The pain of ripping CD Collections

Bert Schurink posted:

 

 So is there anybody who has solved this, or will I have to keep on going through this pain every time I buy a box....

I think you have to keep going through the pain...

On the other hand, have you tried MusiCHI?

Bert, I have done some box sets, but yes classical is a pain ( seconded, by 1990's cd singles... ) 

1. I have found, if nothing in the 1000x1000 + range comes up in dbpoweramp, Google search, followed by searching Amazon - surprising how many 1400 pics they have.  Next option is scanning the cover yourself...

2. The problem with box sets is sometime you get the duplicates, so how to alter the title to differentiate...   I do find I use the Comments for info on the boxset, i..e. "11 - Mozart 111 Masterworks - DG"  as I tend not to use the 11/55 in the CD number as these are 'individual' albums.

3. Back in the day, this was the only way, at least it's only a small percentage, but I have found if I get a 'new' box set, generally a year or so later the the information is better as  a) depending on the ripper settings, it's minimum one month before the information 'sends home' to help others.  b) multiple other folks have taken the time and the information is better.

4. Worse case, original release date...

5. Google translate :-)  Seriously though Google Chrome is the best for auto translation of webpages. Very useful.

One word of caution, If you have set all tagging to be saved, I have recently found if dbpoweramp pre populates a cd, and you go and re select / update it to the correct - the 'hidden' tagging information from the original cd is retained/saved on rip. And you only find this when you go to the saved files and right click/Edit ID Tag and info like Album Sort, Style, Composer sort are shown.  To mitigate this, make sure you have the 'ID Meta' screen at the bottom (left of Album art) showing and check through that data too... 

Good luck

Bert, I agree with your observations. Perhaps the ripping pain can be alleviated, up to a certain extent, by adopting a tagging scheme that makes browsing a collection less dependent on cover art.

Since I am using MinimServer, I have introduced a "work" (I have also seen it called "composition") index. For instance, I have following "work" values for Shostakovich:

• Klavierquintett in g-Moll, Op. 57 (1940)
• Sieben Puppentänze für Klavier solo (1952–1962)
• Sinfonie Nr. 01 in f-Moll, Op. 10 (1924-1925)
• Sinfonie Nr. 10 in e-Moll, Op. 93 (1953)
• Sinfonie Nr. 15 in A-Dur, Op. 141 (1971)
• Sonate für Cello und Klavier in d-Moll, Op. 40 (1934)
• Streichquartett Nr. 2 in A-Dur, Op. 68 (1944)
• Violinkonzert Nr. 1 in a-Moll, Op. 77 (1947-1948)

Introducing suitable "work" values or tags can still be a pain but is conceptually straightforward. More importantly, it can be done offline and does not rely on artwork that might be unavailable.

The "work" index has very much changed the way I am now browsing my library. I tend to find the music I want to listen to following a "composer -> work" browsing idiom rathen than a "composer -> album" browsing path. With this practice, errors in cover art are still annoying but have little effect on the browsing experience.

Best, nbpf

nbpf posted:

Bert, I agree with your observations. Perhaps the ripping pain can be alleviated, up to a certain extent, by adopting a tagging scheme that makes browsing a collection less dependent on cover art.

Since I am using MinimServer, I have introduced a "work" (I have also seen it called "composition") index. For instance, I have following "work" values for Shostakovich:

• Klavierquintett in g-Moll, Op. 57 (1940)
• Sieben Puppentänze für Klavier solo (1952–1962)
• Sinfonie Nr. 01 in f-Moll, Op. 10 (1924-1925)
• Sinfonie Nr. 10 in e-Moll, Op. 93 (1953)
• Sinfonie Nr. 15 in A-Dur, Op. 141 (1971)
• Sonate für Cello und Klavier in d-Moll, Op. 40 (1934)
• Streichquartett Nr. 2 in A-Dur, Op. 68 (1944)
• Violinkonzert Nr. 1 in a-Moll, Op. 77 (1947-1948)

Introducing suitable "work" values or tags can still be a pain but is conceptually straightforward. More importantly, it can be done offline and does not rely on artwork that might be unavailable.

The "work" index has very much changed the way I am now browsing my library. I tend to find the music I want to listen to following a "composer -> work" browsing idiom rathen than a "composer -> album" browsing path. With this practice, errors in cover art are still annoying but have little effect on the browsing experience.

Best, nbpf

Thanks for the tip. I also have a tagging scheme which I use. I am just a perfectionist. I don't want to have to see 25 or 50 covers which are the same and then constantly to search only by tags. visual is just ten times faster.

This is also an issue just the same if you use Unitiserve. But you get the added fun of sometimes the album being recognised as being something else altogether and the US being what it is, you have to let it rip and then go painfully through and edit everything using n-serve.

I have also found that Japanese discs are a bit hit and miss.  I don't have many, but one helpfully found the album title and track names in English, whereas another which plays fine in the CD player and even has an English album name on the sleeve simply is rejected by the US as "No CD present".

I find album art is often wrong, I think because classical music albums are often released with revised art work. But searching in google image search with "composer name of work cover" usually provides the right cover picture on the first page of hits. 

I think the most important thing though is to sort it out as you go along otherwise the task rapidly gets too big and off-putting. I also put a small sticky label on the back of the CD case so that I know this is one I have ripped, otherwise it is all too easy to rip it a second time which on the US just over-writes all the careful edits!

best

David

www.albumart.org is an excellent resource for finding album art. You can type in an album title, some of it, or the UPC or EAN code.

I use PerfectTunes from the makers of dbPoweramp. This has an album art module which is very effective. 

likesmusic posted:

www.albumart.org is an excellent resource for finding album art. You can type in an album title, some of it, or the UPC or EAN code.

Great link - thanks

There is a nice little program on the sourceforge called album art downloader it searches several sites for cd covers

Would love to download CD boxes rather than ripping for hours. Guess music industry needs to improve quantity. Me think of buying a PS3 for ripping SACD because some content is still not available as download...I hate it!

I find boxes to be the hardest. Right know, I have a box as one album. But lately I am thinking to let go of the idea of disks and boxes and group a piece as one 'album'. For instance, a cd with two Dvorak symphonies ripped as two albums, one for each symphony, but with ten same cover art.

if I search to play, I search by composer or piece, so for me that would be a more logical approach.

jmtennapel posted:

I find boxes to be the hardest. Right know, I have a box as one album. But lately I am thinking to let go of the idea of disks and boxes and group a piece as one 'album'. For instance, a cd with two Dvorak symphonies ripped as two albums, one for each symphony, but with ten same cover art.

if I search to play, I search by composer or piece, so for me that would be a more logical approach.

I am not sure I userstand what you mean by "ripping a cd with 2 Dvorak symphonies as 2 albums"? You could ...

  1) ... tag the tracks of one symphony with an "album" value and the tracks of the other with another value,

  2) ... tag all tracks with the same album value but store them in different folders,

... or do something else. In the beginning, I have used 2) for albums that contain works from different composers. This allowed me to have all the works of a given composer in subfolders of a folder named after that composer. But I never split albums according to 1). This would be, I believe, is a mistake.  

Meanwhile I have adopted a scheme which I find more practical than 2) and at the same time more flexible. I keep all the data in a "data" folder. This is organized straightforwardly with one cd (or cd collection) per folder. Thus, for instance, I have three folders

    data / Takács Quartet / Britten | String Quartets Nos 1, 2 & 3 (2013)

    data / Takács Quartet / Janáček & Smetana | String Quartets (2014)

    data / Takács Quartet / Schubert | Death and the Maiden, Rosamunde (2006)

for the three cds by the Takács Quartet that I have recently bought. At the same time, I have a "composer" folder. This folder contains only symbolic links to items in "data". These are organized in a composer / work / interpretation hierarchy. Thus, for, instance, I have folders like

    composers / Smetana, Bedřich (1824-1884) / String Quartet No. 1 in E Minor / Pavel Haas Quartet (2015)

    composers / Smetana, Bedřich (1824-1884) / String Quartet No. 1 in E Minor / Takács Quartet (2014)

    composers / Smetana, Bedřich (1824-1884) / String Quartet No. 2 in D Minor / Pavel Haas Quartet (2015)

with sysmbolic links to the corresponding items in "data". Thus, for example, in

    composers / Smetana, Bedřich (1824-1884) / String Quartet No. 1 in E Minor / Takács Quartet (2014)

I have 6 symbolic links to files in

    data / Takács Quartet / Janáček & Smetana | String Quartets (2014)

with 4 symbolic links pointing to tracks and two to artwork and liner notes. This organization is also the basis for the indexing scheme that I have adopted. Thus, all tracks of

    data / Takács Quartet / Janáček & Smetana | String Quartets (2014)

have

    album = Janáček & Smetana | String Quartets | Takács Quartet

But the Janáček tracks are tagged with

    composer = Janáček, Leoš (1854-1928)

    work = String Quartet No. 2 "Intimate letters"

while the "Smetana" tracks have, among others, the

    composer = Smetana, Bedřich (1824-1884)

    work = String Quartet No. 1 in E Minor

tags. There is obviously no "right" way of organizing a collection of musical contents (althoughthere are many ways which are obviously wrong!) but I have found the scheme outlined above to work very well for me. A flat album view (say, for classical, jazz and a few more "genre" values) very much reflects the way I had ordered my CDs in my shelves. The "composer" folder provides a structured overview of all the works (and, for a work, of the different interpretations) that I have of a given composer by just taking a look at my library files. This makes it easy to manage my collection remotely (for instance, from my laptop during spare time) and without the need of a control point. It also makes it very easy to discover and correct tagging errors and inconsistencies.

One final remark: I would strongly advise against adopting a method for organising files and folders (or indexes and tags) that is motivated by the usage of a particular software or program, be this a control point, a UPnP server or a tagging software. This seems to be a rather common practice (for instance, using a "performer" index to carry information about conductors because a particular system not supporting a "conductor" index) but is, in my view, a major mistake: our way of organizing a collection should be informed by specific interests and needs, not by the programs that we currently use.

Best, nbpf

nbpf: I do, will and want to rely on tags. That's how UPnP servers index and serve music and that's how I browse a vast collection of music.

Indeed, there is no right or wrong, only a shortage of implementation of the full ID3v2 implementation. Sadly. That would solve our daily pain. 

I am thinking about use 'album' as the tag to indicate the 'piece', like 'Mahler: Symphony no.9', even when the origin is a Box set with all Symphonies by Mahler from Bernstein or Chailly. If the cd has a symphony of Sibelius combined with a piece from Nielsen, I would separate the two pieces. I don't care they came originally on one CD, I listen by piece, not by album.

So, it is more geared toward my listening habits than anything else.

jmtennapel posted:

...

I am thinking about use 'album' as the tag to indicate the 'piece', like 'Mahler: Symphony no.9', even when the origin is a Box set with all Symphonies by Mahler from Bernstein or Chailly. If the cd has a symphony of Sibelius combined with a piece from Nielsen, I would separate the two pieces. I don't care they came originally on one CD, I listen by piece, not by album.

As I explained above, I do not see any obvious reason for (mis)using the "album" index to carry information that could more properly be represented by values of a specific index. I think that the value of "album" should more or less faithfully describe what is on the cover of a CD. In other words: album name and album cover art should help identify a specific album or  CD, be this part of a box or not. I also would expect one to use a namegiving scheme for album names that allows one to browse through albums in roughly the same order in which one would have the corresponding CDs on shelves. Again, why don't you simply use the album index to describe albums and discriminate between the two (or more) pieces with a "piece" ("composition", "work", etc.) index? This seems a very natural approach to my and complies with a "call things by their name" principle which I consider to be essential. Just my two cents, of course. Best, nbpf

It all comes down to the way that you think and is hard-wired into us. I prefer to choose my album by selecting the cover picture. It's only if I can't remember exactly which album something is on that I would dream of searching. In fact k would rather select a likely album and look at the track listing than search using words. Personally I would never spend hours (or even minutes come to that) arranging things systematically by sophisticated tagging. The right way to arrange things is the way that works for you personally.

best

David

David Hendon posted:

It all comes down to the way that you think and is hard-wired into us. I prefer to choose my album by selecting the cover picture. It's only if I can't remember exactly which album something is on that I would dream of searching. In fact k would rather select a likely album and look at the track listing than search using words. Personally I would never spend hours (or even minutes come to that) arranging things systematically by sophisticated tagging. The right way to arrange things is the way that works for you personally.

best

David

It goes without saying that most of what we are discussing in this threas (and, by large, in this forum) are personal preferences. We try to explain the approaches (and the schemes and rules) that we find useful for ripping, tagging or for naming files and folders in the hope that these approaches could be useful for others as well.

If you do not care much about ripping and tagging schemes and you are fine with browsing your collection after albums and track names, it's fine. This thread is likely not going to be very interesting for you.

On the other hand, if you are not fully satisfied with your current ripping and tagging and files and folders namegiving practice -- be this because you are a perfectionist, because organising a collection of musical contents and providing flexible searching and browsing functionalities to different users is your job or for whatever other reason -- you might actually find this thread interesting.

It might also be that case that you just come to thinking that you could perhaps apply a given scheme but you are not sure about the consequences of this approach. This seems to be the kind of question at the root of  jmtennapel's post who is thinking about splitting albums with works by different composers by introducing different values of the "album" index for tracks by different composers.

As I mentioned, I think that this approach is questionable. The rationale is that the same goal can be achieved with a more canonical approach and that giving different tracks of the same album different "album" index values seems to go against a principle -- that of naming things by their names -- which I consider to be worth to try to comply with. Of course, one can use the "album" index (and, in fact, whatever index) to encode whatever information one likes.

Once again, this is my personal view on a subject that is possibly irrelevant to many of us. This is not a problem, I guess.

If your starting point is that whatever is put together on a cd is the truth, that is fine. I see cds as random carriers of sound. Even when I had it all organized on shelf, I had to find what I was looking for by memory (knowing which was on what album). It was a single index entry into a library. Thanks to digitalization and computers we can have multiple indexes on the same content and as such, the organization of ones music is too much stuck to to the shelf way of organizing in the capabilities we have.

Proof: try to locate an exact piece in Spotify or Tidal. Horrendous!

Bert,

For Pop/Jazz/Blues music I keep the original album. For classical music however I define an "Album" as "A composition/piece/concert/symphony", and the field Artist I use for composer. Simply because classical music was written by piece. it does not make sense to have one album containing more than one piece. Also CD boxes I break down in separate albums, each containing a "Composition". This can be a hell of a job but it pays off once your collection is growing.

This is not unambiguous, for instance what to do with Schubert songs? Several songs are grouped, such as "Winterreise", but what about the many separate songs he wrote? Piano pieces from Ravel can be grouped in one composition "Gaspard de la Nuit", others are separate small pieces. Well, I combine those with some subjective logic.

A Philips CD box of Alfred Brendel playing Schubert I broke down in separate compositions (Sonatas/Impromptus etc). A lot of work but it pays of in the end.

As cover art I like to use the original album cover, which I search on Googles Images. To find the cover of the original LP you can add the word 'vinyl' . E.g. "Schubert Brendel vinyl". You can also search on eBay or Amazon on vinyl albums to find the original LP cover.

Good luck!

Went through the same pain when I digitized my collection (1500+ discs), except that while doing it I ended up writing an app to help me along.

So if you are ripping on  iTunes/Mac then iBatch for iTunes will help massively.

It grabs complete album descriptions/artwork from discogs.com, by search or ID, and offers unparalleled batch processing. Current version also offers, archiving and music library backup and recovery.

jmtennapel posted:

If your starting point is that whatever is put together on a cd is the truth, that is fine. I see cds as random carriers of sound. Even when I had it all organized on shelf, I had to find what I was looking for by memory (knowing which was on what album). It was a single index entry into a library. Thanks to digitalization and computers we can have multiple indexes on the same content and as such, the organization of ones music is too much stuck to to the shelf way of organizing in the capabilities we have.

Proof: try to locate an exact piece in Spotify or Tidal. Horrendous!

I do not know what you mean by "being the truth",  but looking at albums as random carriers of sound and not caring about whatever is put together on a cd is fine, I think.

But then, I would expect you also not to care about "album" values, not to purposely use such values to encode information that could be encoded more naturally with another index.

Of course, that's your choice. And if you do not plan to share your collection with others, it certainly does not matter whether you use "album" values to encode information about compositions or "date" values to encode information about conductors or, in fact, whatever other scheme.

Best, nbpf

nbpf posted:

jmtennapel posted:

If your starting point is that whatever is put together on a cd is the truth, that is fine. I see cds as random carriers of sound. Even when I had it all organized on shelf, I had to find what I was looking for by memory (knowing which was on what album). It was a single index entry into a library. Thanks to digitalization and computers we can have multiple indexes on the same content and as such, the organization of ones music is too much stuck to to the shelf way of organizing in the capabilities we have.

Proof: try to locate an exact piece in Spotify or Tidal. Horrendous!

I do not know what you mean by "being the truth",  but looking at albums as random carriers of sound and not caring about whatever is put together on a cd is fine, I think.

But then, I would expect you also not to care about "album" values, not to purposely use such values to encode information that could be encoded more naturally with another index.

Of course, that's your choice. And if you do not plan to share your collection with others, it certainly does not matter whether you use "album" values to encode information about compositions or "date" values to encode information about conductors or, in fact, whatever other scheme.

Best, nbpf

As such I find the concept of organizing around what you find an interesting way, while it only works for classical......but I still would belief that sometimes the connection of some aspects on an album also has a meaning. Very often the review is also about the pairing of pieces and the thought behind the pairing.

as my collection is too big I anyhow have to stay on an album level to stay consistent. In the meanwhile I have the Brendel job behind me and I am happy with the end result.