IS EVERYTHING DEBATABLE, REALLY?

[kumakuma]

54 minutes ago, Audiophile Neuroscience said:

 

Yesterday IIRC you said his core claim was something else. Today what you are claiming is a misrepresentation. i believe the issue is about identical checksums, one way a file is identical and therefore if different it has to be due to something else.Now I'm guessing it will be said that if checksums are identical the file is identical in every possible way known to science. Whatever the case why bang on about it unless Alex brings it up?

 

Your memory fails you.

 

I listed five interrelated claims things yesterday, only one of which (#3) Alex disputed as a misrepresentation of his claims.

 

#2 was the same as what I am calling his "core claim" because this is the one that repeats most often. Note that this is my term, not his.

 

Identical checksums is not the core issue. 

 

Identical checksums simply means that two files are identical with a very degree of probability, depending on the algorithm used. Instead of calculating checksums you can also compare two files bit-by-bit to see if they are identical. The files that Alex has sent me in the past are bit identical and also, of course, have the same checksums.

[beerandmusic]

37 minutes ago, sandyk said:

In the capacitance multiplier section of the John Linsley Hood designed PSU add-on that I use there are 2 parallel 4700uF electro's.When the original was designed , Low ESR types were never mentioned. If a Low ESR type is used in parallel with a normal type of the same capacitance, the effective bandwidth of the original design is markedly increased. However,if both types used are Low ESR types, analogue audio sounds too detailed with accentuated treble and sibilance.

You wouldn't normally use very low ESR types as used in computers ,especially not in parallel in Analogue Preamplifiers etc. However , using a selected low ESR type in parallel with a normal type of electro in the JLH normally results in a more balanced SQ as verified by the 100s of people who have used the JLHs in Pre and Power Amplifiers.

It has been shown elsewhere in this thread , that even the waveform saved on a HDD is actually an ANALOGUE waveform with all it's vagaries before being processed.

Alex

false....

main difference being one signal (analog) is continuous with infinite values ....

in contrast, a digital signal, inre music, can only provide a value at time Tx where there are unlimited samples with analog.

 

I am not going to try and convince anyone, but facts are facts.

 

Audio engineers may try to design a high speed digital circuit as if it were analog (as stated in the article that some may misinterpret), but it is stored as a digital file on a hard drive....sorry...i won't bother responding to any more about how music file is stored on a hard drive...you either get it or you don't.

[pkane2001]

Just now, beerandmusic said:

false....

main difference being one signal (analog) is continuous with infinite values ....

in contrast, a digital signal, inre music, can only provide a value at time Tx where there are unlimited samples with analog.

 

I am not going to try and convince anyone, but facts are facts.

 

Audio engineers may try to design a high speed digital circuit as if it were analog (as stated in the article that some may misinterpret), but it is stored as a digital file on a hard drive....sorry...i won't bother responding to any more about how music file is stored on a hard drive...you either get it or you don't.

 

Not that it matters, but let me throw this in: there's no such thing as an analog signal with continuous infinite values. Charge (and current) is quantized in units of one electron. So, underneath it all is still a digital signal using unary system 

 

[sandyk]

22 minutes ago, beerandmusic said:

false....

main difference being one signal (analog) is continuous with infinite values ....

in contrast, a digital signal, inre music, can only provide a value at time Tx where there are unlimited samples with analog.

 

I am not going to try and convince anyone, but facts are facts.

 

Audio engineers may try to design a high speed digital circuit as if it were analog (as stated in the article that some may misinterpret), but it is stored as a digital file on a hard drive....sorry...i won't bother responding to any more about how music file is stored on a hard drive...you either get it or you don't.

Despite the evidence posted showing an actual waveform stored on an HDD, you still refuse to accept that it doesn't become a binary waveform without quite a bit of further processing.  

 

You will never get it !

[Audiophile Neuroscience]

 

 

41 minutes ago, kumakuma said:

 

Your memory fails you.

 

It often does but apparently not in this instance.

 

Quote

 

I listed five interrelated claims things yesterday, only one of which (#3) Alex disputed as a misrepresentation of his claims.

 

#2 was the same as what I am calling his "core claim" because this is the one that repeats most often. Note that this is my term, not his.

 

 

Yes you listed 5 claims, none identified as core.

 

22 hours ago, kumakuma said:

 let me list some of the things that Alex has claimed over the years.

 

1. Two songs from CDs ripped on different machines may sound different when played back in the same environment and these differences remain when the files are transferred to a different storage media or the files are sent over the Internet even though both files are bit for bit identical.  In other words, there is some intrinsic difference between identical digital data that is external to it's bits and bytes.

 

2. Transferring a file over the Internet causes a deterioration in sound quality even if the digital data at both ends is identical.

 

3. Simply opening a file in a program and closing it again with no changes to the file causes a deterioration in sound quality although no data has changed.

 

4. Converting a file between lossless formats (WAV to AIFF and back to WAV) causes a deterioration in sound quality although digital data before and after this conversion is identical.

 

5. Zipping a file before sending it over the Internet protects it from sound deterioration although there is no difference between a "naked" file and a file that has been zipped and unzipped.

 

 

In a separate post yesterday you *did* identify the core claim...

 

21 hours ago, kumakuma said:

 

At it's core, Alex's claim is that noise can be captured in digital data outside the bits that make up the data and that noise travels with this digital data when it copied, transferred over the Internet, or written to the CD.

 

and today a different core claim

 

2 hours ago, kumakuma said:

 

His core claim is that he can create two identical files that sound different, send these files to someone over the Internet, and have the recipient hear the same differences between these two identical files.

 

 

Your statement as "core" , 1."Alex's claim is that noise can be captured in digital data outside the bits that make up the data " IMO is different to your statement 2. as "core" "he can create two identical files that sound different"

 

Statement 2 is IMO, and as said, a misrepresentation.

 

The underlined part in statement 1 is yours, not mine.

[fas42]

1 hour ago, sandyk said:

In the capacitance multiplier section of the John Linsley Hood designed PSU add-on that I use there are 2 parallel 4700uF electro's.When the original was designed , Low ESR types were never mentioned. If a Low ESR type is used in parallel with a normal type of the same capacitance, the effective bandwidth of the original design is markedly increased. However,if both types used are Low ESR types, analogue audio sounds too detailed with accentuated treble and sibilance.

You wouldn't normally use very low ESR types as used in computers ,especially not in parallel in Analogue Preamplifiers etc. However , using a selected low ESR type in parallel with a normal type of electro in the JLH normally results in a more balanced SQ as verified by the 100s of people who have used the JLHs in Pre and Power Amplifiers.

It has been shown elsewhere in this thread , that even the waveform saved on a HDD is actually an ANALOGUE waveform with all it's vagaries before being processed.

Alex

 

Just to note, that paralleling capacitors should not be done lightly - there are precise combinations of values and types that should be fine, and others that can get you into trouble. The reasons are purely technical - the parasitic characteristics of the caps' constructions, such as ESR, and how they're mounted can mean there is a resonant impedance - the combo will ring, and this will most likely degrade the SQ.

[beerandmusic]

10 minutes ago, sandyk said:

Despite the evidence posted showing an actual waveform stored on an HDD, you still refuse to accept that it doesn't become a binary waveform without quite a bit of further processing.  

 

You will never get it !

 

my statement of fact is that a music file stored on a hdd is a digital file....nothing will ever change that fact.

[kumakuma]

1 minute ago, Audiophile Neuroscience said:

Your statement as "core" , 1."Alex's claim is that noise can be captured in digital data outside the bits that make up the data " IMO is different to your statement 2. as "core" "he can create two identical files that sound different"

 

Statement 2 is IMO, and as said, a misrepresentation.

 

Both of my statements refer to the same thing.

 

The only difference between them is that yesterday's statement included the mechanism (embedded noise) that Alex has repeatedly said is what he believes is the cause of the difference that he hears between the two identical files.

 

Statement 2 is not a misrepresentation of Alex's claims. He has said that ripping CDs on different computers with different power supplies will create identical files (rips) that sound different. In other words, he is creating these files by ripping CDs.

[sandyk]

17 minutes ago, fas42 said:

 

Just to note, that paralleling capacitors should not be done lightly - there are precise combinations of values and types that should be fine, and others that can get you into trouble. The reasons are purely technical - the parasitic characteristics of the caps' constructions, such as ESR, and how they're mounted can mean there is a resonant impedance - the combo will ring, and this will most likely degrade the SQ.

Ceramic capacitors are particularly troublesome in this respect.

I don't use extremely low ESR types specifically designed for Motherboard use .

Neither do I use Panasonic FM types that John Swenson appears to like. I don't have a problem with Panasonic FC, just not 2 FCs in parallel.

[opus101]

1 hour ago, sandyk said:

In the capacitance multiplier section of the John Linsley Hood designed PSU add-on that I use there are 2 parallel 4700uF electro's.When the original was designed , Low ESR types were never mentioned. If a Low ESR type is used in parallel with a normal type of the same capacitance, the effective bandwidth of the original design is markedly increased. However,if both types used are Low ESR types, analogue audio sounds too detailed with accentuated treble and sibilance.

You wouldn't normally use very low ESR types as used in computers ,especially not in parallel in Analogue Preamplifiers etc. However , using a selected low ESR type in parallel with a normal type of electro in the JLH normally results in a more balanced SQ as verified by the 100s of people who have used the JLHs in Pre and Power Amplifiers.

 

 

Before attributing sibilance to the use of low ESR caps I'd want to take a look at the grounding of such caps. Given their lower impedance above 3kHz the end result is much more sensitive to layout.

 

In my experience when sibilance is associated with caps, its never been the cap itself rather I've used a sub-optimal attach point in the circuit for it. I can and do use Nichicon HZ in my line-level circuits to achieve the lowest supply impedance above 3kHz.

[Audiophile Neuroscience]

27 minutes ago, kumakuma said:

The only difference between them is that yesterday's statement included the mechanism (embedded noise) that Alex has repeatedly said is what he believes is the cause of the difference that he hears between the two identical files.

 

Statement 2 is not a misrepresentation of Alex's claims. He has said that ripping CDs on different computers with different power supplies will create identical files (rips) that sound different. In other words, he is creating these files by ripping CDs.

 

This is probably the crux of the issue. Logically, two identical anythings cannot by definition be different. They can be identical in some or most ways but not all ways. I get what you are saying and accept you were not misrepresenting Alex per se. I apologize.

 

I still think that leaving out the "cause of the difference" is misleading. Two truly "identical" files cannot sound differently (barring perturbations and anomalies of the playback system or listener).That is why I have understood that Alex has specified, or at least meant, checksums are the same with different ripping methods. He has also said he doesn't know why they sound different, speculates about possible reasons, and that is a perfectly scientific and rational stance. It may not be probable as believed by others but I don't think it is impossible from what has been said.

 

Yes, Alex did play me two different rips which had the same checksums and yes I did (think) I heard subtle differences on that occasion.This is hardly scientifc proof and I perhaps more than most would not entertain it as proof.

 

Edit- cue the wgscott insults if the above is quoted.

[kumakuma]

6 minutes ago, Audiophile Neuroscience said:

 

This is probably the crux of the issue. Logically, two identical anythings cannot by definition be different. They can be identical in some or most ways but not all ways. I get what you are saying and accept you were not misrepresenting Alex per se. I apologize.

 

I still think that leaving out the "cause of the difference" is misleading. Two truly "identical" files cannot sound differently (barring perturbations and anomalies of the playback system or listener).That is why I have understood that Alex has specified, or at least meant, checksums are the same with different ripping methods. He has also said he doesn't know why they sound different, speculates about possible reasons, and that is a perfectly scientific and rational stance. It may not be probable as believed by others but I don't think it is impossible from what has been said.

 

Yes, Alex did play me two different rips which had the same checksums and yes I did (think) I heard subtle differences on that occasion.This is hardly scientifc proof and I perhaps more than most would not entertain it as proof.

 

I'm glad we are finally on the same page.

 

I don't think that including a "cause of the difference" makes sense as none of the probable causes that Alex has come up with are technically possible based on the facts that he has laid out.

[Speed Racer]

40 minutes ago, sandyk said:

Despite the evidence posted showing an actual waveform stored on an HDD, you still refuse to accept that it doesn't become a binary waveform without quite a bit of further processing.  

 

You will never get it !

 

OMG!! LOL!!!! No, no, no!  Waveforms are not stored on the hard drive. All files are stored as a bunch of bits on the hard drive and whether the bits are ones or zeros is controlled by whether or not they are magnetized.

 

Are you really that ignorant of how data is stored?

 

 

[Audiophile Neuroscience]

21 minutes ago, Speed Racer said:

 

OMG!! LOL!!!! No, no, no!  Waveforms are not stored on the hard drive. All files are stored as a bunch of bits on the hard drive and whether the bits are ones or zeros is controlled by whether or not they are magnetized.

 

Are you really that ignorant of how data is stored?

 

 

 

Now dont crucify me here.I keep on saying I am not an engineer. How can mathematical abstractions of a 1 or a 0 be magnetized? Something is magnetized - isn't that that something an analog representation, albeit binary (either this or that),of 1s and 0's?

[davide256]

2 hours ago, sandyk said:

In the capacitance multiplier section of the John Linsley Hood designed PSU add-on that I use there are 2 parallel 4700uF electro's.When the original was designed , Low ESR types were never mentioned. If a Low ESR type is used in parallel with a normal type of the same capacitance, the effective bandwidth of the original design is markedly increased. However,if both types used are Low ESR types, analogue audio sounds too detailed with accentuated treble and sibilance.

You wouldn't normally use very low ESR types as used in computers ,especially not in parallel in Analogue Preamplifiers etc. However , using a selected low ESR type in parallel with a normal type of electro in the JLH normally results in a more balanced SQ as verified by the 100s of people who have used the JLHs in Pre and Power Amplifiers.

It has been shown elsewhere in this thread , that even the waveform saved on a HDD is actually an ANALOGUE waveform with all it's vagaries before being processed.

Alex

Sandyk, magnetic tape can be analog but a HDD has no mechanism to capture an analog signal... data is always binary when recorded on a HDD, has to be decoded for time domain and amplitude

[sandyk]

6 minutes ago, Speed Racer said:

 

OMG!! LOL!!!! No, no, no!  Waveforms are not stored on the hard drive. All files are stored as a bunch of bits on the hard drive and whether the bits are ones or zeros is controlled by whether or not they are magnetized.

 

Are you really that ignorant of how data is stored?

 

 

You are deliberately misrepresenting what I have said about the e.e times link that I posted. The illustration shows the RECOVERED electrical waveform before being processed.

Did you even  bother to read the link that I provided ?

 

[kumakuma]

12 minutes ago, Audiophile Neuroscience said:

 

Now dont crucify me here.I keep on saying I am not an engineer. How can mathematical abstractions of a 1 or a 0 be magnetized? Something is magnetized - inst that that something an analog representation, albeit binary (either this or that),of 1s and 0's?

 

[sandyk]

8 minutes ago, davide256 said:

Sandyk, magnetic tape can be analog but a HDD has no mechanism to capture an analog signal... data is always binary when recorded on a HDD, has to be decoded for time domain and amplitude

I am not disputing that. I am talking about the RECOVERED electrical waveform before processing.

Low battery warning.

[Audiophile Neuroscience]

4 minutes ago, kumakuma said:

 

 so I'm still seeing an analog representation of digital, as seen by the read/write heads, even when stored. Is that not fair to say. Digital is an abstraction.

[kumakuma]

4 minutes ago, Audiophile Neuroscience said:

 so I'm still seeing an analog representation of digital, as seen by the read/write heads, even when stored. Is that not fair to say. Digital is an abstraction.

 

You've been hanging out with the wrong crowd here on CA.

[gmgraves]

On 11/5/2017 at 10:36 AM, beerandmusic said:

Does the Phasure do native DSD?

imho, no dac can claim they are the best if they don't.

 

We've had this discussion before. DSD, while batter than LPCM, is a minuscule portion of the streaming and music download market. Most DACs that have it have it because they are delta-sigma or 1-bit DACs. companies that decide to include the feature, do so because many delta-sigma chips have the feature built in and to break it out and offer it as a feature is fairly trivial. Many processor boxes that use delta-sigma chips with a DSD input pin, choose not to incorporate the feature in their design because of minuscule program material support. If DSD ever takes off and DSD program material becomes more available, I'm sure more processor manufacturers will go back and design the feature into their next generation processors. But for the here and now, when people compare digital processing boxes, they are talking about LPCM and not DSD. So, my advice is keep that in mind when you see someone hailing this DAC or that DAC as the worlds best. 

[gmgraves]

On 11/5/2017 at 3:23 PM, fas42 said:

 

Any system of reasonable quality, that has sufficient headroom with regard to the amplifier combined the particular speakers - you can't combine a SET with ribbons, say, and expect the SPLs to happen - has the potential. As said earlier, the very lowest bass notes would be more difficult to perfectly mimic - but for 99.99% of recorded piano works there wouldn't be a problem.

 

At the moment, it is almost impossible to buy a combo that is good enough in raw form to get such a result - how to transition to it being more common I'm not sure. I suspect it will be a learning process; people will need to understand what areas to focus on, to get the job done. For myself, I have a background such that I've been happy to hack cheap gear to get what I'm after - expensive, bling components being fiddled with would make all concerned nervous, including myself!

 

Not quite sure what you mean by "equating near perfect and horrible" ... if a system can produce the necessary SPLs, but it sounds very sharp and unpleasant to the ears doing this then the signs are good; something that is terribly, terribly polite while playing everything is going to be far harder to "fix".

 

One would have to find a solo piano recording where the recording engineer did not put the microphones on a piano extension bar inside the piano before one could try such a test. The way pianos are miked nowadays, they end up, in playback, 10 (or more) feet wide with the treble end of the keyboard coming out of the right speaker and the bass end from the left speaker. First of all, no real piano sounds like a piano miked from inside the piano, and secondly, nobody is going to be fooled by a room-wide piano! Perhaps if one put right and left speakers together, next to one another...

 

https://www.sweetwater.com/store/detail/PianoMicSys

[pkane2001]

2 minutes ago, gmgraves said:

 

We've had this discussion before. DSD, while batter than LPCM, is a minuscule portion of the streaming and music download market. Most DACs that have it have it because they are delta-sigma or 1-bit DACs. companies that decide to include the feature, do so because many delta-sigma chips have the feature built in and to break it out and offer it as a feature is fairly trivial. Many processor boxes that use delta-sigma chips with a DSD input pin, choose not to incorporate the feature in their design because of minuscule program material support. If DSD ever takes off and DSD program material becomes more available, I'm sure more processor manufacturers will go back and design the feature into their next generation processors. But for the here and now, when people compare digital processing boxes, they are talking about LPCM and not DSD. So, my advice is keep that in mind when you see someone hailing this DAC or that DAC as the worlds best. 

 

Don’t forget that there are those who prefer to upscale their PCM content to higher rate DSD. That does produce an audible difference with DACs capable of native DSD, IME. Whether or not you prefer that difference is obviously subjective. I tend to go back and forth. I prefer PCM with orchestra and hard rock, but like DSD on vocals, guitar, and chamber music.

[gmgraves]

4 minutes ago, pkane2001 said:

 

Don’t forget that there are those who prefer to upscale their PCM content to higher rate DSD. That does produce an audible difference with DACs capable of native DSD, IME. Whether or not you prefer that difference is obviously subjective. I tend to go back and forth. I prefer PCM with orchestra and hard rock, but like DSD on vocals, guitar, and chamber music.

 

 

While true, I think bringing that up confuses the issue. First off, upscaling PCM to DSD adds no information to the music. Many sigma-delta DACs do that anyway. I have often said that my Sony SCD XA777ES SACD player was the best sounding REDBOOK stand-alone CD player I've ever heard and it converted CDs to DSD before outputting them as audio. 

[pkane2001]

9 minutes ago, gmgraves said:

 

 

While true, I think bringing that up confuses the issue. First off, upscaling PCM to DSD adds no information to the music. Many sigma-delta DACs do that anyway. I have often said that my Sony SCD XA777ES SACD player was the best sounding REDBOOK stand-alone CD player I've ever heard and it converted CDs to DSD before outputting them as audio. 

 

The reason I brought this up is that when considering a DAC, the support for native DSD might still be important even if one has no DSD content.