POW-r is a noise shaping scheme, not a sample rate convertor. L2 is a mastering plugin, and present documentation does not mention SRC.
That leaves WaveLab's internal SRC, and according to http://src.infinitewave.ca versions 5 and 6 perform somewhere between attrocious and sloppy.
Many mastering engineers claim tangible benefits from 88.2k and 96k, but about just as many mastering engineers have been using sample rate convertors of a blatantly incompetent design. Until recently it was almost like the more expensive the tool, the worse the SRC. And this is surprising, as the recipe for correct SRC has been known for decades.
Splendid link Werner, once the spurious 'i' has been weeded out of the URL. Glad to see Audacity's SRC performing so much better than the expensive stuff. 🙂
In my book Audacity's SRC is only borderline acceptable.
There is absolutely no argument for not demanding perfection in a software SRC. Hardware is a different thing, though ...
There is absolutely no argument for not demanding perfection in a software SRC. Hardware is a different thing, though ...
Bits are bits. Digital audio's theoretical basis is rock-solid. And even auditory perception is pretty stable now, even if not complete.
When many people report deviant sound quality, be it from CDs different from glass masters differing from masters, or hi-res sounding superior where theory suggests that it wouldn't manner, well, then it is probably so.
But more often than not a detailed investigation into what's going on will reveal that someone somewhere is doing things the wrong way (see incompetent SRC software, for example). This is a far cry from the fundamentals being wrong. Any new technology goes through a learning phase, and for digital, for many many people, this is still going on. It's not like analogue was perfected a long time ago, not? Else show me the perfect turntable!
Two weeks ago I spent some time converting a 88.2kHz master with some horrible (i.e. no emotions attached) but impulsive music to 44.1kHz and back to 88.2kHz. The resultant files went onto a DVD-A and were listened to through an Apogee DAC and AKG K-400 headphones. I assure you that, with this system at least, I could not discern the least bit of difference. The validity of this experiment is small. I need more suitable music samples and I also need to build my new dedicated listening room and go through the loudspeakers (ESL-63s, if they don't die on me before the room is done!) But at the very least it indicates that, when all is done properly, the differences between CD and hi-res are less than vast.
I suspect you're not kidding here, but really...😛 Visual perception is way more advanced than auditory perception in my estimation, and even that is not yet complete.
If I'm wrong and out of touch here, do you know of any book (or work) on auditory perception which covers the same ground as Hoffman's Visual Intelligence - How we create what we see ?
Having hunted around on Amazon on and off for a few years, the nearest I've found (and recently acquired) is Bregman's Auditory Scene Analysis: The Perceptual Organization of Sound.
Cool, so just post up the number of the post where I said what you've claimed.
If you need that kind of guidance, I can't help you 😉.
You can do it again for me though.
Tell me what is the problem with a 44.1kHz sample rate.
Tell me the problem with using 16bits.
Tell me about CD bit errors and correction.
When you've finished explaining how there is nothing wrong with any of those (again), you will have you answer of exactly why you consider the CD to be perfect.
QED 🙂
Oh, the guidance isn't to help me, its to help you 😀
I'm willing to help you when you're willing to help yourself. Not before 😛
)
Well, some people demagnetize CDs and claim to hear differences.
@Globulator: Since you've attributed this comic book statement to me, please indicate where I said that.
This happens so often ...
Audiophile Type (AT) proclaims that analogue/LP is/sounds better (the difference is significant, but few seem to understand) than digital/CD.
So far so good. Hearing is fundamentally subjective so who's to argue?
AT continues by attributing the experienced differences to "fundamental faults" of digital, faults Engineering Types (ET) are arrogantly ignoring or just plain ignorant of, and to the obvious (but hitherto uninvestigated, let alone scientifically proven) superiority of the human auditory system. If the thread runs long enough names and anecdotes connected to Rupert Neve, Oohashi, Kunchur, ... are being dropped.
This is where the fun begins as the above mainly reveals a total or partial lack of understanding of signal theory and the sampling theorem in particular. Any scientific reports quoted are from the select list of discredited or never-reproduced efforts.
ETs then kindly (or less so, if they've been doing this almost twenty years) point this out, and try to rectify some misconceptions. Some ETs urge the interested reader to look up Shannon's proof, a piece of math so beautiful it should be on display at the Louvre (or at least Musee d'Orsay).
The ET continues by exposing a number of implementation issues that might pose problems with digital when not done right, issues that are very real due to practical constraints (not very often) or human error/ignorance/incompetence (all too often, check out DAW software programmers or mastering engineers).
AT stubbornly ignores this and essentially refuses to be taught anything new at all. A number of volleys go back and forth, after which suddenly the ETs are accused of stating that digital is perfect.
Bored with this the ET withdraws and does something useful, such as designing a super-advanced anti-alias filter or cueing a record on his $10k turntable.
Audiophile Type (AT) proclaims that analogue/LP is/sounds better (the difference is significant, but few seem to understand) than digital/CD.
So far so good. Hearing is fundamentally subjective so who's to argue?
AT continues by attributing the experienced differences to "fundamental faults" of digital, faults Engineering Types (ET) are arrogantly ignoring or just plain ignorant of, and to the obvious (but hitherto uninvestigated, let alone scientifically proven) superiority of the human auditory system. If the thread runs long enough names and anecdotes connected to Rupert Neve, Oohashi, Kunchur, ... are being dropped.
This is where the fun begins as the above mainly reveals a total or partial lack of understanding of signal theory and the sampling theorem in particular. Any scientific reports quoted are from the select list of discredited or never-reproduced efforts.
ETs then kindly (or less so, if they've been doing this almost twenty years) point this out, and try to rectify some misconceptions. Some ETs urge the interested reader to look up Shannon's proof, a piece of math so beautiful it should be on display at the Louvre (or at least Musee d'Orsay).
The ET continues by exposing a number of implementation issues that might pose problems with digital when not done right, issues that are very real due to practical constraints (not very often) or human error/ignorance/incompetence (all too often, check out DAW software programmers or mastering engineers).
AT stubbornly ignores this and essentially refuses to be taught anything new at all. A number of volleys go back and forth, after which suddenly the ETs are accused of stating that digital is perfect.
Bored with this the ET withdraws and does something useful, such as designing a super-advanced anti-alias filter or cueing a record on his $10k turntable.
The CD player is definitely more convenient because your typical unit comes with a remote control weapon with a powerful feature: the "skip to next song" button.
The CD, itself, is a PIA with its high security jewel case that explodes if the wrong person tries to open it, and that *%&%# labeling that takes a magnifying glass to decipher.
The CD, itself, is a PIA with its high security jewel case that explodes if the wrong person tries to open it, and that *%&%# labeling that takes a magnifying glass to decipher.
Yes. Not to mention a professional safecracker to remove those damned sticky anti-theft seals, which are as effective at thwarting the purchaser as they are any potential thieves.
Many feel that not all sparkling new production shows a clean pair of heels to the performers of yesteryear. This is a case for playing LP's as the magnetic tape masters do degrade(as far as I am aware they are not kept in a vacuum protected from all magnetic influence). How this influences the sound depends on the actual instruments themselves; a jazz ensemble seems to suffer not so much but massed orchestral strings can sound rather tired and congested in LP remasterings even after as little as about 10 years. I am,of course, aware, of all the fuss about tranferring with original vacuum tube tape recorders etc, which is a separate issue. The case for playing original 78's is rather less strong since the CD's are tranferred from the shellac discs(or more uncommonly original metal parts). I do play 78's-I find it fun!-and at least I am in charge of equalisation and filtering(or rather,not filtering). I completely accept what engineers say about the digital technology per se. Whether or not it is applied skillfully at every step in the process is a different matter.
I have a cute little CD opener that was a gift at an audio show. Love it.
@ RONGON. Thanks a million for your post. It's an interesting story and comparing the 2 files would be great. It may be just one data point, but it's a start. I know there are other recording engineers on the forum, getting them to do the same thing would be great.
If the files ripped from your master CD-R and the pressed discs are the same, then we will at least know that EAC is doing its job. The next step would be to record the spdif output of CD player where you do hear a difference - and hope that the spdif recorder isn't "fixing" things.
Yes the digital foundation is good but the practical side is another matter. After spending years a projectionist, I tell film makers "Assume that when your film is shown anything that can be done wrong will be done wrong." Most of us are human.
@ RONGON. Thanks a million for your post. It's an interesting story and comparing the 2 files would be great. It may be just one data point, but it's a start. I know there are other recording engineers on the forum, getting them to do the same thing would be great.
If the files ripped from your master CD-R and the pressed discs are the same, then we will at least know that EAC is doing its job. The next step would be to record the spdif output of CD player where you do hear a difference - and hope that the spdif recorder isn't "fixing" things.
Yes the digital foundation is good but the practical side is another matter. After spending years a projectionist, I tell film makers "Assume that when your film is shown anything that can be done wrong will be done wrong." Most of us are human.
I believe you are giving the durability of magnetic tape short shrift. I have heard many LPs of large orchestral pieces that have been remastered from tapes as old as fifty years and they have a more realistic sound than many recent releases on CD. If you try Decca titles remastered by Speakers Corner you'll see what I mean.
John
- Status
- Not open for further replies.