Comparing CD ripper outputs ?

[Gidorra]

For the past "30 years" or thereabouts, I ripped with CDeX. It was only recently that I started ripping with EAC - and now I am wondering whether I have to re-rip my entire collection for peace of mind Have you guys ever checkummed the rips and compared? Is there an easy way to compare the rips of e.g. 20 CDs to get an idea whether my entire past efforts were rubbish, or what shall I do? I am guessing ca 700-1000 CDs total here, so I would not do that light-heartedly...

 

[Petrushka]

For the past "30 years" or thereabouts, I ripped with CDeX. It was only recently that I started ripping with EAC - and now I am wondering whether I have to re-rip my entire collection for peace of mind Have you guys ever checkummed the rips and compared? Is there an easy way to compare the rips of e.g. 20 CDs to get an idea whether my entire past efforts were rubbish, or what shall I do? I am guessing ca 700-1000 CDs total here, so I would not do that light-heartedly...

I ripped 1300 CdS with EAC. My only regret is I did not spend more time up front cleaning up the metadata. I used freedb, which is user supported, and not consistent.

I think the most efficient way is to use a fast rip, and only use the slower methods if a rip fails CRC. I believe dbpoweramp has the same features, but easier to use.

 

[Gidorra]

I ripped 1300 CdS with EAC. My only regret is I did not spend more time up front cleaning up the metadata. I used freedb, which is user supported, and not consistent.

I think the most efficient way is to use a fast rip, and only use the slower methods if a rip fails CRC. I believe dbpoweramp has the same features, but easier to use.

I insta paid 7 bucks for that GD3 database for EAC because the reason I switched to begin with is because CDeX seemingly would not connect to the free databases anymore, and punching in the details for some 300 CDs was not an option... I would recommend paying that for lifetime access. I am just ripping the 330 CD Karajan Box and so far all the CDs were in the GD3 database after I am ca 70% done here.

 

[DVDdoug]

If it sounds OK it's OK. There could be some inaudible errors but I wouldn't worry about it. If you have clicks, pops, or skips you can re-rip (and hope they come-out OK the 2nd time). I've had a few occasions where EAC reported errors but it sounds OK. It didn't take long to forget which CDs had errors.

If you have WAV (or FLAC) files you MIGHT be able to check them with CueTools and AccurateRip.

EAC supports AccurateRip but I've only used it while ripping and I'm not sure if it works on previously ripped files.

Is there an easy way to compare the rips of e.g. 20 CDs

foobar2000 bas a bit compare feature and I[m pretty sure it only checks the audio data and ignores metadata.

 

[Apesbrain]

FLAC by default writes a checksum to the files it creates. You can verify checksums using foobar2000:

To verify FLAC checksums in foobar2000, use the built-in "File Integrity Verifier". Highlight your FLAC files, right-click, select Utilities, and then Verify integrity to check for audio corruption, which compares the embedded MD5 checksum against the decoded audio.

 

[krabapple]

If there was anything really wrong, you'd probably hear it. It would not be something like a subtle EQ change

 

[kemmler3D]

Agree that if you can't hear any problems, you shouldn't borrow trouble. Passing CRC with every bit in its place is something people who are aiming for something like archival use care about. Ironically from what I've seen, mostly pirates who intend for their illegal copy to be a truly perfect one.

 

[Gidorra]

OK.. Friday night, innit... lonely hearts I just ripped a CD with CDEX and EAC, compared the md5s with MD5Checker for Windows, and only 1 track came up with the same checksum

 

[Vincent Kars]

MD5Checker

Probably not a good idea. It is the MD5 of the entire file not the audio part. Rippers might reserve some space in the header to avoid rewriting the whole file if you changes a tag. If one uses PADDING=100 and the other PADDING=101 you sure get a different MD5.

Better do what DVDdoug suggested. Use the Foobar binary comparator, it compares the audio part only, not the tags.

 

[Daverz]

For the past "30 years" or thereabouts, I ripped with CDeX. It was only recently that I started ripping with EAC - and now I am wondering whether I have to re-rip my entire collection for peace of mind Have you guys ever checkummed the rips and compared? Is there an easy way to compare the rips of e.g. 20 CDs to get an idea whether my entire past efforts were rubbish, or what shall I do? I am guessing ca 700-1000 CDs total here, so I would not do that light-heartedly...

I second the recommendation of cuetools. You can have it check the entire collection against the cuetools database and also the AccurateRip database, and in some cases it can even repair rips.

CUETools - CUETools

cue.tools

 

[d3adf1sh]

yeah when i was ripping a bunch of cd's i started to notice that using EAC, Foobar, and Windows Media Player all gave me different sounding results. what actually got me to start experimenting was i had one cd that was HDCD encoded and it was ripped by EAC but sounded like trash compared to listening to the disc. it's one i'd had for decades so i know what it's suppose to sound like. so i tried ripping the disc with WMP being microsoft owns HDCD and the rip actually sounded right! but then other rips done with WMP didn't sound as good. so i did more digging.

one thing i found out was when ripping with EAC and it checks the accuarate rip database and all that it's checking is ONLY on the wav extraction NOT the final FLAC file. and come to find out that all the programs were using different versions of FLAC and read somewhere, that someone was doing some tests, at one point anything over FLAC compression level 2 was degrading sound quality during playback, i'm assuming it still could decode back to the original file? but the only way they could get it to sound right was using compression level 0 - 2. and i believe the guy said he just continued to just use FLAC 0 (uncompressed) being it still was a bit smaller than WAV and had tag info.

personally ever since then i've just used EAC to rip to WAV and do the conversion to FLAC myself separately using FLAC Frontend at level 0 and to my ears is the closest thing to the actual CD of any ways i've tried. and i know that FLAC Frontend uses an old version of FLAC i think it's version 1.3.0 but hey, if something aint broke, don't fix it. i also know there have been tons of flac versions released since then and almost every release seems to have "bug fixes" who would've know there were so many bugs in such a seemingly simple program to begin with?

so if you're not happy with what EAC or whatever program is giving you, you could always give my way a shot and see what you think?

 

[RandomEar]

yeah when i was ripping a bunch of cd's i started to notice that using EAC, Foobar, and Windows Media Player all gave me different sounding results. what actually got me to start experimenting was i had one cd that was HDCD encoded and it was ripped by EAC but sounded like trash compared to listening to the disc. it's one i'd had for decades so i know what it's suppose to sound like. so i tried ripping the disc with WMP being microsoft owns HDCD and the rip actually sounded right! but then other rips done with WMP didn't sound as good. so i did more digging.

There's nothing to dig. EAC + AccurateRip DB is bit perfect if it tells you so. If anything else sounds different - which is unlikely enough and not proven by sighted testing - it's ripped incorrectly. Most likely, you were tricked by your brain, your rippers for some reason encoded different replay gain settings into the files or some other odd difference in settings.

one thing i found out was when ripping with EAC and it checks the accuarate rip database and all that it's checking is ONLY on the wav extraction NOT the final FLAC file. and come to find out that all the programs were using different versions of FLAC and read somewhere, that someone was doing some tests, at one point anything over FLAC compression level 2 was degrading sound quality during playback, i'm assuming it still could decode back to the original file? but the only way they could get it to sound right was using compression level 0 - 2. and i believe the guy said he just continued to just use FLAC 0 (uncompressed) being it still was a bit smaller than WAV and had tag info.

FLAC is lossless, there is no difference in the output to WAV. Unless "that guy" you read about somewhere did capture the actual ouput and compared it bit by bit and found some error in his specific FLAC decoder on his PC, this is all just hearsay. People who don't understand how tech works claim all kinds of nonsense about stuff. I have never seen or heard about such errors in FLAC and a quick search also brought up nothing relevant.

personally ever since then i've just used EAC to rip to WAV and do the conversion to FLAC myself separately using FLAC Frontend at level 0 and to my ears is the closest thing to the actual CD of any ways i've tried. and i know that FLAC Frontend uses an old version of FLAC i think it's version 1.3.0 but hey, if something aint broke, don't fix it. i also know there have been tons of flac versions released since then and almost every release seems to have "bug fixes" who would've know there were so many bugs in such a seemingly simple program to begin with?

so if you're not happy with what EAC or whatever program is giving you, you could always give my way a shot and see what you think?

There is no "closest" with bit perfect rips. They are perfect, or they are not. Also, the idea that there might be some "small degradation" somewhere you can just barely hear with things like bit errors in audio files is misguided. Did you ever go into the editor of you choice and changed a bit in a WAV? Try it and listen - if you hit one of the higher bits, you will hear that immediately. It's not subtle, like, at all. And if there actually were bit errors in decoded FLAC files, you would statistically hit the most significant bit every 16 errors. So it's not like the errors could just change the lowest bits and you would never notice. This just isn't how digital formats work.

 

[krabapple]

yeah when i was ripping a bunch of cd's i started to notice that using EAC, Foobar, and Windows Media Player all gave me different sounding results.

Pretty darn unlikely, if you used them right. And compared them right.

what actually got me to start experimenting was i had one cd that was HDCD encoded and it was ripped by EAC but sounded like trash compared to listening to the disc. it's one i'd had for decades so i know what it's suppose to sound like. so i tried ripping the disc with WMP being microsoft owns HDCD and the rip actually sounded right! but then other rips done with WMP didn't sound as good. so i did more digging.

HDCD??? Of course HDCD can sound different decoded vs undecoded. That has nothing to do with ripper quality per se, it has to do with whether the ripper can decode HDCD or not. HDCDs are a very niche case.

For many HDCDs, btw, the only differences between decoded and undecoded is an overall level boost/cut. "Proper" HDCDs actually decode to audio with an expanded dynamic range (i.e., they had the 'Peak Extend" option enabled during encoding).

Back in the day, any digital audio that came from a Keith Johnson HDCD ADC could light up the 'HDCD' indicator at home, regardless of whether any of its options were used.

one thing i found out was when ripping with EAC and it checks the accuarate rip database and all that it's checking is ONLY on the wav extraction NOT the final FLAC file. and come to find out that all the programs were using different versions of FLAC and read somewhere, that someone was doing some tests, at one point anything over FLAC compression level 2 was degrading sound quality during playback, i'm assuming it still could decode back to the original file? but the only way they could get it to sound right was using compression level 0 - 2. and i believe the guy said he just continued to just use FLAC 0 (uncompressed) being it still was a bit smaller than WAV and had tag info.

This is all utter nonsense.

 

[d3adf1sh]

yeah think what you want. i'll refer you to the title of this thread so i'm not the only one that's hearing a difference between rippers. and what got me digging in the first place was because listening to the disc, in most cases, sounded better than the flacs. i'll repeat again that accurate rip checks the wav extraction NOT the final flac. and different crc's means they are differences in the final flac files.

you can think what you want, i was just giving the op the solution i found to better sounding flac rips. if you want to take my advice or don't doesn't bother me one bit. i will link this one article and if i can find the one i was originally referencing i'll update the post.

--flac's dirty little secret

FLAC’s dirty little secret

First — and briefly, don’t worry ! — some personal historical background. My journey in “computer audio” started many years ago with Logitech’s Squeezeboxes. From the very beginning, I looked for the best and more versatile file format. In those times, FLAC was not an option on Mac so I settled...

dcs.community

let's not forget the "bug fixes" they've implemented in certain versions of flac so there have obviously been problems

FLAC - Changelog

A free, open source codec for lossless audio compression and decompression

xiph.org

 

[rolfrander]

I ripped 1300 CdS with EAC. My only regret is I did not spend more time up front cleaning up the metadata. I used freedb, which is user supported, and not consistent.

I think the most efficient way is to use a fast rip, and only use the slower methods if a rip fails CRC. I believe dbpoweramp has the same features, but easier to use.

I have also used EAC for all of my collection, because of the claim that it tries really hard to detect and correct any reading errors. Since the goal would be that the bits in the file is equal to the bits on the disc, I guess it should be possible to compare an EAC-generated wav-file with the same track ripped by other tools, but I've never been interested enough to try.

However, the point I really wanted to make was regarding metadata. In my collection (which, admittedly is not that large, a few 100s, not in the thousands), I have taken special care with the metadata. I started with data from online databases, but cleaned it up before writing to flac. I did this using custom scripts, so not an approach I can recommend unless you are familiar with programming. A couple of choices I made that makes my library clean and easy to navigate, but which creates trouble with many streamers, mediaplayers and the like:

1. I normalize "the" and "a", as is common when sorting english alphabetically. So, I don't have "The Police", I have "Police, The"
2. for classical music, I organize according to composer, not disc. For example, it is not uncommon for example to have one disc with a symphony by Tchaikovsky, and also a shorter piece by the same composer, and possibly something by another, to fill the disc. In my library, this is split up in three levels: composer/work/movement (as opposed to artist/disc/track). That means: in my classical collection, I have lots of directories (= "disc") with just one movement (track).
3. the disc (or work) directory is prefixed by year, so the 1995 "live!" album from The Police is cataloged as "Police, The/1995-live" which makes it easy to navigate the discography chronologically.

For my ability to curate a listening session, this works fine.

For iTunes, Volumio, Plex et al to try to guess the artist and disc to find images of the artist and cover art, this is a disaster.

 

[RandomEar]

yeah think what you want. i'll refer you to the title of this thread so i'm not the only one that's hearing a difference between rippers.

Yes. People also claim to hear differences between DACs or amps and then continue to fail level-matched blind tests. Nobody disputes that you are convinced to hear differences. However, as explained, this does not mean they actually exist. Human hearing is easily tricked and unreliable. Perception bias is strong. Unless you've got controls for those effects in place, any listening test is pretty much useless.

Familiarize yourself with these problems and then conduct proper, blind and level-matched ABX tests. With digital files, this is luckily trivial. If those tests still show differences, you're welcome to report back with the data. Of course, it would be much easier to just decode that level 6 compressed FLAC back to WAV and just compare it to the uncompressed WAV...

and what got me digging in the first place was because listening to the disc, in most cases, sounded better than the flacs. i'll repeat again that accurate rip checks the wav extraction NOT the final flac. and different crc's means they are differences in the final flac files.

Yes, it checks the extracted WAV. You said yourself you need to check the FLAC and members have explained why a simple CRC is insufficient for a format which does not only carry audio but also meta data. A single additional space in a song name or a different time stamp will change the CRC. You cannot compare audio data by using CRCs on compressed files with metadata.

you can think what you want, i was just giving the op the solution i found to better sounding flac rips. if you want to take my advice or don't doesn't bother me one bit. i will link this one article and if i can find the one i was originally referencing i'll update the post.

The problem is that you were not giving a solution for "better sound", but - as it stands right now and without sufficient proof - you were unknowingly spreading misinformation. There's lots of people out there who do not understand the technical details behind stuff like audio compression and posts like yours perpetuate the silly idea that somehow, despite the fact that the compression is 100% lossless - less compression would "sound better" somehow. Which, again, it does not. Because it can't sound better if it's literally identical

--flac's dirty little secret

FLAC’s dirty little secret

First — and briefly, don’t worry ! — some personal historical background. My journey in “computer audio” started many years ago with Logitech’s Squeezeboxes. From the very beginning, I looked for the best and more versatile file format. In those times, FLAC was not an option on Mac so I settled...

dcs.community

That source is a true gold mine! It's even got the wife in it, though sadly not from the kitchen

My wife and I, we did some listening and comparisons with AIFF and WAV files

The author claims to hear a difference on his specific DAC/streamer, which seems to include a FLAC decoder. Also, he does not provide any evidence whatsoever that what he claims is true. He just "heard it", like all the other disproven audiophile claims. And there's ample pushback to his claims in the same thread...

 

[MrBlueSky4470]

yeah think what you want. i'll refer you to the title of this thread so i'm not the only one that's hearing a difference between rippers. and what got me digging in the first place was because listening to the disc, in most cases, sounded better than the flacs. i'll repeat again that accurate rip checks the wav extraction NOT the final flac. and different crc's means they are differences in the final flac files.

you can think what you want, i was just giving the op the solution i found to better sounding flac rips. if you want to take my advice or don't doesn't bother me one bit. i will link this one article and if i can find the one i was originally referencing i'll update the post.

--flac's dirty little secret

FLAC’s dirty little secret

First — and briefly, don’t worry ! — some personal historical background. My journey in “computer audio” started many years ago with Logitech’s Squeezeboxes. From the very beginning, I looked for the best and more versatile file format. In those times, FLAC was not an option on Mac so I settled...

dcs.community

let's not forget the "bug fixes" they've implemented in certain versions of flac so there have obviously been problems

FLAC - Changelog

A free, open source codec for lossless audio compression and decompression

xiph.org

No, you can not hear differences between rippers if they do their job right. Make a bit comparison and you'll see that the files are identical. There are no "pretty" 0s & 1s and "ugly" ones.

You can also always use CUEtools on existing rips to verfify them against the AccurateRip database. If it gets verfied as "accurate" then it is. No matter what ripper was used.

 

[d3adf1sh]

The problem is that you were not giving a solution for "better sound", but - as it stands right now and without sufficient proof - you were unknowingly spreading misinformation.

if you don't believe me then try it for yourself. i was mainly suggesting a way for the OP to get better sounding flac rips if he, like me, hear's differences between rippers. and give it a break with the "misinformation" crap. let me guess, you're a CNN guy?

Why Do WAV And FLAC Files Sound Different? Article By Dr. Charles Zeilig And Jay Clawson

Why Do WAV And FLAC Files Sound Different? Article By Dr. Charles Zeilig And Jay Clawson Of HIFICRITIC

www.enjoythemusic.com

 

[d3adf1sh]

You can also always use CUEtools on existing rips to verfify them against the AccurateRip database. If it gets verfied as "accurate" then it is. No matter what ripper was used.

did you even read ANY of my other posts? accurateRip ONLY TESTS WAV EXTRACTION not the resulting flac file.

 

[RandomEar]

if you don't believe me then try it for yourself. i was mainly suggesting a way for the OP to get better sounding flac rips if he, like me, here's differences between rippers. and give it a break with the "misinformation" crap. let me guess, you're a CNN guy?

Why Do WAV And FLAC Files Sound Different? Article By Dr. Charles Zeilig And Jay Clawson

Why Do WAV And FLAC Files Sound Different? Article By Dr. Charles Zeilig And Jay Clawson Of HIFICRITIC

www.enjoythemusic.com

I'm sorry, but that article is absolute pseudoscience rubbish. Instead of doing a trivial bitstream comparison, they invented their own measurement of "height":

Height estimates were made (using a tape measure hungfrom the ceiling behind the speakers) by listening to a specific chord, repeated twice on a harp passage during a 6s segment (39-45s into track1).

They then did some supposed single blind testing to evaluate that imagined parameter and in the end, they conclude that the problem was that metadata/cover art was included in the FLAC file and that degraded the audio. Of course it was! Like... come on man, 1st of April was six days ago

What is it with audiophiles and inventing their own measurement systems, physics and pseudoscience and then adding 100 complex thingamajigs to their test setup for quantities which could be measured using a simple ruler?