Pure Vinyl

[jcbrum]

 

Here's a tip which I'd almost forgotten about, .... the original AVI ADM9 USB digital loudspeakers were designed to allow you to digitize your records. All you needed were the LS and a computer with a usb port. No external dac's, adc's, or sound cards at all, ... nothing except the computer and the ADM9's .... how about that !

 

I think the new Neutron 5.2.1's do the same trick, but better ask Ashley.

 

Regards JCBrum.

 

 

 

[tfarney]

I'm not archiving any vinyl, JC, but I stilll found that very interesting.

 

Tim

 

[babybear]

Thanks very much for your input JC. The recording levels was one of the items was not sure about at all - ie, how hot to record. I am sure that will take some experimenting to get right. I'll take a look at Audacity as well. Thanks again

 

[Roseval]

Anybody tried wood glue?

http://www.audiokarma.org/forums/showthread.php?t=99837

 

[The Computer Audiophile]

Great link Roseval! I'm not that familiar with wood glue, but it seems like there would be some sort of residue left on the vinyl and that would kill a stylus.

 

[PeterSt]

If you indeed jost your job ... keep up the self-confidence.

It may nog help, but loosing your job these days is not a shame. It really isn't you know.

 

Peter

 

[myco]

Personally I think it's a pretty pointless thing to do since to capture entirely the frequency and dynamic range of a vinyl lp requires much less than 24/192 offers. IMO 24/48 is more than adequate.

 

There is some misunderstanding here... see: http://www.channld.com/vinylanalysis1.html and: http://www.channld.com/pure-vinyl_support_faq.html#anchor39537

 

...

 

 

another serious error is to use digital riaa correction whist making the master recording. The cut and boost levels require about 40db of headroom which detracts directly from the overall dynamic range available.

 

Error is a pretty strong word to use... This is also a misunderstanding, as explained and proven in this Audio Engineering Society presentation paper: http://www.channld.com/aes123.pdf

 

Also, the Lynx L22 has full 192/24 ADC and DAC capability; the card is only limited to 24/96 for digital I/O. That can be improved to full 192/24: Lynx offers an optional digital I/O module for the L22 that will provide 192/24 digital I/O via AES/EBU dual-wire. It’s an extra-cost option, but at least that option is available.

 

 

 

[jcbrum]

Myco, have you done much digital transcription from vinyl yourself ?

 

You might not realize that the articles you quote were written or supported by the people who sell PV.

 

We have already dealt with the L22 issues.

 

I assure you there is no misunderstanding of the subject on my part.

 

You are of course entitled to your own beliefs, but they cannot affect the findings which are derived from my own original work, which I prefer, compared to information gleaned by reading dubious third party websites.

 

Regards JCBrum.

 

 

 

[jcbrum]

On the other hand Myco, it is even possible that since this is your first post, you are in some way connected with Pure Vinyl ?

Again, my good wishes, JCBrum.

 

 

 

 

[Blu]

Thank-you myco for the spectrograms showing how 16/44.1 digital from a CD cuts off the frequencies at 22.5 kHz compared to 24/192 digital, which showed that it was capable of recording frequencies up to 96 kHz, which were present on the bell percussion sounds from the vinyl LP (long play) album of "Crime of the Century" by Supertramp.

 

Very interesting objective evidence that you will only get all the sound recorded on a vinyl album, by recording to at least 24/192 digital.

 

But then you need to consider some other questions.

1) Can we hear frequencies above 20 kHz?

I have read that frequencies above 20 kHz have an effect on lower frequencies that we can hear, and because of this effect, higher frequencies do have an effect. My apologies but I cannot find the source at present.

 

2) Is your HiFi system of such high resolution that it can play sounds up to 96 kHz or above?

 

Isn't that just science for you, you think you have solved a problem only to have it raise more questions.

 

[jcbrum]

Blu, you say, "frequencies up to 96 kHz, which were present on the bell percussion sounds from the vinyl LP "

 

1. Even if these sounds were present on the LP (they could be an artifact of the test set), it doesn't mean they were present in the original performance. They might simply be distortion artifacts (ringing) appearing from the recording equipment.

 

You say. "I have read that frequencies above 20 kHz have an effect on lower frequencies that we can hear, and because of this effect, higher frequencies do have an effect."

 

2. Frequencies higher than the fundamental note do exist as multiples of the original, and are called 'harmonics'. Frequencies lower than the original are called sub-harmonics, and don't exist in reality. They are a figment of imagination.

 

I think your so called "objective evidence" is seriously flawed. It couldn't be anything to do with a sales pitch could it ? Aren't the people who are creating and supporting this "evidence" the same ones as are selling product on the back of it ?

 

I think that you cannot hear anything above 20khz, or any harmonics above 20khz, and your hifi can't produce anything useful above 20khz except maybe distortion which might stop it from working properly.

 

The combined efforts of the best brains from Sony and Phillips working together, produced the consumer audio CD specifications, which was widely held to be a truly excellent system, and which displaced vinyl almost entirely. - MAYBE - If you'd been there you could have help them out a bit with your ideas ? huh ?

 

The biggest audio quality gain that computer audio files have over CD comes from the use of 24 bit resolution rather than 16 bit.

 

You can use 24/192 recording if you want, - it's intrinsically ok, with good modern kit, - just make sure that the band from 20khz to 96khz isn't full of spurious crap and distortion which you can't hear, but is actually adversely affecting the working of your equipment, and is visible on pretty pictures from salesman's analyzers.

 

Regards JCBrum.

 

 

 

 

 

 

[DavidJPettifor]

jcbrum wrote:

Songs by the Andrews Sisters, (ask your grand-pa) are a good example. They sold more records than the Beatles, and Elvis Presley who bridged 78's and vinyl, especially 45's.

 

There was a great documentary on BBC Four recently. Still available for viewing on iPlayer I believe;

 

http://bit.ly/vE5Zu

 

OK, back on topic now...

 

--

djp

 

 

 

[Roseval]

showing how 16/44.1 digital from a CD cuts off the frequencies at 22.5 kHz compared to 24/192 digital

This is the brick wall filter (in)famous from the first generation CD players (and probably the current NOS DACs)

A DAC using oversampling can use a filter with a much gentler roll off

 

frequencies above 20 kHz have an effect on lower frequencies that we can hear

What we can't hear, we can't but I can imagine that a tweeter producing a 18 kHz tone only sounds different when producing a 18 kHz and a 30kHz tone at the same time. "Different" is of course not equal to better.

 

your hifi can't produce anything useful above 20khz except maybe distortion

Questionable, a amp like the Densen B-150 (http://www.densen.dk/index.php?page=products&produkt=b150) claims a frequency response (+0 -3db): 2-500.000Hz

Don't know what modern tweeters can do.

But if you see your dog taking a stroll along the ceiling you know your system is capable of tones in excess of 20 kHz.

 

You can use 24/192 recording if you want, - it's intrinsically ok, with good modern kit

Maybe, but I have heard claims that some DACs simply sounds better at a lower frequency.

If I remember correctly, the Benchmark DAC 1 does accept 192 but internally resamples everything to 110 K simply because that frequency yields best sound quality in this design.

 

 

 

[Ashley James]

Roseval

 

I've explained all this numerous times. Modern 1" tweeters will reach anything from 20 kHz - 35 KHz, hard diaphragm ones with massive resonances at the top end, however because their diameter dictates dispersion, they don't put much into a room at over 10-12 kHz. This doesn't matter because there isn't much music above about 5 kHz and your ears are rolling off by then anyway. They are down 20 dB at 20kHz, which is by a factor of 100 below their mid band sensitivity.

 

In our DAC we settled on 24/96 and used SRC to do it to get rid of the "J" word and because it seems to be the Industry standard. We also found the DAC ran much hotter at 192 kHz and thought that a bad thing. It will play anything up to 192 but resampled to 24/96.

 

Ash

 

[jcbrum]

quote Roseval, .. " -frequencies above 20 kHz have an effect on lower frequencies that we can hear

What we can't hear, we can't but I can imagine that a tweeter producing a 18 kHz tone only sounds different when producing a 18 kHz and a 30kHz tone at the same time. "Different" is of course not equal to better." - endquote.

 

Don't imagine it, .... try it ! ... I have.

 

If you use amps (easy), and transducers (v difficult), with a clear available passband of say 200khz, and inject two signals at equal level, - one at 18khz and the other at 30khz, you will 'hear' (detect on an analyzer) principally four tones, plus some other mush (sidebands), which are 12khz, 18khz, 30khz, and 48khz.

 

In practice, you will only hear 12khz, because the others are inaudible to nearly everyone.

 

This 12khz was never there in the original performance, and only occurs because of non-linearities and inadequacies in the electronics. It represents distortion and false artifacts, which could be present on the vinyl record, if one were to be made of this 'performance'. (i.e. the vinyl record could display signals of 48khz and beyond !)

 

This example is a simplification, but it illustrates the need for 'roofing filters' and such-like in electronics. A subject which is not well understood by many 'audiophile journalists' and other 'commentators'.

 

I am absolutely sure that designers of high calibre, (such as M. Grindrod of AVI and Peter Walker of Quad) are fully aware of these matters, and it rubs off on people such as Ashley James and others, who have wide real-world experience of the recording and electronic sound re-production industry.

 

My personal 'hero' is Alan Blumlein, largely un-sung, who 'invented' stereo and the recording techniques to go with it.

 

I'm not so sure about Clive Sinclair (now Sir,) and computers though, ;-) , although he is undoubtedly a brilliant man. (perhaps that's an anglophile anecdote !)

 

Regards JCBrum.

 

[jcbrum]

This may be of passing interest, .......

 

http://www.alanturing.net/turing_archive/pages/blumlein/index.html

 

http://www.doramusic.com/blumleinmainmenu.htm

 

JC.

 

[Ashley James]

It is some years now since I was a regular visitor to the two largest Recording Studios in Europe where Vinyl Disc Mastering took place. They were CTS and Abbey Rd. I went to lots of others but these are the most impressive.

 

I seem to remember being told that the maximum frequency that could be applied to the outside of a record was about 9 kHz and on the inside, about 5 kHz, but that the shaft that drove the cutting head resonated at 3.5kHz and although there was a passive suck circuit to neutralise it, it was audible and sounded nasty, so they would add lots below this frequency and not bother with much above because it would be removed with the first pass of the stylus anyway.

 

As for hearing much above these frequencies, you don't. I've given the facts before in previous posts and I'll email anyone hearing measurements who's interested. Human hearing is typically 20 dB down at 20 kHz.

 

Ash

 

 

 

[BEEMB]

Ash,

 

The DAC in the AVI's can accept up to 192 - but I wasn't aware you can actually pass this over an optical connection anyway? I thought it was limited to 24/96.

 

[Ashley James]

Seems to suggest that you can, regardless we offer an optional coaxial socket for professional users.

 

Ash

 

[BEEMB]

 

Okay .. I've just noticed that so many firewire interfaces appear to limit optical output to 24/96 at the same time as allowing higher rates over coax.

 

m.

 

[The Computer Audiophile]

This is an operating system limitation. Apple told me several times that the optical interface in its devices is capable of 24/192. Nobody from Apple has ever been able to get it working on any of my machines while in OS X. I have been successful using Windows Vista 64 bit via Bootcamp on my Macs sending bit perfect 24/192 via optical.

 

[RHA]

As a non technical type this has been a great thread to learn about that of which I know little/nothing. Thank you to all who have contributed. Unfortunately it has brought up a couple more questions which I hadn't thought of before. Perhaps someone can clarify.

1. If, as the Stereophile article says, Pure Vinyl sounds better does that mean that the software imparts its own sonic signature on the bits? I had thought of the software as basically an envelope to capture and hold the bits from the ADC in their pristine form and allow you to manipulate them in the editing process. If the same equipment is used for both recording and playback would there be a sonic difference depending on the software used? Does this difference in sound quality mean that the software somehow distorts or manipulates the original bits received from the ADC chip?

2. Do the various ADC and DAC chips have unique sonic signatures? I've read of different DACs using the same chips having totally different sound signatures. I attributed this to the differences in implementation of power supply, noise/jitter reduction, analog output etc. but what do I know? How does this all work? Is there a factor of primary importance?

 

I'd like to sneak in another question not directly related to Pure Vinyl... the use of normalization in the recording process. Given the wide variety of gain on different LPs I have not been as diligent as others here who monitor and set the gain for each recording. (God bless 'em). In general I've hit on a "one size fits all" approach that will generally give me a signal of .5 dB showing on the waveform in Audacity. After editing, tagging, etc I normalize to try and get a uniform output level when playing back in iTunes. Comments, good, bad? I have my helmet on. Hit me!

Thanks to all for the wealth of info here. Cheers, Rod

 

[WoodsDweller]

Thanks especially to jcbrum for his detailed comments. I have saved them to a document in case I ever get around to ripping my old vinyl. This was just the level of advice I was looking for.