[71 dB]

2. The "Monty's video" is the follow up of that article (i.e. to support "no staircases" and perfect reconstruction of sine wave using 44.1/16 source). He used these as supporting facts for his claim ("192kHz music files make no sense")

The highest sampling frequency that would make sense in consumer PCM digital audio is about 60 kHz. Even 96 kHz sampling frequency is overkill and 192 kHz is just ridiculous unless you are a bat. As I have suggested before, 58.8 kHz/12 bit consumer audio with shaped dither would make sense as an alternative to 44.1 kHz/16 bit (exactly same bitrate) allowing "easy" band-limiting filtering and enough perceptual dynamic range, but we have 44.1 kHz/16 bit digital audio for historical reasons. Fortunately it works just fine as digital technology has matured about 30 years ago to the point allowing perceptual transparency.

 

[gregorio]

To me, it is pretty clear to me that there is no question asked too that 1.001 is not the same as 1.000

I notice you failed to answer my question, what is the SPL?

If you did, maybe you would understand the conditions under which 1.001 is identical to 1.000, that you are currently either ignorant of, or just deliberately lying about!

The setup I'm talking about....live musician with an acoustic guitar with mics/amp to a set of speakers. When he takes a break he plays recorded music through the same setup.

Doesn't sound the same.

If they do sound different it’s because either it’s a different recording or the playback is in some way different (EG. A different level, different EQ or some other processing).

G

 

[eq1849]

The highest sampling frequency that would make sense in consumer PCM digital audio is about 60 kHz. Even 96 kHz sampling frequency is overkill and 192 kHz is just ridiculous unless you are a bat. As I have suggested before, 58.8 kHz/12 bit consumer audio with shaped dither would make sense as an alternative to 44.1 kHz/16 bit (exactly same bitrate) allowing "easy" band-limiting filtering and enough perceptual dynamic range, but we have 44.1 kHz/16 bit digital audio for historical reasons. Fortunately it works just fine as digital technology has matured about 30 years ago to the point allowing perceptual transparency.

What is more accurate?
A) a 44.1khz signal upsampled by the DAC to its max 16fs of 705.6khz
B) a native 705.6khz signal fed into the DAC, bypassing the DAC's interpolation algorithm/up sampling

 

[gregorio]

What is more accurate?
A) a 44.1khz signal upsampled by the DAC to its max 16fs of 705.6khz
B) a native 705.6khz signal fed into the DAC, bypassing the DAC's interpolation algorithm/up sampling

Most probably they would be audibly identical. If they’re not, then the answer is almost certainly “A” (assuming a typical/optimal filter in the DAC of course).

G

 

[eq1849]

Most probably they would be audibly identical. If they’re not, then the answer is almost certainly “A” (assuming a typical/optimal filter in the DAC of course).

G

How is "audibly identical" determined?

by "typical/optimal" do you mean a perfect filter?

 

[Davesrose]

To me, it is pretty clear to me that there is no question asked too that 1.001 is not the same as 1.000

They are both 1.00 if rounded to the hundredth.

If your answer is "they are the same" for Question 1, you have, I believe, an engineering mindset based on practical usage. Your claim is not objective as your claim is based on your own belief that "1.00000000000001 is same as 1.00000000000000"

It is objective from a computer engineering perspective to say they are the same if the intended value of a function is rounded (decimal values often don't go that far...and for applications that do need it, they tend to be represented as float values).

 

[sunjam]

I notice you failed to answer my question, what is the SPL?

If you did, maybe you would understand the conditions under which 1.001 is identical to 1.000, that you are currently either ignorant of, or just deliberately lying about!

If they do sound different it’s because either it’s a different recording or the playback is in some way different (EG. A different level, different EQ or some other processing).

G

90 dB SPL ==> 89.999999999725328 dB SPL

 

[sunjam]

They are both 1.00 if rounded to the hundredth.

It is objective from a computer engineering perspective to say they are the same if the intended value of a function is rounded (decimal values often don't go that far...and for applications that do need it, they tend to be represented as float values).

Yes, the keypoint is they are not the same even they are very close

 

[sunjam]

Hi-Res is closer to the original analog audio signal when compared to CD (even the difference is small)

 

[sunjam]

I believe that some can hear it and some cannot.

If you are still thinking about hearing the frequency higher than 20kHz with Hi-Res, you are fooled by the Hi-Res marketing materials.
Hi-Res is for the better reconstruction of the final audio signal.

 

[Davesrose]

Yes, the keypoint is they are not the same even they are very close

My key point was they are the same if the intended output is a hundredths decimal. That's the way computer science works (and is not subjective). Same also with your example: 90 dB SPL ==> 89.999999999725328 dB SPL. Well beyond the limits of human hearing, and may be the same in computer programming if the function rounds ^10.

 

[bfreedma]

Yes, the keypoint is they are not the same even they are very close

It’s situational.

When calculating the trajectory of a rocket launching a payload intended to reach a distant comet, the variance may matter.

When discussing virtually any topic related to audible audio reproduction, the difference is completely irrelevant so can be considered the same.

You need to stop trying to make universal statements and understand relevance to the topics being discussed here.

 

[gregorio]

How is "audibly identical" determined?

If the difference is outside audibility. For example, if the only difference is in a frequency range higher than 20kHz they would be audibly identical.

by "typical/optimal" do you mean a perfect filter?

Yes, an audibly perfect filter (to be clear). IE. The type of fast roll-off linear phase found in almost all DACs for the last 25+ years.

90 dB SPL ==> 89.999999999725328 dB SPL

That’s your answer to what SPL it would be is it? That explains a lot, 120dBSPL below 90dBSPL is not 89.9999… dBSPL! How ironic that you question my math but get such ridiculously simple math so wrong yourself, that’s what we’ve come expect from you now though!

Want to try again and get the correct answer or just ignore the facts and continue with the BS?

G

 

[eq1849]

If the difference is outside audibility. For example, if the only difference is in a frequency range higher than 20kHz they would be audibly identical.

G

HOW are you determining that?

 

[eq1849]

Yes, an audibly perfect filter (to be clear). IE. The type of fast roll-off linear phase found in almost all DACs for the last 25+ years.

G

Sampling Theorem requires PERFECT filters.