Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Submitting multiple rips of the same CD (Read 1633 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Submitting multiple rips of the same CD

EAC's behaviour of submitting rip results to AR and CTDB is a bit opaque, I'm wondering what happens when you rip a CD more than once.

Numerous times with some problem CDs I have made multiple rips, buffing out the CD or changing settings each time until it rips cleanly. So there might be 5x wrong results and one correct result, what happens in these situations, do they all get submitted?

Re: Submitting multiple rips of the same CD

Reply #1 –
AccurateRip submissions are based on individual tracks. Submissions go to a holding area first where multiple submissions from the same IP and hardware are eliminated before adding them to the database.

The CUETools Database will initially accept unique submissions based on the CRC32 hash of the whole CD. There's a cleanup script that's supposed to run periodically to eliminate [multiple] unconfirmed confidence 1 submissions from the same user for the same title.
http://cue.tools/wiki/CUETools_Database#What_information_does_the_database_contain_per_each_CD.3F
Quote
What information does the database contain per each CD?
[...]
* Submission log, including user's ip addresses and CD drive model, which is used to eliminate erroneous submissions, virtual/faulty drives, and to verify if a submission by one user is independently confirmed.

korth

Re: Submitting multiple rips of the same CD

Reply #2 –
Do you know how AccurateRip decides which result to keep?

Re: Submitting multiple rips of the same CD

Reply #3 –
That would depend on whether or not the hash confirms an existing record for that track already in the database and the confidence.
If there is more than one hash for a track, AccurateRip holds confidence 1 hashes outside of the database until confirmed by an additional submission by a different user.
korth

Re: Submitting multiple rips of the same CD

Reply #4 –
Sounds like both are pretty well hardened against this sort of thing then, and I should just rip away and not worry about it. I just didn't want to be contaminating the DB with bad rips, and there was no way to select which ones to submit.
 

Re: Submitting multiple rips of the same CD

Reply #5 –
Quote from: Dracaena on
I just didn't want to be contaminating the DB with bad rips
Actually, in a way there will be more misinformation in contaminating the database with apparently-good rips.
First, AR verification isn't supposed to be a vote on how many have this result - it is supposed to be "does anyone have the same as yours?" It is about finding a soulmate to love, it isn't about all the idiots you are never gonna marry :-)
Now suppose you go from this computer to that computer with your bad CD. 12 of 13 tracks are OK, but that damn track 09 is damaged. You try to at least get it down to a tolerable-sounding error. What happens if all are submitted? If you rip it on 3 computers, it will look like 3 CDs have been ripped by three independent people. Three "random rips", on the other hand (that is track 09), will not match anything and not lead to that misinformation.
Well not quite so simple, AFAIUnderstand, and the CUEToolsDB behaves a bit different from the AR.

Those numbers will be (mis-?)interpreted as a "vote" though. Say, in case of for example two masters with the same CDTOC (that quite often happens with CD singles), or for example someone applies dithering once more from file to glass master - and also in the past without cross-pressing validation it would happen quite often for different pressings - but the interpretation of that vote typically isn't "which one is more accurate", rather it is "which pressing sold the most?".
(Yeah sure, sometimes it can lead you to track down a manufacturing error, I've actually seen that happen. With the LED light release, speak of the Devil.)