RMIR v2.15.0 pre-release for testing

Post by **mathdon** » Sat Apr 29, 2023 7:04 am

HamburgerHelper1 wrote:Here is the link to the RMIR with a negative realtime macro
http://www.hifi-remote.com/forums/dload ... e_id=26717

I cannot reproduce this. If I look at the realtime Special Functions, the Functions column shows Hold(0.00) and if I edit it then the Hold for value is again 0.00. This is true also with the realtime ones that have no Hold value on the Functions column. They too show Hold for (0.00) when editing. Not a minus sign in sight.

I have tried this both in Windows 10 and in Ubuntu 20.04 so perhaps it is a quirk of your Linux version. The value is stored as floating point and the internal representation of floating point numbers is, by its very nature, approximate. There is no mathematical difference between +0.0 and -0.0 but an absolutely tiny change when stored as floating point can result in something that reads back as -0.0 rather than 0.0. I suggest you simply ignore it, as I can't test a fix for something that I cannot reproduce.

HamburgerHelper1 · Post by **HamburgerHelper1** » Sat Apr 29, 2023 7:24 am

I was planning on just ignoring this quirk.
Perhaps Someday I'll see what it does on a different machine running 'Windows OS
For now I'm happy with RMIR and it looks like we might finally be done finding all of the
features the 3660 has to offer

davecs · Post by **davecs** » Sat Apr 29, 2023 9:00 am

HamburgerHelper1 wrote:I was planning on just ignoring this quirk.
Perhaps Someday I'll see what it does on a different machine running 'Windows OS
For now I'm happy with RMIR and it looks like we might finally be done finding all of the
features the 3660 has to offer

I think the "-0.00" thing is a legacy problem from 2.14.x and will not survive editing. See my previous comment. Get the RDF's right in the 3660 family and we're good to go!

HamburgerHelper1 · Post by **HamburgerHelper1** » Sat Apr 29, 2023 10:00 am

RDF's for some of the version's of this remote need Fine Tuning in the phantom button section

Insignia NS-RMT8D21 has 3 Phantoms
3660 has 8
3680 has 3
3661 has 8

HamburgerHelper1 · Post by **HamburgerHelper1** » Sat Apr 29, 2023 4:27 pm

I assume all versions of this remote can have 8 Phantom buttons but I do not know enough about them to say for sure .
Some of them start at a different location and some have allmacrobind as well
insignia = Phantom1=$3D 3 phantoms and no AllMacroBind
3660 = Phantom1=$3F 8 phantoms and AllMacroBind
3660 v2 = Phantom1=$3D only has 3 phantoms and no AllMacroBind
3661 = Phantom1=$3F has 8 phantoms and AllMacroBind,
3680 = Phantom1=$3D 3 phantoms and no AllMacroBind,

Post by **mathdon** » Sun Apr 30, 2023 4:00 am

HamburgerHelper1 wrote:RDF's for some of the version's of this remote need Fine Tuning in the phantom button section

Insignia NS-RMT8D21 has 3 Phantoms
3660 has 8
3680 has 3
3661 has 8

I don't know why there is a discrepancy between these remotes. I have always regarded phantom buttons as a bit of a mystery. They are names assigned to unused but valid keycodes, but many remotes have such keycodes and those with some of them assigned phantom names often have further unnamed and unassigned ones. In this case, those with 3 phantoms have their keycodes starting a $3D, those with 8 have them starting at $3F. To avoid the possibility of existing setups failing if any named values are removed, while also bringing uniformity to these remotes, I suggest they all get 10 phantoms starting at $3D. Is this OK?

Edit: I posted this before seeing Randy's latest post. AllMacroBind means that macros cannot be bound to this key. Since a phantom can only be accessed by being included in a macro, and macros cannot be nested in these remotes, it makes no sense to bind a macro to a phantom. So all should have AllMacroBind.

HamburgerHelper1 · Post by **HamburgerHelper1** » Sun Apr 30, 2023 5:18 am

10 Phantom Buttons for the entire 3660 Family Sounds Great

davecs · Post by **davecs** » Sun Apr 30, 2023 8:01 am

If I remember correctly, somewhere in the 3660 thread, Rob said that he'd put 3 phantoms into an RDF in response to someone asking why there weren't any.

Then the 8 phantoms were put into the original 3660 and 3661 RDF after testing by Graham and Randy HH1. If the 3 and the 8 all actually work, only overlapping on one value, then there must be at least those 10 available.

davecs · Post by **davecs** » Mon May 01, 2023 3:05 am

Just one thing re phantom buttons. I've made use of a small number. Will the re-numbering of them affect anything I've done?

Post by **mathdon** » Mon May 01, 2023 9:02 am

davecs wrote:Just one thing re phantom buttons. I've made use of a small number. Will the re-numbering of them affect anything I've done?

It should not do so as both the remote and a .rmir file refer to buttons by their keycode. The most that should happen is that you may see some name changes when you load a .rmir file or download a remote, but neither will make any functional difference.

davecs · Post by **davecs** » Mon May 01, 2023 10:50 am

mathdon wrote:It should not do so as both the remote and a .rmir file refer to buttons by their keycode. The most that should happen is that you may see some name changes when you load a .rmir file or download a remote, but neither will make any functional difference.

OK - I edited the RDF for the 3661 for the phantoms to start at $3D, and for ten of them. I then added Discrete Power Off, Discrete Power On, Discrete Unmute and Discrete Mute codes to my Audio amp, on Phantom1, Shift-Phantom1, Phantom10, Shift-Phantom10 in the Device Editor, noting that all the Phantom entries had been shifted up by 2.

Then I made 4 multimacros on the Audio TVbtn, which is unused, and uploaded to the remote. Each press worked as expected, in turn, Off, On, Mute, Unmute.

Is it safe to assume that if Phantoms 1 and 10 work, those in between should be OK?

Post by **mathdon** » Mon May 01, 2023 11:44 am

I have replaced RMIR v2.15.6 in the RMIR Development folder with v2.15.7. The changes from v2.15.6 concern XSight remotes and the RDFs for the URC-3660 series. The XSight changes were needed to make some recent additions compatible with these remotes, the RDF changes create the 10 phantom buttons for the URC-3660 series as discussed above.

davecs wrote:Is it safe to assume that if Phantoms 1 and 10 work, those in between should be OK?

Yes.

davecs · Post by **davecs** » Mon May 01, 2023 4:24 pm

It all looks fine to me, but are there any specific tests I could do, on a 3661 or a 6440?

Post by **mathdon** » Tue May 02, 2023 4:27 am

davecs wrote:It all looks fine to me, but are there any specific tests I could do, on a 3661 or a 6440?

Testing on the 6440 would be very helpful. It is some time since I tested on that. It, and its US equivalent, are unique in that they are the only remotes with a segment structure that have an extender, possible only because of their use of a settings.bin file. So it may have unique bugs, too. No specific tests, just make sure that it seems to work OK, especially extender features.

davecs · Post by **davecs** » Tue May 02, 2023 6:33 am

I need to re-organise the devices on my 6440, hopefully I can make a start later today. During this testing cycle, on 23 April, I did make a small change to it, which I successfully uploaded, to add a DKP on the Mute button, so that single press mutes/unmutes the amp (as before), and a double press sends a DiscreteON and DiscreteUnmute to the amp, and that uploaded just fine, and works! I also did the "header change" in the RMIR file and uploaded that to the OARUSB04G, and that's working too.

JP1 Remotes