How Good Can Vinyl Sound?, Anyone got some sample clips? Options

[sshd]

I do not own a turntable and have never heard the "great vinyl sound" coming from a good turntable handled correctly. I am however sligltly interested in purchasing one, as there are some interesting albums out there unavailable on CD.

Recently I found a small clip on the net: A lossless needledrop of Santana's debut album on the MFSL label. Since I own the CD on same label I had something to compare with: I was very disappointed with the needle drop. There were no pops or clicks, but the sound was dead.

Can anyone provide some sample clips of very good needle drops?

[2Bdecided]

See this thread for recent discussion on this topic...

http://www.hydrogenaudio.org/forums/index....mp;#entry633881


There's a link to a needle drop here...

http://www.hydrogenaudio.org/forums/index....st&p=632247

...I can't remember the details, but mere mortals couldn't afford that turntable.


You can afford this one though...

http://www.hydrogenaudio.org/forums/index....mp;#entry633478



Cheers,
David.

[bryant]

http://www.hydrogenaudio.org/forums/index....showtopic=67018

[DVDdoug]

I do not own a turntable and have never heard the "great vinyl sound" coming from a good turntable handled correctly.

Well... I do own a turntable, and I've never heard that "great vinyl sound" either! I find the "snap", "crackle", and "pop" distracting and annoying. The preamp noise bothered me a little, but not as much as the "vinyl defects".... All of my records seem to have "developed" defects.

And, back in the "vinyl days", most records were not produced with high-fidelity in mind. It was a rare treat to find a record with great audio quality. Even with most of the ticks & pops removed, most of my vinyl transfers don't sound like "CD quality" to me. (If the CD is available, I'll buy it.)

My turntable is medium quality (Technics), with a high-end Shure cartridge. (The cartridge was probably Shure's top-of-the-line when I bought it.) I only use it for vinyl-to-digital transfers... I haven't "played" a record in many years.

If I had to replace the set-up, I'd probably get something similar. I'd get an inexpensive turntable (but not bottom-of-the-line cheap) and a medium or good (but not "audiophile") cartridge. I'm no longer in search of the best (analog) sound I can afford... My theory is, "If you want perfection, go digital!"

This post has been edited by DVDdoug: May 15 2009, 21:00

[Arnold B. Kruege...]

I do not own a turntable and have never heard the "great vinyl sound" coming from a good turntable handled correctly.

I have owned a number of turntables, and was a long-time audiophile back in the days when vinyl was pretty much all we had. I don't know what that "great vinyl sound" is, other than hype. And while we're talking about hype, "perfect sound, forever" (CD when it first came out) was hype, too.

I've had any number of vinylphiles demonstrate their "great vinyl sound" to me. Still, no cigar!

Back in the day, those of us who were scientific about getting the most out of vinyl were also intimately familiar with the inherent limiations of vinyl. I mean besides the obvious tics and snaps, there are very basic problems that are varying degrees of insoluable. And, that's just the ones that are audible!

For me the biggie is the inherent speed accuracy of the digital formats. LP pianos often sound wattery to me.

I am however sligltly interested in purchasing one, as there are some interesting albums out there unavailable on CD.

Been there done that, too. I currently have a Rega turntable with Shure and Grado cartridges for transcribing LPs. Doesn't get a lot of use. Still wish I had my last system from the days of - a TD125 with SME arm and V15 cartridge.

Recently I found a small clip on the net: A lossless needledrop of Santana's debut album on the MFSL label. Since I own the CD on same label I had something to compare with: I was very disappointed with the needle drop. There were no pops or clicks, but the sound was dead.

I've done a number of ABXs of time, level and spectrally-matched needle drops of LPs and CDs that seem to be very similarly mastered. I can still nail 'em based on all those other things than tics and pops that are wrong with the LP format.

Can anyone provide some sample clips of very good needle drops?

Others have. Some pretty good transcriptions have already been linked.

[Glenn Gundlach]

I do not own a turntable and have never heard the "great vinyl sound" coming from a good turntable handled correctly.

I have owned a number of turntables, and was a long-time audiophile back in the days when vinyl was pretty much all we had. I don't know what that "great vinyl sound" is, other than hype. And while we're talking about hype, "perfect sound, forever" (CD when it first came out) was hype, too.

I've had any number of vinylphiles demonstrate their "great vinyl sound" to me. Still, no cigar!

Back in the day, those of us who were scientific about getting the most out of vinyl were also intimately familiar with the inherent limiations of vinyl. I mean besides the obvious tics and snaps, there are very basic problems that are varying degrees of insoluable. And, that's just the ones that are audible!

For me the biggie is the inherent speed accuracy of the digital formats. LP pianos often sound wattery to me.

I am however sligltly interested in purchasing one, as there are some interesting albums out there unavailable on CD.

Been there done that, too. I currently have a Rega turntable with Shure and Grado cartridges for transcribing LPs. Doesn't get a lot of use. Still wish I had my last system from the days of - a TD125 with SME arm and V15 cartridge.

Recently I found a small clip on the net: A lossless needledrop of Santana's debut album on the MFSL label. Since I own the CD on same label I had something to compare with: I was very disappointed with the needle drop. There were no pops or clicks, but the sound was dead.

I've done a number of ABXs of time, level and spectrally-matched needle drops of LPs and CDs that seem to be very similarly mastered. I can still nail 'em based on all those other things than tics and pops that are wrong with the LP format.

Can anyone provide some sample clips of very good needle drops?

Others have. Some pretty good transcriptions have already been linked.

Not to be too big of a jerk but how can one post a clip of that 'analog sound' when your clip is digital (and maybe even compressed) ?? BTW VERY respectable disc playing gear. Mine was a Dual 721 with a Shure V15 type V.

G²

[krabapple]

First, all digital ends up as 'analog sound' when it comes out of the DAC and ultimately your loudspeakers/earbuds/headphones

Second, your objection presumes that a good digitizing of an LP somehow changes the sound. Why would you think that?

[Arnold B. Kruege...]

Not to be too big of a jerk but how can one post a clip of that 'analog sound' when your clip is digital (and maybe even compressed) ?? BTW VERY respectable disc playing gear. Mine was a Dual 721 with a Shure V15 type V.

When we say that digital can be a lot more accurate than legacy analog media, we aren't kidding!

By most accounts, a well-done needle drop is a sonically accurate representation of the playback of the LP.

If you take a good digital recorder and compare a recording it makes, to the signal that was applied to it to make that recording, nobody can reliably detect an audible difference. This experiment has been done many times by many people, going back into the late 1970s. It has been done many times with high quality live recordings.

[boondocks]

As someone that grew up with vinyl, who had 3 copies of "favorite albums", one to play anytime, one to record and shelve, and one for backup, I can tell you I see no advantage, IMO, Only, for vinyl.
SQ is entirely dependent upon YOUR hearing.
My albums were played on a Thorens TD125, with a Shibata stylus for my CD4 quad stuff. (yeah, I'm old)

I guess the bottom line is, try vinyl. If it's good enough for you, well, that's the end of it.....
and I would not fault you one bit.

If it's not, then join another convert to digital.

Besides, keeping vinyl in good shape forever, IMO again, is impossible, no matter what the guru mags say with
there snake oil remedies and expensive cleaning machines. Been there, done that. Vinyl wears out, but less so
if the turntable is properly set up and kept that way.

My .02

[Axon]

Many people get that "great vinyl sound" out of extremely inexpensive systems - in fact, they get it from systems that other people abhor for their poor sound quality.

I posit that "great vinyl sound" really doesn't have that much to do with the actual audio involved, but is more a reflection of the emotions of the listener, and the tactile response of the medium. Those are very real factors, and it's not particularly useful to ignore them, insofar as they reflect real aspects of the enjoyment of music. But you shouldn't confuse them with more substantive aspects of sound quality, and you shouldn't necessarily buy into vinyl always expecting to get that level of enjoyment. It's a largely subjective phenomenon.

[C.R.Helmrich]

First, all digital ends up as 'analog sound' when it comes out of the DAC and ultimately your loudspeakers/earbuds/headphones

Second, your objection presumes that a good digitizing of an LP somehow changes the sound. Why would you think that?

I've recorded a song from an 80's LP on my PC in 24bit/96kHz once and noticed that the bandwidth extended to around 30 kHz (true, the roll-off started at 15 kHz or so, but still, it reached till 30). So to be fair, one should record at 16bit/96kHz to get a "good digitizing of an LP", I think. Correct me if I'm wrong, but all above links offer only 16bit/44kHz samples. One FLACs was even processed with LossyWav!

Chris

[Arnold B. Kruege...]

I've recorded a song from an 80's LP on my PC in 24bit/96kHz once and noticed that the bandwidth extended to around 30 kHz (true, the roll-off started at 15 kHz or so, but still, it reached till 30)

There's a very fair question to ask about LP tanscriptions whose spectral analysis shows content > 20 KHz:

Is it music or is it distortion?

The LP format does have a lot of inherent nonlinear distoriton at high frequencies due to geometric mismatch between the path of the sharp, chisel-shaped cutter, and the round or elliptical stylus.

Also, there is considerable dynamic deformation of the groove itself. which mostly springs back shortly after playing. Point being, the stylus deforms the groove in a nonlinear way while its being played.

And of course there's the other issue about which so much relevant evidence has been collected that there is hardly a controversy over it any more. Can we hear the removal of audio > 22 Khz? The answer is generally no, not at all.

[C.R.Helmrich]

Is it music or is it distortion?

The LP format does have a lot of inherent nonlinear distoriton at high frequencies due to geometric mismatch between the path of the sharp, chisel-shaped cutter, and the round or elliptical stylus.

Also, there is considerable dynamic deformation of the groove itself. which mostly springs back shortly after playing. Point being, the stylus deforms the groove in a nonlinear way while its being played.

Good question. Maybe this distortion has a similar effect like the "added warmth" of a tube amp? Maybe it's the reason why some people prefer the "smooth sound" of an LP? (To me, the mentioned 24/96 recording also sounded sort of "smooth and warm"). Surely such distortion must also be present in the audible band, I guess?

And of course there's the other issue about which so much relevant evidence has been collected that there is hardly a controversy over it any more. Can we hear the removal of audio > 22 Khz? The answer is generally no, not at all.

I read a scientific article a few years ago demonstrating how in blind tests, subjects were able to hear a 24-kHz tone and that for all of them, the threshold of hearing did not increase between 20 and 24 kHz but actually stayed on a rather constant level (at 80 dB SPL, true, but still...). 26-kHz tones, however, were inaudible to all, IIRC. I'll try to dig up that paper again.

Edit: Here is the paper. Did not quite remember it correctly, but see for yourself.

Chris

This post has been edited by C.R.Helmrich: May 15 2009, 23:11

[Arnold B. Kruege...]

I read a scientific article a few years ago demonstrating how in blind tests, subjects were able to hear a 24-kHz tone and that for all of them, the threshold of hearing did not increase between 20 and 24 kHz but actually stayed on a rather constant level (at 80 dB SPL, true, but still...). 26-kHz tones, however, were inaudible to all, IIRC. I'll try to dig up that paper again.

Edit: Here is the paper. Did not quite remember it correctly, but see for yourself.

While it is not a slam dunk that tests involving pure tones and music will have signficiantly different results, this is no doubt one of them.

Since there are very very few musical instruments with fundamentals > even 10 KHz, anything above 20 KHz is going to be a harmonic of some fundamental below 20 KHz. Unless the content > 20 KHz is very, very strong, its going to be masked by fundamentals and/or lower harmonics below 20 KHz. Or its simply going to fall below the limit of audibility, which is very high at frequencies this high.

This would be a good time to post a spectral analysis of one of your 24/96 files... HA has a forum for doing this, so you don't need to line up space on your own. You could even post excerpts for others to listen to and/or analyze.

If you look in the scientific section of HA, there's even a paper about perception that has some masking curves in it. If you scale them up to the 8-10 KHz range, they do a pretty good job of wiping out all but the strongest responses > 10 KHz.

[C.R.Helmrich]

While it is not a slam dunk that tests involving pure tones and music will have signficiantly different results, this is no doubt one of them.

Since there are very very few musical instruments with fundamentals > even 10 KHz, anything above 20 KHz is going to be a harmonic of some fundamental below 20 KHz. Unless the content > 20 KHz is very, very strong, its going to be masked by fundamentals and/or lower harmonics below 20 KHz. Or its simply going to fall below the limit of audibility, which is very high at frequencies this high.

This would be a good time to post a spectral analysis of one of your 24/96 files... HA has a forum for doing this, so you don't need to line up space on your own. You could even post excerpts for others to listen to and/or analyze.

If you look in the scientific section of HA, there's even a paper about perception that has some masking curves in it. If you scale them up to the 8-10 KHz range, they do a pretty good job of wiping out all but the strongest responses > 10 KHz.

I'd really like to post a FLAC file of the LP here, but I recorded that 4 years ago at on a relative's turntable, and deleted it shortly afterwards (if only I knew back then... ). Anyway, as I said, the high-frequency roll-off ended at 30 kHz. So the energy was already down quite low at 22 kHz. Yes, agreed, most likely below the hearing threshold. Just thought I'd add this comment since the original discussion was about accurate representation of Vinyl sound. If you don't digitize the entire spectrum "coming out of a vinyl", then it's not an accurate representation, I'd say.

I know about the masking curves at high frequencies. I posted some of them the other day Yes, if you consider masking, it's even more likely that anything beyond 20 kHz on an LP is inaudible. Still, there's something there The next time I get a hold of a turntable, I'll post a 24/96 FLAC.

Chris

[greynol]

The next time I get a hold of a turntable, I'll post a 24/96 FLAC.

While you're at it, convert it to dithered 16/44.1 and provide some ABX results. Just make sure the clips are no longer than 30 seconds.

[krabapple]

First, all digital ends up as 'analog sound' when it comes out of the DAC and ultimately your loudspeakers/earbuds/headphones

Second, your objection presumes that a good digitizing of an LP somehow changes the sound. Why would you think that?

I've recorded a song from an 80's LP on my PC in 24bit/96kHz once and noticed that the bandwidth extended to around 30 kHz (true, the roll-off started at 15 kHz or so, but still, it reached till 30). So to be fair, one should record at 16bit/96kHz to get a "good digitizing of an LP", I think. Correct me if I'm wrong, but all above links offer only 16bit/44kHz samples. One FLACs was even processed with LossyWav!

Chris

Just because the bandwidth extends to 30kHz doesn't mean it's musical information; nor does it mean the reproduction is distortion-free or the level significant (notice that roll-off?). And that's all besides the paucity of evidence for audible perception above ~24 kHz max, (most people's top limit being well below 20 kHz, much less 24). Finally, if you actually 'need' 30 kHz, then an 88kHz digitization will do nicely.

[greynol]

I think the implication is that the information above 22 kHz is what is partly responsible for making vinyl more "danceable".

[krabapple]

I know about the masking curves at high frequencies. I posted some of them the other day Yes, if you consider masking, it's even more likely that anything beyond 20 kHz on an LP is inaudible. Still, there's something there The next time I get a hold of a turntable, I'll post a 24/96 FLAC.

Chris

No one's said there's nothing there. The question is what 'it' is, and whether it matters.

I have some FLACs that were originally 192/24 (ripped from DVD-A) , 96/24 (same) , some SACD 'laserdrops' that are 88/24, and I'm sure I've got some LP captures that are 'hi rez' too. I would never claim that the high sample rate makes an audible difference, though. if anyone wants to hear clips, I can post some.

[WonderSlug]

I think the implication is that the information above 22 kHz is what is partly responsible for making vinyl more "danceable".

Especially to cats and dogs.

[C.R.Helmrich]

No one's said there's nothing there. The question is what 'it' is, and whether it matters.

I have some FLACs that were originally 192/24 (ripped from DVD-A) , 96/24 (same) , some SACD 'laserdrops' that are 88/24, and I'm sure I've got some LP captures that are 'hi rez' too. I would never claim that the high sample rate makes an audible difference, though. if anyone wants to hear clips, I can post some.

Actually, yes, that would be great. Sorry, off-topic, but do you have something with significant high-frequency content? Like solo guitar, or percussion? I'd like to do a 16/44 vs. 24/96 ABX because I never did (heard a comparison once at an AES conference, blind, but not double-blind and not random-order and not etc. etc.)

Chris

[krabapple]

I looked at spectra pf my complete repertoire of stuff in my archive that is still at 'hi rez' rates, and it was pretty amusing how most of it is obviously lowpassed at or below 22 kHz....some with extreme brickwall in place -- even when the delivery format is >44.1 kHz. Still there were some that were truly 'full range' at the highrez sample rates, but actual musical material above that frequency is scant --- typically it looks like noise up there with some transient musical material.

No needledrops included in this round (I have to dig them up), just DVD-A rips or SACD transfers. It also appears that I converted all my 'classical' hirez captures to 16/44 at some point, probably to save space at the time. So all I have are some vintage (1960s-70s) recording to show right now -- clips mostly of DVD-A rips. The one SACD representative -- a gorgeous Muddy Waters recording from the 1960s --was captured from the stereo analog out of an Oppo 970 HD player, into an M-audio 2496 card, with Audition as the recording software, set to 88.2/24. All of the 16/44 clips were resampled and reformatted from their highrez counterparts using Audition, with dither (triangular pdf, no noise shaping, dither depth 1, pre and post filtering on)

Apparently each of these can only be downloaded 10 times (rapidshare rule) but if there's persistent demand I can repost them. THese are clips, not complete tracks.

Fleetwood Mac 'Dreams' from the Rumours DVD-A
http://rapidshare.com/files/233532367/Drea...p9624.flac.html
http://rapidshare.com/files/233532368/Drea...p1644.flac.html

Yes 'Five Percent for Nothing' from the Fragile DVD-A
http://rapidshare.com/files/233533239/Five...g2496.flac.html
http://rapidshare.com/files/233533240/Five...g1644.flac.html

Muddy Waters "Feel Like Going Home" from the Folk Singer SACD
http://rapidshare.com/files/233534243/Feel...e2488.flac.html
http://rapidshare.com/files/233534244/Feel...e1644.flac.html

Billy Cobham "Quadrant 4" from the Spectrum DVD-A
http://rapidshare.com/files/233534770/Quad...-2496.flac.html
http://rapidshare.com/files/233534771/Quad...-1644.flac.html

Neil Young "Heart of Gold" from the Harvest DVD-A
http://rapidshare.com/files/233535638/Hear...24192.flac.html
http://rapidshare.com/files/233535640/Hear...d1644.flac.html

[cliveb]

Just because the bandwidth extends to 30kHz doesn't mean it's musical information

Some "audiophile" LPs might have a tiny bit of musical information >20kHz. And of course CD4 quad LPs have an FM carrier signal with the demodulation info up around 34kHz (or is it 35kHz?).

But for the overwhelming majority of LPs (like >99% of them), there's a very good reason why any signal above about 18kHz that comes off a mainstream LP isn't musical information: a low pass filter is deliberately applied at the cutting stage, primarily to prevent the RIAA-boosted high frequencies from overloading (or even burning out) the cutter head.

[C.R.Helmrich]

Thanks a lot for the uploads, krabapple! Yes, the lowpass at 22 kHz is obvious. Plus the recordings are from the 60s or 70s. I don't think I will be able to ABX those against 16/44, so I won't

I uploaded a 16/96 recording of a digitally controlled analog synth in this thread. Found this on a CD-R from 2004. It contains a lot of "musical content" above 22 kHz, up to 48 kHz. I might ABX that against the 44-kHz downsampled version soon.

Cheers,

Chris

[Soap]

I fricking hate rapidshare and I doubt I am alone. Rather than complain, though, I have taken the liberty to rehost krabapple's samples (and added the entire collection in one .zip) at http://cleansoap.org/HA.
I haven't tested these files, I assume they downloaded and uploaded without error. Please PM me if you have any problems.

PS - the all.zip will not finish its upload until ~09:30 EDT.