Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: CAN we re-create the "vinyl sound" (Read 35441 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

CAN we re-create the "vinyl sound"

Reply #50 –
Quote
[...] When you listen to string vibrate you can hear allot of things that come from other vibrating objects around the instrument and this is not necessarily a bad thing. You can also "feel" the air vibrating around the string, and it does sound like kind of a "dust" around the sound. [...]

If anything, it is down to the microphone to pick this up, not the recording medium. The recording medium will only record what is fed to it (from source to mic to preamp etc.), and naturally we prefer this to be done with little or no degradation of the signal.

Something that is generally forgotten and/or confused in discussions about bitdepths and sample rates is the difference between producing audio recordings and reproducing them. Recording a cello at 24 bit might make sense, for various reasons, but with a properly treated signal, there's hardly any listening environment around that will make a (perhaps dithered) 16 bit signal insufficient. The air flowing around and into the microphone will make more noise than the playback medium, in this case CD, introduces, as will the blood circulating in your ears - not to mention your amplifier and speaker.

For more in-depth information - 100% void of myths and hype - I can recommend reading stuff written by Nika Aldrich (who happens to be an audio engineer recording classical music), such as his paper on dither, or browse through his forum over at ProSoundWeb. Very informative. Two brief quotes related to this topic, found in a thread here:

Quote
[...]there are only 4 characteristics of waveforms that we look at for the sake of discussing sound waves: frequency, amplitude, phase, and dynamic range (the amplitude of the noisefloor). That's all there is. There are no other characteristics of waveforms. There is no characteristic called "resolution." The function you are calling "resolution" is a made up way of describing an effect on a waveform based on visual observation that manifests itself in some other way - based on the characteristics described above. We need to relate the visible concept that you call "distortion" to an actual characteristic of the waveform.

Example, "distortion" is not a characteristic of a waveform. Instead, we say that the function of "distortion" adds frequencies at certain amplitudes to the already present waveform. [...]

and
Quote
The only thing that is gained between the 16 bit version and the 24 bit version is the increase in dynamic range.
.

This guy has certainly not bought into the warm and fuzzy feeling® myth, or the hype created by manufacturers of recording/playback equipment. Happy learning!

-- Uosdwis

CAN we re-create the "vinyl sound"

Reply #51 –
Positive ABX tests prove that they can hear the difference.
Negative ABX tests prove that it can't be proven to hear a difference.

If you lie on purpose, or guess in an attempt to be correct (because you don't want to fail) you are only fooling yourself. That's ABX for an individual.

As more people do ABX tests, the likelyhood on average of people cheating the system goes lower.

Although it proves nothing if 1000 people ABX and they all lie.  But out of 1000 people, why will they all lie?

Audio is analog. Audio is subjective. Nothing can be proven but statistics can be given. If instead of speakers we have strings vibrating, tubes blowing and piano keys ... (well, we've all seen grand pianos playing by themselves...) you could get a much more realistic reproduction of the sound.

No one will argue. But I sure would like to be able to afford something like that.


Everyone pretends to see the emperor's new clothes ... but there will always be that child who will speak up.

EDIT: I remember now. It's called the law of large numbers. What is the chance that you roll a 6 on dice? Theoretically it is 1:6. In practice, you don't roll a 6 1 time out of 6. But if you roll it 600 times your number is close to 100. If you roll it 6000 times your number is close to 1000. It will rarely be an exact multiple of your theoretical ratio but approximates it much better the more people do it.

ABX test have a ratio for individual people, something like 12/16. If you lie, you only fool yourself. In a group such as this forum, every individual that contributes to the test increases accuracy. On very obvious samples you don't even need an ABX to prove it's different. (Scientifically you can't prove it, but you'll be the odd person.)

On subtle differences it may not prove anything, but it shows the probability that you can't hear the difference.

When someone offers you a delicacy from their culture, eat it first. Then ask what its made of. Sometimes you'll surprise yourself. Hey, they're not dying and they've probably been eating it all their lives.

CAN we re-create the "vinyl sound"

Reply #52 –
Quote
Quote
The only thing that is gained between the 16 bit version and the 24 bit version is the increase in dynamic range.
.

Avoiding rounding errors and this about greater dynamic range are the only things I have read in this thread so far that are using fact to describe the benefits of 24/96 over 16/44.1.  The rest is arguing over opinions and personal perceptibility, or the usual "ABX the two and get back to me" except this time this only goes further by making it a requirement that both be sampled from the same original master.  Now don't get me wrong, I am not a proponent of either, I am simply trying to seperate the facts from the opinions.

FWIIW, to me it all comes down to how much one cares about the storage space containing the higher quality recording and whether or not he or she is well-off enough in this world to afford the equipment that MIGHT show an audible difference in the two.  Yes, from the statements in this thread, it is obvious that some of us believe that the recording industry is only trying to make more money with these new formats, which is morally wrong to many people apparently, that is, if the quality is not significantly better.

Besides that fact that they want to make more money off of us, what makes a 24/96 recording WORSE than a 16/44.1?  Is there anything wrong with buying a higher quality recording if you can afford it, or are there problems with 24/96 recording that make it a bad choice over 16/44.1?

Here are my questions again:  What are the factual advantages besides what I already mentioned that make 24/96 better than 16/44.1?  What are the advantages and the reasons against buying higher quality recordings than 16/44.1?


Edit: added a line.
WARNING:  Changing of advanced parameters might degrade sound quality.  Modify them only if you are expirienced in audio compression!

CAN we re-create the "vinyl sound"

Reply #53 –
I would like to go back to the to the LP discussion for minute.
I doubt it we will ever agree on whether there are audible differences between 24/96 and 16/44.1, I suppose the final answer will be: "it depends on the equipment used to record, the equipment used for playback and on the listener's ear". I don’t mean to disrespect anyone, but please believe that the ear of someone that has fine tuned an instrument everyday for 15 years has a little more ability to distinguish then others.

I agree with mmortal03 that eventually it all comes down to money, if you can afford the storage and the equipment they why not…

But I would like to get back to the source, this was not intended to be a theoretical discussion, I have a large collection of LPs and in fact LPs are very cheap to buy these days.
The interesting question is how good a signal can be extracted from a phono, it may not be "EXCATLY" the same as the source, but those who actually had the pleasure of listening to tube amps connected to Thorens turntable with proper speakers probably noticed it sounds "different" then even a very good CD transport. The question here is not weather it sounds better then a 16/44.1 CD, the question I want an answer for is do I have a chance to reproduce this same "nature" using higher sample rates and wider dynamic range.
Perhaps 24 bit is an over kill perhaps 96khz is also an overkill, but the fact of the matter is that to my ears a 16/44.1 CD does not sound the same as an LP.
What I want is the comfort of digital transport (and that is all it is, comfort) and the nature of an LP.
Since 24/96 recording are still very rare, and some material will never be, as the artists have died long ago, I believe LPs are as close as I can get.
 
Uosdwis R. Dewoh, I will read the article you posted, it seems to be very interesting.

Thank you all for a very interesting discussion, I will go to sleep now

BTW I am thinking of getting the M-Audio USB Audio Duo or the USB Quattro for that purpose, comments will be welcome.
I am also opened to suggestions about more expansive equipment.

http://www.midiman.com/products/m-audio/duo.php
http://www.midiman.com/products/m-audio/quattro.php
"La vengeance est un plat qui se mange froid."

CAN we re-create the "vinyl sound"

Reply #54 –
Quote
I doubt it we will ever agree on whether there are audible differences between 24/96 and 16/44.1, I suppose the final answer will be: "it depends on the equipment used to record, the equipment used for playback and on the listener's ear". I don’t mean to disrespect anyone, but please believe that the ear of someone that has fine tuned an instrument everyday for 15 years has a little more ability to distinguish then others.

But I would like to get back to the source, this was not intended to be a theoretical discussion, I have a large collection of LPs and in fact LPs are very cheap to buy these days.
The interesting question is how good a signal can be extracted from a phono, it may not be "EXCATLY" the same as the source, but those who actually had the pleasure of listening to tube amps connected to Thorens turntable with proper speakers probably noticed it sounds "different" then even a very good CD transport. The question here is not weather it sounds better then a 16/44.1 CD, the question I want an answer for is do I have a chance to reproduce this same "nature" using higher sample rates and wider dynamic range.
Perhaps 24 bit is an over kill perhaps 96khz is also an overkill, but the fact of the matter is that to my ears a 16/44.1 CD does not sound the same as an LP.
What I want is the comfort of digital transport (and that is all it is, comfort) and the nature of an LP.
Since 24/96 recording are still very rare, and some material will never be, as the artists have died long ago, I believe LPs are as close as I can get.

The problem I'm having is that the tube amp reproduces the sound inaccurately, and for the life of me I can't see why anyone would willingly want the wrong sound if they value the accuracy of reproduction at all. That's the point of psychoacoustic compression: sure, it may not be the same file, exactly, but it sounds identical (or damn close) to the original source.

That being said, someone should do something like this, take a sound, run it through a tube amp, and construct an impulse response table/curve/descriptor for it so the people who like the sound of tube amps and "high quality" equipment like this can have some facsimile of it. That'd make FB2K perfect. 

As for distinguishability, can you trust the people of the board who have been training their ears to hear psychoacoustic artifacting and loss in broad ranges of audio that their ability to distinguish any kind of loss in music conversion is more tuned than others? I'd trust you with your instrument, but your training ends there. It does not translate into training to distinguish broad classes of audio loss.

CAN we re-create the "vinyl sound"

Reply #55 –
Quote
Perhaps 24 bit is an over kill perhaps 96khz is also an overkill, but the fact of the matter is that to my ears a 16/44.1 CD does not sound the same as an LP.

[paranoias]
...Music is ART, not MATHS.
[/paranoias]

Vynil is never exact because it's analog. Unlike CD, how will it sound EXACTLY cannot be calculated.
Have you tried to sample the output of vynil sourced sound with the Hi-fi's headphone output connected to the soundcard's Line In at 32bit 48kHz with a editing program like Cool Edit pro, then downsampling to CD quality?

CAN we re-create the "vinyl sound"

Reply #56 –
Quote
Vynil is never exact because it's analog. Unlike CD, how will it sound EXACTLY cannot be calculated.

Well unless the CD audio is digitally beamed into your skull, it's gonna have to be converted to analog data at some point... and then you get uncertainties. 

(That's assuming the digital data is being properly transported in the first place...)

CAN we re-create the "vinyl sound"

Reply #57 –
@bryant
Have you thought over if you should use APE tags for Wavpack?
it'd be a horrible loss if you didn't.... relying on ID tags would be horrible...

CAN we re-create the "vinyl sound"

Reply #58 –
Quote
Well unless the CD audio is digitally beamed into your skull, it's gonna have to be converted to analog data at some point... and then you get uncertainties.  

What?  I like listening to 1s and 0s, don't you too?

Ok, sure, but vynil will always be less accurate...

CAN we re-create the "vinyl sound"

Reply #59 –
Properly made vinyl has dynamic range higher than non-dithered CD!

But plastic properties (it's an overcooled fluid, similar to glass)
make it useless for long term storage.
(at least in temperatures higher than absolute zero)
It just flows (very slowly, but still... gets thicker in lower parts, thinner in the upper parts).
After about 15 years vinyl is no longer accurate.
And that's not even counting dust and scratches!

CDs/DVDs are immune to this (at least pressed ones), because data is written in metal.
If you store as a file, you get even more error protection.
After 25 years they will still be bit accurate (if not too dirty or damaged).

SyletH, you forgot that 90% of recordings are emphasized
and they should be deemphasized them by applying a special filter or you won't get real dynamics.

We got off-topic...

And don't start tube amps vs. transistor amps war here, same goes for vinyl vs. CD!
Just use what you want, I might only give you some pointers.

<edit>
Some modifications.
</edit>
I've changed only because of myself.
Remember, when you quote me, you're quoting AstralStorm.
(read: this account is dead)

CAN we re-create the "vinyl sound"

Reply #60 –
Quote
That being said, someone should do something like this, take a sound, run it through a tube amp, and construct an impulse response table/curve/descriptor for it so the people who like the sound of tube amps and "high quality" equipment like this can have some facsimile of it. That'd make FB2K perfect.  :D

Something like that does exist... but it isn't free.
It's called iZotope Ozone... has heck-of-a-lot of effects.
Compressor, configurable limiter, reverb, delay, parametric EQ
and another thing:
Quote
Harmonic excitation based on analog tube saturation models. Multiband excitation with adjustable crossovers to add extended bass, sizzling highs, or anything in between.

If you want crackles and dust, then try iZotope Vinyl. (But who sane likes that?)
If someone would make DirectX audio plugin interface for foobar2000,
then it could be used with full quality.
http://www.izotope.com
I've tried it once for winamp, but I think I prefer normal, unprocessed sound.
I've changed only because of myself.
Remember, when you quote me, you're quoting AstralStorm.
(read: this account is dead)

CAN we re-create the "vinyl sound"

Reply #61 –
Quote
Properly made vinyl has dynamic range higher than non-dithered CD!

I don't think so, by a wide margin

CAN we re-create the "vinyl sound"

Reply #62 –
Quote
CDs/DVDs are immune to this (at least pressed ones), because data is written in metal.
If you store as a file, you get even more error protection.
After 25 years they will still be bit accurate (if not too dirty or damaged).

But alluminium rots faster (20yrs estimated) than gold (100yrs estimated). Too bad the gold discs are not so common anymore. Don't know a thing about the materials used for DVD-R. I saw a "Smart and Friendly" DVD-R blank the other day, Smart and Friendly is a cheap brand in CD-R media... And there is also those stupid drives spinning discs too fast... When the CD format spec was made, Phillips stress tests sated that discs will physically break between 50x to 60x speeds. Oh why nobody ever reads the specs these days..., please provide methods to make those cheapo cdroms not spin so fast, about 40x at max would be ok with me.
She is waiting in the air

CAN we re-create the "vinyl sound"

Reply #63 –
Yes, it oxygenates, but if not exposed to air it won't 'rot'.
Good CDs have plastic all over data layer, even back.
The thicker, the better, but also slower.

R/RW discs are no good for storing anything for long...
organic dyes used in them last about 1-10 years.

Some CDs are so bad that they break when read.
Plastic used in them isn't temperature resistant
and breaks on sudden temperature change.
I've changed only because of myself.
Remember, when you quote me, you're quoting AstralStorm.
(read: this account is dead)

CAN we re-create the "vinyl sound"

Reply #64 –
Quote
@bryant
Have you thought over if you should use APE tags for Wavpack?
it'd be a horrible loss if you didn't.... relying on ID tags would be horrible...

I agree.

The WavPack format actually allows any kind of tags as long as they go at the end of the file, so the only thing that would have to change is having the winamp plugin look for APE tags. I'll look into this and drop it in if it's not too hard.

CAN we re-create the "vinyl sound"

Reply #65 –
Quote
Quote
Properly made vinyl has dynamic range higher than non-dithered CD!

I don't think so, by a wide margin

It has dynamic range of a microphone used for recording if clean and new.
But it quickly loses this property, due to dust, wear and scratches.

I didn't state that it's more accurate, only that it has higher dynamic range.
Frequency properties of vinyl are very far from what could be called accurate.

<edit>
I'm leaving this topic... if you want to discuss abot storage media just open a new one.
</edit>
I've changed only because of myself.
Remember, when you quote me, you're quoting AstralStorm.
(read: this account is dead)

CAN we re-create the "vinyl sound"

Reply #66 –
Quote
Quote
Vynil is never exact because it's analog. Unlike CD, how will it sound EXACTLY cannot be calculated.

Well unless the CD audio is digitally beamed into your skull, it's gonna have to be converted to analog data at some point... and then you get uncertainties. 

(That's assuming the digital data is being properly transported in the first place...)

This has been done for a long time using the MATRIX technology.
--  Frank Klemm

CAN we re-create the "vinyl sound"

Reply #67 –
this was a joke, right?!
"La vengeance est un plat qui se mange froid."

CAN we re-create the "vinyl sound"

Reply #68 –
No. It's true. I myself am not typing. I am just thinking the words onto the applet. Good old rj-11 hooked right into my implant.
r3mix zealot.

CAN we re-create the "vinyl sound"

Reply #70 –
Quote
About MLP, I have read that when MLP is unable to meet the target bitrate it drops into a lossy mode that is probably identical to WavPack's in principle.

There's some bit-buffering, and there's never been a "running out of bits" problem with a disc yet.

However, if you send 6 channels of 24/96 white noise to the MLP encoder (or any other signal that it can't encode losslessly into the required bitrate), it simply tells you that you can't encode this signal onto DVD-A, and suggests that you either reduce the bit-depth of some or all channels, or downsample some or all channels. Since you only need 6-bits to represent white noise (supposedly any more make would be inaudible) it's hardly an issue for this most agressive test case. For real music, I guess converting the rear channels to 22-bits is hardly going to ruin the experience!

Whatever, in the (currently unheard of situation) where there aren't enough bits, it's the mastering engineer's call as to what happens - you won't drop into lossy coding without knowing about it, or having a choice.

Cheers,
David.

CAN we re-create the "vinyl sound"

Reply #71 –
Quote
[span style='font-size:14pt;line-height:100%']Theory:[/span]
64 kHz is the maximum measured harmonic frequency coefficient generated by any instrument.
(mostly brass) (I've read it in some old book... can't quite remember which was it, so don't quote me)

A study on this found 100kHz harmonics from a trumpet - and that wasn't the limit of the trumpet's harmonics - just the limit of the equipment in the study.

http://www.cco.caltech.edu/~boyk/spectra/spectra.htm

see also

http://www.tgina.com/professional/pdf/wp_wideband.pdf

Cheers,
David.

CAN we re-create the "vinyl sound"

Reply #72 –
Quote
Quote
Properly made vinyl has dynamic range higher than non-dithered CD!

I don't think so, by a wide margin

It could, at some frequencies, under particular circumstances. The poster didn't specify how much distortion is allowed; and remember the "dynamic range" value extends down into the noise - it's the "signal to noise ratio" doesn't. So, what he says is possible.

However, digital audio should always be dithered down to 16-bit for creating a CD. Truncating (and hence loosing or distorting everything below -96dB) is just stupid. Though some people occasionally like the harmonic distortion it adds. Whatever, with dither, the CD wins this particular numerical comparison I think, unless someone knows better.

D.

(sorry about so many posts!)

CAN we re-create the "vinyl sound"

Reply #73 –
Quote
Quote
That being said, someone should do something like this, take a sound, run it through a tube amp, and construct an impulse response table/curve/descriptor for it so the people who like the sound of tube amps and "high quality" equipment like this can have some facsimile of it. That'd make FB2K perfect.  

It's very difficult to add perfect analogue harmonic distortion using digital processing. Not impossible, but not trivial. It's certainly easier to do in the analogue domain! In the digital domain, the process is plagued by aliasing.

CAN we re-create the "vinyl sound"

Reply #74 –
Quote
Quote
[span style='font-size:14pt;line-height:100%']Theory:[/span]
64 kHz is the maximum measured harmonic frequency coefficient generated by any instrument.
(mostly brass) (I've read it in some old book... can't quite remember which was it, so don't quote me)

A study on this found 100kHz harmonics from a trumpet - and that wasn't the limit of the trumpet's harmonics - just the limit of the equipment in the study.

http://www.cco.caltech.edu/~boyk/spectra/spectra.htm

see also

http://www.tgina.com/professional/pdf/wp_wideband.pdf

Cheers,
David.

There's is NO limit.
It is silly to think there's a hard limit.
The higher the frequency the lower the amplitude becomes.
There's a typical maximum in amplitude for a given frequency and then a rolloff between
6 dB/oct for undamped instruments and a much higher rolloff for damped instruments.

The problem are not the instruments, but the mechanic of the human ear. The
sound transport works well until 10...12 kHz, above this frequency the damping of
the ear raises dramatically by 6...10 dB/kHz.

So you need a very high SPL to get some signal at the cochlea at frequencies above
22...25 kHz.  Above 25 kHz nonlinear processes in the air becomes more audible than
the original sound itself.

Die meisten Fragen und Antworten in diesem Forum zeigen, daß kaum einer
der hier Anwesenden auch nur eine blasse Ahnung davon hat, wie kompliziert
die Fragen zu beantworten sind, die gestellt werden. Und ich glaube auch kaum,
daß er an einer exakten Beantwortung der Frage interessiert ist, weil das nicht
stammkneipenkompatibel ist und damit nicht in die heutige Zeit paßt.

Wo viele als Antwort eine Zahl erwarten (96 kHz, 32 bit oder was auch immer),
besteht eine technische korrekte Form vielleicht aus einer einsemestrigen Vorlesung,
die nur verstanden werden kann, wenn technisches Hintergrundwissen vorhanden ist.

Ansonsten sind in Luft bei normalem Druck Schallwellen bis etwa 10 MHz überhaupt möglich,
diese hohen Frequenzen haben aber so hohe Dämpfungen, daß sie sich ein paar Mikrometer
weit ausbreiten können.

Im makroskopischen Bereich sind nur Frequenzen unter 100 kHz interessant, und selbst bei
100 kHz verreckt Schall nach ein paar Metern (ca. 1 dB/m Dämpfung).
Wenn man in einem Konzert 30 Meter entfernt ist, sind das 30 dB Dämpfung
durch Verluste in der Luft (diese Dämpfung hat nichts mit der 1/r²-Dämpfung durch
die Schallverteilung zu tun).

Das nur mal als Bemerkung zur Problematik: Ausbreitung von der Schallquelle
zum Ohr. Alle wichtigen Informationen zum Verständnis von guter Musikwiedergabe sind
mittlerweile im Web zu finden, für den der sucht.
Man sollte nur einen großen Bogen um die Worte "Hifi" und "Highend" machen, da geht es
mit hoher Wahrscheinlichkeit nicht um hohe Qualität, sondern eher um hohe Geldbeträge.


--  Frank Klemm