[ethebull]

Interesting. I’m surprised that you would allow that any CD recording process could yield perceptible flaws. Your comment indicates you’re aware of differences in sound cards and D to A and A to D converters then, Yes? Why would this be? Aren’t they all perfect in practice? Or so beyond human perception as to be irrelevant?

Sure, Sony wants to sell players, Cds, recording and mixing EQ, etc. in support of SACD, though it has basically been a bust in terms of market penetration and success - more to do with the “who cares” attitude of the listening consumer than it’s irrelevance. Your view that it has no impact on the quality of the playback experience in one’s home confirms my lack of interest or respect for your opinions.

I’m sure you’re heart is broken. (not)

 

[bigshot]

I usually judge the quality of an argument by how it's supported, not whether I already agree with it.

The fact is, I've done the legwork to know what I'm talking about when I speak of the differences (or lack thereof) between SACD and redbook. I spent a month doing comparisons between the formats and kept finding that the differences I was detecting was due to different mixes and mastering, not the quality of the format itself. I finally found a label that only produced hybrid SACD with DSD recordings. I went to great pains to set up a direct A/B comparison between the redbook layer and the SACD layer of one of their recordings, and on my own equipment and the equipment of a professional sound mixer, I couldn't detect any difference whatsoever. If there is a difference between formats, it is small enough to be totally insignificant.

As to digital conversion error...

The conversion process from analogue sound to digital is the most important stage in terms of technical sound quality. I've used several myself, from USB capture dongles on up to a 24 bit ProTools workstation. There is a level of quality that you reach where for normal home listening purposes, you don't need any better. My Macintosh is perfectly capable of capturing the sound of an LP and putting it back out with no perceptible loss of quality.

When it comes to converting from digital back to analogue, it's much less of a problem. The mass produced DACs in typical CD players and DAPs are perfectly capable of a high degree of fidelity. There are other things that affect the sound MUCH more. Chief among these is sloppy mixing and mastering, or poor creative choices by the sound engineers.

See ya
Steve

 

[ethebull]

I agree with everything you've stated above except the bit about SACD DSD performance.

I read in another thread about your professional sound experience and I take your judgments at face value. My experience was that SACD had great potential, but it was never realized.

I have 12 years in audio retail, fairly high end, mostly in store management. When Sony introduced SACD, the store I managed was chosen by them to host the kick off release of the $5k flagship player for the WDC metro area. They insisted on a full Sony set-up and broke down my hi-end room to install the overly bright stuff they made. I bit my lip when asked about the player and all. The good news was the player stayed after they left, and I really enjoyed using it in our top shelf set-up for the next couple years. (B&W 801's, Levinson pre&amp) The Sony player significantly improved on the Levinson CD player it replaced, both for SACD and regular discs.

In the long term, I personally never bought into SACD as a music collector. As I said before, I don't usually buy re-releases, and there's only been a handful of new music released on the format that would be of interest to me.

In the end, we both find SACD irrelevant, but for different reasons.

 

[bigshot]

The disk I tested was Pentatone's Stravinsky chamber works by Jaarvi. The reason I think it it's a good one to test with is because Pentatone only puts out SACD hybrids. I would bet that more than 80% of their sales are to classical music buyers who don't even own an SACD player. That makes them completely unmotivated to hobble the redbook layer.

It is extremely difficult to do a direct A/B comparison of the layers of a hybrid SACD. My player has a six or eight second gap as it switches layers, and the level of the SACD layer is always hotter than the redbook. I had to make a CD-R of the redbook layer and rack that up in a separate CD player with balanced line level. I doubt many people go to all that trouble.

See ya
Steve

 

[memepool]

Quote:

Originally Posted by Tarkovsky /img/forum/go_quote.gif It's hard to claim doing this accurately reproduce the mix yes, but if people like the sound of valves and high end roll I don't see the argument against using them when recording. Next you'll be complaining they are using these electric guitars instead of a respectable tangelwood!

you miss my point. I'm not saying there is anything intrinsically wrong with listening to valves, although personally they're not my cup of tea, rather that adding valve input or output stages to digital capture or playback is being done because so many people are unhappy with the "digital" sound of the upper frequencies.
If everything was rosy in the garden there shouldn't be a "digital" sound to the upper frequencies and this widesperad perception is mainly down to flaws in redbook which are now associated forever associated with digital reproduction, whether erroneously or not.

 

[tfarney]

Quote:

Digital and analogue are both capable of reproducing sound with a high degree of fidelity. The differences between the two is the difference between the way they artifact. Good sound is a result of good recording and engineering, not whether it's recorded on film, tape or digital. By the way, when you talk about "unlovely top end" and "chrome plated harshness" you aren't describing sound. You're applying a value judgement to a conclusion based on your observation. What exactly is inaccurate about the reproduction of the high end in digital (ie: 3kHz - 10kHz) that might be causing that? Distortion? Frequency imbalance? What specific aspect of sound relates to the descriptive words you are using and what makes you think that that particular aspect is the cause?

What he said. Analog can be lovely. Digital can be lovely. Both can be ugly. This age-old argument is about people for whom preference is not enough, they must be right to be secure. And none of this - digital vs analog, cd vs vinyl, even high-end vs good midfi, has a fraction of the impact on the listening experience as the quality of the recording and mastering does. And I find the substitution of poetic adjectives in the absence of supporting data to be pretty funny, though neither "unlovely top end" nor "chrome plated harshness" are my favorite. That honor would go to "musical."

Music is musical. Our equipment is hardware that endeavors, imperfectly, to reproduce it.

Tim

 

[nick_charles]

Quote:

Originally Posted by tfarney /img/forum/go_quote.gif This age-old argument is about people for whom preference is not enough, they must be right to be secure.

Good observation

 

[nick_charles]

oops

 

[eweitzman]

I've never seen a cartridge with channel separation spec'd at better than 30db. At best, you'll always have more than 0.1% channel mixing. Add in alignment issues and mechanical transmission and the crosstalk gets worse from there.

I attended a talk by Keith Johnson a few weeks ago. He suggested that the spacious sound of LPs was due to phase anomalies caused by the pinch effect inducing vertical stylus motion.

- Eric

 

[tfarney]

Quote:

Originally Posted by eweitzman /img/forum/go_quote.gif I've never seen a cartridge with channel separation spec'd at better than 30db. At best, you'll always have more than 0.1% channel mixing. Add in alignment issues and mechanical transmission and the crosstalk gets worse from there. I attended a talk by Keith Johnson a few weeks ago. He suggested that the spacious sound of LPs was due to phase anomalies caused by the pinch effect inducing vertical stylus motion. - Eric

There is a big discussion going on over on the Steve Hoffman forums (hi steve) about this, that supposes that surface noise, crosstalk and all manner of vinyl anomalies may contribute to the space illusion of vinyl. I don't think it is too far-fetched. When I switch on the crossfeed circuit of my Headroom amp, the field expands. When I drop the needle onto the most pristine vinyl, I get that characteristic "woosh" of surface sound which sounds, to me, a lot like the ambient noise of a big empty room. The most chauvinistic vinyl devotees are not likely to accept, however, that such artifacts and "errors" are creating their preference. Only some mysterious, as-yet-unmeasured superiority will do.

Tim

 

[eweitzman]

The idea of random phase anomalies creating the spaciousness was put into my head by someone on some forum years and years ago, the joelist maybe. I was glad to hear Keith attribute it to a definite mechanism at this talk. And it was just a tangential comment in response to a question, totally off topic!

I think this could be tested. A spherical stylus would be subjected to more vertical motion by the pinch effect than stylii with other profiles, so it should sound more spacious. Other things being equal of course.

- Eric

 

[bigshot]

The best way to see how vinyl artifacting affects sound is to look at 78s. They have everything that vinyl has, except more. I've done a lot of study and experimentation with surface noise, and I've found that the biggest effect is the fact that the music is in phase while the low level noise has no phase at all, being completely different in both channels. With mono recordings, this creates a soundstage in the middle with a wide spread of surface noise. Summing the channels cancels out a lot of the out of phase noise, making the signal stand out clearer.

With stereo LPs, you can't sum the channels obviously. That means that the no phase noise can't be cancelled out. You always have a certain amount of totally different sound in each channel. Some folks might interpret that as sounding like a "wider soundstage", but it isn't. It's just more out of phase noise.

When you apply crossfeed, the first thing to get affected is the totally out of phase noise. That's why a light level of crossfeed makes vinyl sound better.

But ultimately, all these things are fixes for error. It's always preferable to not have to fix errors if you don't have to.

A slightly larger spherical stylus will increase the contact with the groove wall over a smaller elliptical stylus. This results in more surface noise, but at a lower level and evener texture overall. It's much easier for digital noise reduction to smooth out the noise from a spherical stylus than with an elliptical. I stumbled across this concept by experimenting with various stylus shapes and sizes, but I've never heard anyone else talk about it.

See ya
Steve

 

[nick_charles]

Quote:

Originally Posted by bigshot /img/forum/go_quote.gif A slightly larger spherical stylus will increase the contact with the groove wall over a smaller elliptical stylus. This results in more surface noise, but at a lower level and evener texture overall. It's much easier for digital noise reduction to smooth out the noise from a spherical stylus than with an elliptical. I stumbled across this concept by experimenting with various stylus shapes and sizes, but I've never heard anyone else talk about it. See ya Steve

How do these stylus shapes relate to the shape of the cutting lathe ?, I always naively assumed that the best stylus would be the same shape and thus be able to track the cut groove better ?

 

[bigshot]

The idea isn't that the needle fits the groove precisely. That would cause a tremendous amount of friction. The stylus tip rides partway up the groove wall, never touching the bottom. Imagine a ball bearing riding along in a V shaped groove.

A spherical tip is round and an elliptical is roughly wedge shaped. The wedge makes a more focused contact with the groove, while the spherical makes a more generalized contact.

Generally, one would use elliptical for transcription work and spherical for everyday listening. An elliptical stylus is harder to keep in align and puts more wear and tear on the records, especially if it is a little out of align. The advantage of the elliptical is that it has a slightly better high end frequency response and by varying the size, one can choose the part of the groove wall to make contact with, avoiding wear patterns. I find that spherical works better for transcription work when you plan to do digital noise reduction.

See ya
Steve

 

[eweitzman]

Quote:

Originally Posted by bigshot /img/forum/go_quote.gif The best way to see how vinyl artifacting affects sound is to look at 78s. They have everything that vinyl has, except more. ... A slightly larger spherical stylus will increase the contact with the groove wall over a smaller elliptical stylus. This results in more surface noise, but at a lower level and evener texture overall. It's much easier for digital noise reduction to smooth out the noise from a spherical stylus than with an elliptical. I stumbled across this concept by experimenting with various stylus shapes and sizes, but I've never heard anyone else talk about it.

Steve,

I assume you're talking about 78 playback with a stereo cartridge, probably not with a round cactus needle? I'll have to think about what you say some more.

My point about the spherical stylus relates to vertical motion, not noise. Keith Johnson's point was that the pinch effect causes vertical motion, which causes opposite amounts of phase change in each channel.

The frequency of the phase change occurs at twice the frequency of the signal. Here's an example with purely made up numbers for the phase change. I have no idea what is realistic -- maybe someone can work out the numbers from typical cutter/stylus geometry?

If playing back a 100Hz sine wave, the pinch effect causes the stylus to rise and fall twice per cycle, or at 200Hz. Let's say the rise/fall causes the phase of the left channel to be advanced by one degree and the right channel is dumb by one degree. So we have the phase of the 100Hz signal changing 200 times per second by a total two degree difference between the channels each time. If this was slow enough and big enough to be perceptible, we'd hear the sound moving back and forth from speaker to speaker a slight amount. But it's too fast for that, so it just results in a little image blurring or widening.

Just to be clear about the pinch effect (this is a headphone forum, after all

) : Because of the cutter's shape, the groove cut at the peaks of the 100Hz wave is much wider than at the zero crossing. Play this back with a stylus shape that is close to the cutter's shape and there will be little vertical motion. But if you use a spherical stylus, the narrower parts of the groove at the zero crossings will lift (or pinch) the stylus upwards, and it will fall back further down into the grooves at the lateral extremes where the groove is widest. So there will be more vertical motion and therefore more phase distortion with a round stylus. With a round stylus, there might be two degrees per channel, or four degrees overall, compared to the first example. The phase will still be changing at 200Hz for the 100Hz signal.

Now to get the whole picture, the groove width changes with the deflection of the cutter in response to the complex signal, not just this pure sine wave. The amount of deflection (there's a more common word for this but it escapes me) will determine the vertical motion, but it is the result of the complex composite signal, so each frequency that's present will get a different phase offset. The whole thing seems to get pretty random, but there may be correlations that might have other effects. Is the phase offset like group delay, keeping the fundamental/harmonic relationships intact, or does it change those relationships?

- Eric