RMIR learned signal display suggestion

[The Robman]

In the cases where a signal currently shows up as "Multiple" but in reality there's only 1 clean decode and an additional garbage entry, could we have RMIR display the clean decode? There are several examples of this in the RMIR file that Capn Trips just posted. I think this would make things easier for users.

Here's the file:
https://www.hifi-remote.com/forums/dload ... e_id=26000

Also, is there a way to display the code summary in RMIR like you could in IR.exe?

[mathdon]

The file at that link contains just two learned signals, neither of which have decodes, so I think you have linked to the wrong file.

Also, is there a way to display the code summary in RMIR like you could in IR.exe?

Have you tried File > Summary >Create Summary? If that isn't what you want, could you explain what it is that you would like?

[The Robman]

Sorry, the Capn must have overlayed his file after I posted, here's the original file:
https://www.hifi-remote.com/forums/dload ... e_id=26003

And, regarding the file that you did download, I had forgotten that I had switched RMIR to use DecodeIR for the ict tests that I did, so there is something that I'd like you and Barf to look at. That file contains clean learns in the NEC2 format, and these decode correctly using DecodeIR but not with the IRTrans thingy program, any idea why?

I had never noticed that Summary option before, any chance we could have a button for it in the learned tab, like there is in IR.exe ?

[mathdon]

In the cases where a signal currently shows up as "Multiple" but in reality there's only 1 clean decode and an additional garbage entry, could we have RMIR display the clean decode? There are several examples of this in the RMIR file that Capn Trips just posted. I think this would make things easier for users.

I have now looked at the original file, and with IrpTransmogrifier as decoder, all 33 learns have no decodes. Most of them do with DecodeIR, some with multiple decodes, but the garbage entry in those multiple ones is a Gap-xx-xx-xx catch-all entry that has no equivalent in IrpTransmogrifier. DecodeIR is retained in RMIR as an optional legacy decoder and neither DecodeIR itself nor its use as a decoder in RMIR is still supported, so I do not see any reason to change its behaviour from what it has always been.

As to why the signals do not decode with IrpTransmogrifier, it seems to me that the reasons are as follows, but Bengt may need to correct me on this. There are two protocols in the signals that decode with DecodeIR. The NEC2 signals have a frequency of 41451 or 41667. The standard frequency for all NEC variants is 38400 and IrpTransmogrifier has a default tolerance of 2000, so these are outside its accepted frequency range. The NECx1 signals have a frequency of 37383, which is within tolerance, but their repeats are invalid. In IRP notation the NECx1 repeat is (8,-8,D:1,1,^108m) and these signals have D=8 so D:1 is 0. These signals send a 1 rather than 0 at that position.

I had never noticed that Summary option before, any chance we could have a button for it in the learned tab, like there is in IR.exe ?

The Learned tab seems a strange place for a Summary that is nothing specifically to do with learns. I could put a button on the toolbar, if I can think of a suitable icon for the button. What do you think of that alternative?

[The Robman]

The code summary button in the learns tab of IR.exe just displays the codes of all the learned buttons.

[Barf]

I have now looked at the original file, and with IrpTransmogrifier as decoder, all 33 learns have no decodes. Most of them do with DecodeIR, some with multiple decodes, but the garbage entry in those multiple ones is a Gap-xx-xx-xx catch-all entry that has no equivalent in IrpTransmogrifier. DecodeIR is retained in RMIR as an optional legacy decoder and neither DecodeIR itself nor its use as a decoder in RMIR is still supported, so I do not see any reason to change its behaviour from what it has always been.

Yes, that is basically it. DecodeIR was obviously developed with the aim to decode as much as possible, also some signals that should not be decoded, "false positives".

The reason why DecodeIR was entirely removed from IrScrutinzer 2.*.* was that it requires so much cruft, cruft that needs to be maintained for "all platforms". Still, IrScrutinizer 1.4.3. is available and contains DecodeIR 2.45, and, to some extent, is supported (for this reason).

As to why the signals do not decode with IrpTransmogrifier, it seems to me that the reasons are as follows, but Bengt may need to correct me on this. There are two protocols in the signals that decode with DecodeIR. The NEC2 signals have a frequency of 41451 or 41667. The standard frequency for all NEC variants is 38400 and IrpTransmogrifier has a default tolerance of 2000, so these are outside its accepted frequency range. The NECx1 signals have a frequency of 37383, which is within tolerance, but their repeats are invalid. In IRP notation the NECx1 repeat is (8,-8,D:1,1,^108m) and these signals have D=8 so D:1 is 0. These signals send a 1 rather than 0 at that position.

All coincides with my observations. Let me add that the length of the captures is wildly varying, so it appears fairly random.

I had never noticed that Summary option before, any chance we could have a button for it in the learned tab, like there is in IR.exe ?

The Learned tab seems a strange place for a Summary that is nothing specifically to do with learns. I could put a button on the toolbar, if I can think of a suitable icon for the button. What do you think of that alternative?

I like popup, "right-mouse" menus. There are a number of such in IrScrutinizer. They are easy to program, supports disabling, accelerators, mnemonics, cascaded menues, tooltip texts, and take up no screen real estate. Drawback can be that users do not find them.

[The Robman]

I like popup, "right-mouse" menus. There are a number of such in IrScrutinizer. They are easy to program, supports disabling, accelerators, mnemonics, cascaded menues, tooltip texts, and take up no screen real estate. Drawback can be that users do not find them.

I like the idea of making them default to ON, letting the user know they exist, and telling them how to disable them. That way they know they're there in case they need to turn them on again for some reason.

[The Robman]

As to why the signals do not decode with IrpTransmogrifier, it seems to me that the reasons are as follows, but Bengt may need to correct me on this. There are two protocols in the signals that decode with DecodeIR. The NEC2 signals have a frequency of 41451 or 41667. The standard frequency for all NEC variants is 38400

Sounds like they are really "Pioneer" signals, which makes sense.

[Barf]

I like popup, "right-mouse" menus. There are a number of such in IrScrutinizer. They are easy to program, supports disabling, accelerators, mnemonics, cascaded menues, tooltip texts, and take up no screen real estate. Drawback can be that users do not find them.

I like the idea of making them default to ON, letting the user know they exist, and telling them how to disable them. ....

Rob, I do not think you are talking about the same thing I am talking about. A popup menu is, if used, an integral usage element, and can not be "turned on and off" at will, just as little as the steering wheel of an automobile can be turned on and off.

[The Robman]

Maybe so, I assumed you were referring to the same type of pop-up items that RMIR uses, and in most cases, I turn them off whenever possible. The most recent example is the one where RMIR wants the user to chose from 8 or more executors for simple NEC1 signals, where 99% of the time, just picking the NEC1 executor is the right think to do.

I see you we actually talking about right-click menus, which of course, would only appear if you actually right-click.