Core ripping classical

Ripped another one

Misidentified by ROVI. Most of the tracks are wrong. It has decided that the composer is Vaughn Williams which is correct (but uneditable) and the conductor is Sir Adrian Boult. Which it isnt. And I cant change either.

Now I could fix up the dozen or so tracks which are all listed as "Symphony No 2 in G Major ("A London Symphony")" 

I could use the iphonex keyboard for this.

Or I could chainsaw my leg off, cos that would probably be preferable.

Now there’s a thought????

Jon, what a pity to still see all those issues with metadata ... I know that I keep on repeating myself: but a NAS offers so much more flexibility (at least for my needs) ... maybe I'm too picky when it comes to metadata editing :-) 

This issue is well known, the May 5th Uniti progress thread Naim were pretty specific about what they were NOT going to enable. One of them was editing of composers and conductors. Nothing seems to have changed their mind since as far as I can see?

Gazza posted:

This issue is well known, the May 5th Uniti progress thread Naim were pretty specific about what they were NOT going to enable. One of them was editing of composers and conductors. Nothing seems to have changed their mind since as far as I can see?

But the 27th June update said that "metadata for composers and conductors" is in the release planning. Of course it doesn't say what that means, but the indexing does now take account of composer and conductor so presumably editing those pieces of metadata will follow in due time.

best

David

Thanks a David ,let’s hope so for those interested in classical music

Have you tried ripping the same specific  CDs on another platform (e.g. dBPoweramp, and do they come out correctly as expected that way? Just wondering if the fault is indeed Core or something to do with those CDs. 

When I ripped my CD collection I had A fair few problems with incorrect or missing data. Many still not fixed because O didn’t know then to check each rip  immediately after doing, so just rattled through the lot. the word offenders are classical, though not universal. And my ripping was with dBPoweramp ..

jon honeyball posted:

Now I could fix up the dozen or so tracks which are all listed as "Symphony No 2 in G Major ("A London Symphony")" 

I could use the iphonex keyboard for this.

Or I could chainsaw my leg off, cos that would probably be preferable.

Don't do that, it'll make that bike somewhat harder to ride

TallGuy posted:

jon honeyball posted:

Now I could fix up the dozen or so tracks which are all listed as "Symphony No 2 in G Major ("A London Symphony")" 

I could use the iphonex keyboard for this.

Or I could chainsaw my leg off, cos that would probably be preferable.

Don't do that, it'll make that bike somewhat harder to ride

It's ok he would still be able to ride his hobbyhorse.

David Hendon posted:

Gazza posted:

This issue is well known, the May 5th Uniti progress thread Naim were pretty specific about what they were NOT going to enable. One of them was editing of composers and conductors. Nothing seems to have changed their mind since as far as I can see?

But the 27th June update said that "metadata for composers and conductors" is in the release planning. Of course it doesn't say what that means, but the indexing does now take account of composer and conductor so presumably editing those pieces of metadata will follow in due time.

best

David

The other question is whether it is possible to set the Core to save rips to the downloads folder instead of saving the rips to the music folder and, if not, why so.

Saving rips to the downloads folder would allow one to edit no matter which indexes one wants to edit using any third-party software. By contrast, if one opts for saving the rips to the music folder, one is at the mercy of Naim's metaediting software that might improve over time or perhaps not improve at all.

Also, relying on a proprietary tool for storing metadata seems a very bizarre idea in 2017. One can conveniently store metadata in .flac or .wav files. By contrast, a proprietary tool for managing metadata locks one in that format. This basically means that one gives away interoperability for nothing. Why would one want to do so?

Have to say from my very limited experience this has always been the case with ripping classical.  

I have found that editing metadata for consistancy takes longer than ripping the d@mn disc. Just how many tracks can there be entitled “001 Allegro” ?!

So I’ve reverted to old technology.........getting out of the chair, walking to the study*, getting the CD out of its box, walking back to the living room/office/conservatory/etc and stuffing it into the CDX/Karik/CD3.5/Meridian 200.........

*and don’t the medics say that us old folks should walk a lot to stay fit?

I really would like to know why Naim has decided not to implement manual metadata editing of composer and conductor tags with v2.5. Is it only a question of time and complexity for another release or is there a fundamental issue with this?

Chag -

Most classical stuff seems to tag, not as I'd like it.
For example, if I want to hear "Beethoven's 5th", I want to search by "Beethoven" not the conductor, whose name I never know, nor the orchestra who performed it as I want to listen to "Beethoven".
I have an album of Chopin but it's listed under artist "Alan Kogoswoski" - not very helpful when I want to listen to Chopin.
Tchaikovsky anyone? Better search under "Andrew Litton". Where else?

Some titles ripped using dbpowramp but most ripped on my HDX. I hadn't realised there was an issue until I'd ripped over 100 Classical CD's so I can't be bothered to try to "correct" them to my liking.

Genres: many are listed as "Pop - other" which to my mind means "pop music" rather than "popular classical"....

It just takes forever to find what I want play when listening to Classical.

This is one of the library frustrations I have using Audirvana as my renderer, so certainly not unique to Core.

Interestingly, when I had an ND5XS there was a solution that worked yet was so simple that I didn’t even realise my metadata was an issue until I moved to Audirvana: My filing system was very simple: folders for each genre (set quite broadly, by how I tend to think ‘what type of music do I fancy’ classical, opera, rock..), nested within them folders for the artist (composer, surname first for clasdical). And within those folders sat each album as a folder,  the individual tracks within all commencing with the track number (with a leading zero below 10) and double/triple albums numbered with a prefix of1,2,3 so 101, 102...

With Logitech Media Server on my Zyxel NAS, and subsequently Serviio on Mac Mini acting as NAS, the Naim app controlling ND5XS unfailingly navigated perfectly.

blythe posted:

Most classical stuff seems to tag, not as I'd like it.
For example, if I want to hear "Beethoven's 5th", I want to search by "Beethoven" not the conductor, whose name I never know, nor the orchestra who performed it as I want to listen to "Beethoven".
I have an album of Chopin but it's listed under artist "Alan Kogoswoski" - not very helpful when I want to listen to Chopin.
Tchaikovsky anyone? Better search under "Andrew Litton". Where else?

Some titles ripped using dbpowramp but most ripped on my HDX. I hadn't realised there was an issue until I'd ripped over 100 Classical CD's so I can't be bothered to try to "correct" them to my liking.

Genres: many are listed as "Pop - other" which to my mind means "pop music" rather than "popular classical"....

It just takes forever to find what I want play when listening to Classical.

I have two coping strategies that I have adopted with classical music on the Core.

1) I always edit the album name to be composer surname - musical work. Then I use album view to scroll through and find what I want. This is ok for the 500 or so albums I have, especially as you can jump to the first letter of the album name easily, but I can see that it might not work as well for 5000 albums.

2) If the Core rips an album with a non-obvious or incorrect album name - a common occurrence with classical music - then I find the album after it has been saved to the Core by looking in "Recent CDs" at the most recent. Then I immediately edit the album name and cover artwork to put it into a more sensible place.

best

David

David Hendon posted:

blythe posted:

Most classical stuff seems to tag, not as I'd like it.
For example, if I want to hear "Beethoven's 5th", I want to search by "Beethoven" not the conductor, whose name I never know, nor the orchestra who performed it as I want to listen to "Beethoven".
I have an album of Chopin but it's listed under artist "Alan Kogoswoski" - not very helpful when I want to listen to Chopin.
Tchaikovsky anyone? Better search under "Andrew Litton". Where else?

Some titles ripped using dbpowramp but most ripped on my HDX. I hadn't realised there was an issue until I'd ripped over 100 Classical CD's so I can't be bothered to try to "correct" them to my liking.

Genres: many are listed as "Pop - other" which to my mind means "pop music" rather than "popular classical"....

It just takes forever to find what I want play when listening to Classical.

I have two coping strategies that I have adopted with classical music on the Core.

1) I always edit the album name to be composer surname - musical work. Then I use album view to scroll through and find what I want. This is ok for the 500 or so albums I have, especially as you can jump to the first letter of the album name easily, but I can see that it might not work as well for 5000 albums.

...

I think that the choice of a naming scheme should not be informed by the (absence of) features of a specific software of a specific device. But starting an album name with a composer's surname (or with the surnames of the composers) is, I believe, a very good guideline. I tend to fill in the "Album" field with what is printed on the album's cover and the composers' names very often come first. Still, it is clear that, in order to browse sizeable collections in an effective manner, one needs the features offered by state-of-the-art servers like MinimServer, Asset, etc. If I had to rely on what the Core's server has to offer for classifying music contents, I would rather stay with my unripped CD collection. 

One reason I haven't relegated my Classical CD's to the cupboard upstairs. I keep them by my main hifi just in case... I wish I'd paid more attention whilst ripping them and sorted the tagging then..

It sure sounds like Core ripping classical is not satisfying.  Honestly, my own rips and tagging using Asset or MinimServer are not satisfying either. I have yet to figure out just what system of tagging to adopt to make it easier for me to find what I want to find.

I typically browse by Album/Artist for rock, jazz, etc.  I think that I need to get the Composer tags right and when I want to listen to classical, browse by Composer. (99% of the time that is my search criterion.)

I've posted about this before ... and been challenged/criticised for my approach but, for me, I find the best way to tag Classical (and Soundtrack) works is to use the Composer's name as the Album Artist ... and then use the Conductor, Soloist, Orchestra, Ensemble ... as the Track Artist.

Using this method, e.g. Beethoven's Triple Concerto in C, Op.56 [Oistrakh/Knushevitsky/Oborin/Sargent/PO] (1958) can be located by:

Album: name as above

Album Artist: Beethoven,Ludwig van

Artist / Album: any of the 4 names (but not Philharmonia Orchestra as this is the fifth name)

Composer: Beethoven,Ludwig van

Year: 1958

(It may be possible to find it under Conductor, too, but I don't use this facility)

If you keep to the Composer's name only then works such as Music for the Funeral of Queen Mary, Z.860 [Christophers/Sixteen] (1994) will be awfully brief ... Purcell's contribution was about 1/3 only of the playing time.

This approach is not perfect and does take time getting the tags right ... but it seems to work without being complicated.

Mitch

Bart posted:

It sure sounds like Core ripping classical is not satisfying.  Honestly, my own rips and tagging using Asset or MinimServer are not satisfying either. I have yet to figure out just what system of tagging to adopt to make it easier for me to find what I want to find.

I typically browse by Album/Artist for rock, jazz, etc.  I think that I need to get the Composer tags right and when I want to listen to classical, browse by Composer. (99% of the time that is my search criterion.)

If you use MinimServer you can use the indexes that you please to classify your music. For classical music, I find Composer, Work, Conductor, Ensemble, Period, Genre, Form, Artist and Album particularly useful. In most cases I pick up an album or a track by browsing via

  Composer > Work >

But sometimes I also use

  Period > Composer > Work >

  Form > Composer > Work

or perhaps

  Ensemble > Composer >

I have found out that I very rarely use Artist, Genre and Conductor but that might change if I change my habits, of course. Generally speaking, I think that being able to search after Composer, Work, Ensemble and Conductor (and Director if you have opera albums) is mandatory.

Musicmad posted:

...

If you keep to the Composer's name only then works such as Music for the Funeral of Queen Mary, Z.860 [Christophers/Sixteen] (1994) will be awfully brief ... Purcell's contribution was about 1/3 only of the playing time.

...

Not necessarily, it depends on the UPnP server. If you use MinimServer and search after Composer > Work you will first see only the tracks of  Music for the Funeral of Queen Mary, Z.860 [Christophers/Sixteen] (1994) for which the composer is Purcell (as one would expect) but you will be also given the opportunity to select the whole album.

For me the simplest tagging scheme is to put in Album the text which is on the cover of a CD and fill in Composer, Work, Conductor, Ensemble, Artist, Period, Form, etc. with their respective values or leave them blank.

It goes without saying that a track/file can be associated to more than one artist and, in fact, to more than one conductor (parts of War Requiem are often performed with two or three conductors, for instance), to more than one composer (typical for transcriptions) and, especially in opera works, to more than one ensemble (orchestra and chor) and to more than one work ("Das Rheingold" is part of  "Der Ring des Nibelungen").

This complexity implies that, if you try to compensate for the weaknesses of a UPnP server by using a clever tagging scheme you very likely will get into troubles. If you call things by name and use Composer for composers, Conductor for conductors, Work for works, etc. you can hardly make anything wrong.

The problem I have is that the UnitiServe did such a good job with classical music. I used the PC program to correct minor error in the metadata to get the Album name, composer etc right. Mostly it messed up box set or multiple CD, and one needs a consistent Album title convention including composer at the beginning and the CD # at the end. The US then did the cover art better. I found the App then sometime later synced with the right album cover in most cases, but if it did not the cover art search was good and allowed one to edit the Google image search screen. Sometimes with box sets different cover art came up and I corrected it to be consistent throughout the set or to have the same opera/symphony cover used for each opera/symphony within a box set.

I inherited my cousin’s opera and classical collection and had very few metadata problems. Here’s an example below. Can’t think why the Core did not build on the really good search algorithm of US. Alan G Melville is not particularly well know! 

Phil

blythe posted:

Most classical stuff seems to tag, not as I'd like it.
For example, if I want to hear "Beethoven's 5th", I want to search by "Beethoven" not the conductor, whose name I never know, nor the orchestra who performed it as I want to listen to "Beethoven".
I have an album of Chopin but it's listed under artist "Alan Kogoswoski" - not very helpful when I want to listen to Chopin.
Tchaikovsky anyone? Better search under "Andrew Litton". Where else?

Some titles ripped using dbpowramp but most ripped on my HDX. I hadn't realised there was an issue until I'd ripped over 100 Classical CD's so I can't be bothered to try to "correct" them to my liking.

Genres: many are listed as "Pop - other" which to my mind means "pop music" rather than "popular classical"....

It just takes forever to find what I want play when listening to Classical.

To avoid what you describe in the first paragraph, Audirvana creates a field named "artists", filled with the names of all the artists in the different fields: album artist, track artist, conductor, soloist, composer, etc.  I suspect that when I search by Bee.. they look at this field and find the album, track and artist no mater in what "artist field" was Beethoven originally. Also allows you to correct different spellings for artists names creating an "artists data base" which is very easy to modify.

I have not used dbPoweramp, but Metadatics that is very easy to use to modify metadata and file names in your folders with little typing.