Jump to content
IGNORED

Lies about vinyl vs digital


Recommended Posts

Just now, mansr said:

There are many ways that noise can corrupt a digital data transfer. That's not the issue here. Alex claims there is "something" different in the files even when they have been verified to be bit for bit identical, i.e. no errors occurred during the transfer.

That claim can be a subject of interpretation also.   In chatting with Alex, it seems to be more finely defined as a difference in the behavior of electronics (CPU, noise, etc) when playing flac vs. wav.   Most .wav files are naturally more simple (e.g. most are 16bit signed integer) and require little processing.  There could even be a difference in floating point .wav files because of powering up the AVX sections of the CPU chips causing differing low frequency noise patterns and different coupled & ground noise effects.  Flac files, of course, require much more processing and will have even more different noise patterns in the supply, ground, coupled effects.


I don't expect this effect to be significant in most cases, but I can imagine anecdotal situations & conditions where it is true.

 

On the other hand, the same file, with exactly the same data, being passed through different data paths, both ending up on the same filesystem for ultimate playback using the same software -- DO sound identical.  Minor difference in file wrappers (most .wav files are simple wrappers of signed integer or floating point data. with very minor amounts of metadata) don't make any practical audible difference at all.

 

There is a syndrome, often missed by people who have set-beliefs about hearing -- they over estimate their ability to remember sequences of sound longer than a few seconds.  I do expect that there are a lot of anecdotal claims about the same data sounding different from playback to playback.  Of course, that 'sounding different' of identical files might be true, but the difference is more in the listening apparatus and brain of the individual, not so much in the transmitted acoustic signal.

 

Myself, I cannot remember sound sequences longer than about 12 seconds before my comparisons become inaccurate.   In effect, that limits me to comparing two sequences of 6 seconds each.   Any claim that two 3minute recordings sound the same or different (I mean, technically, in detail) are specious when it concerns my own human wetware.  (I have huge experience in doing accurate comparisons, as inaccurate comparisons waste lots of my time and embarass me with bad results.)

 

John

 

Link to comment
Just now, marce said:

We are talking about digital data stored in a buffer or memory, not data copied to a CD, ergo how can the differences be stored with the data.

I have an open mind about how things really work, but the fact is bit identical files store the same information and unlike analogue data storage mediums (tape etc.) the noise is not stored with the data... 

 

Well, if the files are identical, even if some minor changes in metadata, will sound the same.   Otherwise, the human auditory/human audio perception matters become significant.

 

I made the mistake once of doing an ad-hoc comparision -- that was my final straw on ANY trust of my hearing except for super short sequences as I noted above.

 

As I mentioned above -- any comparison other than very short sequences are suspect -- PERIOD!!!   There are probably some savants who can do long comparisons accurately, but even as someone who had perfect pitch as a young person, I cannot do long comparsions...

 

Whenever someone makes a claim, I know that 99.9% of the people do not lie, but there must be explanations for their claims.  That is why I have a very open mind, and actually investigate their claims.

 

John

 

Link to comment
9 hours ago, mansr said:

Remember that Alex claims an SMPS used while ripping a CD (or more generally writing a file to storage) will poison that file such that it can be heard when played back regardless of what power supply is used then. I think most half-way reasonable people would disagree with that.

 As is so often the case, you are either deliberately misquoting me, or leaving parts out . What you hear both at the Ripping and Play stages is governed to some extent by the quality of the PSUs  used,whether with internal storage or external storage , and numerous members in other areas of the forum will agree with me .

This also includes the many members, and others world wide who have fitted Uptone Linear PSUs in their Mac Minis.

The PSU in my PC IS SMPS, but the critical areas such as internal SSDs and the internal writer have additional voltage regulation, but not the HDDs .

 I have also stated that it is possible to REGENERATE a file to improve it's Signal Integrity, and this can also be done using an Ether Regen as is being currently demonstrated.

 Unfortunately, you, and several others, are not interested in finding these things out for yourself despite the numerous times I have offered to provide proof, even to the extent of previously sending comparison BR Music Video discs to Dennis in the USA, and on a few occasions making direct offers to some members.

 

 Neither is this the thread to discuss this further , as it is well off topic.

 

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment
4 hours ago, John Dyson said:

Well, if the files are identical, even if some minor changes in metadata, will sound the same.   Otherwise, the human auditory/human audio perception matters become significant.

 

I made the mistake once of doing an ad-hoc comparision -- that was my final straw on ANY trust of my hearing except for super short sequences as I noted above.

IF the files were truly IDENTICAL , you would not hear any differences.

You have already demonstrated that it is possible to hear differences of as little as +- 0.1dB !!!

 You are also well aware that a recipient (no names please) of your Supertramp files was recently able to hear, and accurately describe clear differences between your original version and a warmer version that I created , yet we both still preferred the immediacy of your original version.

 Again, this IS off topic and has no place in this thread.

Please take it to PMs

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment
8 hours ago, John Dyson said:

There is a syndrome, often missed by people who have set-beliefs about hearing -- they over estimate their ability to remember sequences of sound longer than a few seconds.  I do expect that there are a lot of anecdotal claims about the same data sounding different from playback to playback.  Of course, that 'sounding different' of identical files might be true, but the difference is more in the listening apparatus and brain of the individual, not so much in the transmitted acoustic signal.

 

 

The trivial example, that I am reminded of here with my current laptop, of how the sound does change from playback to playback is that the prior history of the chain dictates some of the qualities that are heard; IOW, the amount of warmup and conditioning figures. Anyone with the slightest interest in motor sport and tyres knows that the behaviour of the tyres is a constant variable, especially when pushing them to their limits, it's always a balancing act - unfortunately, the same is true for audio, if one is pushing for the best that can be delivered.

Link to comment
6 hours ago, John Dyson said:

Well, if the files are identical, even if some minor changes in metadata, will sound the same.   Otherwise, the human auditory/human audio perception matters become significant.

 

I made the mistake once of doing an ad-hoc comparision -- that was my final straw on ANY trust of my hearing except for super short sequences as I noted above.

 

As I mentioned above -- any comparison other than very short sequences are suspect -- PERIOD!!!   There are probably some savants who can do long comparisons accurately, but even as someone who had perfect pitch as a young person, I cannot do long comparsions...

 

 

 

The best technique I have found if I want to nail that there is something audibly different actually encoded between two, say, files - rather than select which is least damaged, which is what I am normally focused on - is, ummm, the rhythm method, 😁. Load both into something like Audacity, make sure they are in sync, select a very short, likely section - and solo that section in repeat, for the first track. This is no longer music, it's now a mantra of sound, a rhythmic pattern - do this long enough that it becomes 'hypnotic'; and then switch, without interruption, to the other track. The slightest variation is quite obvious - because the audible pattern is minutely disturbed.

 

This of course tells you absolutely zero about "which is better" - it merely allows one to say that there is an audible difference.

Link to comment
9 hours ago, mansr said:

There are many ways that noise can corrupt a digital data transfer. That's not the issue here. Alex claims there is "something" different in the files even when they have been verified to be bit for bit identical, i.e. no errors occurred during the transfer.

 

It's called Bit Infection and is caused by a corona-virus that affects the brain.

Link to comment
1 hour ago, sandyk said:

 I have also stated that it is possible to REGENERATE a file to improve it's Signal Integrity, and this can also be done using an Ether Regen as is being currently demonstrated.

 

 

Which indicates you understand that integrity of the signal as digital data is never lost, Alex. What you unfortunately often imply is that somehow the precise method of processing the data actually infects the integrity of that data - which can never happen, if done without error.

 

IOW, if you don't like the quality of some file while playing it from some copy made however, then all one has to do is transfer it to some other form of storage, REGENERATE if you like that word. The problem is that you constantly suggest that the history of that information somehow taints the signal when viewed purely as data - which it can't.

Link to comment
On January 6, 2020 at 8:39 PM, AudioDoctor said:

Today has been a weird day, and its about to get weirder...

 

I agree with you GUTB. I think Digital has to be taken to a pretty extreme level to match analog. Right now, I appreciate my analog more than the digital and that is probably due to the equipment.

 

Probably... But more likely due to the mastering of the albums in question.

Link to comment
44 minutes ago, marce said:

REGENERATE

An interesting word, with digital you regenerate the file every time you play it, through multiple buffers on its journey from source to speaker, you'd be surprised if you followed the signal on its physical path.

LP's every play is DEGENERATE, it gets smoother the more you play it....

 Nevertheless, quite a few members are able to even hear differences between different types of RAM.

 So much for your numerous buffers .:D

 N.B.

 I haven't made that claim, I am simply reporting it.

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment
7 hours ago, marce said:

but the fact is bit identical files store the same information and unlike analogue data storage mediums (tape etc.) the noise is not stored with the data... 

 According to John Swenson who has worked in HDD fabrication, that is not correct.

Noise IS stored along with the Data, albeit at a level he believes should not affect the results

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment
10 minutes ago, sandyk said:

 According to John Swenson who has worked in HDD fabrication, that is not correct.

Noise IS stored along with the Data, albeit at a level he believes should not affect the results

 

You see, you're confusing the analogue mechanism that actually stores the data, with the concept of data. As John Dyson points out, there can be audible consequences, of 'noise' when retrieving that data, no matter how tenuous - but all you have to do is completely separate the retrieval process, from the conversion and analogue amplification process, by some means to completely eliminate any issues.

Link to comment
7 minutes ago, Ralf11 said:

Nevertheless, quite a few members are so affected by confirmation bias that they think they are able to even hear differences between different types of RAM.

The problem is not just cofirmation bias, but mostly related to each us (everyone except a prodigy) cannot accurately remember TECHNICAL details about long sequences of audio signals.  Like a lot of aspects of our senses, we over-estimate our ability becuase of various illusions.  Even with the true perfect pitch when younger (I mean, I was damned good), I cannot accurately remember all of the details of long sequences of audio.   I know and knew this all along, but even I can be fooled by my own mis-estimation of my hearing.

 

All of this is why I have a hard and fast rule of distrust of my own senses for long sequences of data.  In fact, becuase of a lot of self-verification, I quickly notice impairments in my hearing -- because if I start getting results that are inconsistent with known technical reality, then it is time to clean-out the ears or go to the doctor!!!

 

John

 

Link to comment

The very helpful aspect about what I call convincing sound is that the trigger is welded into place inside one's head - no happy how happy or sad you are, the amount of alcohol previously consumed, how much you hate or love the look of the gear you are sighting ... it either switches on, or off, completely dependably.

 

So, having to remember "what it sounded like" is completely irrelevant - all that matters is how close you are, at that moment, to hurdling the SQ barrier ...

Link to comment
11 hours ago, John Dyson said:

If you are talking about the CD recording stability/quality (not audio signal itself.) It is all about the strength and timing of the recorded signal on the CD itself.  So, if there are larger variations in the timing or variations in the strength, there can be different errors.  This is definitely manifested on a larger scale basis using entirely different media and different recorders.   When a voltage or noise level varies, the resulting behavior of the record electronics (naturally analog, even if in a digital circuit) can vary.   Powering a laser diode is based upon an analog voltage and a driver with analog requirements.

 

I am not claiming that this effect is all that common, but that keeping an open mind about how things REALLY work can help to explain the exceptional occurances that are often anecdotally reported.

 

John

 

I wanted to clarify my comment about 'errors' on CD.  The data is digital -- definitely.  The quality of the digital recording can cause digital detection errors, as these are not analog level errors.  Digital detection errors are hidden by the multi-level protection mechanism, but you want to use those protections as seldom as possible.  If these digital errors gang-up on the correction software/hw, then eventually it will give-way and create a glitch in the sound.

Analog style errors (like Laserdisk has) just don't happen on CD -- but equally bad things can happen.  With all of the protection, all but the worst CDs and marginal detection design and marginal recording design, mostly the signal is pristine from a CD.

 

John

 

Link to comment
13 minutes ago, AudioDoctor said:


Every. Single. Album?

 

 

Regarding mastering...  If you have a CD that doesn't sound 'quite right', and the equivalent vinyl...  Esp if the CD is from before about 1994 or mastered back then, I can probably come fairly close to the sound of the vinyl with my software.   I need perhaps 2-3 significant snippets from the CD, perhaps 55seconds each.  The snippets have to have good amount of HF detail of different types.  I can usually guestimate a reasonable set of decoding parameters, and then the result will come 'close' to the vinyl.  On the other hand, if I have the entire CD and a rip of a good portion of the album -- spending about 15-20minutes trying the various common EQ settings -- the result will probably be similar to the vinyl from before it was ever played.

 

The CDs have basically all of the information that was on the vinyl, but it is a bit mangled.   The key is to de-mangle it.

 

Differences in mastering is where many of the CDs don't sound quite as good as the vinyl.

We have had a pre-historic version of MQA all along -- just didnt know it :-).

 

John

 

 

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...