JP1 Remotes

[Juppers]

1. Device: CaptiveWorks 3000HD
2. Type of device: PC based Media Center
3. Year: 2007
4. JP1/UEI Remote model: N/A
5. Do you have a JP1 cable? Yes
6. Still have original remote? Yes
7. Checked the file section? Yes
8. Checked Pronto file section (at R/C)? Yes
9. Partially working setup code? Yes
10. Learning remote question? I capture the code and the decoded CCF doesn't come back with a protocol. Trying to create a codeset to load to my Slingbox. Sample of what decode CCF comes up with is below

Protocol Device SubDev OBC EFC Misc Decode UDB Function Sequence Key Panel DevName TopFreq Once Rept
0000 0070 0040 0000 0031 002C 0011 002C 004E 002C Learned 1 panel Dish 37.0102
0000 0070 0027 0000 0030 002C 0012 002B 0012 0069 Learned 2 panel Dish 37.0102
0000 006F 0027 0000 0031 002D 0012 002B 0050 002B Learned 3 panel Dish 37.3437
0000 006D 002B 0000 0031 002E 0011 002E 0011 002D Learned 4 panel Dish 38.0289
0000 0070 002B 0000 0030 002D 0011 002C 0030 004B Learned 5 panel Dish 37.0102
0000 006F 002F 0000 0031 002D 0011 002C 0012 002C Learned 6 panel Dish 37.3437
0000 0070 002B 0000 0030 002C 0012 002B 0012 0069 Learned 0 panel Dish 37.0102
0000 006F 0026 0000 0030 002D 0012 002C 004F 002C Learned ENTER panel Dish 37.3437
0000 006D 0027 0000 0031 002E 0012 002C 0012 006C Learned 7 panel Dish 38.0289
0000 006F 002D 0000 0031 002E 0011 002C 0011 002C Learned 8 panel Dish 37.3437
0000 006D 002D 0000 0031 002F 0011 002D 0012 002D Learned 9 panel Dish 38.0289

[johnsfine]

[Juppers]

I do have a JP1 remote that I rarely use anymore. It is a Sony RM-Y809. I also have a pronto which is what I used to capture the codes above. The CCF is at http://www.hifi-remote.com/forums/dload.php?action=file&file_id=5000

Some of the buttons are from other devices, but the majority are the CW3000HD. All numbers, arrow pad, some menu type functions, and transport keys. Basically, the stuff I needed for the slingbbox.

[johnsfine]

[johnsfine]

I took a closer look. Menu is part of the device you're interested in and the middle of its learn is OK.

The others all look like something is wrong with the beginning of each learn. But whatever may be wrong is suprisingly consistent in the way it is "wrong" (except for Menu). Most learning problems that trash the beginning of a signal would do so less consistently.

So it is hard to be sure the beginning really is wrong, rather than just a strangely defined protocol.

All the signals, including Menu, have reasonable looking middle parts. I haven't decoded the middle yet, but I'm sure Rob or I could.

I think the correct whole signal is just repeats of that middle part. I'm nearly sure the end part is just a failure by the Pronto to detect/store the repeat pattern. I think the beginning is some learning error, but I'm not sure.

[Juppers]

Yes. You are correct on the *, LiveTV, tv/vid and Tivo buttons are left over from whatever I last learned, looks like a dish receiver.

When I learned them into my Harmony, I had to get support to fix them up to work correctly there as well. Some of the codes I had to learn from the harmony, because learning from the native remote caused massive repeats of the command for the pronto. hitting up arrow once would be like hitting it 30 times. After learning those from the harmony, they behave better, but I can still tell there is a bit of junk in them by the delayed response.

[Juppers]

I have updated the download above to remove, I think, all of the commands not for the device, as well as relearn some of the commands such as menu, and I added some commands I had missed earlier that I needed.

[johnsfine]

I halfway remember seeing this protocol in some previous JP1 thread, but I don't know how to search for it.

Maybe Rob remembers and/or understands better than I do now.

But if I end up coding a JP1 executor for this, I'd like to first get some better info on what the correct signal looks like.

[Juppers]

Makes sense to me!

I have updated the file to be all learns from the native remote.

[johnsfine]

That changes everything!

I was completely misunderstanding what those "half frames" were because of the extra crap Harmony added.

Now I understand.

There is no repeating part to the signal. There are just two frames, press and release. The repeat understood by the receiver is simply the length of the gap between the press and the release.

So in Pronto Hex, there is no way to encode a correct variable length press. Instead, learning a short press (as you did for those) will only encode the actual length learned. So a use on the Pronto would duplicate the press duration of the learn rather than representing the press duration of the use.

In a JP1 executor, it isn't very hard to encode correct press/release behavior so that a use would have actual duration.

But a Slingbox has no button duration, so for a Slingbox, we might need a simpler executor that gives some fixed duration to every button press.

I'm still a bit currious about the full rules for long button presses. Could you learn a long press into the Pronto. If no additional rules come into play, that would capture a single "press" and never get the release (the Pronto gives up too soon), so it should be half of the Pronto hex of a short learn.

[Juppers]

Updated the file to have the arrow keys be long presses.

[johnsfine]

I think that means the press signal is repeated at a very slow rate during very long key presses.

What I see is:

Up: Press frame, then long gap, then press frame, then moderate gap, then garbage such as you would expect to see when the original remote batteries are dying (which they probably are. A long press can magnify the symptoms of weak batteries).

Down: Just a single press frame (what I originally expected a long press to learn as).

Left: Press frame, then long gap, then press frame.

Right: Just a single press frame.

[The Robman]

[johnsfine]

It sure is close to the protocol in that other thread, probably close enough to use, maybe exactly right.

[Juppers]

Dang. I was ready to load that one and give it a shot, but I realized my new laptop doesn't have a parallel port. Doh!