JP1 Remotes

The Robman · Posted: Tue Jul 28, 2009 10:15 am Post subject:

Once we properly understand a signal format, as long as the learned signal has captured enough of the signal for us to determine which protocol it is, it's not important for DecodeIR to be able to capture all three parts. As long as it correctly displays the protocol data and as long as the executor that we write re-creates all three parts, we should be good to go.

As to the question of whether the UEI format will or won't capture all three parts of a signal like this, it's all a matter of how the signal was learned. UEI will always capture the first part OK, but if the user holds the button down indefinitely, it won't capture the 3rd part. It will recognize the repeating structure and stop capturing. It's only when the user releases the button before the UEI process has finished that it will capture the 3rd part.

In this case, the one full capture has told us enough for us to build a good executor, but if we still had questions, we might ask the OP to re-learn the signals and we would give them specific instructions on how to learn in order to get what we're looking for. It's not uncommon for me to ask someone to learn the same button over and over so I can see which bits are toggle bits, for example. If the checksum was different in the 3rd part and I didn't have enough data to determine how to calculate it, I might ask the user to re-learn using shorter button presses.
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help!

The Robman · Posted: Tue Jul 28, 2009 10:22 am Post subject:

The Robman · Posted: Tue Jul 28, 2009 10:30 am Post subject:

I have updated the files based on what I just described, so try them now.

Here's my (untested) attempt at an upgrade...
http://www.hifi-remote.com/forums/dload.php?action=file&file_id=6980

And here's the commented PB source...
http://www.hifi-remote.com/forums/dload.php?action=file&file_id=6981
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help!

The Robman · Posted: Tue Jul 28, 2009 12:23 pm Post subject:

mathdon · Posted: Tue Jul 28, 2009 2:24 pm Post subject:

As you can see, Rob, I'm slowly working backwards through the Protocol Decodes forum, ordered by date of last posting. I was wondering if there were any others that you were interested in seeing in DecodeIR. And yes, I've already found that you've done the IODATA one twice Exclamation

I'm working on this one from 2005 at the moment, with an extremely weird checksum formula that you managed somehow to decode. There's a 72-bit one that I've deliberately skipped for now, but my intention is that I've done everything with any later posting than this 2005 one, as quite a number of threads don't show up a new protocol. Though I must admit there are threads I haven't read to the end if it didn't look as if they involved a new protocol. So please shout if I've missed one.
________________
Graham

The Robman · Posted: Tue Jul 28, 2009 3:06 pm Post subject:

Whenever you see a thread in this forum (like this one) where there is an IR file but I didn't post anything, please let me know and I'll take a look. There have been times when I was on vacation or busy and I've missed some.

Plus, if there are any where you have questions or want confirmation about something, please post something and I'll respond.

If you find any threads here that really don't deal with a new protocol, please let me know and I'll move them elsewhere (which I do anyway when I spot them).
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help!

vickyg2003 · Posted: Thu Jul 30, 2009 12:32 pm Post subject:

Rob,

I finally got around to testing your upgrade, and it sends the three parts just like you said it would.

I thought I had a basic understanding of how repeat and toggles worked, but with this upgrade I was having trouble understanding how this protocol was going to work, since this third part of the signal was going to send the orignal signal. I was wondering how the equipment would not interpret this as a whole new keypress.

I thougt toggles worked in one of three ways;
1) between every other keypress like the RC5
2) between every other frame
3) between first frame and repeats

When I loaded up the learns and shoot them at the scope, I am not seeing this THIRD part. I'm wondering if you happened to decode a button where OP's finger mis-stepped. I haven't checked ALL the keys, but I think the third part is not necessary.

The Robman · Posted: Thu Jul 30, 2009 1:14 pm Post subject:

Many of the buttons had the beginnings of a 3rd string, so I expect that it really is there. As to why the OEM decided to format the signal this way, I have no idea, but a release string is often used to make sure that the repeating stops.
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help!

vickyg2003 · Posted: Thu Jul 30, 2009 1:58 pm Post subject:

The Robman · Posted: Thu Jul 30, 2009 2:33 pm Post subject:

I'll drop Digital Larry a line o see if he still has access to the Canal in order to test the upgrade. I doubt he still has access to the Canal remote, but if he does we could ask him to relearn some of the signals using shorter button presses, which might help the JP1 remote to capture the break signal.

This 3 part structure is unusual for remotes, it's much more common on infrared keyboards.
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help!

vickyg2003 · Posted: Thu Jul 30, 2009 4:48 pm Post subject:

I put together protocol builder files for 2 and 3 part signals with both HCS08 support and that other processor.
http://www.hifi-remote.com/forums/dload.php?action=file&file_id=6991

The keymaster file for 2 part signal signal with both processors.
http://www.hifi-remote.com/forums/dload.php?action=file&file_id=6992

The keymaster ffile for 3 part signals updated for both processors
http://www.hifi-remote.com/forums/dload.php?action=file&file_id=6993

mathdon · Posted: Tue Aug 11, 2009 7:39 am Post subject:

Hey, I've just had my first success in finding a UEI executor for a new protocol! This CanalSat protocol is PID 018C, which seems only to be in HCS08 remotes at the moment.

The UEI executor doesn't give an "extra" frame, it just has the initial frame with a toggle of 0 followed by repeats with a toggle of 1. The evidence for the "extra" frame in the learned signals was somewhat doubtful, anyway. It also has, in effect, a very slightly different IRP from mine. I had

CanalSat {55.5k,250,msb}<-1,1|1,-1>(1:1,D:7,S:6,T:1,F:8,-89m)+

but UEI has

CanalSat {55.5k,250,msb}<-1,1|1,-1>(1:1,D:7,S:6,T:1,0:1,F:7,-89m)+

The difference is that UEI forces the top bit of "my" 8-bit function code to be 0 and treats it as a fixed bit, making 16 fixed bits and 7 variable ones instead of my 15 fixed and 8 variable ones. I've amended my DecodeIR accordingly. Since the 7 bits of the function code go into the left-most bits of the function byte, this has the effect of shifting the Hex code left by one place.

Here is a "front end" for the existing PID 018C entry in protocols.ini that handles the CanalSat fixed bytes:

mathdon · Posted: Tue Aug 11, 2009 7:57 am Post subject:

Curiously I find that Rob has a spreadsheet with S3C8 code for PID 018C even though it doesn't list any S3C8 remotes as supporting it! Anyway, adopting that code, the protocols.ini entry becomes

gfb107 · Expert Joined: 03 Aug 2003 Posts: 3411 Location: Cary, NC

I'll include that in the next release of RM
_________________
-- Greg
Original RemoteMaster developer
JP1 How-To's and Software Tools
The #1 Code Search FAQ and it's answer (PLEASE READ FIRST)

alanrichey · Expert Joined: 24 Mar 2008 Posts: 3529 Location: UK/USA

Did anything come of this ? I just had IRSCOPE decode a protocol as 'CanalSat', but that isn't included in the Protocols for RemoteMaster.