CD matrix / CD hub:
You can most often find the manufacturer from there:
https://reference.discogs.com/wiki/cd-manufacturing-companies-from-matrix
3 Likes
Thanks! Learned again. Now this should be more prominent announced! I rather seldom have seen such scans, though it would be easy to scan the hub too. I haven’t scanned it until now.
But that’s the problem for inexperienced users: you have an interface with thousands of option, but you don’t no where to start, so you decide to add only required information and some details you think they’ll be useful, but no idea what would be really valuable.
1 Like
I picked the next CD and stumbled into the next DiscID jungle.
St. Germain - Tourist
My CD has a perfect match on this one - as far as there is information available - scan images¹, Cat#, barcode and 1 of 17 DiscIDs². Should I remove the other 16?
¹) I suppose it’s no separate release, if a sticker is directly attached to the jewel case cover, even if it’s pointing to a specific event where it was sold.
²) … but the track times for this release do not match those of my CD.
I’d say in theory we would expect usually one disc ID per release. Of course there will be exceptions to this because reality has no obligation to be sensible, but the most common case shoud be one disc ID per release.
IMHO it would be safe to remove disc IDs that are different to the one you can verify and also are linked to other releases of the release group. But I strongly believe you should not remove a disc ID that is not also linked to a different release.
If what I stated above makes sense and can be agreed upon by others than you could remove all disc IDs except the one you can verify and these three that are only assigned to this release:
5 Likes
Yeah, but that’s the problem: there should be not more than one and that would be ZsnJ4fcUVy7zmvyoCaitDxCpxJQ- because that’s my release.
And this fact has implications: the track times do not match as I’ve found out (… after adding some additional scans including matrix codes 
). These belong to a probably different release.
I should better cancel and add a new release - though I will probably reuse the front, back and medium images as they match exactly and the quality of my own scans would not be an improvement after all.
3 Likes
But there could be more than one. If you really want to remove any other you have to look really closely at the times.
Disc ID “yLHJizspP1LxEuWVFmKthByJOj4-” - MusicBrainz very definitely has to stay, it is identical to yours except for the very last sector.
Disc ID “vh8tcKanWqvXqjYbzXhMYHfs3RU-” - MusicBrainz has the 8th track very off, probably should be removed.
Disc ID “8x4hP5LU66i3.TdiG57csx2S8.A-” - MusicBrainz is similar enough that it could be a valid disc ID for this release group, but unless it could be said for sure I would keep it. If it can be shown that it fits better for a different release it could be moved.
2 Likes
No, thanks, but I’ve overlooked the track times. This release should remain untouched. I add another European release as it was distributed when they toured Europe. I cancel!
And I remove my Disc ID from this one, because if there would be a match, it would be the new one.
3 Likes
This release was added in 2002 with these track times without source. Long before releases where split properly. As you say you have exactly this release you should really update it and not duplicate it. It does not make sense IMHO to duplicate this just to leave the mess around. The next logical step if you add a duplicate would be to merge it anyway 
When the releases got split with the NGS release all the split releases got the same list of disc IDs and the same time as there was no way to tell them apart. If you actually have the CD in hand we should take the chance and clean it up. Discogs also agrees on the times of your disc.
Yes, there very likely is another release with the times currently present. But this is one of the other releases in the release group then.
4 Likes
Have added it minutes ago! 
But it’s not this exact release, because there are different track times, so I reuse its matching cover images, remove the Disc ID and this will be still one of the others.
Look at Discogs: there are loads of different releases. All with slightly different track times. So they will all have different Disc IDs.
1 Like
Is this subtile enough? Compare the two scan images:
- my release
- scan from the seemingly matching release
There no variation in text. These are different releases. I will have to remove at least the incorrect Back+Spine!
2 Likes
It’s a bit like one of those “find 5 differences between the two cartoons” puzzles.
I did eventually see that the text formatting on the left is different for “Guest Ernest Ranglin, Conductor Ludovic Navarre”. So yes, these should be two different releases, and you should leave only your own scans on the release you added.
Regarding the track time discrepancy, the discogs entry may be based on the back cover rather than the actual disc. I see that both back covers show track 2 as 5:41.
4 Likes
The point is that we don’t know the track times of the existing release because we have no verified source. These track times where not entered specifically as the times of the European release but could represent any other release. They were also not set from any of the disc IDs. We have no way of verify this other than having someone with an actual disc that matches this release. You have one and it would be totally valid to use your disc ID, which is also set on the existing release, to finally set verified track lengths. Basically we could pick any of the disc IDs for which we know they belong to a release matching the other criteria and set them. But right now your disc ID is the only one for which we know it. As a bonus it also matches the linked discogs entry.
3 Likes
Actually there are only three that could be for this release, and they agree on the times:
https://www.discogs.com/St-Germain-Tourist/release/533682
https://www.discogs.com/St-Germain-Tourist/release/6795056
https://www.discogs.com/St-Germain-Tourist/release/1943223
2 Likes
About vinyl track times
The best, if no big mistake is made on the package (in which case an annotation comes useful), is when the times are taken from this edition’s vinyl label, back cover or leaflet.
About CD track times
On MB, when you have only one Disc ID and you have set track times from it, it is the best and you also have milliseconds (visible with a user script only).
In other cases, it does not necessarily mean much, and same on discogs, it is just some text that might be copy pasted from another edition (including through the MB/discogs clone function, reuse existing medium or whatever it is called).
5 Likes
It is what I used to do, repurpose a bogus cluster release into my specific edition.
But only if
- it is not in anyone’s collection
- there is no mention to a different edition in the big edit history or, if so, these editions already exist in the release group, or I create them
- all Disc IDs I am about to remove are already attached to another release (or contain an edit note stating which edition it is, in which case I also created that edition if new)
- etc.
I did many checks like this and it was too much time consuming.
So now I split/clone and leave the ill defined releases available for people who just want to tag track titles and don’t care about release edition details anyway.
But moreover, I want my edition clean in my MB collection.
And quick.
1 Like
Yes and to add to the issues.
Disc IDs attached with older versions of Picard or some other tagger would also include some bug that results in different TOC being calculated.
Only CD with data tracks were affected, though.
This is a valid reason to have 2 Disc IDs in these cases.
1 Like
LOL. Picked up the digipak version of this album two weeks ago. I will be joining you into release version limbo soon. 
1 Like
Here is a link to one of my “hub” scans. https://musicbrainz.org/release/a2b24809-226e-4728-a469-54eb07ba2bbe/cover-art @jesus2099 gave you some appropriate links for matrix info. I scan the hub [center] of the cd or vinyl and add any other info that is not easily visible in order to help the next Ed with identification. Matrix info may be found on both sides of the medium. Some hub info is hard to read [tiny]. I actually bought a used microscope to aid me in my compulsion. White LED light is the best I have found to highlight the etchings or mirrored image.
Having said that, if you have the same release with barcode, artwork, Disc ID, cat #, etc, you may notice some of the hub markings may be slightly different. Some of these differences are called Variants. Again, since MB does not have an official place in the software to enter matrix info, I would add a scan of your hub and include the differences in the comment section if they were not visible on your scan. This is not MB protocol, just something I do. Releases can have MANY different variants. Some are as minor as to show who was running the pressing for “Alvin” while he was on vacation.
4 Likes
It is!!! 
You’re right and I learned again.
Yes, but it has these times now and is a container for DiscIDs from unknown releases. Maybe I should remove this Disc ID from the container, although it’s not impossible, that the source of this ID was in fact another release.
No longer, as I’ve gone into details of Mastering SID Code and other codes on the matrix 
In fact the third release (St Germain – Tourist) is very close to mine. It seems to have 19 variants on Discogs.
Mine would be No 20: Matrix / Runout: EMI UDEN 5251142 @ 3 3-1-14-NL
Variant 7 has EMI UDEN 5251142 @ 3 3-1-9-NL 
1 Like