1) Audition was used, not Audacity
2) It does not prove that EAC does a good rip.
It proves that an EAC rip (which can be assumed to be perfect) gives results identical to playing the actual CD on an actual (old) CD-player while recording its output over a crippled SPDIF link into a computer interface.
So if EAC's rip was accurate, then so was the CD player reading the disc and so was the SPDIF interface transferring the data.
This to counter an earlier claim that CD players make significant amounts of errors.
So then Cool Edit Pro should work just as well, being that Adobe purchased Cool Edit Pro from Syntrillium and basically rebranded it. I would imagine the basic "engine" (the original code at the heart of the program) lives on essentially unchanged in Audition. The older two-channel version, Cool Edit 96, probably also shares this engine. If the basic code is good, then the processing should be good. Cool Edit always had a good rep. I used it happily way back when.
By "good rip" I meant "accurate rip" as opposed to a bad rip being a rip with inaccuracies like bits missing. So yes, EAC does a good rip, right? 🙄
OK, so your point is that neither Audition (Cool Edit Pro), EAC, the CD player or the S/PDIF interface introduced any data errors? If everyone agrees on that then we have a good baseline to start from.
So any perceived differences in playback would have to be attributable to differences in DACs (jitter?) or analog circuitry?
--
In other news -- I found the backup discs and a couple of copies of the final CDs from that session. One of the CDs is still in the wrapper. The backup discs don't include a CDR master, unfortunately. There are three CDRs of WAV file backups, including the CD layout file. The silver lining is that we can compare the original WAV files to the final CD tracks from the dupes. Didn't have time to do any work on this last night, and may not for a little while. Botanical pursuits this weekend and have to rebuild a DAW (I don't have one working at the moment -- as I might have mentioned, I don't do that kind of work anymore). But I'm interested to do the bit-for-bit comparison between original WAV files and final CD dupe tracks. However, I have a feeling there won't be any bits not in common.
Let's agree on some fair tests, shall we? First I'll pick a short track to compare.
Proposed Experiment 1)
- Rip CD track to uncompressed WAV file using EAC with all bit mining/error correction measures enabled.
- Compare ripped track to original WAV file from backup. I won't divulge my results until end of experiment.
- Create new redbook CDR with both ripped track and original WAV track as separate audio tracks.
- Listen to new redbook CDR and see if I can tell difference between two tracks.
- Send redbook CDR to others and see if they can hear any difference. Everybody then reports back to me via PM. I'll compile the responses and we see what we get.
Experiment 2)
- Rip CD track to uncompressed WAV file using EAC with all bit mining/error correction measures enabled.
- Copy ripped track and backup WAV file to a USB thumb drive.
- Play back both in a computer with decent audio interface from USB thumb drive. Compare. Keep my observations to myself until experiment is over.
- Send thumb drive out to others to listen and compare. Everybody then reports back to me via PM. I'll compile the responses and we see what we get.
What speed should I do the rip from CD at? Slowest possible? 8X? 16X? Faster?
Anybody have other proposals for a valid comparison? As long as it's not too expensive or time-consuming, I'm sure we could figure out a reasonable group experiment.
--
Last edited:
Right on rongon. Some one willing to take it to the next step, actually doing a test. Looks like your on the right track.
The step of burning a red book CD may be a problem, because that adds a step that is not part of the regular CD manufacturing proccess and has some inherent flaws not associated with commercial CDs. If you want to compare the master files and the CD, leave the rips as wave files (use the thumb drive)
The step of burning a red book CD may be a problem, because that adds a step that is not part of the regular CD manufacturing proccess and has some inherent flaws not associated with commercial CDs. If you want to compare the master files and the CD, leave the rips as wave files (use the thumb drive)
Last edited:
I agree with CBDB.
There are too many variables when ripping to a CD-R, and the subsequent playback. The files should be saved directly to a good quality USB pen.The pen could be passed around for direct playback from the PC, or the recipient may choose to transfer the .wav files to a quality CD-R.
Some of the USB pens are as slow as a wet week when writing, however they seem to be able to play back .wav files properly.
SandyK
I have a suggestion - since the space on a CD-R is large compared to a couple of tracks, why not fill out the disk with a randomized sequence of the first two tracks? In other words, everyone knows the identity of the first two tracks, but the identities of the remaining tracks is known only to you. Then ask people to say whether tracks 3,4..... are the same as either track1 or track2.
One stumbling block I see is that you don't have the CD-R that was sent to the pressing plant. Or did I misunderstand? If your original waves do not match the rip of the final CD, we won't know exactly why.
If they do match, then all is well.
If they do match, then all is well.
If they do match, then all is well.
And no further testing is required.
And if the data from the master and the commercial CD are found to differ, then the CD manufacturer has been shown up for doing things, but we'll never know what and why.
And no further testing is required.
And if the data from the master and the commercial CD are found to differ, then the CD manufacturer has been shown up for doing things, but we'll never know what and why.
Correct, we may never know. In this case anyway. But there may be others who have masters and stamped CDs who are willing to test.
Rongon. You do not have the master that was sent to the CD plant?
Rongon. You do not have the master that was sent to the CD plant?
Correct, I do not have the original master CDR that was sent to the plant. I have the exact same *files* that were on that master that was sent to the plant.
I apparently did not save the master CDR because I knew at the time that I could always reconstitute the project from the CD layout files and text files I left on the data CDR backups. I never envisioned using the project as a research project! To me, the files are the originals, not the CDR master disk. I assumed the technology for making a master disc would improve, not just fade away. (Silly me...)
However, I think there is something to be learned from comparing the CD track to the original file from which that CD track was made. Or am I missing something here? (Always possible, and often probable...)
--
Oh yes, for sure! Ultimately that's what we want to know. But if they don't match, we don't know who to blame, your old Plextor SCSI burner, or the CD plant, or both. 😉
Still, what you and most of use want is for the stamped CD to be an exact match of your master file. In this case, we have a chance to find out. That's great.
Still, what you and most of use want is for the stamped CD to be an exact match of your master file. In this case, we have a chance to find out. That's great.
Amazing thread. Every time another vinyl fan comes in he makes a series of statements that are flat out wrong. And states them as fact! Then knowledgeable people patiently put them right. And get fought every inch of the way, usually with a new stream of nonsensical ‘statements’.
CD *always* sounds better than LP, from the same master tape. Assuming both transfers are done competently. LP is *incapable* of matching the source tape or file, analog or digital, to the degree CD does.
The *only* way LP sounds better is when the master tape is better. Unfortunately for fidelity, this happens too often. Fortunately for classical music fans, it hardly ever happens to them. It’s the pop music fans who are in trouble.
Apologies, there is one other way: when the “audiophile” is not a true audiophile and actually prefers higher levels of noise and distortion.
PS the above conclusions ONLY apply to unsighted listening. Sighted “listening” is not listening.
Arg
CD *always* sounds better than LP, from the same master tape. Assuming both transfers are done competently. LP is *incapable* of matching the source tape or file, analog or digital, to the degree CD does.
The *only* way LP sounds better is when the master tape is better. Unfortunately for fidelity, this happens too often. Fortunately for classical music fans, it hardly ever happens to them. It’s the pop music fans who are in trouble.
Apologies, there is one other way: when the “audiophile” is not a true audiophile and actually prefers higher levels of noise and distortion.
PS the above conclusions ONLY apply to unsighted listening. Sighted “listening” is not listening.
Arg
Pretty bold statements, Mr. Arg. Can you back them up with real facts or documentation, or is this simply your opinion?
That's where you lose me. CD will sound more like the master tape (if done competently, it will sound identical to the master tape), but that may or may not be perceived as "better."
Yep. That's the basic problem "sounds better" are weasel words. They don't mean much as they are really in the ear and mind of the beholder. And of course there is a certain amount of noise and distortion that do no harm.
Exactly.
There are two relevant but entirely different questions:
1) is CD better than LP?
2) does CD sound better than LP?
The whole stereo chain from recording to delivery in the living room is fundamentally flawed and uncapable of fulfilling its task. It goes even so far that the actual task of this chain is not even well defined. Do we want the orchestra in the living room? Or do we want a living room window onto the music venue? Whose living room anyway?
This whole chain contains a number of links. Deciding whether a link is better means assessing its transparency in its particular position: does the signal/sound change when it passes/does not pass through this link?
Deciding whether something sounds better concerns the whole chain, from source to sink (and that includes our auditory perception in the specific state of hanging in the couch with a good glass wine). We know that the chain is broken (well, some of use know and dare to admit it, to others 'stereo' is more of a dogmatic religion). It may well be that substituting a link that is worse leads to a chain that sounds better.
I have no problems whatsoever with the above notions and keep running a system that sounds good both with CD and LP.
After listening to hundreds of different CDs, hundreds of LPs and dozens of master tape dubs, my weasel words are that really well done LPs come closer to the quality of the sound of a master tape than really well done CDs. I think that's what the original poster wanted to know.
The system that I listen to and on which I base these statements belongs to a friend of mine. The tape player is a Studer A810, the phono is an Airtight Supreme cartridge on an SME 312S tone-arm mounted on a Technics SP-10 MkIII turntable. The phono pre-amp is an Allnic H3000, the linestage is an Allnic L3000 and there is a tube based active crossover that feeds four VTL 750 amps that power a pair of Dali Megaline speakers. In other words, a really good highly developed system. I only wish I could afford such a system instead of mustering whatever I can build myself.
Certainly in theory a digital file should exceed analogue in accuracy but at this point CD doesn't seem to sound as realistic as LP.
John
And yet, no-one seems to ever be able to detect by ear the A/D-D/A at 16/44 when inserted into a signal chain and set to appropriate levels (see the Tiefenbrun-Lipshitz test as an example of why this qualification is important). So it may indeed be hard to swallow, but that's where the evidence points. Anyone who disagrees is free to set up a good blind test and demonstrate the contrary.
I've yet to see documentation on an adequately designed experiment (with independent duplicated results) that prove any such thing. At the same time there are reports of a (no formal documentation) of a blind test (the people in the test had no clue they were being tested) conducted in conjunction with a London AES (conducted by Meridian IIRC), which clearly showed preception of differences.
The jury is still out.
dave
The jury is still out.
dave
- Status
- Not open for further replies.