phn said:
Let's just say "it depends"
on
1) the album
2) the surface condtion (whether LP, tape or digital disk)
3) my mood
4) the conjunction of the planets
Any of these could cause me to say that
1) CD
2) SACD
3) LP
4) DVD
5) my dog barking
is better. And this list isn't longer only because I no longer have any VHS audio tapes, a reel-to-reel, or a working cassette deck.
But then, we're all such a contrary bunch anyway - where's the latest cable debate
?The world of high-end audio is full of topics where it's very difficult to prove one's position beyond all doubt. This thread is an example. I remember that way back before the advent of the digital era we used to groove on MUSIC without paying too much attention to the quality of reproduction. With less and less of really good music we tend to dissect sound like quacks cut frogs - into bits (albeit analog or digital).
That problem appears only with spherical styluses and (to a lesser extent) with plain ellyptical styluses.
It does not appear with line contact/hyperelliptical styluses, and even more advanced shapes are available. Since a hyperelliptical stylus costs about $40, and the SAS laser-shaped micro-ridge stylus about $100, there's no reason to use spherical styluses anymore.
Actually is one of the best ways to experience analog: Frequency response from 20Hz to 50,000Hz or more, usually very flat between 20 and 20,000KHz, no digital artifacts, signal-to-noise ratio can be good enough for all purposes (>66dB), and tics and pops are not an issue if you really care for your records.
Modern magnetic tapes are very resilient to degradation. BTW, Reel-to-reel is the best way to experience analog. Or any audio for that matter.
I hadn't really thought of that. DVDA comes much closer to vinyl than CD does at its best, that's for certain, although most of what I've heard even from SACD and DVDA hasn't made me want to discard my vinyl collection in any way.
But to save wear and tear on my stylus and have the ability to eq DVDA copies of vinyl that has been less than optimally equalized is something to consider.
But how significant is this 'continuous degradation'? I have a number of LPs that I've owned for as long as 30 years, some of which have been played hundreds of times, and almost all are still clean sounding and nearly as tick and pop free as the day I first played them. It's all in how you take care of your collection. If they are never cleaned and the cat is allowed to sit on them, they won't fare too well.
'Bop Till You Drop' was the recording that turned my initial high hopes for digital sound back around '79 to deep disenchantment which has never left as long as CD quality audio and worse has been inflicted upon me. The recording was thin, anemic, tinny, missing huge amounts of musically significant detail, and sounded as if it was infested with a peculiar vinyl affliction known as 'non fills' that were actually digital artifacts.
The record degradation happens when you use a spherical stylus coupled to a low-trackability pickup and, say, 3g and more of vertical tracking force. The RIAA has already published the typical record wear to be expected in those cases.
If you use a good quality line-contact stylus in a good quality hi-fi cartridge with around 1,5-2g tracking force, expect the records to last almost forever!!
CD, on the opposite, gives you PE-PE-PE-PERFECT SO-SO-SO-SO-SO-SOUND FO-FO-FO-FO-REVER...
Really, i find even more difficult to take care of CDs... if they have not been correctly manufactured, put them on a humid environment (my city is very humid) for some years and then the data layer starts to peel out!!! where as in LP records, i can just remove fungus using the glue cleaning record, and then it's like new again.
Last edited:
Seems that the "digital sound" was caused by the strong anti-alias filters used before the A/D conversion stage; they were elliptical/Cauer brick-wall filters of 9th order or more, with lots of ringing on the high frequencies and phase deviations!! Cutoff frequency=22.05KHz.
Now you can just use much more sampling rate on the ADC (say, 192KHz), so you don't need such filters, and the filter cutoff frequency can be moved farther away from the audible spectrum. Then you can downsample in the digital domain to 44.1KHz. Sounds much better.
BTW i bought a George Duke record some months ago, "Dream on"(1982), i put it on my turntable and then i notice something strange with the sound... the record was like-new and totally clean, the sound was clean, detailed but... artificial... it sounded "digital"... so i started looking around the record sleeve for indications of the recording technology used... until i find, in very tiny script:
"Mastered using Mitsubishi Digital Audio Systems"
There. I was able to "hear" the "digital sound" beforehand!! Of course i mean the "badly implemented digital sound". This was a 1982 record, remember.
Last edited:
Elliptical styli, not filters
What you post is the case, but my early experiences definitely removed any expectation for 'perfect sound forever' from the Redbook standard - also interesting that Philips originally promoted CD for convenience, not SQ a la the audio cassette but then Sony got involved, upped the standard word length from 14 to 16 bits, hatched the 'Perfect Sound Forever' ad campaign and the die was cast.
There were a few, very few, decent sounding early digital recordings such as by Telarc which to me highlighted how susceptible 16 bit <50Ks/s PCM was to the FUBAR factor. And now, after 30 years, we're just starting to figure out how to get the decidedly mediocre 'best' out of CD audio on a somewhat consistent basis. Talk about audio's lost generation.
What you post is the case, but my early experiences definitely removed any expectation for 'perfect sound forever' from the Redbook standard - also interesting that Philips originally promoted CD for convenience, not SQ a la the audio cassette but then Sony got involved, upped the standard word length from 14 to 16 bits, hatched the 'Perfect Sound Forever' ad campaign and the die was cast.
There were a few, very few, decent sounding early digital recordings such as by Telarc which to me highlighted how susceptible 16 bit <50Ks/s PCM was to the FUBAR factor. And now, after 30 years, we're just starting to figure out how to get the decidedly mediocre 'best' out of CD audio on a somewhat consistent basis. Talk about audio's lost generation.
How about this - I replaced my original pretty well worn first Van Halen album twice - first with a WB digitally mastered copy, and recently with an all analog 180 gram version. You couldn't tell the early bifet op amp sound thumbprint of the electonics used in making the recording at all from the digitized LP but there it was just as big as life on the 180 gram all analog reissue. Both are very good pressings, btw.
Hmm... In _theory_, with 44.1K sample rate and 16bit depth, you can reconstruct any signal from 0 to 22050Hz, and with >90dB of signal-to-noise ratio. One of the significant sources of degradation was that the A/D conversion was also done at 44.1K, so you needed horrible brickwall filters at the input otherwise the even more horrible effect of aliasing will set in...
... but if you use higher sampler rates on A/D (let's say 192KHz), then the filter cutoff frequency can be moved far higher, so you can use a more natural sounding anti-alias filter. Then you downsample to 44.1KHz when mastering the CD. This can be HEARD as much superior sound, and, IMHO, is the reason for some of the latest, well-mastered CDs to sound pretty much fantastic.
On the D/A side, the converters have improved a lot. There is no reason for the CD not to give outstanding sound quality...
...BUT the LP(vinyl) medium can record and play frequencies beyond 22000Hz, sometimes as far as 50000Hz. Maybe this has something to do with the superior sound. There are some studies that say that the brain DOES show changes in neuronal activity when exposed to sounds beyond 20KHz.
I still think, if we don't consider signal-to-noise ratio, that the vinyl record gives a better sound.
I'd be a lot happier with 144 db (what 24 bit would theoretically supply). IAC, the main reason that some CDs are recently managing to sound better than they were is that the professional recording industry has abandoned the inadequate low level resolution that is a part of 16 bit PCM, even with noise shaping. But even the best sounding CDs fall far short of a good SACD or DVDA in this regard, let alone a good LP. I've always considered that the bottom half of the vaunted 90dB CD dynamic range makes CD often unsuitable for serious listening due to the quantizing losses that destroy low level tonality, inner detail and imaging. I almost always wonder, when listening to a CD, what the recording really would've sounded like with the missing information.
The human ear has a dynamic range of >100dB, but once it hears very loud sounds, it cannot hear the very soft sounds again, until you wait for a while.
For listening to more than, say 66dB of sound level difference, you need to be on a totally soundproofed (silent) room... Also, as soon as the loudest volume level (let's say, 100dB above the minimum listenable level) sets in for more than, say, 30 seconds, your ear will enter into "self-preservation mode" so you won't be able to hear those whisper-silent sounds again until you wait for a while...
I'm happy with around 66dB. "It is not just 'good', it is 'good enough'"
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.