SH Spotlight What sounds just like the analog master tape: CD, Vinyl, SACD or a 1:1 analog Reel tape copy?

2 completely different procedures. You can burn a new disk without ever leaving the digital domain. To do your test you would need electronics that are completely neutral and transparent. Since CD players don't all sound the same, it makes sense that output and input stages don't sound the same. If that is the case (and very few would argue that) then it would almost be a miracle IF the copy sounded the same as the original.

 

Of course it's ridiculous that my hypothetical "CD to vinyl and then back to CD" would sound anything like the original CD. That is part of my point when I state that I think it is almost literally impossible that vinyl could offer the most accurate sound reproduction of all the options currently available. The media and equipment alter and degrade the sound. Just the simple act of playing a tape or spinning an LP permanently destroys a part of what you are trying to preserve every single time you play it. You have to worry about temperature, humidity, whether your tape heads are clean or your needle is in decent shape. On the other hand, once you're in digital you can keep spitting out perfect and exact copies of your recording from now until the end of time entirely unburdened by the archaic, costly and unreliable Jules Vernesque machinery of the 20th century. Nothing need ever be lost, though you can, of course, spin down lower resolution copies for enhanced portability. Just as a music fan, I think that I would be much happier knowing that there are hundred or so exact copies of my favorite music from which a CD or better quality file could easily be derived instead of just the one completely irreplaceable and unreproducible "master tape" locked in a warehouse somewhere next to the Ark of the Covenant.

 

This is much like the data truncation that happens when a file is downsampled to 44.1/16 from a higher resolution. Hence the need, as in photography, of dithering to 'soften the edges.'

 

You take a 2x2 real film photo and can scan it at 1000 dpi and looks good and pretty large ... on the contrary a digital photo will bring the ugly squares very soon! A huge Billboard can be made of 35 mm or even 16 mm film but with digital.... one has to take an analog picture of the digital one... to be able to somewhat pull it off! Imo... Yep! digital sux but sure comes handy for non pro work! ..kidding!

 

I think this thread has lost it's way.

 

Well..put it back on track..a
After 192 posts ( great thread) things can veer a little and its up to some to steer it back. If it is needed imo.

 

Why not? Ever consider the possibility that in a few years you might be remastering old classics again in 192/24 lossless for the future of itunes style downloads?

I can see you getting jobs like that within the next decade. And you know what-- I'd be thrilled.

 

[I've tried to read all posts in this thread over the last few days - both my kids had birthdays over the last few days and time for the forum has been short!]

From what I gather, this comparison was not a rigourous scientific test but more a bit of fun for Kevin and Steve since they had the DSD equipment around.

What do Steve's observations tell us? They tell us that the set-up at AcousTech is such that if you send a signal to the cutting lathe then the acetate that is cut will play back with the same tonality as heard while monitoring the master tape. That is, the cutting chain and the playback chain have been set-up in such a way as to match as closely as possible the tonality of the signal going into the lathe.

Equally, the 16/44.1 playback has also been set-up in such a way as to match tonally with the master tape monitor signal.

As for the DSD set-up - the playback tonality doesn't match. Then again, it's not as if AcousTech had the DSD system in the room for any extended period of time to get it set-up perfectly in this instance. With time and effort there are a number of ways to iron this discrepancy out. We know this since Steve has managed to produce SACDs without any tonality discrepancies by applying his "tricks".

So, tonality-wise, this test seems merely to suggest that without careful attention to the set-up then playback tonality will not automatically match input tonailty for a particular recording format. If I brought a different lathe into AcousTech and Steve and Kevin cut an acetate using it that afternoon then I guarantee the tonality would be off.

As for the resolution issue - sure, the acetate and the DSD recording have a head start over basic 16/44.1 record/playback. However, introduce 24/176.4 upsampling on playback and the differences would shrink dramatically.......

 

Steve,
The CD you made to compare-- Was it a standard disc or did you use the HDCD encoder? How much difference would you have heard between the two had you cut two CD's- one encoded and one not?
Thanks for this thread.

Doug

 

Creepy stuff.

 

Okay gotta call you on this one....a LOT of billboards ARE made from digital cameras......most NOT super high resolution, and even as low as 6 megapixel ones are used.....A 12 MP is the ideal and some professional film cameras are STILL used depending on the application, but MOST are digital, because at the distance a driver is from the sign and in a moving car, the actual billboard is NOT a solid PRINT, but its an approximation of a picture, sorta like a newspaper with the dots, its NOT a huge PHOTO, just so ya know...that comparsion isnt the one you wanna use...

They are being looked at from approximately from 50 feet or more when you are actually READING them or scrutinizing the picture in a moving car.... UP CLOSE the film ones look bad also...its just the nature of a billboard...Maybe saying a POSTER compared to a 4x6 print would be the better analogy..! If you look CLOSE at a billboard, the dots making up the "picture" are like the size of a golf ball with empty spaces all around, its not a REAL photo in the sense of a print we are all used to seeing.

 

Glad you did call me on that one ..cause I know its my biz..everything advertising(plus music,film)..but after I read.(and thought more or less same as you)..though...t maybe.no one will notice? LOL... And of course the Poster to 4 x 6 is better analogy...the point did go through though! I hope. PS. Seen some dots like basket balls!

 

I'd believe this if we were talking about a true 24/176.4 recording, not a 44.1/16 source upsampled on playback (even if the sound gets smoothed by it).

 

I think upsampling is underrated Mettoo!

 

Upsampling certainly does help increase the resolution we can hear from a 16/44.1 source.

Of course, it's not adding any new detail - but the upsampled data enables us to hear more of the low level detail that is captured by 16/44.1 but usually masked by standard 16/44.1 D/A conversion methods.

The difference is quite clear on my set-up - reverb trails, for example, are there in all their glory after upsampling while they tail off earlier without.

The overall effect of upsampling is a more accurate reproduction of the original analogue recording. I prefer upsampled 16/44.1 to vinyl

 

This is almost like saying, "There is something about digital that I truly like, but I can't put my finger on it."

 

Personally I feel that digital better reflects the emptiness in my cold, dead heart.

 

Mal,

By definition, upsampling is just guessing at the extra points on the curve. That is never as good as original sampling at the recording.

 

Any process used to record and reproduce sound introduces some kind of change ("distortion") to the original signal.

The real irony is that the analogue disc cutting process, which requires a big EQ adjustment (per the RIAA curve) to allow a lathe to put it on to disc and the opposite EQ adjustment required in playback, is said to provide a "more natural" sounding method of reproduction than a straight to 16/44.1 with no EQ.

Quantitatively, the digital process does chop up the original signal - there is a loss of information, and the introduction of a "different" kind of "distortion." However, it doesn't bring with it the rumble, hiss, pops and scratches, or the tracking distortion of a worn cutter tip or playback stylus, or the wow & flutter introduced by the analogue disc process.

I wonder --- if digital technology came first before the introduction of analogue, would most of our observations be stated in reverse? Is it that we who have been around long enough to grow up on the analogue mediums are just used to those kind of distortions, and therefore see them as sounding more "natural?"

 

I have heard some great acetates over the years. A friend has a set of acetates cut from the Sound of Music. They were cut at 33 1/3, but play back only with a standard needle. we were stunned by the sound.

One of my prize possessions is a one of a kind acetate of a neighbor playing in a string quartet. He's playing a jazzy arrangement of Lady Be Good. The acetate sounded horrible upon the first playback. The second (and last) time I played it, I adjusted the eq settings. with the correct eq, the recording came alive. I transferred it to a CDR and have listened to the disk ever since. It is a great performance.

 

Man, you can tell people really take these observations personally. People need to realize the differences were extremely SUBTLE and that to the untrained ear most people couldn't tell the differences anyhow. To feel that you're superior because you listen to analog or inferior because you listen to digital is just silly.

 

The point isn't that upsamling will add information - it's that it enables retrieval of more information captured in the original 16/44.1 PCM encoding.

For this reason, I only endorse integer upsampling - the more complex upsampling routes actually take us further from the originally encoded analogue wave to my ears. For example - going from 16/44.1 to 24/192 improves the sound in some ways (cleaner top end, more definition in the bass, etc..) but you also loose a little harmonic detail. In contrast, upsampling from 16/44.1 to 24/176.4 gives improvement across the board - including better retrieval of subtle elements of the encoded signal such as harmonics, transients and reverb trails.

At least, these are my findings.

 

Funny!


The experiment as I understand it suggests that Redbook is more or less acceptable for tonality, but hurting for resolution, something 192/24 or 96/24 might vastly improve. Even though no DVD-A experiment was done, based on the results of the Redbook and DSD tests, a reasonable guess could be made that if 192/24 or 96/24 had been included in the test, it would of beaten the CD and also fared very well and maybe outperformed the DSD.

It's a shame that a dual-layer DVD-A/CD was unable to be developed successfully and the best that could be done was the non-Redbook conforming double-sided Dualdisc. (Though I myself have had no problems with my Dualdiscs).

Speaking of Dualdisc, maybe it's more of a shame that Sony further hurt the creditability of the format by only providing a low-resolution version on the stereo DVD-A side (48/16 is it?). It makes me wonder if Sony got into making Dualdiscs just to stick more stakes in the DVD-A format's heart, for the sake of their own SACD. Whatever their intentions, they didn't do the audiophile community any favors with their Dualdiscs.

 

Not according to Nyquist - according to the sampling theorem everything below the half-sampling frequency (ie. 22.05kHz for CD) will be fully captured. It is extremely misleading to say that "the digital process does chop up the original signal".

Of course, implementation of the theory in the real world is not perfect. However, the problems associated with the implementation of PCM encoding cannot be summarised as "loss of infomation" since this concept is based upon a fallacy.

 

I know exactly why I like it - it sounds GREAT!!