Haydn Piano Trios


I have a complete set of Haydn Piano Trios, most of which are Hob. XV . The works are differentiated as Hob XV: 19 etc

Roon misses the plot and puts them all as Keyboard Trios H.15

I have corrected the Movements externally which has made no difference. This makes the album nonsense, it looks like 9 discs of the same composition


Managed to fix by editing he track names manually and splitting off the Movement name


I have the same issue. Did you rename each and every track? That would be too much effort for me. How does it look now?

It was dodgy meta data on disc 2 and 3 , it renamed about 20 tracks thats all

It looks OK now except for 3 trios where the Hob no is not Hob.15 etc


Thanks Mike.

I wonder how the Roon team will deal with those inconsistencies in work definition in their metadata. Grouping all Haydn Trios as Hob XV is an extreme example, but I’ve seen lots of inconsistent defintions with other works. Some examples that come to my mind are:

  • the well-tempered piano, where I have seen a mix of individual “works by BWV” (which is the logcial choice for me) and “works by book”, i.e. all book I grouped as a single work.
  • smaller piano pieces like Brahms op.117, 118 etc.
  • Vivaldis 4 seasons

It is already not easy to agree among classical music collectors on how to group those kind of works, but in Roon it should be consistent. I have no idea how this is going to be adressed on long term. This is one of the main reasons why I need to touch and edit my albums in Roon a lot. My work settings are consistent in terms of file tags.
I would expect to see some inconsistency between identified and non-identfied albums, but what I very often see is already a mix of ways of grouping among identified albums… :frowning:

I must admit I didn’t expect to do corrections to existing track names to get Roon to recognize. To be fair the original metadata was dodgy.

I ripped these direct from CD so that’s what it originally found

I looked up the disc on Discogs and manually corrected , fortunately it wasn’t too much after that it ID’ed OK and fixed the rest

I suppose Haydn is not quite std with the H IV:65 format etc


Sorry Correct that it still doesn’t ID , its giving me Brendel Liszt suggestions – Bizzare


@Mike_O_Neill Agreed: there is no breakdown of Hob. 15 into the constituent compositions whatsoever. I’ve asked Rovi to remedy this. Please note however that they may choose to do so in a way which Roon has not implemented (for very good reasons) at this time.

@Klaus_Kammerer1 The simple answer is that, until we have a better solution (which, of course, is metadata dependent one way or the other), we take what Rovi provides. We are well aware of the problems associated with the examples you quote above and we’ve had internal team discussions about how to solve this in a 1st class way.

You say that individual “works by BWV” is the logical choice for you. We respect that, but say we split The Four Seasons into the four individual concerti, would you really want to lose the context (and metadata) of the grouping?

We want to do a better job here. For example, if you had a recording of Vivaldi’s entire Opus 8, it would be great to be able to access metadata for the constituent works at all levels (Op. 8, Op. 8/1-4 (The Four Seasons), Op. 8/x (where x is any of the 12 constituent concerti). The same applies to the WTC. But that’s quite a bit of work that needs working through thoroughly; and, as usual, it’s a question of prioritization.

Anyway, I wanted to respond to let you know that these are issues that we acknowledge, are thinking about, and intend to solve; but it’s not the Roon way to implement a half-baked solution.

Cc: @mike @brian


thanks a lot for your reply. It’s very much appreciated to get this feedback. I’m aware that all this takes time and I don’t want to be nagging too much.

That’s a fair statement. You cannot do very much right now, can you. It will be interesting to see where you will end. I have lots of respect for your approach and I am in no way blaming the current metadata situation on you, just to be perfectly clear.

What’s logical to me might not be logical to someone else. I’m ok with that. Much of my current frustration with the current metadata situation for works is with the inconsistency. I am going into Roon with my consistent view on how granular I maintain my works and I’m offered a crude mix of different approaches. That’s what’s annoying me. I would be willing to adapt my tagging to what Roon offers me, but at the moment I often do not know how to best modify my tags to fit in.

What IS the right way to go with Brahms op.118 for example. I find both variants in Roon for identified albums… If I want to have a consistent library I need to decide… It’s not really solvable without someone to decide on what the master strategy should be.

Your comment on the Four Seasons is a very valid one and I am ready to admit that I did not make my mind up yet on how to approach this for myself. The four seasons are a very good example of this dilemma. Everybody knows them as a “Work”. I’m no musicologist and I do not know whether Vivaldi intended to compose it as a separate cycle within “il cimento dell’armonia e dell’inventione” which comprises of 12 concerti, as you also point out.

Just like “L’estro armonico” and “La Stravaganza”, by the way. In addition to that, each of those concertos in these collections have a separate RV catalogue number…
So if I start with the one work by catalogue number paradigm (and prefer RV over op.), I’ll have 12 works for each collection. Or should I go with “one work by collection”? Both ways are valid but both ways would not properly cover our current way of presenting these 4 concertos know as the 4 seasons… I don’t think that you can solve that without an additional hierarchy layer like “collection”.
In my custom file tags I had a tag called “work group” that contained the name of the grouping attribute. But I still didn’t know how to maintain the 4 seasons properly with that approach…

So I am waiting for someone to tell me :smiley: :smiley: :smiley:

Fully acknowledged, But I’m sure you know that in the end you cannot please everyone. As I said before. I for myself am willing to adapt and adjust if I know where we are heading.

And my apologies to @Mike_O_Neill for hijacking his thread…:flushed:

1 Like

No need to apologize the more debate the better. Thanks for the repsponse @Joel

Four seasons and WTK are very good examples , being such “Big Hits” they tend to get singled out and named only by their nicknames.

My view is that they should follow their proper Op or RV Number but with a Nickname that the search engine will pickup

So … Op.8 No 1 “Four Seasons - Spring”

Who refers to the Moonlight Sonata as Op. 27 No.2 , I suppose a few , so offer both alternatives , and we will all be happy

The definition of an intellectual – "Someone who can listen the Overture to William Tell without thinking of the Lone Ranger:

The bottom line is consistency and the ability to EASILY find the piece you want in our VAST libraries




Just a thought, how easy would it be to allow Composition and Movement to selectable Prefer File

I have 95+ % tagged but they are not used by Roon

It would allow for a level of consistency, I could ensure that outside Roon



you already can do that in Library settings

However, I had mixed experiences with that setting, because it does not really do what I would have expected. see also the following threads

Hi Klauss

I’d missed that. Am I right that is a global setting not even at Album level, Album level would be first prize, or is that the Muli part Composition metadata preference in the album settings

I assume if you set that all composite and part tags must be in place or else

I have desire to wreck what I have to date merely to tweak it


It’s available at both Library Import Setting and per-album levels. :1st_place_medal:


Is that the Muli Composition Part setting ?


Yes. #10chars