Analog: Still Better?

[John Dyson]

4 hours ago, Rexp said:

Supertramp, you know it will sound good on vinyl or tape. 

 

Oh man, you would have to show a 'Supertramp' album.  It has been a bad habit for me to use their recordings for some of my testing purposes.   Looks like another few hours to look at Supertramp again!!!   (I mostly use 'Breakfast', 'Crime', 'Quiet' and 'Crisis' for testing.)    It really IS habit forming to try to make sure that something works super well, then knowing that there is a problem -- fixing it, then trying again -- ad-infinitum.   Some of the supertramp recordings are a special challenge.

 

I did just listen, and ignoring the delay distortions and other problems because of micing the system -- it SOUNDS like that the recording just MIGHT be a pure one, and not the FA that I hate so much.

 

[John Dyson]

1 hour ago, fas42 said:

Pretty easy  ... the CD version is miles in front - this is the one that presents a much greater sense of being where the microphones were; the acoustic around the trumpet is so much better defined. And when the full band comes in, on the LP, the vinyl distortion is just too obvious.

I agree, your analysis seems correct, and correctly not distracted  by the the apparently 'wider bandwidth' of the LP version.

 

1) The horizontal bands above 20khz are signals leaking into the analog before digitization.  Could be on the original LP or could be noise encroaching locally.   Such bands are often manifest on 'high res' versions of recordings also.   Some DSD stuff has massive random noise at higher freqs also.

2)  The apparent varying signal above 20kHz are most likely artifacts from traditional analog noise reduction used on older recordings.   Much (definitely not all) of the excursions above about 15kHz are noise modulation and even modulation distortions depending on the NR system used.   There are boutique recordings that do have substantial true HF content, but not usually pop rock.   Even a lot of classical/jazz have little true HF content above 15-16kHz.

 

It is very possible that there is real HF (at 15kHz or above) signal along with the modulation distortion or vice versa.   In fact, the modulation distortion and noise modulation actually happens also at lower frequencies, but the signal strength is usually so pronounced that the negative effects are hidden.

 

[John Dyson]

1 hour ago, hopkins said:

Some noise is introduced by the recording of the vinly - the cable going from my preamp to the Tascam recorder acts as an antenna and picks up RF. I need to get a cable with better shielding. Obviously there are also some clicks and pops, but it is not so bad. 

 

What strikes me is the missing bandwidth on the vinyl rip - that's clear from the spectrum analysis, and you can hear it as well. As a result, the piano is "brighter" on the CD version (listen to the first few seconds of the track), and the trumpet has a wider range. I'll listen to the album again on my turntable to check that it is not the recording process that is to blame. The mix seems slightly different as well on the two versions. Otherwise, I would not say that the recorded vinyl version has significant "distortion".

 

I usually only purchase vinyl when the content is unavailable in digital format. I bought this album out of curiosity, to compare it to the CD version. I want to digitize some of the albums I have, in order to "document" them in my music collection app, and be able to listen to them on my computer. While the quality can certainly be improved, this first attempt does not seem to be all that bad. 

 

I'll prepare another recording of the Vegh Quartet (Beethoven string quartets) as I have the original LPs from the 1972-74 recordings and a CD version to compare. 

 

 

 

 

I listened to the CD and LP versions.   They somehow appear to be a slightly different mix.  If the recording was three track, then very possibly could have been.  However, they also appear to be the same session.

 

The differences that you hear are true, and both versions have unfortunately been compressed, so the dynamism of the original recording is weakened (but not totally lost.)

Please be patient with my request for you (and others) to listen to the following analog to your CD version....   The version that I present might be both counfounding, confusing,

prettier, uglier and very different.   However, more than likely, it sounds pretty close to an original session.  YMMV though.

 

I bothered to do this because the original of your example seemed very close to an unadulterated, natural quality recording.

 

https://www.dropbox.com/s/8nokrt1oeut84lx/01 CD-FADECODED.wav?dl=0

 

(Here is a 24bit version for those who sometimes have trouble with FP .wav files):

https://www.dropbox.com/s/6b9sp0qqd0l064k/01 CD-FADECODED-24bit.wav?dl=0

 

 

[John Dyson]

9 minutes ago, hopkins said:

The example does illustrate the difficulty in comparing different releases. 

 

Keep in mind that the recording "equipment" used here is really basic (the Tascam costs < 100$).

 

Concerning the bandwidth, I'll record the CD version with my Tascam recorder. This should be interesting, to me at least, to see if the spectrum looks better.

 

@John Dyson your file does not fix the bandwidth issue, but the dynamic range may be wider? It sounds different (quick listen on my PC, not my speakers). 

 

 

 

Bandwidth and dynamics are intimately intertwined.   This is especially true because APPARENT bandwidth is not the same as measured with a simple spectral analysis.

The decode of the CD will actually appear to DECREASE the bandwidth if you look at averages, even though the peaks will likely be greater.

 

The recording sounded like it has 'promise as being of good quality at its base', and it really does.   The matter of the vinyl having inferior quality in several ways might be interesting, but there isn't much to worry about beyond that fact.   The CD by itself is NOT a good representation of the original recording (instruments don't even sound real AT ALL), and the LP is even worse.   Sadly, the recording is really GOOD.

 

The reason for the investigation is that I was hoping that the LP had not been compressed like the CD was.   Unfortunately, the bandwidth issue is out of your control as it is an artifact of the recording and the way that it was processed, just like the compression applied to both versions -- the version used to create the CD and the version used to create the LP.

 

There are some older LPs that are 'virgin' recordings, alas -- that LP is not one of them.   Myself, I find it interesting and a challenge to find recordings that sound like they should as they are being mixed.   I had hopes for the LP until I analyzed it.  ( My abiility to hear frequency response balance issues is poor -- but distortions are obvious to me, so I use other techniques to determine quality.)

 

If the LP was a 'pure', unadulterated recording, it would definitely seem to have less bandwidth by some measures, but more by others.

My investigation shows that the difference is so trivial as to be almost unimportant -- the LP is simply a different mastering of the same (probably 3trk) tape, and sadly of poor quality by todays standards.

 

 

 

 

[John Dyson]

1 hour ago, fas42 said:

 

Yes. Unfortunately the noise originates in many areas, which is why there is the craziness of the power cord mattering. And, until the very last source of significant noise is tamed then there is every chance, in fact it can almost be guaranteed, that a digital playback chain will be disturbing, or irritating, or boring to listen to, for a large number of recordings.

 

Satisfying digital is magical stuff. But we are still some way from companies producing components which are intrinsically not susceptible to the noise factors - until that time, DIY and tweaking are the best value for money solutions for extracting the true potential of playback from digital source.

All I can say is that I have an anecodte of listening to a really good copy of ABBA, then I found it to be a rip.  No matter what, the noise IS worse than my equivalent digital copies, even though ripped at high res.

 

On the other hand, many original recordings from the 60's and 70's had hiss 'added' in the 1980s by the 'digital sound' processing.   I can get rid of the hiss very nicely, then sounds more simlar to the original releases.  Frustratingly, rips of such ancient material is seldom found.   Mostly, all we have now is the current mediocracy 🙂of the music distributors.

 

The same distortions (stereo image wobbling around, twisted sibilance and cymbals impact where the HF peak is moved to the incorrect temporal position.)   This is all 'distortion', ubiquitious and  is manifest on both vinyl and digital copies nowadays.

 

If you want the 'good stuff', most of it is on vinyl from before the 1980s before the 'digital sound' was added to CDs, and a few years later to vinyl itself.

 

For evidence, listen to the hiss on Carpenters recordings or some of the ABBA stuff -- the hiss is MUCH worse than tape from the late 1960s -- it is just that our standards are low, and EXPECT the hiss.  The same thing that produces hiss also adds dynamics modulation distortion to the rest of the recording.

 

So, from the standpoint of damaged mastering, nowadays vinyl and digital are similar.

My own experience is that pristine vinyl isn't too different from digital, but has a bit more distorition, surface noise and sometimes a bit of rumble.   I am not accomodated to vinyl, so I hear the distortions pretty rapidly.  Also, if my headphones didn't go almost all the way down to DC, I probably wouldn't hear the rumble.   Many headphones and speakers pretty much rolloff around 40-50Hz, but if your speakers do respond very low, and the AMP/Preamp is transparent, you can see the speaker cones wobble in and out a little.

 

 

[John Dyson]

15 hours ago, fas42 said:

The electrical noise that's insidious, John, doesn't result in any sort of audible hiss, IME - what it does, when present, is suck the life and vitality out of the presentation of the music; what you get is a somewhat dead, grey version of what say an LP rendering would be like ... sound familiar, to anyone, ?

 

What's amusing is that I have just experienced this, only 10 minutes ago: current rig lost the DIY isolation transformer, from internal shorting most likely; part of that tweak is still working - and I'm using that. Which seemed to be doing most of the lifting - yet, just before I wasn't happy, and as an experiment unplugged a not used cord to a lamp, in that circuit ... ah, hah!! That was the villain - plugging and unplugging it clearly demonstrated that the length of cable was acting as an antenna, introducing enough noise on that circuit to do damage to the SQ. What this means, is that the now defunct transformers are needed; or, some other alternative that provides the necessary filtering.

 

The bit of noise that you don't realise is there, can make a night and day difference to the listening; this is the trap in digital reproduction, not getting on top of the causes.

The compression on most recordings can result in magnified hiss.  Just listen to the beginning of ABBA SOS on most original CD versions -- that hiss is NOT natural, but the tape hiss as amplified by the normally used DRC on most CDs.   Often the hiss is not noticeable on more recent recordings, simply because the tape hiss might be 10dB less than older recordings.   The level of hiss on SOS is on the order of a tape cassette.

 

Many other older recordings have the magnified hiss problem, along with other issues such as smushed transients even in more recent recordings.  For example, the strke of piano strings are often diffused , additionally other portions of the percussion transient being modifed in time.  That transient is sometimes made more audible because  it no longer happens simultaneously with the rest of the piano sounding.*   Many people don't notice it because the total energy is almost the same.   Perhaps it is most audible with a closely miced piano where the location of the sounds still somewhat matches the keys, but is also spatially spread more wide than should be.   Also, sometimes the effect is slightly obscured by room (auditorium) effects, but still it is mostly  noticeable.   (The piano temporal distortion is only one artifact.  Cymbals and other transients are sometimes also smushed.)

 

* Paraphrasing the comment in 'Amadeus' -- the listener needs the 'tada' at the end to notify that it is the end.  Similary, the time distortion of the key sound sometimes preceeds the actual keypress, thereby informing the listener that there is going to be a keypress :-).

 

Earlier on, the vinyl didn't have this problem, but as the digital techniques were moved to Vinyl, then that medium also ended up having the same troubles.

This all being said, given good digital technology and *excellent* vinyl technology, the results are similar.   Vinyl degrades, but digital does not -- therefore, for me, digital wins hands down.,

 

I can give hiss examples fairly easily, but damaged image and smushed sound really need to be carefully done.   I am notoriously bad and examples, but the hiss is pretty easy to do.