"Lossless" duplicating. Is it for real? Seriously....?

As if there aren’t already enough threads about this subject, here's one more you can laugh at…or put into a serious discussion and thought within your own audio journey.

I have never really read up on why, how and all of the other crap that goes on with a CD and its "lossless" counterpart. Why? Because I trust my ears and I do not rely on technical data, spec's and what other so called reviewers have to say about things. My ears tell me that they are not the same, so why bother reading up on it and learning about it until I literally hear otherwise?

I have on the other hand read about, here and there, certain things about how a CD is compressed and somehow magically brought back to life as "lossless". Here are some observations I have always kept in the back of my head that might just explain not what I am hearing, but maybe why I am hearing what I do when it comes to lossless. Hear me out.

One's are One's and zero's are zero's, right? OK, I can concede to that. However, when you take those one's and zero's and arrange them into groups of 1's and 0's to represent "samples" of music [that may not altogether represent the original sound and locational cues, as you will read on about later] and add that to the fact that original CD's are pressed and not ripped?

You somehow change the overall outcome of the end result as to what hits your ears.

BillD has mentioned this….

Because masters are stamped, whereas CD-Rs are burned and can produce different shapes of the pits in the CD (the burnt CD-Rs being sharper). This can modulate the laser differently in a player

OK, I have to be honest. If 1's are 1's and 0's are 0's, then whether a CD is burnt or stamped shouldn't really make any difference....they are after all 1's and 0's, right? Well, as stated from above, the laser may read the 1's and 0's differently.

That right there may cause why I hear what I hear but folks still seem to want to argue that 1’s are 1’s and 0’s are 0’s and that now, it all boils down to the DAC. Here's where I'm confused. If the laser is reading the information differently between the burnt copy and the original, how does the DAC magically correct the misreading or representation and bring you back a perfect bit-by-bit snapshot?

Never mind, don't answer that. I'm just thinking out loud.

OK. so far we have established that there already is a difference [due to the reading of the laser] between lossless and the original. I can concede to that as well. It only makes sense.

Now, on to my other thoughts on the matter. Let's talk about these "samples" that are turned into these 1's and 0's, shall we?

From what I have always understood, a sample is exactly that. Take, for example a simple chime. You have a chime that is recorded from the drum set that runs R to L on the sound stage. It is located 6' above the floor of the sound stage and located 17' back from the microphone. In analog, you get the full waveform and you should be able to hear it as intended**. In digital, you get samples. How many samples are taken, depend on the format used. We all know that.

Now this chime has to be broken down into samples and translated into 1's and 0's to be recreated, using other samples [or 1 sample from the recording itself] that match as closely as possible the original sample taken. That, from what I understand is what saves space and makes compression possible in the first place.

Who’s to say that the sample is still located 6' above the floor of the sound stage and located 17' back from the microphone? If the sample was taken off of another microphone or some of that sample was picking off reflections of the wall from another microphone location, as opposed to the original, that will smear the image. It may be small and not discernable to some….but to me; it’s audible enough to notice.

Thinking logically about the samples, let's go back to the chime. The sample taken and repeatedly used to represent the first chime sound [out of the 25 individual chimes as they are going off individually]. It may have also been sampled by a different [or many] microphones, other than the one used in a particular passage within a song using only one microphone. Who’s to say that the sample was hit with the same impact and resonance as the sample used either? When a drummer hits the chime(s), they do not always resonate the same every time, nor do they have the same impact. Just a thought, but that may explain why the imaging differs on the ripped CD.

Let's now look at the drum. How about the kick drum? A drummer can hit the drum at varying speeds and forces. This will produce different reverbs and fade out levels until the next kick. It can also produce different Db levels, weight and impact….many times at incredible speed but on the same token, many times at a minutely different tempo or varying force. When you take samples of drum kicks that do not sound exactly the same, your sampling and approximation thereof makes it more as if they were the same. The samples "round out" the original, if you will.

In talking with one of the editors of Stereophile magazine last year, he had relayed to me that more modern bands are not even playing live at a recording session anymore to help alleviate that very thing. They are sat down and told to play a riff 20 or 30 times until the engineers are satisfied that they have one that they can sample over and over throughout the song. Then they move on to the next sample. This makes error correction and recordings sound better. Uh-huh. The reason that they are doing this is due to the fact that they can use less “samples” and that the given samples are more “accurate” than the live recorded samples that are taken and spliced together. While that may be true [for BS recordings and Hip-Hop] , it makes for a rather bland recording to audiophiles because the same drum beat is exactly that. The same fucking drum beat, over and over and over again. No variation, no change in impact. No change in reverb or weight...just the same fuckin' drum beat.

It has been said by Bob Katz concerning compression…

One sure way to destroy the depth in a recording is to compress it too much. Compression brings up the inner voices in musical material. Instruments that were in the back of the ensemble are brought forward and the ambience, depth, width and space are degraded.

Now, if my thought process and the recording studio's thoughts are on the same page then could it even be possible that this is maybe why a lossless copy will not be the exact same thing as the original? Kind of like an LP recording will not be the same as any CD? One uses bits and pieces; the other is the entire waveform. Well, with an original CD that has been sampled even further and then possibly read wrong by a laser because of the edges being sharper, could this all added together possibly be why I’m hearing that lossless isn't really lossless? Food for thought…..

Y’all have a good evening. Love ya’, mean it, bye.

Tom






Note:
**unless the LP has already been compressed and heavily sampled prior to the final stamping.

This is a lot to digest and as you put a lot of thought in to it, I should not address it after having a bottle of beer. So I will do that tomorrow unless we have another depressing day of rain. I like rain, just wish it would all fall at night.

This is an excellent post.
It will not inflame like the one you put up telling people their ribbons are not on your level.

Much on the internet is crap.
Also, last year's web info is OK for analog, but old hat for digital.
If you read things from 2007, they no longer apply.

These days one nut (not you) puts up something on the web, a hundred with no original thoughts of their own repeat it, an then it becomes fact.
That really pisses me off on the garden forums I play on.

We will clear it all up at CarverFest, but this is an excellent post.

It will not inflame like the one you put up telling people their ribbons are not on your level.

...it did not inflame me. Tom has an opinion. It should be respected. On the other hand, others have theirs. I believe that the ribbon speakers are some of the clearest voicing ones I have heard. Myself and others believe they present a unique sound stage. Of course, Tom's Tylers sounded awful good when we threw the 75's at them and used sonic holigraphy. Tom says this distorts the sound stage...myself and others believe it defines it. Who is right? An argument no one will win. I truely believe what my signiture statement says...how can I hear music like a person of 20? Too many rock concerts and too many years. If Tom is right and I am wrong simply because I can't hear what he hears does it make a difference. I say hell no! He's not the one enjoying MY system. That's all that matters...not what Tom says he hears, but what I DO hear!!!

The only way I can respond to that is....whatever sounds good to you? That's all that matters, quite frankly.

To each there own. Each and every one of us have a different listening style, we are all on our own point on on our own audio journey. Does that make me wrong or right in my thoughts and observances? Absolutely not. I'm not a snob. I'm a realist.

The entire point of this thread was to point out where I'm at....at this point in my own audio journey regarding "lossless" formats. Agree or disagree? I don't care. What I do care about is how your own personal audio journey may conflict or agree with my own and your thoughts on the subject.

Nothing more, nothing less.

Hi Tom!!

Not to throw water on your whole line of thought, but the Bob Katz quote is referring to dynamic compression, not bit rate reduction kind of stuff.

If your referring to FLAC playback as decoded by a player versus by a computer versus a redbook playback, remember that the big difference is the need to continuously stream the audio. If it's buffered through memory, then that unit has an advantage. Though it still has to get the timing right (jitter).

From what I've read, the difference between playback of a manufactured disc versus a ripped and reburned disc also seems to come down to jitter and readability of the reflective layer by the laser.

Having aligned the laser on a profession playback CD deck a lot of times and having given thought to the whole readability of that stream of data, I seem to think differently of the whole reading of the pits and how jitter influences the finer detail of the music than some people.

Some past Stereophile articles that explore how digital can at all sound different can be pretty good info on this whole topic!

Mark

Hello Mark! You introduce jitter now to the topic. As if it hasn't been here before [on this forum]...For those who don't know him? Do not knock him as he has forgotten more than most would ever know...

I'm not taking about jitter. I'm talking about the basics of recordings and how it may correspond to the reality of the reproduction of music. How would jitter affect the aspects aforementioned beyond what I may know?

Hey Mark...

Give us your professional opinions on this.

If a redbook CD is taken and made into a raw WAV file at full bit rate, and placed onto a hard drive...
If (using digital cables only) you run the CD and DS (Digital Server) into the same DAC, shouldn't the output to the amplifier be identical ?


I think the term "sample" Tom is using is being used for 2 different meanings... a "bit sample", and a "Hip-Hop audio sample" are not one and the same...

I'm talking about the same sample that's used for the phone, TV's and music. Not just a "hip-hop audio sample", which doesn't exist by the way.

This is wildly speculative as recordings which are that pure are few & far between. Also, room reflections (room being the absolute worst part of most systems) being what they are can change said "air" and pinpoint imaging.

That said, I'll try & find some recordings I think do a good job with imaging/soundstage

Hey Mark...

Give us your professional opinions on this.

If a redbook CD is taken and made into a raw WAV file at full bit rate, and placed onto a hard drive...
If (using digital cables only) you run the CD and DS (Digital Server) into the same DAC, shouldn't the output to the amplifier be identical ?

I get paid for professional opinions.... Personal opinion on the other hand....I can't give away!

That's simply the ago old "a bit is a bit" arguement. If it's a Word Doc then yes. If it's music, then no. It still comes down to the absolute accuracy of the timing of the bits. IE...jitter. Both methods of transport in your example are dependant on the quality of the output of that data and how it arrives thru the connection into the DAC. Attempts at buffering and reclocking in the DAC certainly help but also have some level of sonic signature, it's said. Anyone that's played with PLL's (phase lock loops) can kind of get how that comes into play at a circuit tries to "hold" an incoming signal.

If you go back and read the earliest columns on digital, they were surprised at transports making a difference. They were surprised at SPDIF cables making a difference...etc...etc... And I thought Tom's point was talking about the difference between ripping and reburning the same CD and comparing. There was shock at the first trials that this came to light in.

The format of both SPDIF, and the balanced companion AES/EBU, has the clock run along with the data, which is inherently flawed. That's a big part of the problem of trying to implement a really true transmission line thru the cable and connectors.

This is wildly speculative as recordings which are that pure are few & far between. Also, room reflections (room being the absolute worst part of most systems) being what they are can change said "air" and pinpoint imaging.

That said, I'll try & find some recordings I think do a good job with imaging/soundstage

Gary, try any of the Chesky, anything from Mapleshade, any of the Clarity Records. The Reference Recordings are also well recorded.

I have a Chesky where two gals are singing just a little inside the stereo mic pair image. At about the end of the song, you can hear one of them turn her head while singing.

This is wildly speculative as recordings which are that pure are few & far between. Also, room reflections (room being the absolute worst part of most systems) being what they are can change said "air" and pinpoint imaging.

That said, I'll try & find some recordings I think do a good job with imaging/soundstage

Gary, try any of the Chesky, anything from Mapleshade, any of the Clarity Records. The Reference Recordings are also well recorded.

I have a Chesky where two gals are singing just a little inside the stereo mic pair image. At about the end of the song, you can hear one of them turn her head while singing.

Radioengineer is right on regarding his assesement on Chesky recordings. Bits, Bites and whatever are all over my head. However, I can vouch for the Chesky recordings. They are top notch from a sound quality standpoint. Years ago, (early to mid 90's) Boston Acoustics were promoting their speakers and offering free high resolution recordings produced by Chesky just for the asking. Picked up just about every one that was available. Never heard of many of the artists nor did I care for some of them.

However, if I wanted to demonstrate my system to a friend, I would reach for one of these discs first. Sound incredible. Brought a couple to Carverfest last year to demonstrate on Jame's System. The Sara K and Kenny Rankin tracks sounded very good as some of you will remember, especially once Roy dialed down the bass a tad.

... And I thought Tom's point was talking about the difference between ripping and reburning the same CD and comparing. There was shock at the first trials that this came to light in.

radioeng2,
I'd like to learn more, are these trials documented online? Did they figure out why there is a difference?

I was thinking last night that maybe you could set up a null test between the two analog outputs to display differences.
Steve

... And I thought Tom's point was talking about the difference between ripping and reburning the same CD and comparing. There was shock at the first trials that this came to light in.

radioeng2,
I'd like to learn more, are these trials documented online? Did they figure out why there is a difference?

I was thinking last night that maybe you could set up a null test between the two analog outputs to display differences.
Steve

I think the original discussion was not if an instrument can find a difference, or even if smart guys like BillD can find one with math..

I can't hear any difference, Tom can.
At the end of the day, that is what matters.

But Jesse wants you to prove you can't hear the difference!