[bigshot]

I know this thread has gone to the dogs, but come on, guys.

Do you have a dog?

 

[pinnahertz]

Do you have a dog?

I love them all. I'd want to devote to them all the time they deserve, can't do that right now, so no I don't.

There certainly must be a dog forum, though.

 

[analogsurviver]

The facts presented in the above do not support the conclusion. You clearly presented that vinyl playback is quite limited in channel separation, even implying that greater channel separation is desirable, and worth spending extra to get. But then this: "Even if we assume channel separation of "only" 35 dB @ 1kHz and deteriorating to say no worse than 20 dB @ 20 kHz, any phono cartridge that exceeds the 20 kHz by another 20 or so kHz ( provided this extended response is not riddled with peaks and walleys, but is smooth ), it will trounce any CD ever made for width and localization.

So, you're implying that width and localization are not related to greater separation? So either greater separation is better, or it's not, it can't be both. Which is it?

Yes, they are related. Greater separation is better. But after a certain point, mere better channel separation , beginning as low as approx -30 dB, becomes next to meaningless if the rest of the spec of whatever analog or digital system does not support frequency response up to very least 40 kHz as far as imaging is concerned. In plain words: a properly set up phono cartidge with 30 dB channel separation over the range of at least 100Hz to 10 kHz while being reasonably flat out to at least 40 kHz will image better than anything RBCD - regardless of measured channel separation for the digital device with brick filtering just above 20 kHz.

The you conclude with this, "So, the answer whether analog record can outstrip CD for the sense of space ( width of stereo image, but also depth ) is undoubtedly YES - if and when all the required conditions are met. And, unfortunately, at a price." That implies that a record can improve over the CD with greater sense of space, width, depth, etc, following with the implication that you have to pay up to get that. With your statement earlier that the cart with the greatest separation was extremely costly, are we to understand that greater separation is worth paying for and that it improves image width and depth? And that the best of the best carts in that regard, though much worse than any digital system in terms of separation, are still better at achieving effects that depend on greater separation?

Please see the above answer . In phono cartridges with extended response at least to 40 kHz (preferably MUCH more ), channel separation exceeding 30 or so dB does not get one better localization - but something exactly opposite. It is hard to believe at first, yet true - the thing that gets improved the most by having better channel separation than about 30 dB is - the reproduction of an instrument or voice dead center in the stereo image. Why ? Channel separation can also be regarded as a form of non linear distortion - and is perceived as such. -20dB is 10 % distortion, -30 dB is 3%, -40 dB is 1% and -60 dB is 0.1% distortion.

Again, which is it? It can't be both.

For the third time - localization is not dependent solely on channel separation - but channel separation in conjunction with overall frequency response. It is the balance of the two that has to be optimized for the best performance in localization. Phono cartridges, being rather imperfect analog devices, offer the whole diapason of this balance combinations - and after spending entire adult lifetime with them I can safely conclude that.

And would you agree or not that anything that changes the original signal (and reducing channel separation would be included) is introducing a form of distortion? And, if that's true, would then simply adding interchannel crosstalk to any digital recording would improve width and localization. I would say no to all of that, but you made the claim, care to back it up?

Yes, I agree that anything that changes the original signal is introducing a form of distortion. The pinch effect in phono styli that occurs towards inner grooves on the analog record whenever the curvature of the scanning radius of the stylus exceeds the radius of the recorded signal, resulting in unwanted vertical movement of the stylus when there should be none, thus giving false depth perception, should be avoided at all costs. Better stylus tip geometries have reached the position to correctly read the information in the groove to approx 40 kHz even in the innermost grooves at 33 1/3 RPM - 35 or so years ago, But, yes, claiming that analog record has better localization than digital when played back by a conical stylus in the innermost grooves of the record is false. It can be pleasing, but it is not true to the original.

I hope that above answers your questions.

 

[old tech]

Yes, I agree that anything that changes the original signal is introducing a form of distortion. The pinch effect in phono styli that occurs towards inner grooves on the analog record whenever the curvature of the scanning radius of the stylus exceeds the radius of the recorded signal, resulting in unwanted vertical movement of the stylus when there should be none, thus giving false depth perception, should be avoided at all costs. Better stylus tip geometries have reached the position to correctly read the information in the groove to approx 40 kHz even in the innermost grooves at 33 1/3 RPM - 35 or so years ago, But, yes, claiming that analog record has better localization than digital when played back by a conical stylus in the innermost grooves of the record is false. It can be pleasing, but it is not true to the original.

I hope that above answers your questions.

I presume an optical turntable would be best for vinyl playback with regard to the above and with cross-talk?

I had one of those T/Ts many years ago and it was a PITA as I spent more time cleaning records than playing them, such was its sensitivity to dust. I ended up selling it a great loss. What I liked about it though, was the amount of information it could retrieve from worn or scratched records compared with a stylus.

 

[71 dB]

So, the answer whether analog record can outstrip CD for the sense of space ( width of stereo image, but also depth ) is undoubtedly YES - if and when all the required conditions are met. And, unfortunately, at a price.

CD is very transparent. If you copy a vinyl on CD, one would think the sound is "identical" (meaning you can't hear the difference) to the source vinyl and so is the imaging. I have burned dozens of vinyls on CD and they sound as if they are vinyls. You can store vinyl sound on a CD, but you can't store "CD sound" on vinyl.

 

[71 dB]

Yes, I posted it in post #11 of this thread, "Vinyl doesn't have better imaging in and of itself. Any perception of that comes from either specific mastering choices or the extremely powerful expectation bias coupled with the owning, handling and playing of vinyl."

Oh, yes. Thank you. I agree with this partly. There's so much going on with vinyl. Vinyls aren't completely flat discs, so the stylus travels over very flat "mountains" which modulates at least vertical L-R signal. Who says the gain of L+R and L-R aren't frequency dependent? Maybe L+R has greater gain at 200 Hz, but smaller at 1000 kHz? Separation is frequency dependent. Then there is RIAA - correction, which affects all the noise and distortion present in the playback. The noise floor is pretty "brown", which sounds more pleasing than pink or white noise. The sound image of vinyl "fluctuates" a little like acoustics fluctuates in a concert hall due to air flows. Vinyl imaging can't be pin point sharp, because L+R and L-R signals don't "match" exactly. This blurring probably contributes the sensation of "wider" imaging?

 

[pinnahertz]

You can store vinyl sound on a CD, but you can't store "CD sound" on vinyl.

Absolutely true, but you'd be surprised at how close you can come. I've actually done that comparison, cut lacquer and pressed vinyl from the same master that was sent to the CD plant. The only clearly audible vinyl signature is the noise floor (which is masked by louder signals) and slightly higher distortion on some peaks. At least for the first playing, then it's all downhill from there. The test did not include a widely separated, though, so separation reduction wouldn't be obvious. Testing was done on speakers, not headphones.

 

[pinnahertz]

Oh, yes. Thank you. I agree with this partly. There's so much going on with vinyl. Vinyls aren't completely flat discs, so the stylus travels over very flat "mountains" which modulates at least vertical L-R signal. Who says the gain of L+R and L-R aren't frequency dependent? Maybe L+R has greater gain at 200 Hz, but smaller at 1000 kHz?

Ok, I'll say it: L+R and L-R gains are not frequency dependant, at least not at the cutter head and stylus. Theoretically they should be identical because they are both cut with the same exact mechanism with precision matched L and R channels. L+R and L-R are not handled independently. Those signals don't exist that way in the lathe. They're the result of cutting L and R. The crosstalk issue is mostly confined to the cartridge with physical alignment properties both in the card, and with the cart and arm. There's a bit of capacitive crosstalk in there too.

Separation is frequency dependent. Then there is RIAA - correction, which affects all the noise and distortion present in the playback.

RIAA EQ (it's not a correction) is complimentary, though. The play curve reverses the and has little to do with crosstalk.

The noise floor is pretty "brown", which sounds more pleasing than pink or white noise. The sound image of vinyl "fluctuates" a little like acoustics fluctuates in a concert hall due to air flows. Vinyl imaging can't be pin point sharp, because L+R and L-R signals don't "match" exactly. This blurring probably contributes the sensation of "wider" imaging?

Didn't we already get past this? They have to match very, very closely just to get the separation they have! It doesn't change dynamically, except with frequency. Vinyl images are pin point sharp, just not as extreme. But if any of that were not true, and vinyl worked as you say, how on earth would you get wider imaging by narrowing separation?

I'll answer that too: YOU DON'T. Vinyl does not image wider!!!! If you think you hear that you're hearing the difference in mastering between CD and vinyl, which is HUGE, clearly audible, and unrelated to vinyl as a medium. Or you're experiencing expectation bias.

 

[pinnahertz]

Let's look at it another way. The lateral groove component has more headroom than the vertical. If you hold down L-R relative to L+R, even on a peak only basis, you're reducing separation radically. Continue to reduce it by reducing L-R and you eventually get to mono. How's that for a wide image?

 

[71 dB]

Ok, I'll say it: L+R and L-R gains are not frequency dependant, at least not at the cutter head and stylus.

Of course they aren't supposed to be frequency dependent, but in practice they are more or less. You can't have analog circuits and mechanical cutter heads operating at 100 % accuracy. 99.9 % perhaps but not 100.00 %

Theoretically they should be identical because they are both cut with the same exact mechanism with precision matched L and R channels. L+R and L-R are not handled independently. Those signals don't exist that way in the lathe. They're the result of cutting L and R. The crosstalk issue is mostly confined to the cartridge with physical alignment properties both in the card, and with the cart and arm. There's a bit of capacitive crosstalk in there too.

Of course only signals L and R are fed to the cutter head making the head move at perpendicular 45° angles. If we observe movements as combinations of vertical and lateral vectors, we end up the result, that those movements are represented by L+R and L-R witch are real movements in rotated coordinate system. In this rotated coordinate system L+R the L-R we have directions, which are maximally different from each other mechanically, a major point of mine in this debate.

RIAA EQ (it's not a correction) is complimentary, though. The play curve reverses the and has little to do with crosstalk.

Not a correction? Without it the frequency response is completely wrong, so RIAA de-emphasis is a correction, an EQ to correct the frequency response. Bass is emphasized compared to higher frequencies, so any mechanical/electric inaccuracy is emphasized at low frequencies as well. That's why the background noise is brown-like (personally I call it red noise).

Didn't we already get past this? They have to match very, very closely just to get the separation they have! It doesn't change dynamically, except with frequency. Vinyl images are pin point sharp, just not as extreme. But if any of that were not true, and vinyl worked as you say, how on earth would you get wider imaging by narrowing separation?

I'll answer that too: YOU DON'T. Vinyl does not image wider!!!! If you think you hear that you're hearing the difference in mastering between CD and vinyl, which is HUGE, clearly audible, and unrelated to vinyl as a medium. Or you're experiencing expectation bias.

What is wide imaging? It is not synonym to large separation. Play music on left channel only. The separation is "infinite", but the imaging isn't wide. It's at your left ear. Flat and stupid. Listen to a binaural-like recording and it has width (and depth), because the spatial cues are strong. However, the separation isn't that large on binaural recordings. A few decibels at low frequencies and no more than 25-30 dB at high frequencies.

It seems that vinyl causes "fake" spatial effects due to the electromechanical difference of lateral and vertical movements of stylus and these artifacts create the illusion of wider imaging. Separation is compromised, but the width of imaging isn't the same thing.

Let's look at it another way. The lateral groove component has more headroom than the vertical. If you hold down L-R relative to L+R, even on a peak only basis, you're reducing separation radically. Continue to reduce it by reducing L-R and you eventually get to mono. How's that for a wide image?

After a critical separation point the width reduces and the sound becomes more mono, but before that critical point reducing separation can actually increase the illusion of width.

 

[pinnahertz]

Of course they aren't supposed to be frequency dependent, but in practice they are more or less. You can't have analog circuits and mechanical cutter heads operating at 100 % accuracy. 99.9 % perhaps but not 100.00 %

Your numbers are off.

Of course only signals L and R are fed to the cutter head making the head move at perpendicular 45° angles. If we observe movements as combinations of vertical and lateral vectors, we end up the result, that those movements are represented by L+R and L-R witch are real movements in rotated coordinate system. In this rotated coordinate system L+R the L-R we have directions, which are maximally different from each other mechanically, a major point of mine in this debate.

Yup.

Not a correction? Without it the frequency response is completely wrong, so RIAA de-emphasis is a correction, an EQ to correct the frequency response. Bass is emphasized compared to higher frequencies, so any mechanical/electric inaccuracy is emphasized at low frequencies as well. That's why the background noise is brown-like (personally I call it red noise).

RIAA eq is applied in cutting, the inverse is applied in playback. They are fully complimentary. It's not a correction of anything, it's fully complimentary. The noise spectrum of a record varies with vinyl formulation, a lot. There's no need to invent a new term for it, though.

What is wide imaging? It is not synonym to large separation. Play music on left channel only. The separation is "infinite", but the imaging isn't wide. It's at your left ear. Flat and stupid. Listen to a binaural-like recording and it has width (and depth), because the spatial cues are strong. However, the separation isn't that large on binaural recordings. A few decibels at low frequencies and no more than 25-30 dB at high frequencies.

You've mingled separation with external processes that affect the resulting spacial effects. From a standpoint of the recording and reproducting system, the widest it can possibly get is one channel driven, the other silent.

It seems that vinyl causes "fake" spatial effects due to the electromechanical difference of lateral and vertical movements of stylus and these artifacts create the illusion of wider imaging. Separation is compromised, but the width of imaging isn't the same thing.

I don't know why you cling to this. Vinyl DOES NOT produce fake spatial effects! Not at all. In fact, if anything, they are reduced. If you've arrived at the conclusion you have it can only be by comparison of a digital version and a vinyl version of the same recording. But you're not comparing vinyl and digits! You're comparing two completely different production chains with many, many opportunities for changes in...well, pretty much anything. Your error, and I can't seem to emphasize this enough, is attributing a percieved spatial effect to vinyl itself, while completely ignoring the rather huge and audibly significant decisions made in the pre-lathe process.

After a critical separation point the width reduces and the sound becomes more mono, but before that critical point reducing separation can actually increase the illusion of width.

Nonsense. That isn't supported by any part of spatial hearing. So, give me an example. Prove it.

 

[71 dB]

Your numbers are off.

How much?

RIAA eq is applied in cutting, the inverse is applied in playback. They are fully complimentary. It's not a correction of anything, it's fully complimentary. The noise spectrum of a record varies with vinyl formulation, a lot. There's no need to invent a new term for it, though.

You keep talking to me as if I was an idiot. You want to be THE audio guru on this forum and therefore you try to downplay anyone who knows something. I don't know everything. My knowledge is limited, but so is yours. A learned person recognizes the limitations of his/her knowledge and understanding. It's good if you can correct mistakes by other people, but you can do that respectfully and without downplaying others.

You've mingled separation with external processes that affect the resulting spacial effects. From a standpoint of the recording and reproducting system, the widest it can possibly get is one channel driven, the other silent.

To me (channel) separation is the technical objective measure (e.g. 44 dB) while wideness is a subjective psychoacoustic measure for how wide the sound image appears. Large separation doesn't automatically mean wide sound. feeding the signal to contralateral ear with proper filtering and delay(s) increases the sensation of wideness while decreases separation, because the spatial cues tell the brain that the sounds come from a distance of you in a room instead of being at your ears which happens when separation is very large. How someone with your knowledge and understanding doesn't seem to get this is beyond me and makes me think you are just trying to downplay me.

I don't know why you cling to this. Vinyl DOES NOT produce fake spatial effects!

Oh boy. I feel like trying to explain this to someone with an IQ of 60. I have said several times that vinyl is a electromechanical system that creates distortion differently in lateral and vertical directions meaning a monophonic centered sound will be distorted differently from a side-panned sound meaning "fake spatial effects." People seem to experience vinyl sound wider than CD and something has to explain it. My take on it is these fake spatial effects plus reduction of separation when necessary explains it. You on the other hand keep insisting that large separation equals large wideness which would mean CDs sound wider because they have superior separation. CD is extremely transparent and exposes the crap that was put on it as it is while vinyl creates spatial effects and what not, an illusion that the crap on it is better than it actually is. Vinyl colors all music similarly and some people like this "similarity". Everything sounds "warm" the same way, while CDs sound whatever was put on them from warm to cold. Add the misunderstandings people have about digital audio and it's "inferiority to analog sound" and you have many reasons why some people love vinyls so much.

Not at all. In fact, if anything, they are reduced. If you've arrived at the conclusion you have it can only be by comparison of a digital version and a vinyl version of the same recording. But you're not comparing vinyl and digits! You're comparing two completely different production chains with many, many opportunities for changes in...well, pretty much anything. Your error, and I can't seem to emphasize this enough, is attributing a percieved spatial effect to vinyl itself, while completely ignoring the rather huge and audibly significant decisions made in the pre-lathe process.

Well, for some reasons the production chain is different for vinyl and CD and for the listener it doesn't matter why or in which point of the production chain these differences arouse. What matters is that the differences exist.

Nonsense. That isn't supported by any part of spatial hearing. So, give me an example. Prove it.

See above what I wrote about this issue. If you play left loudspeaker only, the sound is localized to that loudspeaker. On headphones you have annoying sound at your left ear that appears to be very lose. Wide means far in the side and that's not what you get on either cases. Add right channel with spatial cues and now the brain decodes how the sound is originated somewhere in the space, perhaps 2 meters behind your loudspeaker or 1 m away from your head making the sound appear wider ( + deeper).

Look at HRTFs measured from the side of a head. Is there huge separation? No. At low frequencies a few decibels and even at high frequencies not more than 25-30 dB.

 

[ev13wt]

...
So, the answer whether analog record can outstrip CD for the sense of space ( width of stereo image, but also depth ) is undoubtedly YES - if and when all the required conditions are met. And, unfortunately, at a price.[/SIZE]

I think this is a cute tidbit, no?


Vinyl is the inferior format. How is this even discussable? More so in a science forum.


The thing is, vinyl is certainly "good enough".
It also has one thing going for it. Its organic. It has so many tiny "problems" that vary from minute to minute, day to day.
Its interactive. One tap with your finger you hear in the speakers and yer hooked forever!

Vinyl is great. Digital is great. Tape is great.

I record digital playlists to tape. Why? I know it sounds worse. Technics RS-1500.

Watching the reels spin while the music is flowing... The worlds best audio experience for me.


Heres a bad video of it.

 

[pinnahertz]

You keep talking to me as if I was an idiot.

Interesting. I'm not getting that. If that's what you're hearing, it's not my intention. On the other hand, please don't make stuff up either.

You want to be THE audio guru on this forum and therefore you try to downplay anyone who knows something.

Not true. At least, not from my side. I have never once stated that as a desire, nor do I claim to know "everything". On the other hand, nearly 5 decades in pro audio...I may have picked up a fact or two. Even so, I learn something new every day, and I'm glad of it.

I don't know everything. My knowledge is limited, but so is yours.

All true. I'm willing to up my limits and try to do so.

A learned person recognizes the limitations of his/her knowledge and understanding.

True, and I agree, though I detect that as more of a jab than just stating fact. And as such, it may not have landed well here.

It's good if you can correct mistakes by other people, but you can do that respectfully and without downplaying others.

I think you're taking this all very, very seriously. On the other hand, if you can't take correction, then research before you post. I will always be one to separate fallacy from fact, wherever possible.

To me (channel) separation is the technical objective measure (e.g. 44 dB) while wideness is a subjective psychoacoustic measure for how wide the sound image appears. Large separation doesn't automatically mean wide sound. feeding the signal to contralateral ear with proper filtering and delay(s) increases the sensation of wideness while decreases separation, because the spatial cues tell the brain that the sounds come from a distance of you in a room instead of being at your ears which happens when separation is very large. How someone with your knowledge and understanding doesn't seem to get this is beyond me and makes me think you are just trying to downplay me.

Well, I agree with the opinions above, but that's not what crosstalk does.

Oh, and think whatever you like. My goal is truth in audio.

Oh boy. I feel like trying to explain this to someone with an IQ of 60.

Hmmm...Downplaying, 71? Kind of uncalled for and juvenile, don't you think? I don't think I've ever made an evaluation like that of anyone.

I have said several times that vinyl is a electromechanical system that creates distortion differently in lateral and vertical directions meaning a monophonic centered sound will be distorted differently from a side-panned sound meaning "fake spatial effects."

And I disagree, but we can come back to that in a bit.

People seem to experience vinyl sound wider than CD and something has to explain it. My take on it is these fake spatial effects plus reduction of separation when necessary explains it. You on the other hand keep insisting that large separation equals large wideness which would mean CDs sound wider because they have superior separation.

Yes, reduction in separation alone does not widen the presentation at all. That, by the way, is provable.

CD is extremely transparent and exposes the **** that was put on it as it is while vinyl creates spatial effects and what not, an illusion that the **** on it is better than it actually is. Vinyl colors all music similarly and some people like this "similarity". Everything sounds "warm" the same way, while CDs sound whatever was put on them from warm to cold. Add the misunderstandings people have about digital audio and it's "inferiority to analog sound" and you have many reasons why some people love vinyls so much.

Yes, I get there are misunderstandings about this. I don't agree with most of the reasons. The thing is, I've actually tested this. The project was to be released on CD, but after many long discussions with the producer we decided to do a vinyl release as well. I recognized the opportunity to study the "vinyl sound" under controlled conditions, and did so.

I don't expect you to accept my statements of my experience and test results because you haven't heard the evidence. We might have arranged that if I didn't think this entire debate so futile. I have a CD and a vinyl record that I personally supervised throughout each and every step of production, including cutting the lacquer, for which I was present and in charge. My comparison of the CD and the vinyl was carefully calibrated and controlled. My results showed there is no such thing as a "vinyl signature" or "vinyl sound", other than increased surface noise and mistracking distortion of certain high level signals. The stereo perspective was identical, the spectral distribution or "color" was identical. But I did two things that hardly anyone else has ever done: cut the lacquer intentionally to match the CD without changes, and then compared the results knowing both chains were precisely under control and identical. But I do understand that's all hearsay to you, and you wouldn't accept it out of possibly disrespect.

To be clear, I don't care. I proved it to myself and several others. Vinyl is shockingly transparent, does not have a specific sound quality, and doesn't change the width of the (and I hate this term!) "soundstage". What people are hearing is all a result of mastering. All of it.

Now, I still have all of those materials of course, and could get them to you to compare for yourself. But frankly, if I shared them, I'd expect you to take shots somehow at the material, the mix, doubt that I didn't spin the test somehow, who knows what. And that's not worth my time. A had an open mind and wanted answers, that's all. And then we'd probably volley back and forth about how many lacquer masters each of us has ever cut, how many cartridges have we each run full tests on, and on...and on.. and on...

There's really no point to that p***ing match, is there?

Well, for some reasons the production chain is different for vinyl and CD and for the listener it doesn't matter why or in which point of the production chain these differences arouse. What matters is that the differences exist.

Yes, in most cases the production chain is radically different. Even different people are involved, and they make different judgments and decisions. That's reality, if the listener cares or not.

See above what I wrote about this issue. If you play left loudspeaker only, the sound is localized to that loudspeaker. On headphones you have annoying sound at your left ear that appears to be very lose. Wide means far in the side and that's not what you get on either cases. Add right channel with spatial cues and now the brain decodes how the sound is originated somewhere in the space, perhaps 2 meters behind your loudspeaker or 1 m away from your head making the sound appear wider ( + deeper).

Problems: 1. You're using your own opinion as fact, using words like "annoying" and the phrase "very loose". Opinion doesn't fly here in Sound Science. Do you have research that suggests that a population segment agrees with your opinion? 2. Interchannel crosstalk doesn't add spatial cues at all. Crosstalk from one channel to the other is simply crosstalk, in phase, no ITD, no HRTF. That can't possibly widen the image.

Look at HRTFs measured from the side of a head. Is there huge separation? No. At low frequencies a few decibels and even at high frequencies not more than 25-30 dB.

Yes, but spatial hearing is based on those differences, which include ITD (you don't mention that), and a very specific diffraction curve (you allude to but not specifically). You can reduce separation by mixing L into R and vice-versa, and you'll never expand the image width unless you add some form of HRTF or acoustic crosstalk cancellation. And interchannel crosstalk on vinyl doesn't do that.

Now, circling back around, you've been hammering on the difference in distortion of the vertical/L-R component in a groove. Here's an opportunity for both of us to learn. Why don't you post your reference on this? Something that clearly defines and quantifies the difference in linearity between a lateral and vertical component of a stereo record groove. Then I'll share whatever I can come up with, and we might learn if that's really an issue or not, and to what degree. From there we might figure out if the crosstalk signal that may result can in fact widen the image.

Would that be productive?

 

[bigshot]

My advice is to always strive to be nice and helpful. You don't want people to think you're argumentative. If you get to the point where you think you aren't being heard, pouring out a lot more words won't help. Better to answer with fewer more carefully chosen words.