testing one two... 16bit 44.1kHz vs 24bit 96kHz

[scott2000]

I'm confused as to the approach of listening when it's been shown that the math and science is supposed to prove our hearing is maybe psychological in many cases. Are we saying that, if we don't hear a difference( I think I may have today at least), then the science is wrong because there should be differences? Or is there no science to prove there are differences? I honestly don't know. I'm sure I've seen this type of testing turned the other direction to disprove audiophoolish claims....

Interesting....

At least I got my truck running better today....It sounds pretty good I must say. Everyone with older cars and trucks should have a timing light..... Amazing what a couple of degrees can do......

 

[abbey road d enfer]

I'm confused as to the approach of listening when it's been shown that the math and science is supposed to prove our hearing is maybe psychological in many cases.

There are many aspects of audition that are perfectly measurable; even psychoacoustics are modelled more and more precisely every day.

Are we saying that, if we don't hear a difference( I think I may have today at least), then the science is wrong because there should be differences? Or is there no science to prove there are differences? I honestly don't know. I'm sure I've seen this type of testing turned the other direction to disprove audiophoolish claims....

This test subject is "which one you think has the highest resolution", but my answer was "I can't answer this but I can tell you which one I prefer"; that's where I consistently chose the lowest resolution  ;D
Actually there are differences but it turns out the most mathematically accurate is not necessarily the best sounding.

 

[efinque]

I wouldn't personally be able to tell the difference... not me or anyone else I think.

However, the higher sample rate exists for a reason and that is downsampling (=the Nyquist theorem) where you'd record program material and process it through outboard gear (and record it back), which basically halves the sample rate and bit depth.

There's also a technology called DSD (Direct Stream Digital) available in Merging Technologies Pyramix consoles for example which is about 10 times more accurate than any of the digital media listed here, which in return allows for even more editing capabilities for engineers (or how should I know, I haven't used one.. I'm doing fine with my small project studio) but it doesn't really offer anything exciting for end-users since the final products are still printed in 16bit/44,1kHz which is CD quality.

SACD's are another story but I haven't heard of anyone who has a player that's capable of playing back those.

This is just me but I believe that the most digital artifacts reside in the high register of the spectrum and some DAW's/sequencers audio engines can affect those (there's been a discussion on the subject on djforums).. it could be a psychological thing too. I think it has something to do with the layout and color scheme of the software..

EDIT : It's not really a problem in todays EDM for example.. there's usually so much going on that you won't be able to distinguish the artifacts.  Some recordings with material that has an unusually wide dynamic range (I've recorded a choir in a church once..) could possibly reveal those artifacts.

EDIT 2 : I think it's because the audio engine takes (stores) the same amount of samples whether the signal is 30Hz or 20kHz.. 20kHz is 20 000 cycles per second which is apparently more dense in the time domain than it's 30Hz counterpart so from a full wave period you got a different amount of samples for the extreme ends and the 30Hz has more of them, which makes me think why there is no "frequency dependant" audio engine (or is there?)

EDIT 3 : I found a wikipedia article on something called "adaptive bitrate streaming" but it's in hebrew.

 

[abbey road d enfer]

However, the higher sample rate exists for a reason and that is downsampling (=the Nyquist theorem) where you'd record program material and process it through outboard gear, which basically halves the sample rate and bit depth.

?????

 

[efinque]

?????

https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem

 

[abbey road d enfer]

https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem

I am very familiar with the Shannon theorem and the Nyquist frequency, and much more about digital audio.
You seem to misunderstand the real-world implications of it. If converting a signal from digital to analog resulted in halving the frequency response, no professional sound engineer would do it.

 

[efinque]

I am very familiar with the Shannon theorem and the Nyquist frequency, and much more about digital audio.
You seem to misunderstand the real-world implications of it. If converting a signal from digital to analog resulted in halving the frequency response, no professional sound engineer would do it.

I thought resampling was one of the few real world applications of the theorem that could be explained in lay terms.

Another one is the "picture-of-a-picture"

 

[JohnRoberts]

I am very familiar with the Shannon theorem and the Nyquist frequency, and much more about digital audio.
You seem to misunderstand the real-world implications of it. If converting a signal from digital to analog resulted in halving the frequency response, no professional sound engineer would do it.

+1  No need to school Abbey about the basics.... (he knows).

Got to love the WWW....    :

JR

 

[abbey road d enfer]

I thought resampling was one of the few real world applications of the theorem that could be explained in lay terms.

Another one is the "picture-of-a-picture"

Don't hesitate to use scientific terms; most of us here are quite familiar with science.
"picture-of-a-picture" does not convey any meaning for me in the context.

 

[scott2000]

I chose all but one in the lower resolution........

on some old DCM bookshelf speakers I had just hooked up minutes before