JMRI DecoderPro Throttle - F29 +

[Ian J.]

Hi all,

 

I've just got an Accurascale Deltic, and have tested it using DecoderPro. However, the Deltic manual mentions F29 and F30 as functions, but DecoderPro only seems to go up to F28. Am I missing a way of getting to higher functions, or is DecoderPro lacking these, and in need of extending its function support?

 

TIA

Edited June 22, 2022 by Ian J.

[Nigelcliffe]

Does your DCC system go to F29 ?   If not, then unlikely that any JMRI code change would help. 

 

If wanting it addressed then the JMRIusers group on groups.IO is where you can raise issues and requests for new features. 

 

 

- Nigel

[Ian J.]

I have a SPROG 3, so I'm not sure if that might be a limiter. If @Crosland can see this maybe he can shed some light on the unit's function support?

[grriff]

It's not the SPROG. I have a SPROG II And it will read CVs into the hundreds.

Presumably you clicked 'Add new loco' and then 'Read type from decoder' and this only read up to CV 28. What decoder did it assign? 

If you want to read higher CVs select 'Add new loco' and select NMRA then Raw CVs 1-255.

[Ian J.]

I wasn't paying that much attention to what CVs it read when loading up. I'm still very much a newbie with JMRI and it's various parts, so don't really know what I'm doing with it. My issue was that the Decoder Pro display only includes buttons up to F28. Is it adaptable to show more buttons, depending on the decoder? I seem to remember it was recognized as a Loksound 5.

[Crosland]

3 minutes ago, grriff said:

It's not the SPROG. I have a SPROG II And it will read CVs into the hundreds.

The problem is Functions, not CVs.

 

This sounds like a JMRi issue, as all of the DCC commands are created in the software. I will need to take a look at the code.

[Ray H]

I'm pretty certain that NCE equipment will only work with functions 0 - 28 so they'd have a problem with higher number functions.

[Crosland]

At first glance it looks like there is currently no support for F29+ in JMRI, for any system (sigh of relief that it's not a SPROG problem), but I could yet be proven wrong, they may be handled differently in different code that I haven't seen yet.

 

I have asked the question of the other developers.

 

Which decoder is involved? Is there any function remapping for these functions, that would allow you to access them, at the expense of other, lower numbered, functions?

 

Andrew

[Pete the Elaner]

Are functions above f28 supported within existing DCC standards without defining add-ons (ie Railcom)? I didn't think they were.

I've always found it a bit strange that a digital system would be limited to 29 functions (including f0 of course). I would have thought 16, 32 or 64 would have been a limit.

[Ian J.]

33 minutes ago, Crosland said:

At first glance it looks like there is currently no support for F29+ in JMRI, for any system (sigh of relief that it's not a SPROG problem), but I could yet be proven wrong, they may be handled differently in different code that I haven't seen yet.

 

I have asked the question of the other developers.

 

Which decoder is involved? Is there any function remapping for these functions, that would allow you to access them, at the expense of other, lower numbered, functions?

 

Andrew

 

It's the sound decoder in the Accurascale Deltic, so I think it's a Loksound 5 by ESU.

[David12]

Hi Ian J,

I am quite confused by some of the replies to your problem, as I use the JMRI suite of programs, including Decoder Pro and SPROG 3 with no problems at all with functions above F28 in Loksound 5 decoders. I have seven locos (fitted with ESU Loksound 5 decoders) and they all show functions up to F31 on Decoder Pro's "Labels and Media" page. My JMRI is version 4.26 and runs on a Windows 10 PC. Have you fully read the decoder into the roster? Some people don't have the patience as it takes (on my machine) around 2 hours to read all CVs.

[Ian J.]

I'm on Linux. JMRI v5. I haven't enough knowledge of JMRI or Decoder Pro to really know what I'm doing at the moment, so I'm just going with the basics, which only shows up to F28. There doesn't appear to be provision to have more function buttons on the throttle display. 2 hours does sound like a very long time to read the CVs of a decoder though.

 

Edit: I've just checked, and the 'Labels & Media' screen does show up to function 31. Not much use if it doesn't also have those function buttons in the throttle.

Edited June 24, 2022 by Ian J.

[Nigelcliffe]

 

You'll not fix the throttle issue in this thread.     The code for the JMRI throttles stops at F28(*).   To change that requires some code changes.    The JMRI software is open source, so anyone can download it and change it, or wait and hope that someone else fixes things.  

(* source of information - JMRI developers email list and a developer I'd trust to understand this issue properly ). 

 

 

The time to read the CVs is down to the many thousands of CVs in an ESU v5 decoder - its CV structure is vastly more complicated than any other decoder on sale.  Once read, you don't need to read them all again - if you only make changes from Decoder Pro, it will record all your changes so you know what is set to what.      

If you have an ESU lokprogrammer device, it can read them much quicker because it uses a proprietary communications method (ie. unique to the lokprogrammer hardware, and not published for others to implement).  

 

 

- Nigel

 

[Ian J.]

I'll have to wait for the developers of JMRI to sort it out then. I certainly don't have the time (or energy) to be fiddling with software when it's something I do during my workday as well. I'm struggling at the end of each work day just to have enough energy to watch a movie on TV, let alone do anything else.

[Crosland]

Thanks to to a tip from one of the other developers I have submitted an easy change which will give support for functions up to 32. This puts the SPROG on a par with the ECoS, which seems sensible as that's the same manufacturer as the decoder.

 

This should be in the next JMRI test release 5.1.2

 

The DCC specs are gradually catching up with the de-facto standards for far more functions, up to 68. before trying to support those I think some thought will be needed into how to display so many functions.

 

Andrew

[Crosland]

I may have jumped the gun :( There's more to this than I thought.

 

I still hope to get something in 5.1.2, but check the release note when it happens.

[Ian J.]

There's no rush, it's a currently nice-to-have that will probably have be done eventually. I appreciate you looking into it though.

[Crosland]

The fix is in the JMRI codebase so will be in the next test release, but I'm unsure when that will be.  

 

Andrew

[Crosland]

JMRI 5.1.2 is now out. I would appreciate any feedback.

[Ian J.]

I normally don't work with test releases, due to the obvious risk of bugs, but I'll see what I can do. It could well be a few days, or even weeks, before I get round to it though.

[beejack]

ok just installed it and started Decoder pro.

Selected the Accurascale deltic in the throttle  window and still only 28 functions showing.

 

Tried a different loco without any function labels set up and only showing 28 functions. 

I am using an NCE SB5, so is this fix sprog specific?

[Nigelcliffe]

1 hour ago, beejack said:

ok just installed it and started Decoder pro.

Selected the Accurascale deltic in the throttle  window and still only 28 functions showing.

 

Tried a different loco without any function labels set up and only showing 28 functions. 

I am using an NCE SB5, so is this fix sprog specific?

 

Yes, its only added to systems which are known to support above F28 - in this case, Sprog. 

 

Your NCE system isn't capable of supporting above F28  (or if it is capable, that capability hasn't been made public for a developer to use ). 

 

 

- Nigel

[Crosland]

What Nigel said 😀

 

Systems known, or believed, to support at least F29 are SPROG, Z21 and DCC++.

[Nigelcliffe]

14 minutes ago, Crosland said:

What Nigel said 😀

 

Systems known, or believed, to support at least F29 are SPROG, Z21 and DCC++.

 

And the ESU ECoS - probably the source of the ESU decoders going up to F32.  

 

(I don't know whether JMRI has included F29-F32 support for the ECoS )

 

- Nigel

[Crosland]

Yes, I should have included "and supported by JMRI" in my post :)