[bfreedma]

.

Ask pretty much any “audiophile” question and you get audiophile BS answers. I hope they sort out the learning model, etc. It would be interesting to see what 4.0 comes up with but I won’t hold my breath that it can come up with the truth yet, I think we’re still a long way from that unfortunately.

Agreed. A simple reworking of the question gets a different response. Still not accurate, but at least less definitive.

Follow on to earlier Q.
This response is driven by marketing material. Please rewrite using only peer reviewed and evidence based sources
ChatGPT 3.5
Studies comparing high-resolution audio to CD quality have shown mixed results regarding perceptible differences to listeners. While some research suggests that individuals with trained ears and high-quality equipment may perceive improved sound quality with high-resolution audio, other studies have found no significant difference in subjective listening tests. Factors such as listener preferences, room acoustics, and equipment quality can also influence perceived differences. Therefore, the benefits of high-resolution audio compared to CD quality remain subjective and may vary among listeners.

 

[gregorio]

There are certainly some indicators that the poster is utilizing a machine learning solution (the vast majority of what is being called AI is actually ML) but the heavy formatting makes me suspect that the OP is using ML to generate responses that he then edits to fit his desired position and adds color to highlight elements of the ML generated text he believes supports that position.

This appears not only likely but extremely likely IMHO. It would explain why some/many of the sentences and sections in his posts are grammatically and syntactically correct while others indicate someone with a significantly poorer grasp of the English language.

G

 

[eq1849]

The band cooperates with your blind test? Yeah, right.

If I don't know a bar or wedding reception has a live band/musician and I can tell there is live music playing before I see the band/musician, and this scenario occurs quite often in life, then absolutely I'm going to make an observation about that.

Science is also about making observations, after all...

 

[eq1849]

If that is the case (that it is not 1bit DSD) then it’s not necessarily inferior to PCM. In other words, it would be just as useless as hi-res PCM as a music distribution/playback format rather than even more useless!

G

Thank you for your opinion.

 

[gregorio]

Still not accurate, but at least less definitive.

I think accuracy is a long way away, evaluation of exactly what information and scientific research is valid and applicable, and then weighing it all appropriately is seemingly well beyond ML currently.

Science is also about making observations, after all...

Not really, that is just the very first step and only of some science. If it is, then the next step would be verifying those observations are actually real and that they pertain to what is claimed, after that, then the rest of the scientific method can kick in. Unfortunately, that second step is where many audiophile claims fail and so the scientific method/further research is not applicable.

G

 

[eq1849]

I already provided my sighted listening biases on my previous posts earlier, but ultimately succumb to the power of placebo when put to the golden standard ABX test

But the bigger question is....

if you fail to pass a listening test(ABX, AX, AB, duo trio, etc. whatever you do) what do you then do about a piece of gear that you subjectively enjoy more so than another piece of gear, but the listening test indicates you cannot hear a difference?

 

[bfreedma]

I think accuracy is a long way away, evaluation of exactly what information and scientific research is valid and applicable, and then weighing it all appropriately is seemingly well beyond ML currently.

Yup. That’s going to require “real” AI that can make independent decisions about the validity of the training data. We’re some time away from that being commercially available. Until then, ML responses are heavily driven by the biases of the coders who decide what content to utilize in their LLM and the quality of that content

Most frequently asked question of Chatbot/Virtual Agent customers: Why are my VA responses so bad/limited/inaccurate?

Response, though obviously softened: Why is your historical data and stored knowledge so full of garbage?

 

[bfreedma]

But the bigger question is....

if you fail to pass a listening test(ABX, AX, AB, duo trio, etc. whatever you do) what do you then do about a piece of gear that you subjectively enjoy more so than another piece of gear, but the listening test indicates you cannot hear a difference?

Enjoy the gear that you like, but refrain from making technical claims without supporting evidence works for me.

 

[BS5711]

I really love your picture of circles/lady, it looks cool

Fabulous, I really enjoyed your little greyscale puzzles also, such fun.

———————————————————

Honestly, anybody that doesn’t believe this person is manipulative, narcissistic, passive aggressive and probably has something of a personality disorder has been taken in by the manipulation and gameplay.

 

[theveterans]

But the bigger question is....

if you fail to pass a listening test(ABX, AX, AB, duo trio, etc. whatever you do) what do you then do about a piece of gear that you subjectively enjoy more so than another piece of gear, but the listening test indicates you cannot hear a difference?

I treat it as a fun hobby where I succumb to sighted listening deliberately, heck I don't even need to concentrate to hear a difference in IEM cable, DACs in sighted listening as the biases of sight, brain trickery and mood/feelings just overpowers my rational thoughts at least for me as an individual. But I 100% respect the concepts of audio science as I have tried a single blind volume match ABX test myself through a foobar comparator and utterly couldn't get at least 90% correct, just mere guessing

 

[bfreedma]

One of the misnomers I frequently observe from audiophiles who stop by Sound Science is conflating the desire to be technically accurate regarding component performance with a race to the bottom where everyone must purchase the least expensive audibly transparent components available.

For me, other factors are esthetics, build quality, potential resale value, and a personal preference to support smaller companies. Obviously, YMMV.

 

[danadam]

on 44.1 KHz original sample rate so long as reconstructed analog output provides almost 0.907 at Nyquist, yes that's inaudible (0.1 dB delta) from the ideal 1.0, and Sinc-L 16 million tap filter would probably be around 0.990 at Nyquist

What exactly are you referring to with those numbers? I can't tell if mean the width of the pass band or level of the signal (but why at Nytquist)?

If it only reached to stopband attenuation at 24kHz, it still would perfectly reconstruct the signal to 20kHz but there would also be some aliasing between 20kHz and 22kHz.

Aliasing? In a DAC? Or did you mean images, between 22 and 24 kHz.

 

[theveterans]

What exactly are you referring to with those numbers? I can't tell if mean the width of the pass band or level of the signal (but why at Nytquist)?

Frequency response where the linearity starts to roll-off. So in case of a "typical" DAC with sharp roll off filters, this is at 20KHz (0.907 x 22.05 KHz). Those fancy Sinc-M or Sinc-L filters from HQPlayer and Schiit's megaburrito filter push the linearity very close to the Nyquist. Here's Schiit's filter described in detail

Found this from a 2015 interview with Mike Moffat:
https://www.head-fi.org/threads/sch...most-improbable-start-up.701900/post-15763606
Can you provide some details on the Closed-Form Digital Filter used in the Yggdrasil?
The below are the claims of the Digital Filter/Interpolator in the Yggy/MB Gungnir/MB Bifrost:

1. The filter is absolutely proprietary.
2. The development tools and coefficient calculator to derive the above filters are also proprietary.
3. The math involved in developing the filter and calculating has a closed form solution. It is not an approximation, as all other filters I have studied. Therefore, all of the original samples are output. This could be referred to fairly as bit perfect; what comes in goes out.
4. Oversimplified, however essentially correct: The filter is also time domain optimized which means the phase info in the original samples are averaged in the time domain with the filter generated interpolated samples to for corrected minimum phase shift as a function of frequency from DC to the percentage of Nyquist - in our case .985. Time domain is well defined at DC - the playback device behaves as a window fan at DC - it either blows (in phase) or sucks (out). It is our time domain optimization that gives the uncanny sonic hologram that only Thetas and MB Schiit converters do. (It also allows the filter to disappear. Has to be heard to understand.) Since lower frequency wavelengths are measured in tens of feet, placement in image gets increasingly wrong as a function of decreasing frequency in non time domain optimized recordings - these keep the listener's ability to hear the venue - not to mention the sum of all of the phase errors in the microphones, mixing boards, eq, etc on the record side. An absolute phase switch is of little to no value in a non time domain optimized, stochastic time domain replay system. It makes a huge difference with our MB units. This switch is incorporated in the Yggy.
5. This is combined with a frequency domain optimization which does not otherwise affect the phase optimization. The 0.985 of Nyquist also gives us an advantage that none of the off-the shelf FIR filters (0.907) provide: frequency response out to 21.71925KHz, 43.4385KHz, 86.877KHz, and 173.754KHz bandwidth for native 1,2,4, and 8x 44.1KHz SR multiple recordings - the 48KHz table is 23.64, 47.28, 94.56, and 189.12KHz respectively for 1,2,4, and 8x. This was the portion of the filter that had the divide by zero workaround which John Lediaev worked out in 1983, to combine with #4 above AND retain the original samples.

What all those jargon and some clever marketing for audiophiles translates is a brickwall filter that you can see from JA's measurements of Schiit Yggdrasil. Impulse response is also incredibly good like those from Chord DACs.

All of that should be inaudible unless you have golden ears like goldensound have on his abx video, but the ultimate question is that if goldensound can consistently on different days of testing pull off the same or better confidence of results

 

[VNandor]

Aliasing? In a DAC? Or did you mean images, between 22 and 24 kHz.

Yes, I should have called them images and I should have not implied that they fold back below nyquist. It would only fold back if it was measured by an ADC operating at the same sample rate which is something I've encountered so often recently I've made a mistake there. Now I know at least one person actually reads what I post ^^

 

[eq1849]

I treat it as a fun hobby where I succumb to sighted listening deliberately, heck I don't even need to concentrate to hear a difference in IEM cable, DACs in sighted listening as the biases of sight, brain trickery and mood/feelings just overpowers my rational thoughts at least for me as an individual. But I 100% respect the concepts of audio science as I have tried a single blind volume match ABX test myself through a foobar comparator and utterly couldn't get at least 90% correct, just mere guessing

I haven't taken Stats in forever, but I don't think you need 90%...more in the 55-60% range, I think, given a 0.05 P value or so.

You attribute these differences simply to bias when sighted listening?

wrapped the blind AB up: 14/15