Minor crackling and distortion in FLAC files ripped from CD

[ejiggle]

To get a few inevitable questions out of the way:

Ripped using DBPoweramp- Secure (Recover Errors) > Drive Speed MAX > Ultra Secure 2/4/2 > Re-Reads 34, Speed MAX > Drive Read Cache 1024KB- AccurateRip > Query Rupping Results > Cross Pressing Verification- CD ROM > MAX Speed > Offset +6 > Apply De-emphasis

Flac 5, but also tested uncompressed.

Using zero DSP's currently, though the issue was also present when testing ReplayGain (NOT RG-Apply) both in FLAC and ALAC with a mix of album/track settings (advanced settings on default, EBRU, -18, disable clip prevention UNCHECKED)

Tested on Airpod Pro 2's and Q3030i's)

I have played the music back using VLC (Windows 10), iTunes, Apple Music app on iPhone 15 Pro Max (no EQ or soundcheck settings enabled)

100% of the affected files have AccurateRip confidence ranging from 4 to 200.

I have been cross referencing with Spotify and Tidal, quality maxed out with no EQ or soundcheck/normalization settings.

I have also tested a few of the CD's on two different read/writers. The issue persists on playback from disc as well as in the file.

I noticed some obvious crackling and even clipping on a few tracks while shuffling last week, and immediately checked all my playback settings. Everything looked good, so I decided to cross reference with Spotify to see if the streaming track had the same artifacts. None of them did. This spiraled me down a rabbit hole checking random songs on both my PC and iPhone and comparing them to streaming and Youtube versions, and the bottom line is that many of my tracks have the same issue, and all of them are inferior to the versions on streaming services. Obviously concerning, but distressing for me as I just spent weeks ripping 500+ CD's. The internet swears that if AccurateRip is confident, there's no chance my data isn't bit-perfect. The internet swears my CPU (i12900k) isn't the problem. The internet swears my $40 LG CD writer isn't the problem. But there is obviously a problem, and I'm confident enough to say it isn't my playback hardware, as it exists across completely unrelated setups and isn't present on other supposedly identical versions of the tracks.

The best I can figure is some of my 10+ year old cd's have degradation or damage of some kind, introducing minor distortion or clipping artifacts, but they are minor enough that the data checks out with AccurateRip. I guess the other option is that the files are such high quality that I'm hearing things not present in a more compressed version available on streaming services, but that seems...improbable, considering how much worse they are to my ear.

I'm at a loss here.

[schmidj]

It sounds like what you are hearing is part of the audio that is in the CD, not degradation, or you are having a hearing problem or your brain is playing tricks on you. . Hey, find another copy of the CD, for instance on eBay, buy it, and compare it. Many commercial recordings have all kinds of noises that you may think is a problem, that was recorded or mixed or mastered that way when the CD was produced. And check your ears. The brain does funny things when you listen to things, particularly if you are listening on headphones.

Why did you apply deemphasis? The vast majority of CDs did not use preemphasis, applying deemphasis would make them sound muddy.

Be aware that wireless headphones use Bluetooth with quite a bit of compression. They are also potentially subject to various types of interference.

And you have no way of knowing what processing the various streaming services are doing to the audio.

If your saved file or rerip checks OK with accuraterip, and the confidence is anything above 1 (or the number of times you have re-ripped the same CD), your rip is bit-perfectly identical to all those other rips. Even one bad bit out of the megabits ripped will fail the accuraterip test. There is no such thing as accuraterip passing a file with "minor" errors. Conversely, many files with accuraterip errors will sound just fine even to most listeners.

Listen to the CDs (not a rip) on different systems, everything different, different CD player, different copy of the CD, different amplifier, with loudspeakers, not headphones. What are you hearing? Do your friends hear the same issues? if the same, almost certainly it is what was recorded into the CD.

[ejiggle]

Why did you apply deemphasis? The vast majority of CDs did not use preemphasis, applying deemphasis would make them sound muddy.

It's my understanding that this is only applied if there's a pre-emphasis on a particular disc, which doesn't pertain to anything in my collection. The setting is on by default.

Be aware that wireless headphones use Bluetooth with quite a bit of compression.

The crackling is identical on wireless headphones, car speakers, desktop speakers. Same spots in the songs, same volume, etc. Reproducible on PC and Macbook with different players and drives. It is definitively, overwhelmingly there, but not present anywhere else online. I will be ordering another disc and comparing, but I can't imagine why streaming services would have different masters than what made it onto a CD between 2006-2010, nor why accuraterip would be giving me such high confidence in the file. I'd be happy to upload an A/B.

[schmidj]

First, if you have ticked the "apply deemphasis to CD tracks marked with preemphasis" in the ripper options that is the correct setting. You don't want the DSP choice "Audio CD: Deemphasis" which applies it all the time. (but do be aware there was a posting here some time ago about old CDs where the bit on the CD was not set correctly, as I recall)

I would not assume or presume that the major streamers use source material from CDs for the major labels any more. It seems evident that the labels are sending them digital files, probably already encoded as mp3 or m4a to stream. If you look at YouTube you can see where the source material came from, YouTube generally doesn't do their own ripping. For many pop songs you'll see multiple files. At least one or some will be from the record companies that own or are licensed to stream the material, but others sometimes are from private individuals who have uploaded copies. That is more common with stuff from phonograph records than CDs. But don't assume anything about the source of what you stream from the Internet or what processing it underwent.

As many have explained on this forum, that confidence number is strictly how many rips have been reported with the same identical checksum, not at all how good the rip is. Unless you have been ripping the same bad CD over and over again, and the error plays back identically every time (very unlikely but I've had it happen I recall once out of the about 10,000 CDs I've ripped, apparently the error was burned into the CD master, not damage that happened after it was pressed like "disc rot" or damage), a confidence of 2 is just as good as a confidence of 200. The confidence of 2 says two different people ripped the track and got the same checksum. Confidence of 200 says that 200 people ripped the track and got the same checksum. There is no report of the people who got bad rips (with different checksums). And someone here dd the math, the checksum that Accuraterip uses is rugged enough that (I forget the exact number but) there is only one chance in several million of different bit patterns giving the same checksum. More likely that you will be struck by lightning.

There is the possibility of a bad pressing or lot of pressings getting out of the replication house, it has happened, but very unusual. I mentioned the one I had. Another copy of the CD I obtained was fine, the bad one was very obvious, it skipped, but exactly the same way every time it played. Probably an error on playback of the dat or other tape master making the CD master, with lousy QC never listening to it. However having a bunch of different CDs with bad pressings is very, very unlikely.

Don't forget that, particularly with pop music, there has been a "loudness war" that was particularly strong in the early 2000s, producers felt their songs had to be the loudest to get listened to. And recording engineers and mastering houses took liberties with quality to make the master louder. Occasional (or not so occasional) clipping was one common technique. And the re-mastering that has been done for the streamers, who all enforce modern loudness standards, may well have been more conservative and avoided clipping.

There is software out there, I own iZotope's "Restoration", that has modules to "remove" clipping in recordings, it may or may not improve the sound. And you probably won't hear clipping as much in a compressed mp3, particularly at lower bitrates, because one of the compression techniques is often to throw out the higher frequencies, which would include many pops, clicks and the harmonics added by hard clipping.

Crackling? might be from one or more of the instruments being played?? If all you listen to now is pure, synthesized music, you may not recall how "dirty" real musical instruments (and the studios or live venues they were recorded in) are.

You also need to have other friends, ones you trust, listen and see if they hear what you hear. The brain can do funny things to your perceived hearing.