
> assume is something inside a "tar.bz2". Getting stuff from a tar.bz2 will
> involve quite a bit of setup work and learning for me (and I assume for most
> Win users). Is this file available by itself, or zipped somewhere?

You can browse the CVS on SourceForge.

Christoph

RE: Analysis of pulse-space sequences in Mode 2

[Julio C. Martinez <julio@vt...>]

Forgot to ask how to access "remotes/generic/NEC.conf". I guess this is not in 
the database and needs to be obtained from "the distribution", which I assume 
is something inside a "tar.bz2". Getting stuff from a tar.bz2 will involve 
quite a bit of setup work and learning for me (and I assume for most Win 
users). Is this file available by itself, or zipped somewhere?

Thanks!

>===== Original Message From lirc@... (Christoph Bartelmus) =====
>Hi!
>
>Julio C. Martinez "julio@..." wrote:
>
>[...]
>> LIRC home page. I have been successful in learning and using a couple of
>> my remotes, but not the one I am particularly interested in, a Pioneer
>> SR-90 steering wheel remote control. I can't seem to find a config file
>[...]
>> WinLirc insists on the signal not being consistent. However, I copied
>> into a spreadsheet the sequences and they do seem consistent for the
>> same button. I have attached csv files with the data for two of the
>> buttons in the remote.
>
>Looks like standard NEC code (see remotes/generic/NEC.conf in the lirc
>distribution). I don't quite understand why WinLIRC has problems with
>it. Indeed there are Pioneer config files in the database that were
>created just fine using WinLIRC.
>You could try modifying one of the existing Pioneer config files.
>
>Christoph
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by: IBM Linux Tutorials
>Free Linux tutorial presented by Daniel Robbins, President and CEO of
>GenToo technologies. Learn everything from fundamentals to system
>administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click

RE: Analysis of pulse-space sequences in Mode 2

[Julio C. Martinez <julio@vt...>]

Thank you. I found the "technical details" page at WinLirc.SourceForge.net, 
and that was quite useful in trying to figure things out. I also looked at 
some of the Pioneer remotes in the remote database and one of them containing 
raw codes , which I didn't know could be used, was very useful. I got the 
remote working, except that I cannot get to differentiate repeated signals 
from new ones -- the rep count is always 0.

I understand exactly how the Pioneer SR-90 encodes its signals, but am not 
able to express it with the config parameters that are available. Essentially, 
a typical press of a button will have the following sequence:

Large Space - as is expected
Pulse 8575 - header pulse?
Space 4200 - header space?
Pulse 580 - lead pulse?
0x4AB5 regardless of which button
0x26D9 for 10 of the eleven buttons in the remote, 0x16E9 for one oddball 
button.
Pulse 580 - intermediate pTrail?
Space 25500 -intermediate gap?
Pulse 8575 - intermediate repeat pulse?
Space 4190 - intermediate repeat space?
Pulse 580 - intermediate lead pulse?
0x0AF5 regardless of which button
0xXYZT 16 bits that are unique to the button pressed, i.e., the code

On a very short press of the button it all ends here, if repeated signals are 
generated, then it is exactly as above, except that the "Large Space" that 
begins the sequence is 25500 in length.

So whether the large space is 25500 then it is a repeat. If the large space is 
larger, then  it is a new key press.

So there is consistency and suficieny of information to determine which button 
was pressed and whether it is a repeat or a new signal. In the sequence of 
pulses quite a bit is common among all buttons. Is there a way to configure a 
config file to represent this? As I said, I'm currently using raw mode and it 
works, but cannot get repeat counts and sometimes misses sending me a signal. 
If I keep the button pressed, WinLirc will send me signals every 180000 which 
is what should happen (all marked with 0 for repeat), but every two or three, 
the time between signals sent to the socket by WinLirc is an exact multiple of 
this by two and occasionally three (e.g., the time between signals when a 
button is kept pressed might be: 180k, 180k, 180k, 360k, 180k, 180, 360, 180k, 
360k, 180k, 180k, 180k, 540k, 180k, 360k, etc. etc.). Is there a way that 
although retaining raw mode, I could provide info to the config file so that 
at least I dont get the 360k (and definitely not the 540k) gap?, and they are 
all 180k? Does the CONST LENGTH flag have anything to do with this? Because 
occasionally I get the 0.54 seconds between signals that are in fact repeats, 
I have to wait at least that much time to react to button press because I need 
to distinguish between a long press and a short press. This makes interacting 
with the remote very slow. Guaranteed time between signals from WinLirc lower 
than say 0.375 seconds will make it acceptable. Guaranteed <200 would make it 
perfect.

Help is always appreciated. As soon as I get this config working reliably I 
will upload...

Thanks!

>===== Original Message From lirc@... (Christoph Bartelmus) =====
>Hi!
>
>Julio C. Martinez "julio@..." wrote:
>
>[...]
>> LIRC home page. I have been successful in learning and using a couple of
>> my remotes, but not the one I am particularly interested in, a Pioneer
>> SR-90 steering wheel remote control. I can't seem to find a config file
>[...]
>> WinLirc insists on the signal not being consistent. However, I copied
>> into a spreadsheet the sequences and they do seem consistent for the
>> same button. I have attached csv files with the data for two of the
>> buttons in the remote.
>
>Looks like standard NEC code (see remotes/generic/NEC.conf in the lirc
>distribution). I don't quite understand why WinLIRC has problems with
>it. Indeed there are Pioneer config files in the database that were
>created just fine using WinLIRC.
>You could try modifying one of the existing Pioneer config files.
>
>Christoph
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by: IBM Linux Tutorials
>Free Linux tutorial presented by Daniel Robbins, President and CEO of
>GenToo technologies. Learn everything from fundamentals to system
>administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click