Bug Report: [RCA Combo (w/Duration)] - old S3C8

[The Robman]

I was trying to program the RCA VH920 a/v selector into my 15-1994 remote recently and the duration feature wasn't working, so upon further investigation I discovered what's wrong.

This particular executor requires RM to edit the executor code, when you enter a value in the "Duration" box, it's supposed to edit the 2nd to last statement in the executor code from "E6 0D 00" to "E6 0D 05" (assuming you stay with the default value of 5). This works for remotes that use the modern version of S3C8, but it doesn't work for remotes that use the older version of S3C8, like the 15-1994. I suspect this is because RM also has to edit the executor code to convert the last statement in the executor code from "8D 01 46" to "8D 01 33".

I don't know what it would take to fix this in RM, so it might be a better idea to pass the duration value into the code as a fixed byte, then RM wouldn't need to edit the executor code itself. I could do that if that's the best solution.

[The Robman]

The following protocols.ini entry seems to fix it

[RCA Combo (w/Duration)]
PID=01 FF
VariantName=JP1RCAD
OldRefs=0143:JP1
DevParms=Duration:8=5
DeviceTranslator=Translator(0,8,0)
FixedData=05
CmdParms=Device=0:4,OBC=0
CmdTranslator=Translator(0,4,0) Translator(1,8,8)
Notes=S3C8 hacked combo
DefaultCmd=0f 00
CmdIndex=1
Code.S3C80=28 60 12 8B 12 95 45 04 08 00 FA 03 D4 00 FA 01 E0 0F 8C 07 D0 07 BC 20 01 20 10 E4 03 0D 8D 01 46
Code.HCS08=20 15 17 2F 12 95 45 04 08 00 F9 03 E8 00 F9 01 F4 0F A0 07 CF 07 D0 3C AC 3C A9 4E 60 B2 CC FF 5F

[mathdon]

I don't know what it would take to fix this in RM

It should not take much. If you open RM, set the remote to RS 15-1994, select the RCA Combo (w/Duration) protocol and look at the Output tab, the protocol code is both correctly changed to that for the S3C80 and the E6 0D xx byte changes according to the set duration. So somehow it just doesn't get imported into RMIR correctly. I will look into this and try to fix it.

[The Robman]

I did do that and had verified that it was correct, but as you say, the wrong value got loaded into the remote. With the change that I posted before, this becomes irrelevant as this new version of the executor passes the duration into the code as a fixed byte, and I created (and tested) a version for HCS08 remotes too.

If this is the only protocol that requires RMIR to edit the executor, I'm not sure it's worth fixing now.

[mathdon]

There are 8 entries for CodeTranslator in protocols.ini, so a bug in its operation in RMIR was, to me, worth fixing. It is fixed in build 7 of RMIR v2.11 in the RMIR Development folder. It would be helpful if you are still able to check that it works, with the original protocols.ini entry for that protocol.

[The Robman]

It works, thanks Graham. But please let me know if you're going to go with my new version of the RCA Duration protocol, which also has an HCS08 version, just so I can keep my RMIR file in sync.

By the way, you can add the following to [RCA Combo (Official)]

Code.HCS08=20 15 17 2F 02 95 45 04 08 00 F9 03 E8 00 F9 01 F4 0F A0 07 CF 07 D0 3C A9 CC FF 5F

[mathdon]

I have included your protocols.ini changes in development build 8, posted in the usual folder.

FYI The CodeTranslator bug did not only affect S3C8, it affected all remotes that store protocol upgrades in a device-independent section. This means all JP1.3 and earlier remotes, excepting only the device-dependent protocol upgrade sections that some of these remotes have. A CodeTranslator modifies that protocol code when the protocol is assigned to a device upgrade, and that modification was not being made.