96Khz vs 192Khz

[ssully]

"just set it so the hottest LP signal does not clip, and leave it there"

This was the purpose of my experiment. I was trying to find where this point was on the new equipment I had just bought. I tried different levels because I did not know where that point was on the display. It was by pure chance that I noticed that the one level setting sounded like the CD.

The places where I heard this realism effect I liked were between the songs on the record when everything in the sound was low level. During the song, the effect was drowned out by the music. On the CD, there is much less between the songs.

Well, again, something was wrong 20 years ago with either the technology or the procedure. Done right today (or then , frankly), there would be no such difference. It is certainly nothing inherent to CD or to digital recording.

 

[quadsearcher]

I'm fairly certain I know what was going on. The Phillips CDR-200 CD recorder had a flaw. I saw it doing warranty service back then, and I actually bought one knowing it had a flaw because I wanted to confirm it and understand what the heck was going on. Philips had no bulletins on the topic. Its input impedance was apparently very low and it would load down the source. If I recall correctly there were 100Ω SMD resistors across the line inputs, as shown in the service manual. I never tracked down where they were on the PCB. This could only be a mistake or a deliberate crippling of recorded quality.
No solid state preamp I tried would drive it well enough to have the recording sound like the source.
It worked "fine" with a tube preamp.

 

[par4ken]

Well, again, something was wrong 20 years ago with either the technology or the procedure. Done right today (or then , frankly), there would be no such difference. It is certainly nothing inherent to CD or to digital recording.

Here is an experiment for someone to try. Record a selection of music @44.1Khz/16 bit, keeping the level high but not clipping. Record the same selection but 20 or 30 dB lower than the first recording. Then normalise both to a maximum of 0. What Ssully is saying is that both will sound exactly the same, I personally doubt it. I would be interested in peoples' opinion, if you hear a difference fine if you don't also fine but please do respect others opinion on the matter!

 

[jimfisheye]

Try that experiment with 24 bit, not 16 bit!
Or... try it with both to demonstrate the shortcoming of 16 bit along with how it becomes a moot point with 24 bit.

A recording peaking at -40db will be captured with 17 bit resolution with a 24 bit system. The same signal will be captured with 9 bit resolution with a 16 bit system.

That -40db peak signal over a balanced analog audio connection captured with 17 bits (of 24) will still be fully intact. The same signal through an unbalanced analog connection into 9 bits (of 16) will be audibly damaged.

Try it!

You can also do trials of sample rate conversion. Start with your own source. Fool your own lying ears.

Some of the above shenanigans with unbalanced connections with dodgy consumer electronics is exactly what pro or "prosumer" devices with balanced audio connections avoid. And playing around with early generation 16 bit digital devices (that may have been more like 12 bit devices) with unbalanced analog connections... Yeah, that stuff is cheapness. You're not comparing digital technology here! Just different examples of hot garbage.

The last 30 years of advancement with 24 bit AD and DA stages isn't some made up bs. Balanced audio connections are a thing and have been for a lot longer than that. Being convinced that the old 16 bit devices are the end all be all of digital and refusing to entertain the modern digital age would be a mistake. Any modest USB connecting audio interface with balanced analog connections for $300 give or take is a very accurate recording device.

Vinyl is still an expensive PITA. You need a cartridge built by a Swiss watch maker and a dialed to the nth degree preamp to boost that trickle of a signal up to line level without mutilating it. A modern audio interface with a balanced input for line level should not be the challenge here! If something is malfunctioning that grossly... Honestly, just throw it away and start over with just about anything else made in the last 10 years.

 

[jimfisheye]

a nest of uncontrolled variables

Great album name.

 

[atrocity]

Turntable: Collaro TSC-640-1019

Clearly really a Voice of Music. I'm not saying that's bad, I'm just fascinated by how many companies but their names on V-M hardware.

 

[atrocity]

Didn't those standalone CD recorders also limit you to making only a single direct digital copy?

Depends on how the source CD is flagged. If the flag isn't set, then you can make one direct copy but can't copy that copy. If the flag is set, you can serially copy.

I'm working my way through ripping a friend's CD library for him and am surprised at how many I'm encountering with FLAGS DCP ("Digital Copy Permitted") set. It's definitely a small minority of the 1500 or so I've gotten through, but it's probably a couple dozen at this point, including some major label releases (just don't ask me to remember which right now). I assume they were all set that way in error.

It's also surprising how many CDs out there are either quietly flagged as HDCD or--much, much worse--openly advertise themselves as HDCD when they don't actually use any HDCD features at all. There's even one Tracy Chapman album that claims to be HDCD and has a couple tracks that are flagged as using HDCD features, but when the standard 16/44.1 and 20/16.44 files are normalized to the same levels are visually identical.

 

[ssully]

Here is an experiment for someone to try. Record a selection of music @44.1Khz/16 bit, keeping the level high but not clipping. Record the same selection but 20 or 30 dB lower than the first recording. Then normalise both to a maximum of 0. What Ssully is saying is that both will sound exactly the same, I personally doubt it. I would be interested in peoples' opinion, if you hear a difference fine if you don't also fine but please do respect others opinion on the matter!

You know, I wondered when I wrote that post whether I should include a phrase like, of course, recording at too low a level will 'lose' signal, in any medium. But then I figured nah, any reader will realize that writing "something was wrong with either the technology or the procedure" and that 'done right, there would be no such difference", covers it.

Silly me.

 

[MidiMagic]

I'm fairly certain I know what was going on. The Phillips CDR-200 CD recorder had a flaw. I saw it doing warranty service back then, and I actually bought one knowing it had a flaw because I wanted to confirm it and understand what the heck was going on. Philips had no bulletins on the topic. Its input impedance was apparently very low and it would load down the source. If I recall correctly there were 100Ω SMD resistors across the line inputs, as shown in the service manual. I never tracked down where they were on the PCB. This could only be a mistake or a deliberate crippling of recorded quality.
No solid state preamp I tried would drive it well enough to have the recording sound like the source.
It worked "fine" with a tube preamp.

That sounds backwards to me. A tube preamp usually had a 10K output impedance, whereas a transistor preamp usually had a 1K output impedance. My transistor preamp has a 600Ω output. I would think that a lower impedance would drive a 100Ω input better than a high impedance.

I can imagine a 100Ω series resistor on the input, but not a shunt resistor. The series resistor would protect the input from connection to a speaker output.

I never had trouble with input loading, and I am still using it.

Like I said, I used that low signal level only as part of this calibration process, and once to investigate if the same thing happened to Steppenwolf Live that happened to Hot August Night.. After I found the correct setting, I used only that from then on.

I am now wondering if they gain-rode these live albums when making CDs because they sound weird when random track play order is applied to a live album.

 

[jaybird100]

CD recorders were a rather cool device when they first came out. Sadly they were crippled right from the start. They required the use of special music CD's to which a tax was added for the recording industry. Presumably the tax was to help support the "starving artists" but in reality was likely more to supplement the incomes of the successful artist to make up for the supposed losses due to piracy.

The proliferation of home computers with CD burners and soundcards made the CD recorder obsolete very quickly, almost right away. With a computer you could easily edit your recording, de-click vinyl, normalise etc. You could make digital copies of commercially releases CD's as many times as you wanted. Didn't those standalone CD recorders also limit you to making only a single direct digital copy?

The behavior of the recording industry in those days quite frankly disgusted me.

As a user of a CD recorder, I can tell you with some authority that, yes you could copy a disc digitally, but you couldn't make a digital copy of that digital copy.

 

[Wagonmaster_91]

I have a Marantz CDR630 that records on regular CD-Rs. I chose that model specifically because it did not require using 'Music CD-Rs".

 

[marcb]

I have a Marantz CDR630 that records on regular CD-Rs. I chose that model specifically because it did not require using 'Music CD-Rs".

FWIW, and if I recall correctly, the hardware which did not require Music CD-Rs had the “tax” built into the price of the hardware.

 

[Wagonmaster_91]

I thought they tied the tax to blank CD-R media purchases. Been to long to remember.

 

[ssully]

It's also surprising how many CDs out there are either quietly flagged as HDCD or--much, much worse--openly advertise themselves as HDCD when they don't actually use any HDCD features at all. There's even one Tracy Chapman album that claims to be HDCD and has a couple tracks that are flagged as using HDCD features, but when the standard 16/44.1 and 20/16.44 files are normalized to the same levels are visually identical.

In that case it's probably just 'virtue signaling' -- showing that the then-SOTA Pacific Microsonics A/D converter was used for the release, even though none of its optional features were.

There is excellent and comprehensive information about HDCD on the Hydrogenaudio wiki here and here.

 

[MidiMagic]

Clearly really a Voice of Music. I'm not saying that's bad, I'm just fascinated by how many companies but their names on V-M hardware.

Turntable: Collaro TSC-640-1019
VM didn't make anything this good (except maybe the 1585).

 

[marcb]

I thought they tied the tax to blank CD-R media purchases. Been to long to remember.

Again, as I said, I believe they did both.

 

[jimfisheye]

I had an HHB CDR recorder around 1997. Recorded to CDR in real time. Hard to imaging any appeal for such a thing nowadays! Probably spent hundreds on it... I still have a Sony R500 DAT deck from around the same time. Last time I tried to use it it failed me. DAT format... I see head alignment in my future but I digress.

I remember there was a line of CDR burners that used the proprietary non standard pre-formatted CDRs mentioned above. They discounted the machines to try to lure you in. It was a scam product you needed to know to avoid. There's always one in the crowd...

Someone did this with cassettes way back in the early days too. (Slightly different shaped cartridge.)

 

[atrocity]

Turntable: Collaro TSC-640-1019
VM didn't make anything this good (except maybe the 1585).

Yeah, looking again I see I'm completely wrong. The layout is EXTREMELY similar to what V-M did for years, though.

Apparently Collaro, like V-M, also made changers that others put their name on, so I had the entire thing backwards.

 

[atrocity]

In that case it's probably just 'virtue signaling' -- showing that the then-SOTA Pacific Microsonics A/D converter was used for the release, even though none of its optional features were.

There was definitely some of that going on, though I'm still really confused about the Tracy Chapman tracks that claim to use the features but don't. Then again, I'm relying on ffmpeg and have no way of knowing if it's buggy.

I've also seen the opposite situation where discs are not labelled as HDCD but are quietly flagged as such. No doubt that's simply the result of using the PM converters. I have yet to encounter a CD that's silently flagged as HDCD and really does use any of the options.

 

[murrysdad]

I had a wonderful brilliant friend Russ Hamm who in the day wrote a seminal paper at the dawn of the new era, "Tubes Versus Transistors-Is There An Audible Difference" Russ always felt that 192k recording was the closest thing to tube sound that he had ever heard. Do you agree?...s

https://archive.org/details/TubesVersusTransistors-IsThereAnAudibleDifference/page/n1/mode/2up