Teac-K and Kaseikyo protocols im KM

[johnsfine]

Added 'Teac-K' $00BB & 'Panasonic MIX Combo' $001F protocols. (thanks to Rob)

A lot of the planning and design for RM was based on the fact that strange new protocols like this can't be added quickly to KM, but can be added quickly to RM.

I still haven't added Teac-K to RM. You're making me look foolish.

But, are you sure about the way you built Teac-K?

Shouldn't the default for the third byte of fixed data be 7F, not FF? In my decoder, for obscure Kaseikyo forms I follow Kaseikyo specs in treating the fifth nibble as the check nibble on the first four, not as part of the device number, so the device number is just 4 bits (the sixth nibble).

In Panasonic and JVC-48 we continue to misinterpret that fifth nibble for compatibility with everything that was done before we saw the Kaseikyo spec. But for forms with check byte rules different from Panasonic, I see no advantage to that compatibility.

If you had intended device to be 8 bits (I hope not) then the default should have been 1 not zero, AND you included device incorrectly in the formula for the other check byte. (So I assume device should have been 4 bits and only the computation of the third byte of fixed data is wrong).

[The Robman]

Mark's not going to know how to answer this, as I added these two protocols to KM. I guess I mis-interpreted how IR is displaying the data then. The Teac signals are displayed as device code 0 and sub-device code 4 and I treated these as though they "Panasonic" device codes.

I guess I'm going to have to dig out my Kaseikyo spec and re-visit this coding.

[The Robman]

John, am I right in assuming that what you call "device code" in DecodeIR (via IR.exe) is the 4-bit "Div" code, and what you call "sub-device" is the 8-bit "Appliance" code?

[jon_armstrong]

Here is what I thnk is the comparison between the two protocols:

TEAC-K {37.9k,432}<1,-1,1,-3>((8,-4,M:8,N:8,X:4,D:4,S:8,F:8,C:8,1,-100),(8,-8,1,-100)+)
where:
X=M(0..3)^M(4..7)^N(0..3)^N(4..7)
C=D+S(0..3)+S(4..7)+F(0..3)+F(4..7)

Kaseikyo {37k,432}<1,-1,1,-3>(8,-4,M:8,N:8,X:4,D:4,S:8,F:8,E:4,C:4,1,-173)+
where:
X=M(0..3)^M(4..7)^N(0..3)^N(4..7)
C=D^S(0..3)^S(4..7)^F(0..3)^F(4..7)^E

BTW, the ^ means XOR and F(4..7) means the top 4 bits of F

Not relevant to this discussion but to help understand the notation, the Teac has a one-time data frame and a null repeat segment like NEC1. Kaseikyo has the entire frame repeat.

[jon_armstrong]

Rob,

I thik you are referring to this table from Panasonic:

There are 48 bits in a Kaseikyo format. These bits are assigned as follows:

Subgroup # of bits
Manufacturer code : 16 bits (2002 hex)
Parity : 4 bits- calculated value
Division :4 bits ( 8 hex)
Appliance : 4 bits- for TV ( 0 hex)
Control : 4 bits- for TV ( 0 hex)
Function Data : 8 bits
Parity :8 bits

So I think the answer is Yes but S:8 is Control:4 Appliance:4 (you have to reverse the order)