Yes, this is exactly the type of thing I am referencing, a great example.
What occasioned this bit of unpleasantry? 
Why is there, what looks like a slice of strawberry cheesecake next to my username?
Anniversary of you joining the forum
It shows that today is the anniversary of the day you created your account (July 19, 2018). - therefore “cake day”!
Well I guess that’s as good of an excuse as any to go get a slice of cake.
4 Likes
Congratulations!
Yes, cake is in order!
1 Like
Cake tastes better when its edges are coated in frosting!
Even green frosting, though I prefer chocolate.
2 Likes
Yet when I rip a CD with AccurateRip, it takes about 1/20th of the time to rip it than to play it. Think about that. If it can be ripped accurately at 20x, playing the CD is not going to be a problem. All of these alleged errors would slow down the rip.
clean cds rip very fast for me, damaged and old cds take very much longer. it is an oldish cd-rom though
Just about 95% of my CDs I bought used on Discogs, SteveHoffman.tv, or eBay. Only a couple have had any problem accurately ripping and those are always damaged in some way. Sometimes polishing the media fixes it but usually not.
If the CDs are not damaged, there are no significant amount of errors that require error correction during playback as has been implied or even stated here.
Just to go back to the original post, has anyone with one of these Reality Check CDs done the proposed test yet?
Rip both original and processed with some bit perfect ripper and checksummed or otherwise compared the files?
I know the thread has wandered around a bit, which is always cool, but I still want to k ow if the software that the guy used DSPd it in anyway (which might explain the perceived difference in “enjoyability” or quality or whatever).
Enquiring minds want to know, and it may help answer of the points posed throughout the thread 
I think there is a misunderstanding here. As an analog medium, all CDs have errors. These can be single-bit errors, burst errors, and many more. Thus, Redbook accordingly incorporates error correcting codes, and interleaving redundant copies of the data, C2 error pointers, C1 (single frame correction) and C2 (multiple interleaved frame correction), etc. to address these errors. Severe errors caused by damage, grubby discs and the like are addressed through interpolation. Every read of a CD involves error correcting. It is simply the nature of the beast.
These errors can be measured with techniques such as determining the block error rate (the ratio of erroneous blocks to the total number of transmitted blocks).
The issue is not getting an bit accurate stream to the DAC. The issue is how much noise is generated as error correction is taking place, the drive re-read head activity, and more - all to create a bit accurate stream. This generates noise. As we all know, noise negatively impacts sound. The less error correction required and the more efficient correction, the better the sound.
All of this is taking place when ripping as well, but as it is not being listened to in real time, the process will go faster than 1x speed. But ripping programs are also re-reading, de-interleaving, and more to get an accurate data stream.
One can often set the number of re-reads and other data correction in ripping programs. Exact Audio Copy is probably the first program most audiophiles heard of which incorporates re-read until correct. The PWT does the same thing.
Bottom line: as well all know, bits is not bits.
2 Likes
An excellent question. I, too, am curious if the burned CD generated by this process produces different files.
I haven’t, nor am I sure how to go about it.
If anyone in the Denver/Boulder area knows how, I’d be happy to bring both the original and the enhanced CD by.
1 Like
Here is some additional information on CD error detection/correction. The CD error detection/correction technique uses the Reed-Solomon algorithm described briefly here. For additional detail, see this university paper. This algorithm is used in hard disk drives, cellular wireless, CDs, DVDs, etc. because it is extremely roust.
Error detection/correction takes two forms. The first is typically automatic. That is because the Reed-Solomon algorithm is good at detecting and correcting multiple bit errors (think a bit of dust or a fingerprint smudge on the CD) that obscures adjoining bits. This type of error detection/correction is typically built into the hardware/firmware of the transport and is used for playback. Simplistically, most error detection/correction algorithms can detect N errors and correct M errors where M < N. Where M < N, Reed-Solomon will output bit perfect results even though M bits on the disk are inconsistent with the coding. The exact value of N depends on the particular implementation but hopefully you get the idea.
The second form of error correction would only be available to ripper/copier software and would only be invoked with badly damaged disks M >= N. If the Reed-Solomon codes can not be decoded with the standard technique the algorithm can report an error which could allow software to read the bits directly. Imagine a CD with nicks/scratches tangent to the tracks caused by your 3 year old using CDs as frisbees. In this case, the ripping/copying software could read the info say three times (to account for transient read errors) and take votes on each bit (two votes wins). Then by examining previous and subsequent encodings (and the PCM values they represent), an interpolated bit stream could be created. Note that the interpolated stream may not be bit-perfect with the master but we’re talking a badly damaged disk anyway so a carefully constructed algorithm should be able to recover the music with at most imperceptible (sub-millisecond) changes.
Some rippers are better than others. I prefer dBpoweramp because their marketing literature seems to imply they do some variation of both of the above detection/correction methods. It’s marketing literature says “dBpoweramp is able to obtain previously un-achievable levels of error recovery, through re-reading, c2 pointers and AccurateRip v2 (cross pressing verification).” This leads me to believe, together with its DSP capabilities, that it is the best for recovering damaged disks and would always produce bit perfect rips/copies for undamaged disks. dBpoweramp is what I use and I have no complaints.
3 Likes
A generous offer. It would also be fun to learn what others hear when comparing the two versions.
1 Like
I use the linux script abcde (“A Better CD Extractor”).
It calls cdparanoia to read each track, which reports errors (and whether they are fully correctable, or interpolated, or just an outright failure, giving the configurable option to bail, or keep going with an interpolated sample inserted).
It’s pretty robust and actually reports what it is doing, which I always like.
simple enough to use, it comes with a standard config file with “safe” preset config - just type “abcde” in a terminal window and follow the yes/no prompts for recognition of the CD for naming / tags etc.
Various output formats including flac at the lowest compression ratio are available (which is generally what I use), and MP3 etc. for the cloth-eared who don’t care. or only have an MP3 player.
It has served me well
1 Like
This sounds like an interesting program. I, too, like knowing what is going on rather than the typical - “Well, here is your file.”
1 Like
When AccurateRip says you got it, no error correction was involved. None.