[doug s.]

OK, but the question arises how an ultrasonic sound can either "affect" those below 20 khz, and how someone could perceive them (through hearing, that is...I can certainly see how ultrasonics might be otherwise perceived, especially at high energy levels).

As far as I am aware, there are only two ways -- via subharmonics, and via intermodulation distortion products with a result in the audio range.  So, yes, I'd agree that "effects" of ultrasonic sounds might be heard (though not the fundamental frequencies, by definition).

However, those "effects" are ALREADY PART OF THE SOUNDFIELD when recording music!  So a well-made recording has those effects in i,  within the audible frequency range, does it not?

And, so, no oversampling should be required to digitize them.

ultrasonics affect the harmonics of lower-frequency sounds.  if you want the full picture, you want to record and play back the full picture.  if you record all the ultrasonics, but do not play them back, then the playback will be missing that interaction.

doug s.

[Chromisdesigns]

ultrasonics affect the harmonics of lower-frequency sounds.  if you want the full picture, you want to record and play back the full picture.  if you record all the ultrasonics, but do not play them back, then the playback will be missing that interaction.

doug s.

I'd buy that, except HOW are you going to "record the ultrasonics", when studio mikes fall off rapidly after (and often before) 20 Khz?

The only thing related to ultrasonics that is getting recorded is whatever sub-harmonics and intermodulation products  that already exist in the soundfield coming from the instruments.

If you viewed a wide-band spectrum analysis of the mike output, I don't believe you would see much at all over 20 Khz.  Would you?

[doug s.]

chromi, i don't know, re: mic response.  but, imo, what needs to be done is improve the resolution all the way around, not dumb it down to the lcd...

sony/philips could have introduced cd at 24/192 right from the start - ya, it may have delayed the roll out a year, and it may have reduced the profit margin from ~2000% per disc to ~1800-1900% or so, but imagine how much better digital audio would be today, if they'd stardes with 24/192 at the onset, and had ~30 years to perfect it?   

ymmv,

doug s.

[*Scotty*]

I would like to see someone connect the dots between the number of bits and the 1% level of distortion.
 1% THD is the generally accepted point where the amount total harmonic distortion that is present, at least in an amplfier, is considered to have reached an objectionable level. At what number of bits in a 16bit system and a 24 bit system has this threshold been exceeded?
 Also is there a correlation between the 1% THD level and spectral content of the music. I suspect our ability to hear when the music has 1% THD is dependent on frequencies at which 1% THD have been exceeded. I arbitrarily chose a 1% level. It has been reliably reported by a number of researchers that our ability to audibly detect THD is quite a bit lower than a 1% THD level.
 I suspect that the 16bit system's short-comings lie in how rapidly its distortion level increases as the signal level decreases and there are fewer and fewer bits left to describe the waveform.
I have never seen anyone thoroughly discuss the consequences of a recording medium having an inverse relationship between signal level and total harmonic distortion characteristics.
 I think getting distracted by the need to preserve ultra-sonic overtone components that may or may not have been captured due microphone limitations when the recording was made is counter productive. The extra bandwidth beyond 20kHz is helpful because it allows us to place our anti-aliasing filter poles far enough outside the audio-bandwidth to avoid phase shift in the mid and high frequencies of the audio-band.
 This is at least one trickle-down improvement for CDs that is a direct result of 24/96 mastering.
Scotty

[JohnR]

I'd buy that, except HOW are you going to "record the ultrasonics", when studio mikes fall off rapidly after (and often before) 20 Khz?

Earthworks sell mics specified as flat to 50 kHz, presumably others do too.

[doug s.]

I think getting distracted by the need to preserve ultra-sonic overtone components that may or may not have been captured due microphone limitations when the recording was made is counter productive.  The extra bandwidth beyond 20kHz is helpful because it allows us to place our anti-aliasing filter poles far enough outside the audio-bandwidth to avoid phase shift in the mid and high frequencies of the audio-band.
 This is at least one trickle-down improvement for CDs that is a direct result of 24/96 mastering.

i wholeheartedly agree about the importance of hi-rez in the studio.  but...  if the only adwantage of hi-rez recording is to make redbook playback so much better - i.e. it doesn't have any benefits in the playback realm - then doesn't this sort of admit that 16/44.1 is all you need for playback; that hi-rez is only needed for recording to ensure the redbook cd is mastered properly, w/o the problems associated with mastering in 16/44.1?  if, in fact, it is true that mastering at 24/96 or higher allows for perfect undistorted 20hz-20khz discs - which redbook can do - then if there's no impact in the audible bandwidth above 20khz, why does one need hi-rez playback?  (personally, i hope it is not true, as i think winyl is easily superior to redbook; i would like to hear a digital source that is not only enjoyable, but sounds as good as a record...)

doug s.

[AgentOrange]

Funny timing for this. I posted this yesterday on yardcraphome.com

======================================




Playing around on a project I kinda did before. On the treadmill this morning, played "Rumours" on the turntable.

Later I loaded FleetWood Mac's "Rumours" into my playlist on the deck. Actually, I did it with three copies. I have heard this album a million times over the years and know it well. I loaded these:

CD - 14H kbps 44kHz
Flac - 892 kbps 44kHz
24bit -30H kbps 96kHz
These numbers of course change a little on different songs.

All are in a playlist on my WinAmp player going through "exactly" the same system except the CD is in the computer drive and the files are on a Solid State hardrive. All audio is turned off in the computer, no fan and nothing moving except for the CD spinning.
I am able to click a tune in the playlist, hear a few notes and then click the other versions. I can jump between them with one click.

After about 30 minutes I have a few observations. The Flac file is easy to pick as it is about one notch lower in volume - that means nothing. To equalize the test, I have to go one notch up or down on the volume knob when it comes up, it is very close.
The 24 bit sounds better. The tunes sound stronger and the bass is tighter. The singers sound more alive.
Maybe the CD has a slight edge over the Flac, but I don't know why I say that.
All sound very good.

A test means little unless it is a blind test, and I would need help for that. After a lot of concentrated listening, came up with the idea to put them on shuffle and try to sit and guess which version loaded up. After about 20 minutes I got good at it. Not a fair test as the Flac has slightly lower volume level and the tight bass points out the 24 bit. After a little practice, I can guess right because of those factors, so it means nothing. I will try it all again sometime when I have a guest. Then I could do the controls, give him a pad to guess on. I guess a few people would be better. I could move it down stairs as there is only one 'sweet spot' on the deck.

The verdict:
The 24 bit sounds better. I can't really hear much between the other two versions but my nod would go to the CD.
They all sound very good to me.
It would be fun to compare the LP, but it is in the bar and playing on a different (better) system not concerned with digital conversion.

[trebejo]

Fleetwood Mac's first two albums with Lindsey and Stevie have some truly horrible CD versions out there. I keep hoping to find "the good one".

If you are going to do this test, do it like you mean it! Take the very same 24 bit upsampled signal, downsample it properly and channel it through your (good) DAC. Pre-balance the sound level and avoid psychoacoustic effects. Use a trio of fellow audiophiles and don't tell them which is which beforehand.

Hey, sounds like a lot of work! Yep, it sure is, but of course if you're going to obsolete Shannon you should expect it to be a little bit of work.

<sigh> If we keep going this way we're going to end up arguing about whether blondes are better than brunettes...

As long as we're being anecdotal, at the moment I am listening to CCR's Chronicle. It sounds fantastic, better than it ever did on that scratchy vinyl. Hey, it's a flac file, through a great DAC! We could do this anecdotal thing all day.

It seems nobody seems interested in testing like the OP suggested. Alas, my DAC is of a previous generation and lacks that feature.

btw I owned one of the upsampling DACs before this fuddy duddy playing here now. Well the upsampling failed miserably in comparison. The world's full of anecdotes...

[*Scotty*]

trebejo, I purchased The Very Best Of Fleetwood Mac on Rhino Records and I was very pleased with the sound of this 2CD set. It has a lot of material from the two albums you referenced. On this basis
I think you would find the re-masters from Rhino of these two albums satisfactory.
 I have both them of my short list. See Amazon links below.
http://www.amazon.com/Fleetwood-Mac-Deluxe-Edition/dp/B00009RAJH/ref=sr_1_3?ie=UTF8&qid=1331837120&sr=8-3
http://www.amazon.com/Rumours-Deluxe-Fleetwood-Mac/dp/tracks/B00009RAJI/ref=dp_tracks_all_1#disc_1
Scotty

[rbbert]

...
sony/philips could have introduced cd at 24/192 right from the start - ya, it may have delayed the roll out a year...

More like 20 years; microprocessors capable of decoding 24/192 and working in a home player were certainly not developed before 2000.

[trebejo]

trebejo, I purchased The Very Best Of Fleetwood Mac on Rhino Records and I was very pleased with the sound of this 2CD set. It has a lot of material from the two albums you referenced. On this basis
I think you would find the re-masters from Rhino of these two albums satisfactory.

Thanks! This would fill a noticeable gap.

[doug s.]

More like 20 years; microprocessors capable of decoding 24/192 and working in a home player were certainly not developed before 2000.

48khz & 50khz was awailable even before the cd came out.  sony/philips simply assumed 16/44.1 was good enough, thanks to the nyquist–shannon sampling theorem, and to the fact that they could record easily, by using already existing wideo tape and machines.  if they wanted higher rates, they would have deweloped them a lot sooner.  but, they were so sure they already had "perfect sound forever"...

doug s.