How can Vinyl still sound good compared to digital?

[Jud]

4 hours ago, sandyk said:

 Hi Jud

 Are you able to provide information about the actual CDs that are poorly mastered to John Dyson who may have copies of them in his personal collection, and possibly uncover and correct the reasons for their very poor quality.

As a very recent example, John was able to correct with his S/W , a SuperTramp -Crime of the Century" CD where 2 people participating in John's project found that his corrected version then easily outperformed their own copies on Vinyl.

 Kind Regards

Alex

Hi Alex - Tommy and Gaucho, which I mentioned in my post, are 24/96 DVD-A. (There are somewhat better DSD versions, but still not quite as good as the vinyl for me.)

 

A CD I can think of that is badly mastered and doesn’t sound as good as the vinyl to me, is Miles Davis’ Amandla.

 

I can’t think of others offhand. For me the Giles Martin Beatles hi res, for example, sounds better than the vinyl, as do the Steve Wilson Tull remasters and the Steve Hoffman and Mark Linett Beach Boys remasters.

[John Dyson]

3 minutes ago, Jud said:

Hi Alex - Tommy and Gaucho, which I mentioned in my post, are 24/96 DVD-A. (There are somewhat better DSD versions, but still not quite as good as the vinyl for me.)

 

A CD I can think of that is badly mastered and doesn’t sound as good as the vinyl to me, is Miles Davis’ Amandla.

 

I can’t think of others offhand. For me the Giles Martin Beatles hi res, for example, sounds better than the vinyl, as do the Steve Wilson Tull remasters and the Steve Hoffman and Mark Linett Beach Boys remasters.

I have a copy of the Nat King Cole remasters that APPEAR to be DolbyA!?!?!?  I vacilate on that from time to time, but usually when trying to decode material that is NOT encoded -- it isn't a pretty result.

There is a surprising amount of material that loses its last 10% of quality because it is 'carefully mastered 'DolbyA'.  If material starts off REALLY GOOD, then the DolbyA version sounds 'okay', but isn't as good as it can be.

 

John

 

[sandyk]

50 minutes ago, Jud said:

A CD I can think of that is badly mastered and doesn’t sound as good as the vinyl to me, is Miles Davis’ Amandla.

 Hi Jud

 It sure does sound dull and lifeless, although there is no apparent clipping. The 24/192 sounds better, but there is something a little suspect looking with the spectrum analysis .

 

 Kind Regards

Alex

[PeterSt]

5 hours ago, sandyk said:

 As many others have already noted , a " needle drop" digital copy of a Vinyl recording can sound virtually identical to the original Vinyl recording, so it is clearly a case of poor CD Mastering techniques.

 

No. It is clearly a case of how difficult it is to get digital right. You just can't randomly run into a sequence of LPs which sound better than their digital counterparts. However, you *will* run into showrooms with poor systems in them for what ... 19 out of 20 times ? easily. And this shows at the reproduction of digital.

 

Btw, your (yes your, plus Frank's but alas) laptop takes part in those 19 rooms. I know you won't except that because of your PSU's etc., but I too have PSU's etc. ... in a desktop. What ? a workstation. No wait, a plain server. It really really seriously matters, and not a little bit.

 

Crime of The Century remains poor at it. Unless you can discover some punch. And if you can it is all over wrong because it just isn't in there. Again, it is a poor example for this thread and I can count those on one hand. What's sad is that it is such a great album.

And the vinyl version s*cks even more, because it is ... vinyl. Mastering is OK now, but unlistenable for me because it sounds like vinyl (they all sound the same as a grey-ish mash, like mono and without any dynamics at all, uninteresting, dead, super boring after listening to two or so albums).

 

PS: What John did to the current version tends to go towards "punch" but with a large hole somewhere (so it is counter productive because it audibly shows inconsistency). This is quite similar to (finally) having the highs OK (says me), BUT having the S'es underwhelmed. Something with transients at the larger scale which are picked up by the (de)coding. Larger scale transients work out (cymbals, also ride -), smaller scale remain untouched (S'es) and this makes it inconsistent again.

(just added this in an attempt to show that I really listen and not to criticise - I think I'd prefer this over all the versions I have)

 

[sandyk]

8 minutes ago, PeterSt said:

(just added this in an attempt to show that I really listen and not to criticise - I think I'd prefer this over all the versions I have)

 Hi Peter

 With this one, I wasn't a SuperTramp fan, so I just told John that I thought what he had done to correct it was an improvement and asked a couple of my friends who both have this album on Vinyl for comments, and they both agreed that it sounded way better than their copies on vinyl

 

Regards

Alex

[fas42]

2 hours ago, STC said:

 

 I do not deny when you compare a good vinyl and digital playback via loudspeakers, you sense the vinyl sound to be fat and easier on the ears. Subjectively, you can prefer either one and asking audiophiles for such opinions usually tainted with prejudice.

 

How it works is usually that the analogue is too "fat", and the digital is too "lean" - 😜. As one works on sorting out each medium, eliminating the distortions that tend to drag along with each format, the first slims, and the second expands - until they meet in the middle - and are, just right, both of them, 🙂.

 

Of course this won't please those who enjoy their styles of makeup - those who just like to hear the recording, however, may be happy ...

[fas42]

17 minutes ago, PeterSt said:

 

 

And the vinyl version s*cks even more, because it is ... vinyl. Mastering is OK now, but unlistenable for me because it sounds like vinyl (they all sound the same as a grey-ish mash, like mono and without any dynamics at all, uninteresting, dead, super boring after listening to two or so albums).

 

 

 

Have to put the same amount of work into vinyl, as for digital, to get the good stuff. Yesterday's listen of Led Zep I, on original, ordinary vinyl, delivered about 95% of the content - very impressive!

[John Dyson]

35 minutes ago, PeterSt said:

 

No. It is clearly a case of how difficult it is to get digital right. You just can't randomly run into a sequence of LPs which sound better than their digital counterparts. However, you *will* run into showrooms with poor systems in them for what ... 19 out of 20 times ? easily. And this shows at the reproduction of digital.

 

Btw, your (yes your, plus Frank's but alas) laptop takes part in those 19 rooms. I know you won't except that because of your PSU's etc., but I too have PSU's etc. ... in a desktop. What ? a workstation. No wait, a plain server. It really really seriously matters, and not a little bit.

 

Crime of The Century remains poor at it. Unless you can discover some punch. And if you can it is all over wrong because it just isn't in there. Again, it is a poor example for this thread and I can count those on one hand. What's sad is that it is such a great album.

And the vinyl version s*cks even more, because it is ... vinyl. Mastering is OK now, but unlistenable for me because it sounds like vinyl (they all sound the same as a grey-ish mash, like mono and without any dynamics at all, uninteresting, dead, super boring after listening to two or so albums).

 

PS: What John did to the current version tends to go towards "punch" but with a large hole somewhere (so it is counter productive because it audibly shows inconsistency). This is quite similar to (finally) having the highs OK (says me), BUT having the S'es underwhelmed. Something with transients at the larger scale which are picked up by the (de)coding. Larger scale transients work out (cymbals, also ride -), smaller scale remain untouched (S'es) and this makes it inconsistent again.

(just added this in an attempt to show that I really listen and not to criticise - I think I'd prefer this over all the versions I have)

 

Note that I believe that Crime of the Century and the traditional enjoyment of it comes from an original eccentric mastering of the disk a long time ago.  It might have been done on purpose, but the material DOES decode to sound more normal.  (I have an approximation to it on the site -- under differentEQ.)  It does not deliver the expected sound, but shows a major difference that just a bit of EQ before sending to the decoder.   Doing EQ before the decoder does NOT result just in an EQ difference, but really changes the dynamics.

 

I really don't know WHY there is such an appeal for Supertramp, esp Crime of the Century, but I remember liking it back in my audiophile days some 30yrs ago.  This recent decoding effort was partially as looking back on my history, developing some interest from some moderately high profile audio people.  Plus, it has been a fun challenge!!!

 

John

 

[fas42]

Don't have Crime of The Century ... had a quick listen of a couple of versions of Bloody Well Right on YouTube - what's the problem? Should be a brilliant track to listen to - does the sparkle, and tone, of the electric piano disturb people, perhaps?

[tmtomh]

Digital is indisputably the more accurate format - in that sense it has higher fidelity to the source than vinyl does. That does not, however, mean that digital always sounds as good or better than vinyl to many people (personally I tend to prefer digital).

 

As for why vinyl sounds so good to so many people, I believe there are many possible reasons, However, as time as gone on and I have read and listened more, I have become convinced that a major factor - perhaps the major factor - is inter-channel crosstalk produced by phono cartridges. This crosstalk:

• Varies by frequency, cartridge, and LP/cutting/pressing;
• Generally is greater in the midrange and treble; and
• Tends to be out of phase with the primary L and R channel content.

The result is a variable, mechanically/electrically induced bit of midrange and high-end "air"/ambience and "sweetness."

 

Combined with mechanical resonances in the vinyl playback chain that often produce a bit of bass or mid-bass warmth (aka added harmonics), this IMHO goes a long way towards explaining the "organic," "natural" sound signature that many attribute to vinyl playback.

 

Of course the technical limitations of vinyl also sometimes produced sonic differences between LPs and CDs that can make LPs sound better. Levels and compression cannot be pushed as hard on vinyl, so as noted by many already, CDs often feature harsher, more extreme compression and peak-limiting than LPs. Many LP cuttings also have to attenuate the treble slightly, which can produce a less harsh, or a mellower, sound that many prefer and associate with "natural" aka "real" sound. And many LP cutting masters have to have the L-R channel separation reduced, or even summed to mono, in the bass frequencies, which increases DR Meter ratings and can potentially increase the perception of dynamics.

[STC]

1 hour ago, tmtomh said:

Digital is indisputably the more accurate format - in that sense it has higher fidelity to the source than vinyl does. That does not, however, mean that digital always sounds as good or better than vinyl to many people (personally I tend to prefer digital).

 

As for why vinyl sounds so good to so many people, I believe there are many possible reasons, However, as time as gone on and I have read and listened more, I have become convinced that a major factor - perhaps the major factor - is inter-channel crosstalk produced by phono cartridges. This crosstalk:

• Varies by frequency, cartridge, and LP/cutting/pressing;
• Generally is greater in the midrange and treble; and
• Tends to be out of phase with the primary L and R channel content.

The result is a variable, mechanically/electrically induced bit of midrange and high-end "air"/ambience and "sweetness."

 

Combined with mechanical resonances in the vinyl playback chain that often produce a bit of bass or mid-bass warmth (aka added harmonics), this IMHO goes a long way towards explaining the "organic," "natural" sound signature that many attribute to vinyl playback.

 

Of course the technical limitations of vinyl also sometimes produced sonic differences between LPs and CDs that can make LPs sound better. Levels and compression cannot be pushed as hard on vinyl, so as noted by many already, CDs often feature harsher, more extreme compression and peak-limiting than LPs. Many LP cuttings also have to attenuate the treble slightly, which can produce a less harsh, or a mellower, sound that many prefer and associate with "natural" aka "real" sound. And many LP cutting masters have to have the L-R channel separation reduced, or even summed to mono, in the bass frequencies, which increases DR Meter ratings and can potentially increase the perception of dynamics.

In short the sound heard by us need to be too pristine to be perceived as natural?

[John Dyson]

This whole thing of digital vs. vinyl isn't an either/or.  When you look at all of the variables, unless it is bad vinyl -- the difference in recordings grossly overshadows the difference in format.  Which medium is better -- digital can be by far, way way way better.   Of course 12bit digital at 32k samples isn't as good as 16bit at 48k, but at a certain point -- the bit depth and sample rates hit a point of diminishing returns -- but even 44.1k/14bit bit depth gives any vinyl a run for its money - ignoring CD4 or something like that.  All vinyl has rumble, all vinyl will have more noise than properly dithered 14bit digital, and perhaps even 12-13bit digital, but vinyl is often 'good enough', and the difference in mastering can far overshadow the format &or medium.

 

As someone who wouldn't touch a turntable unless being used to rip for experimentation purposes, I do not think that people who use turntables are using the wrong format.

 

It is so unfortunate that there is a disagreement or confusion when there needn't be.  A lot of this confusion happens because of historically terrible mastering* on digital material -- and I have discussed some of the issues over and over again.  Terrible mastering still happens for digital formats today -- but it might be getting better.  The details of the mismastering matter is not the basic point being made here.

 

I have been 'fooled' by ripped vinyl until actually measuring it or hearing ticks/pops, but the quality was 'pretty good'.  In fact, give me pristine vinyl anyday over poorly mastered digital (if I didn't have the pseudo-remastering tools that I do have & have created.)

 

It isn't that 'vinyl' is better or worse than digital -- the best question to ask: where do you get the best copy of a recording?

 

*  Note that the problem isn't mastering per-se, but how the recording is handled after mastering when being sent to duplication/distirbution.

 

John

 

[STC]

Maybe it isn’t about the medium at all. It is about the sound that reaches your ears. I will never go back to vinyl because I believe that digital is far more accurate to be utilized for perfect sound reproduction. 

Unlike digital, there is no one accepted standard for vinyl reproduction. Unlike digital, vinyl is pressed with EQ and again the playback is also used with EQ. Unless one is using the same RIAA for the pressing/cutting and the playback, I do not think you can have flat response as intended in the original recordings. 
 

Then we have subjective preference where some ( and generally virgin audiophile wannabe) preferring the vinyl sound. 
 

For seasoned audiophiles, the digital version that they think most accurately represent the sound as natural are those recorded with stereo microphones far away from the source. In a way, the sound of direct and the ambiance mixture. IOW, the not so pristine sound. 
 

Have you seen a pure sine wave from a vinyl compared to one from the digital medium?  
 

@tmtomh pointed out about the crosstalk. I think what he meant is channel separation. Vinyl does not have good channel separation like the digital medium. Technically this can be considered as a flaw but at the same time it can also be a blessing in disguise because the sound produced by a system with poor channel separation will be more like mono. It will feel fat and large. And you still get the soundstage but only for those really loud at either side which causes less stress to the brain trying to decode the location of the phantom image. It sound more natural for localization. 
 

It is not the accuracy but the correctness of sound perceived of vinyl playback is what makes them still desirable. IMO. 
 

Guys, surely there must be someone here listening to vinyl playback via headphones. Your input is much appreciated.

[semente]

2 hours ago, STC said:

In short the sound heard by us need to be too pristine to be perceived as natural?

 

Yes for me.

 

Distortion draws my attention away from the illusion of instruments playing music and towards the reality of a system playing recorded music.

[semente]

For me, the best vinyl playback sounds more like CD than vinyl. But even the best vinyl players I've listened to fell short.

[STC]

Distortion got many meanings. If you don’t like distortion in any form then most likely you have throw away most of your album because they can have some sorer of distortion one way or another. To me even EQing is distorting the original sound. 
 

 

[semente]

38 minutes ago, STC said:

Distortion got many meanings. If you don’t like distortion in any form then most likely you have throw away most of your album because they can have some sorer of distortion one way or another. To me even EQing is distorting the original sound. 
 

 

 

Distortion in reproduction: a hi-fi system is not producing music, it's not an instrument that makes music, a hi-fi system reproduces recorded music.

I let the guitar player and/or the producer how much distortion the amplifier should be producing at a particular moment of a particular music. I'm perfectly happy with distortion used creatively in music production.

Most of the music I listen to is acoustic anyway, so the lesser the distortion the more real the reproduction.

 

Many people confuse the purpose of a hi-fi system with that of an instrument.

 

 

 

That guy has an incredibly small foot.

[semente]

49 minutes ago, STC said:

To me even EQing is distorting the original sound.

 

To me it depends.

 

EQ'ing can be used to achieve a more accurate reproduction of the recorded signal, by using a notch filter to flatten the response of a mid-woofer at the top of it's passband, a waveguide to control dispersion or to attenuate modal resonance peaks.

 

It can also be used to achive a more accurate representation of the absolute sound, by flattening a response of a microphone or to compensate from an overly distant or overly close mic position.

 

One can argue that EQ'ing is not completely transparent, that it will affect the purity of the signal and have an impact on realism, but in my opinion for most cases the advantages of using EQ will outweigh the disadvantages.

[STC]

35 minutes ago, semente said:

Many people confuse the purpose of a hi-fi system with that of an instrument.

Yes but yet they are still confused. Unless you listen in an anechoic chamber or with headphones all sound that reaches your ears are “distorted”. They no longer represent the same sound that was in the recording. 
 

Guys, any LP headphones listeners here?

[semente]

26 minutes ago, STC said:

Yes but yet they are still confused. Unless you listen in an anechoic chamber or with headphones all sound that reaches your ears are “distorted”. They no longer represent the same sound that was in the recording. 
 

Guys, any LP headphones listeners here?

 

I would replace Unless with Even "if you listen in an anechoic chamber or with headphones all sound that reaches your ears is “distorted”".

 

As I said, EQ'ing can be used as a means to achieve a more accurate reproduction of the recorded signal at the listening spot.

It is an improved representation of "the same sound that was in the recording" than what you'd get if not using EQ.

 

On the other hand, an equipment which adds effect (i.e. something which can be perceived as increased reverb) is no longer representing "the same sound that was in the recording".

 

 

The notch filter example:

 

The beginning of the bass crossover. Red = response with no components.

Green = 0.82 mH coil inserted. Blue = R2021 and C2021 added. As can be seen this creates quite a bump at 1 kHz.

All as expected. To smooth the 600-1500 Hz range a parallel notch filter was added = yellow graph.

 

http://www.troelsgravesen.dk/DiscoveryW18_crossovers.htm

[STC]

1 hour ago, semente said:

On the other hand, an equipment which adds effect (i.e. something which can be perceived as increased reverb) is no longer representing "the same sound that was in the recording".

Without the graph and the explanation you have no idea whether a recording got an intentional or natural boast at those frequencies. 
 

I always believe never never use any kind of EQ or manipulation to the original sound. Having said that I also thing it is not wrong for people with poor room to use the bass or treble or even a little DSP. I personally do not use them. 

[Jud]

11 hours ago, STC said:

 

I am afraid I have to disagree with you on this one. I can't speak for all the LPs out there but I have done comparisons with Tracy Chapman on Clearaudio Statement TT vs Boulder and Esoteric digital. all of them driven by Krell Master Reference monoblocks. In a recent AV show I have also compared Norah Jones LP and digital (but I have to say it was a challenge to distinguish them).

 

I know you like digital with a lot of ambient information. Are you saying this sounds less good than vinyl to you?

[semente]

9 minutes ago, STC said:

Without the graph and the explanation you have no idea whether a recording got an intentional or natural boast at those frequencies.

 

You keep confusing the recording/signal with its reproduction.

 

If the system is to reproduce the recording accurately then the system response using pink noise must be flat.

A flat system will accurately reproduce an "intentional or natural boast at those frequencies" that's what is in the recording.

 

If Gravesen had not added the notch filter to his speakers then the speakers would add a "boast" on top of the recordings' "boast" if there was one, or just a "boast" on top of a good recording, or make a recording that needed the "boast" sound better.

 

Weren't you using Harbeths at some point? They use notch filters to flatten the frequency response.

 

9 minutes ago, STC said:

I always believe never never use any kind of EQ or manipulation to the original sound. Having said that I also thing it is not wrong for people with poor room to use the bass or treble or even a little DSP. I personally do not use them. 

 

A very slight room EQ adjustment below 100Hz (simulation) - attenuated peaks and perceived reduction of decay at those frequencies:

 

 

[STC]

26 minutes ago, Jud said:

 

I know you like digital with a lot of ambient information. Are you saying this sounds less good than vinyl to you?

That is not true. I like music to sound like you are hearing them in the best acoustic and my room’s natural RT is 0.28s. I brought up a little later because it was too dead. 
 

The most detailed solo vocal sound so clean in such room. You don’t like ambiance. You choose the right ambiance. Otherwise, concert halls shouldn’t exist. 
 

I wonder why Audiophile’s got no problem with diffusers, absorbers and all kind of room tuning device but when I use real active ambiance where I can control the perfect ambiance that no audiophile could even dream off, it is becoming a joke. 

[STC]

20 minutes ago, semente said:

 

You keep confusing the recording/signal with its reproduction.

 

If the system is to reproduce the recording accurately then the system response using pink noise must be flat.

A flat system will accurately reproduce an "intentional or natural boast at those frequencies" that's what is in the recording.

 

If Gravesen had not added the notch filter to his speakers then the speakers would add a "boast" on top of the recordings' "boast" if there was one, or just a "boast" on top of a good recording, or make a recording that needed the "boast" sound better.

 

Weren't you using Harbeths at some point? They use notch filters to flatten the frequency response.

 

 

A very slight room EQ adjustment below 100Hz (simulation) - attenuated peaks and perceived reduction of decay at those frequencies:

 

 

I am just telling that your example of the notch wouldn’t be known to exist without the graph. And you would not even hear if it’s adding reverbs or not without the explanation provided by the designer.