Taiwanese IR Protocol

[alanrichey]

OK, this one is a doozy And nothing like I have ever seen. IRScope is giving this for the Power button:

37647 XMP.136.201-180F445800000F00
37745 XMP.137.209-180F445808800F00
38348 XMP.137.149-180F445808800F00
37276 XMP.137.172-180F445808800F00
38208 XMP.137.162-180F445808800F00
38390 XMP.136.198-180F445808800F00

and this for the Number 1

37470 XMP.137.141-180F445806800100
37987 XMP.136.200-180F445806800100
37987 XMP.136.201-180F445806800100
36715 XMP.137.235-180F445806800100
37874 XMP.137.149-180F445806800100
37487 XMP.137.218-180F445806800100

I can see there is a logic there as the last 16 digits seem to be unique to the button, as is the 3rd parameter.

I've uploaded the ICT file for the Power, 1 & 2 buttons at https://www.hifi-remote.com/forums/dload ... e_id=13196 If someone can make sense of them I will do the rest of the buttons.

[3FG]

Al,
You can read about decoding XMP signals in DecodeIR.html, but the short answer is that this is (most likely) a XMP signal, but less than stellar learns. Assuming you got these from a Harmony, the signals sent by the Harmony probably work but are on the ragged edge of correct timing. Then the second generation learning process adds some more timing uncertainty. In a situation like this, an approach to learning which typically works well may not be optimal for these signals.
I recommend that you systematically capture signals, varying the length between the Harmony and the Widget between captures. You're looking for a standard XMP decode which is reproducible at a given separation. As always, be sure to shield the Widget from sources of IR--like a heater or incandescent light.

[alanrichey]

OK, the original user is reading this so I will get him to try that.

Thanks

[3FG]

If he is learning from an OEM remote, then we want him to check that the learned signals do work. Sometimes people forget to do that, and we end up trying to decipher learns which had no chance of being decoded to a working signal.

[SilverSeraphim]

Unfortunately, I can't test the learned signals. The receiver is actually in Taiwan, but my in-laws brought the remote to America.

This is probably an incredibly foolish thing to say, but can we look at the decodes and guess what is the most probable signal based on a collection of multiple learns?

My only way to test it is to upload the best guess codes to the Slingbox and try to change the channel.

One other question. Is there any chance we'd get better decodes with higher quality batteries or something? Maybe this cheap remote barely sends readable signals by default...

[alanrichey]

Update: New batteries, dark room, short key press, different distances make no difference, I am still decoding it the same.

This seems odd, as it is 100% consistent under all scenarios and I am getting identical decodes for each different method. If they were poor learns I would expect to see some inconsistency ?

[SilverSeraphim]

Is it possible that these are in fact correct learns and that the tool is just seeing something outside the scope of what it typically expects for a decode? Perhaps the 100+ inconsistency rating is a false positive.

I see the following in the documentation provided by 3FG:
"Almost all examples we've seen start with "0E0F441A0" or "060F44120". But we've also seen upgrades from UEI for "0D1F441A0" and "0C2F441A0" and "0B3F441A0"."

In our case, both of the XMP codes posted above start with "180F44580". The timescales seem to be consistent as well (136-137).

Can we work from the assumption that this is correct information but in an unusual application? If say the power hex code 180F445800000F00 is indeed correct, how would I go about manually decoding that for import into the Slingbox?

[alanrichey]

You've just moved beyond my area of expertise

[SilverSeraphim]

OK, guys! Good news (I think).

I imported using the Harmony tool created by monochrome88:
https://www.hifi-remote.com/forums/viewt ... 07a4c19324

Here's what I got for the power button:

Carrier clock : 41025 Hz
Total mark/space pairs : 18

+195 -920 +196 -2272 +195 -780 +195 -2816 +195 -1323 +195 -1323 +195 -1459 +195 -1866 +195 -14262 +196 -916 +195 -780 +195 -780 +195 -781 +195 -780 +195 -2816 +195 -780 +195 -780 +195 -524093

Thus:
Signal = One frame. Incomplete signal: start frame with no following frames.
IRP = {41.0k,msb}<-136u|195u,-780u>(<1:-1|1:-2|1:-3|1:-4|1:-5|1:-6|1:-7|1:-8|1:-9|1:-10|1:-11|1:-12|1:-13|1:-14|1:-15|1:-16> (A:32,196u,-14.3m,B:32,196u,-524.1m)) {A=$1B0F4458,B=$10000F00}

Screenshot:
http://i.imgur.com/Ziv3mXw.png

Does that look like what I need?


EDIT:
Played around some more until I got Power, Ch Up, Ch Down all with 3+ frames "including distinctive start frame."

I'm going to test making a basic remote with RemoteMaster and see if I can upload it to the Slingbox.

EDIT 2:
It works! I can now change the channel on the Taiwanese cable TV box via the Slingbox.

Thanks so much for all of your help guys!

[alanrichey]

OK, that is officially amazing. I am VERY impressed I need to check out that software myself.

[vickyg2003]

SilverSeraphim, could you please post your RM and Bin files for this Taiwanese box in our file section so we can help the next guy.

Thanks

[SilverSeraphim]

OK, I finished the whole remote and tested it.

I uploaded the files, hopefully this is correct for how they should look:

https://www.hifi-remote.com/forums/dload ... e_id=13219

https://www.hifi-remote.com/forums/dload ... e_id=13220

[alanrichey]

I normally package them together with some instructions. Put as I said over at the Slingbox forum, just post the RMDU ile there and I will do it for you.

[SilverSeraphim]

Done. Thanks!

[alanrichey]

And the package is now available at https://www.hifi-remote.com/forums/dload ... e_id=13222