Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Vynil or digital? (Read 48177 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: Vynil or digital?

Reply #125 –
Quote from: Funkstar De Luxe on
My point is that I can produce a file that has a (let's say) infinite dynamic range and bandwidth.

Illustrates my point about you being over your head. I'd geuess that a high proportion of our readers know that no such thing is possible with PCM if you are using infinite to indicate large amplitudes. Of course the digital domain allows zeros, so that is a back door way to get infinite dynamic range, but its trivial and useless. The infinite bandwidth part is unconditionally impossible in the real world, because both the number of samples file length  and therefore the sample rate are always finite.

Quote
This can be created and measured digitally. My assertion is that a theoretical file like this is of no practical difference to any other system that offers enough bandwidth and dynamic range to be transparent, be it CD, reel to reel tape, or vinyl. I fully agree that digital audio massively outperforms analogue, but there are in fact plenty of analogue systems that offer transparency.

As usual your ignorance of digital kills your argument even before we get to the well-known (to most around here) fallacy of the excluded middle.   Here's some much needed reading for you:

http://www.webpages.uidaho.edu/eng207-td/Logic%20and%20Analysis/most_common_logical_fallacies.htm

Quote
Now to bring things back on point - when I listen to a record, I am not aware of the playback medium. To me, this means that in some sense it is transparent to me.

Given all the clicks and pops if nothing else, for me to be unaware that vinyl was the playback medium would require deafness.

Quote
If necessary I can offer some ABX results, but it's difficult to verify that I don't hear a difference (i.e. A failed ABX test is easy to fake...)

Please do the scientific thing, and upload file excerpts, and we'll do our own tests.

Only, no fair digitally doctoring the files other than trimming to get an acceptable length.

Maybe you didn't notice that we have an upload forum for this. We do.  Waiting for the running and hiding... ;-)


Re: Vynil or digital?

Reply #127 –
Always Love these analog vs digital debates
Keep it nice & keep it going.
and yes "Funkstar De Luxe", some ABX would be good.

And By God would someone please correct the 'Vynil' spelling

Re: Vynil or digital?

Reply #128 –
I think part of this discussion centers on whether we're talking about "transparency" of a format in an optimistic sense (e.g. transparent on some music that tends to mask problems) or pessimistic sense (is it transparent on the most challenging content?).

For an example of a format that is not normally transparent being transparent, Audiocheck.net has some 8-bit samples (Gangnam Style, Neil Young's 'Rockin in the Free World') that I can't ABX aganist 16-bit: http://www.audiocheck.net/blindtests_16vs8bit.php

Re: Vynil or digital?

Reply #129 –
Quote from: eric.w on
I think part of this discussion centers on whether we're talking about "transparency" of a format in an optimistic sense (e.g. transparent on some music that tends to mask problems) or pessimistic sense (is it transparent on the most challenging content?).

For an example of a format that is not normally transparent being transparent, Audiocheck.net has some 8-bit samples (Gangnam Style, Neil Young's 'Rockin in the Free World') that I can't ABX aganist 16-bit: http://www.audiocheck.net/blindtests_16vs8bit.php
Quote from: eric.w on
I think part of this discussion centers on whether we're talking about "transparency" of a format in an optimistic sense (e.g. transparent on some music that tends to mask problems) or pessimistic sense (is it transparent on the most challenging content?).

For an example of a format that is not normally transparent being transparent, Audiocheck.net has some 8-bit samples (Gangnam Style, Neil Young's 'Rockin in the Free World') that I can't ABX aganist 16-bit: http://www.audiocheck.net/blindtests_16vs8bit.php


Please check out https://hydrogenaud.io/index.php/topic,107570.msg941704.html#msg941704. You might want to try again using the ABX Comparator of your choice (FOOBAR2000 ABX plug in suggested)

Re: Vynil or digital?

Reply #130 –
Quote from: The Irish Man on
"Funkstar De Luxe", some ABX would be good.
Ok, I've seen this enough in this thread.  Funkstar De Luxe owes you people no such thing!

The rule and explanation is quite clear; and despite grumblings from intellectually dishonest or deficient(?) placebophile-sympathizing trolls, the enforcement of TOS8 has been extremely fair consistent:: the people who claim they can distinguish differences are the ones required to submit samples and ABX logs.

Quote from: The Irish Man on
And By God would someone please correct the 'Vynil' spelling
Not going to happen if I have anything to say about it.  :))

EDIT: Punctuation.

Re: Vynil or digital?

Reply #131 –
Quote from: greynol on
... The rule and explanation is quite clear; and despite grumblings from intellectually dishonest or deficient(?) placebophile-sympathizing trolls the enforcement of TOS8 has been extremely fair consistent: the people who claim they can distinguish differences are the ones required to submit samples and ABX logs. ...

Why isn't the rule symmetrical? Why should people claiming no difference not have to submit samples so that their findings can be verified by others?
Regards,
   Don Hills
"People hear what they see." - Doris Day

Re: Vynil or digital?

Reply #132 –
Quote from: splice on
Quote from: greynol on
... The rule and explanation is quite clear; and despite grumblings from intellectually dishonest or deficient(?) placebophile-sympathizing trolls the enforcement of TOS8 has been extremely fair consistent: the people who claim they can distinguish differences are the ones required to submit samples and ABX logs. ...

Why isn't the rule symmetrical? Why should people claiming no difference not have to submit samples so that their findings can be verified by others?

I think you know the answer and just want someone to confirm it.

The situation is naturally asymmetrical.

Negative hypotheses such as "There is no audible difference" are difficult or impossible to prove.

Positive hypotheses such as "There is an audible difference' is far easier to prove when they exist.

For example, to prove the negative hypothesis, I would have to obtain a negative result for every listener, every audio system, every recording, etc,  in every permutation.

To prove the positive hypothesis, all I need is one positive result.

Re: Vynil or digital?

Reply #133 –
Quote from: Arnold B. Krueger on
Quote from: splice on
Quote from: greynol on
... The rule and explanation is quite clear; and despite grumblings from intellectually dishonest or deficient(?) placebophile-sympathizing trolls the enforcement of TOS8 has been extremely fair consistent: the people who claim they can distinguish differences are the ones required to submit samples and ABX logs. ...

Why isn't the rule symmetrical? Why should people claiming no difference not have to submit samples so that their findings can be verified by others?

I think you know the answer and just want someone to confirm it.

The situation is naturally asymmetrical.

Negative hypotheses such as "There is no audible difference" are difficult or impossible to prove.

Positive hypotheses such as "There is an audible difference' is far easier to prove when they exist.

For example, to prove the negative hypothesis, I would have to obtain a negative result for every listener, every audio system, every recording, etc,  in every permutation.

To prove the positive hypothesis, all I need is one positive result.


But what if one particular listener claims transparency relating to a listening comparison and submits no samples or ABX log, but another listener (for an unrelated listening comparison) claims transparency and does submit samples and ABX log, then other members ABX those samples and all hear a difference which their logs can verify.

In that situation one claim of transparency stands unchecked, and the other has been disproved. It seems an unsatisfactory situation if we are seeking the truth.

Please note I am fairly new here but keen to learn.


Re: Vynil or digital?

Reply #134 –
Quote from: cresco on
It seems an unsatisfactory situation if we are seeking the truth.
That's what the real world is like. Disproval only works in mathematics. Whenever there are statistics involved, things only ever become unlikely.

Re: Vynil or digital?

Reply #135 –
Quote from: Arnold B. Krueger on
...
Negative hypotheses such as "There is no audible difference" are difficult or impossible to prove.

Positive hypotheses such as "There is an audible difference' is far easier to prove when they exist.
...

Someone may claim there is no audible difference. They should provide evidence, perhaps some files that can be shown to be different in content but not, to them at least, audibly different.
If others can show that there is an audible difference, perhaps by ABX testing, then the claim is proven false.
Regards,
   Don Hills
"People hear what they see." - Doris Day

Re: Vynil or digital?

Reply #137 –
Quote from: cresco on
Quote from: Arnold B. Krueger on
Quote from: splice on
Quote from: greynol on
... The rule and explanation is quite clear; and despite grumblings from intellectually dishonest or deficient(?) placebophile-sympathizing trolls the enforcement of TOS8 has been extremely fair consistent: the people who claim they can distinguish differences are the ones required to submit samples and ABX logs. ...

Why isn't the rule symmetrical? Why should people claiming no difference not have to submit samples so that their findings can be verified by others?

The situation is naturally asymmetrical.

Negative hypotheses such as "There is no audible difference" are difficult or impossible to prove.

Positive hypotheses such as "There is an audible difference' is far easier to prove when they exist.

For example, to prove the negative hypothesis, I would have to obtain a negative result for every listener, every audio system, every recording, etc,  in every permutation.

To prove the positive hypothesis, all I need is one positive result.

But what if one particular listener claims transparency relating to a listening comparison and submits no samples or ABX log,

Other evidence such as measurements showing relative  freedom from potentially audible artifacts is often acceptable to many. If you show me a RMAA report with all artifacts say 90-100 dB down for conventional audio gear (amps, dacs, etc.) then most are convinced IME.  Perceptual coders are a special case and conventional measurements do not apply.

"It is so because I say it is so", is generally not accepted because of the immense propensity of humans to perceive illusions.

Quote
but another listener (for an unrelated listening comparison) claims transparency and does submit samples and ABX log, then other members ABX those samples and all hear a difference which their logs can verify.

The more instances of people who reliably hear differences that we have, the more dubious the claim of no audible difference becomes. Sometimes one instance of reliably hearing a difference can cast a fatal shadow.

Quote
In that situation one claim of transparency stands unchecked, and the other has been disproved. It seems an unsatisfactory situation if we are seeking the truth.

It is all about the preponderance of reliable evidence.  "It is so because I say it is so" is not reliable evidence, no matter how many times it is repeated. There has to be confirming reliable evidence.

Any day of the week I can trot over to The Computer Audiophile site and find 100+ "It is so because I say it is so" claims related to something that measures  better than all audio signal  artifacts 100 dB down.  Archimago's Blog is full of these and they are like shooting ducks in a row. One is technically more unreasonable than the next, but we have 100, 200, 300 glowing testimonials.

How may 100's of glowing  testimonials can one find for magic power cords? Does anybody with a brain believe any of it?

Re: Vynil or digital?

Reply #138 –
Quote from: Arnold B. Krueger on
Please check out https://hydrogenaud.io/index.php/topic,107570.msg941704.html#msg941704. You might want to try again using the ABX Comparator of your choice (FOOBAR2000 ABX plug in suggested)
I noticed differences in volume from your flac files and the website's ABX interface.
I ABX'ed your flac files with RG on and off, log attached.

The original files in the website are vorbis with very different bitrate and track peak (attached)
http://www.audiocheck.net/blindtests_16vs8bit.php

No wonder the 8-bit file sounded quieter since it is a windows DirectSound "feature".
https://hydrogenaud.io/index.php/topic,114101.msg939763.html#msg939763

Re: Vynil or digital?

Reply #139 –
I saw RG raising its ugly head. Don't assume RG is going to properly level-match in sitations where there may be differences in tonality, or absence of high frequencies. Feel free to search for discussions on the subject.

Re: Vynil or digital?

Reply #140 –
I did read your posts about new vs old algorithm used in foobar2000's RG, but at least it eliminated audible differences of the files for me at this time.

I thought about an experiment long time ago. Upload two files and let different members determine how the two files should be loudness matched by their ears instead of any algorithm, and gather statistics based on real human.

Re: Vynil or digital?

Reply #142 –
Quote from: Arnold B. Krueger on
Quote from: eric.w on
I think part of this discussion centers on whether we're talking about "transparency" of a format in an optimistic sense (e.g. transparent on some music that tends to mask problems) or pessimistic sense (is it transparent on the most challenging content?).

For an example of a format that is not normally transparent being transparent, Audiocheck.net has some 8-bit samples (Gangnam Style, Neil Young's 'Rockin in the Free World') that I can't ABX aganist 16-bit: http://www.audiocheck.net/blindtests_16vs8bit.php

Please check out https://hydrogenaud.io/index.php/topic,107570.msg941704.html#msg941704. You might want to try again using the ABX Comparator of your choice (FOOBAR2000 ABX plug in suggested)

Thanks - good catch about the level difference on the audiocheck.net 8-bit vs 16-bit samples. I managed 16/16 listening for a loudness difference in one of the synth/drum hits early in the sample.

Code: [Select]
foo_abx 2.0.2 report
foobar2000 v1.3.14
2017-07-09 14:13:16

File A: audiocheck 16 versus 8 8 bits.flac
SHA1: 07c37b6bf4d73f790d1b73dd1ed5f496acef1854
File B: audiocheck 16 versus 8 16 bits.flac
SHA1: 1d821150a59baaed4cad2c4c8c8c20a851ba8cc8

Output:
DS : Primary Sound Driver, 16-bit
Crossfading: NO

14:13:16 : Test started.
14:16:11 : 01/01
14:16:25 : 02/02
14:16:39 : 03/03
14:17:14 : 04/04
14:17:23 : 05/05
14:17:28 : 06/06
14:17:38 : 07/07
14:17:48 : 08/08
14:17:55 : 09/09
14:18:00 : 10/10
14:18:11 : 11/11
14:18:22 : 12/12
14:18:31 : 13/13
14:18:42 : 14/14
14:18:52 : 15/15
14:18:57 : 16/16
14:18:57 : Test finished.

 ---------- 
Total: 16/16
Probability that you were guessing: 0.0%

 -- signature -- 
b5b85b1b6079102420e1022011f6d36c533e9a4c

edit: for reference I'm running foobar through wine on macOS 10.12.

Re: Vynil or digital?

Reply #143 –
i also managed to ABX the Gangnam Style samples 9/10 times (under spoiler), the very first hit sounds beefier on the 16 bit sample.  can't pick out the difference on the neil young samples on the audiocheck.net site though.

Spoiler (click to show/hide)

Re: Vynil or digital?

Reply #144 –

Quote
I've written perhaps a half million posts over a period of more than 20 years  directed at people who were silly enough to think that they could correct people like me

Hm. Certainly no-one can in the face of such practiced trolling! No human though is perfect and that includes you, so certainly you have been wrong in more than one occasion, and I can think of a few in very recent times. But its like playing whack a mole (where to start? I stopped counting the falsehoods, personal attacks and outright trolling) because the problem here isn't about the facts, its just about about making others wrong. This is easily seen in the posting history.

Just letting you know I've got your number on this one.

Re: Vynil or digital?

Reply #145 –
Quote from: Atmasphere on

Quote
I've written perhaps a half million posts over a period of more than 20 years  directed at people who were silly enough to think that they could correct people like me

Hm. Certainly no-one can in the face of such practiced trolling! No human though is perfect and that includes you, so certainly you have been wrong in more than one occasion, and I can think of a few in very recent times. But its like playing whack a mole (where to start? I stopped counting the falsehoods, personal attacks and outright trolling) because the problem here isn't about the facts, its just about about making others wrong. This is easily seen in the posting history.

Just letting you know I've got your number on this one.


Interesting that with all of the mistakes that I've  supposedly made, you seem to have posted zero comments that were anything but "This is true because I say it is true", and referenced a professional grade source for your many  exceptional claims. Yes, I've made mistakes, but with you the trick is finding you saying something that is anything but self-serving and incorrect.

Here's a reading prescription for you: 
"Distortion in Phonograph Reproduction" JAES Volume 1 Issue 1 pp. 78-85; January 1953.
"High-Frequency Stylus-Groove Relationships in Phonograph Cartridge Transducers"  JAES Volume 11 Issue 3 pp. 250-262; July 1963
"Sensitivity of Phonograph Turntables to Normal Loads" JAES Volume 16 Issue 3 pp. 354, 356; July 1968
"On Stylus Wear and Surface Noise in Phonograph Playback Systems" JAES Volume 3 Issue 1 pp. 2-18; January 1955
"Record Stylus Pressure" JAES Volume 3 Issue 2 pp. 66-69; April 1955

Note that some of these papers go back to the mid-1950s, but come to think of it, that is about when almost all of  current LP technology was invented.

Re: Vynil or digital?

Reply #146 –
Quote
Interesting that with all of the mistakes that I've  supposedly made, you seem to have posted zero comments that were anything but "This is true because I say it is true", and referenced a professional grade source for your many  exceptional claims. Yes, I've made mistakes, but with you the trick is finding you saying something that is anything but self-serving and incorrect.

Again with the personal attacks? Apparently attacking the argument isn't good enough? That's a workable definition for trolling.

Let's take the Timeline as an example. By your way of posting, it must be terrible, despite the fact that it can detect stylus drag where your method cannot. And does not have the error of the test LP. And allows you to test the machine over hours rather than just the time you set the needle down. Plus being able to see the results almost instantaneously.

I have it so I can know when the lathe is working right and when it needs service (Scully lathes need a lot of attention). Its faster and more accurate. But you don't like for a 'reason'; which your posting history suggests that you didn't think of it first. So you have to denigrate it instead of actually doing some measurements. Also based on your say-so, you won't accept any testimony I might have despite expecting that I accept yours which appears to be no more credible.

Now if you didn't already have to tell someone else here about how amazing your 1/2 million posts are, you might have been more credible; if that is want you want, don't be boasting about how credible you must be because of how many posts you've made...

Technology, even with LP reproducers and mastering systems, has marched on in the 60+ years since some of those articles were written. Stereo, for example...  And as a result, improved the technology. We've been experimenting with class D amplifiers for the lathe, just as an example. The modifications to QRP's pressing machines are another. You can contact QRP (http://www.qualityrecordpressings.com/) if you want to know more. Or Sutherland (that's what I did once I heard about the Timeline). Sutherland is a good guy and builds extremely competent gear. http://sutherlandengineering.com/products/timeline/

Re: Vynil or digital?

Reply #147 –
You see the thing is - measurement wise digital is  superior to analogue in respect to dynamical range, THD etc etc.  However, it seems to be that some people LIKE a bit of second harmonic distortion or whatever - some people seem to like the irregularities and noise, compression etc that analogue recording can introduce. I have a mate - insists on vinyl, and an old Linn speaker - I listened to it - he beamed at me - it was a distorted mess - but I dutifully told him how awesome it all sounded - why ruin his day? But this isnt just due to nostagia or confirmation bias. A recent academic study on preference of analogue and digital summing  (summing is just how you mix all your different audio tracks on to make a song)  showed preference for analogue and digital recording techniques was genre specific. So, in other words, were all the tracks mixed on an analogue desk before being rendered to a stereo (digital) file - or in a DAW (digital audio workstaion - basically in a computer) before being rendered to a stereo digital file. The fact they both ended up being rendered to digital does not make the results irrelevant - its the process of recording that introduces distortion , crosstalk etc etc The results the randomised listening test showed people preferred analogue summing for rock but digital for classical. Interestingly many sound engineers I know (live and studio) who work in rock genres say they find digital almost too clean and prefer analogue for some tasks. Anyway you can see the study here:

https://theproaudiofiles.com/wp-content/uploads/2015/01/Analog_Summing_Tarr_2014.pdf

So the horse we are flogging might not be quite as dead as some may have thought......



Re: Vynil or digital?

Reply #148 –
Quote from: Atmasphere on
Quote from: arny
Interesting that with all of the mistakes that I've  supposedly made, you seem to have posted zero comments that were anything but "This is true because I say it is true", and referenced a professional grade source for your many exceptional claims. Yes, I've made mistakes, but with you the trick is finding you saying something that is anything but self-serving and incorrect.

Again with the personal attacks?

What personal attacks other than yours?   The statement: "you seem to have posted zero comments that were anything but "This is true because I say it is true", and referenced a professional grade source for your many exceptional claims." is a statement of objective fact that of course, you could rebut if there was any true factual rebuttal.

You still keep ranting about my alleged mistakes, but with no proof or even a logical argument,  You persist in this false claim even after I explained why it is false:

Quote
Let's take the Timeline as an example. By your way of posting, it must be terrible, despite the fact that it can detect stylus drag where your method cannot.

As I pointed out, that is a false claim for perfectly logical reasons. As I said before my method duplicates the actual playing of a record which obviously includes the effect of stylus drag. Let me give it to you step by step:

(1) If one plays a test record with a test tone on it and
(2) stylus drag reduces the rotational speed of turntable,
(3) the frequency of the recorded signal is reduced in perfect proportion to the loss of speed.

Quote
And does not have the error of the test LP.

Instead, the Timeline has its own sources of error.  The Timeline has a laser that according to its documentation flashes at the same rate as the record rotates. It must use some internal timing standard to set the rate of those flashes. The precision of that flashing is an obvious source of error.

The test tone method can also accurately determine flutter and wow and display it with high precision including speed variations that may only last for a tiny fraction of a second. The timeline is said to have a read out in the form of an uncalibrated light beam on the wall.  The Test Record method is inherently calibrated and can read out rotational speed down to tiny fractions of a RPM. This gets into another shortcoming of the Timeline, in that its operation being based on the projection of light is dependent on operation with other sources of lighting being controlled to avoid washing it out.  

Quote
And allows you to test the machine over hours rather than just the time you set the needle down. Plus being able to see the results almost instantaneously.

Given that a test record can be played almost indefinitely, the above criticism is obviously false.  The test record method can provide a read out instantaneously if the test tone's frequency is read out in real time, by say, an electronic frequency counter which is a common piece of test equipment, that is even built into common digital multimeters.

Quote
I have it so I can know when the lathe is working right and when it needs service (Scully lathes need a lot of attention). Its faster and more accurate.

False claim, again based on personal say-so.  Here are the manufacturer' specs for the Timeline from the manufacture's web site:
http://sutherlandengineering.com/products/timeline/

Trouble is, there are no accuracy specifications there that I can find!  The user manual gives these items;

Batteries: 2 N size alkaline
Battery life: greater than 800 hours
Laser: 650 nm, less than 2.5 mw output
spot size 1/4 “ at 15 feet
Size: 3” diam, 1 3/4 “ high

Note, there is no accuracy specification at all.

In contrast, the accuracy of the test record method is dependent on the precision of the test record, and the means for frequency measurement, both of which can easily be within 0.005% or better.

Quote
But you don't like for a 'reason'; which your posting history suggests that you didn't think of it first.

Personal attack on my character noted.

Quote
So you have to denigrate it instead of actually doing some measurements.

I didn't denigrate it, I analyzed it objectively and critically.

Quote
Also based on your say-so, you won't accept any testimony I might have despite expecting that I accept yours which appears to be no more credible.

Where have you accepted any of the findings of mine or the peer-reviewed papers I cited. It seems like you know nothing about them and have never read them.

You obviously have no experience with the test record method, and clearly have not a clue about how to do it. or experience with any of the equipment used to perform it. If you did, your statements about it would not be rife with gross errors as I have shown.

You are ignoring the fact that my post was confirmed by another poster. Futhermore, using a test record as I have described is common practice, and it is not my fault that you are so unaware of common testing procedures.  Test records can be obtained for only a tiny fraction of the cost of the $400 Dennison Timeline record weight.

Quote
Technology, even with LP reproducers and mastering systems, has marched on in the 60+ years since some of those articles were written.

Which ones, how, and in accordance with what authority?

Quote
Stereo, for example...

Some of the articles postdate stereo and mention it. You obviously don't know which ones, and obviously you have read none of them.

Quote
And as a result, improved the technology. We've been experimenting with class D amplifiers for the lathe, just as an example.

This would address few if any of the inherent gross shortcomings of the LP that I have documented with these articles.

Quote
The modifications to QRP's pressing machines are another. You can contact QRP (http://www.qualityrecordpressings.com/) if you want to know more. Or Sutherland (that's what I did once I heard about the Timeline). Sutherland is a good guy and builds extremely competent gear. http://sutherlandengineering.com/products/timeline/

Please show how any of these changes have addressed the well known gross technical inherent failings of the LP. I don't see any connection or know of an independent unbiased authority that recognizes them.

Re: Vynil or digital?

Reply #149 –
Quote from: evgenetic on
i also managed to ABX the Gangnam Style samples 9/10 times (under spoiler), the very first hit sounds beefier on the 16 bit sample.  can't pick out the difference on the neil young samples on the audiocheck.net site though.

Spoiler (click to show/hide)

Thank you. I tried to ABX the files I posted but as I have been complaining bitterly for some time, my ears are no longer good enough for this sort of thing. The good news is that I still enjoy listening to music very much!