24/192 music downloads are useless??

[mtrycrafts]

... My perspective comes from more of a philosophical, theoretical background in that we may be selling the software perspective short given our acute limitations where the hardware is concerned. Until we figure out what the limiting component/s is/are in the hardware chain how are we to know whether the bitrate is adequate for true 'you are there' realism - dynamics, and all? Seriously, and I've proposed this in other threads - what is the next big thing that takes us there?...
DJ

To read this article, much of I do agree with, you'd think this guy believes we've gone as far as we can with this from a software perspective. Sorry, I ain't buyin it. We're still a long way off from 'you are there live', and it can't all be hardware related.
DJ

Well, a couple of things here as I see it. Software can be tested with respect to what is and isn't audible. After all, human hearing is finite with well known limits.
So, you are left with other factors to consider.
Do you listen at concert ;levels at home? How loud is that to start with. Can your speakers reproduce such levels cleanly? What else does it take?
Can a few speakers duplicate the number of instruments and the never-ending shits on each recording sessions.
How about the space itself? You have to fool the recording/speakers to somehow give you the same sound-field as live.
If the goal is live, you just about have to go to a live event

 

[djreef]

As far as further improvements, it has to do with getting rid of passive crossovers and the wider introduction of zero phase shift digital crossovers and well as the development of affordable wide band drivers.

Agreed. This is the same reasoning I used in setting up my 2.2 rig, at least in as much as I could for under 4 grand.

DJ

 

[djreef]

Well, a couple of things here as I see it. Software can be tested with respect to what is and isn't audible. After all, human hearing is finite with well known limits.
So, you are left with other factors to consider.
Do you listen at concert ;levels at home? How loud is that to start with. Can your speakers reproduce such levels cleanly? What else does it take?
Can a few speakers duplicate the number of instruments and the never-ending shits on each recording sessions.
How about the space itself? You have to fool the recording/speakers to somehow give you the same sound-field as live.
If the goal is live, you just about have to go to a live event

No, I get all of that, my concern is with getting me as close to the performance as possible from a realism perspective - and yes I do run my system hot at times . I understand that the hardware side is where most of the innovations are going to have to come from in order to make this happen. I do agree with the author on sampling rates - I think it's silly to run them up into the triple digits, but I'm not so sold on bit rate, and word length. My original argument stems from the hardware scenario. If the hardware is the limiting factor (by a long shot), then how do we know if the software side might need to be improved upon? I guess we'll probably only know when we get there.

DJ

 

[mtrycrafts]

... If the hardware is the limiting factor (by a long shot), then how do we know if the software side might need to be improved upon? I guess we'll probably only know when we get there.

DJ

No, because we can test speakers in a mono setting and see if the bit depth and word length is sufficient or not; this is testable now and that is what the article was about.
But, single mono speakers will never get you that 'live' performance, not does 2 speakers or 3 up front. Hard to capture the sound in a auditorium and make it sound at home that it is live. That is not bit depth or word length but acoustic capture and reproduction.

 

[avnetguy]

I do agree with the author on sampling rates - I think it's silly to run them up into the triple digits, but I'm not so sold on bit rate, and word length.

I'd be more inclinded to bump up the sample rate over the resolution to see gains. Now I'm taking this from a pure digital reproduction standpoint and have no basis on the effects on what can actually be heard but in order to reproduce a frequency with reasonible accuracy the sampling rate needs to be about 4 times higher than the highest frequency you wish to reproduce. So if you want to push it to the typical speaker/amp upper testing end of 20kHz you'll need a sampling rate of at least 80 kHz. So the current gear that supports 88.1 and 96 should be more than enough.

I'm sure someone has done this already but I'd really like to see what 12 kHz to 21kHz sine waves, in 1 kHz steps, looks like from DSO captures using a 44.1 kHz source.

Steve

 

[avliner]

Why not trying the program shown below, in order to cross-check data and be sure whether or not there will be an audible difference??

Audio DiffMaker

 

[djreef]

But, single mono speakers will never get you that 'live' performance, not does 2 speakers or 3 up front. Hard to capture the sound in a auditorium and make it sound at home that it is live. That is not bit depth or word length but acoustic capture and reproduction.

Right. Hardware, yet again. I think we agree that this is where the engineers and designers need to make the greatest strides.

DJ

 

[djreef]

I'd be more inclinded to bump up the sample rate over the resolution to see gains. Now I'm taking this from a pure digital reproduction standpoint and have no basis on the effects on what can actually be heard but in order to reproduce a frequency with reasonible accuracy the sampling rate needs to be about 4 times higher than the highest frequency you wish to reproduce. So if you want to push it to the typical speaker/amp upper testing end of 20kHz you'll need a sampling rate of at least 80 kHz. So the current gear that supports 88.1 and 96 should be more than enough.

I'm sure someone has done this already but I'd really like to see what 12 kHz to 21kHz sine waves, in 1 kHz steps, looks like from DSO captures using a 44.1 kHz source.

Steve

Exactly, but my point in all of this is how will we know that lower bitrates and word lengths are sufficient if hardware, in general, is the limiting technology.

DJ

 

[mtrycrafts]

Right. Hardware, yet again. I think we agree that this is where the engineers and designers need to make the greatest strides.

DJ

Hardware in a sense of speakers and rooms but not sure if it is possible ; may get close or realistic enough.

 

[mtrycrafts]

... but in order to reproduce a frequency with reasonible accuracy the sampling rate needs to be about 4 times higher than the highest frequency you wish to reproduce. So if you want to push it to the typical speaker/amp upper testing end of 20kHz you'll need a sampling rate of at least 80 kHz. So the current gear that supports 88.1 and 96 should be more than enough.

I'm sure someone has done this already but I'd really like to see what 12 kHz to 21kHz sine waves, in 1 kHz steps, looks like from DSO captures using a 44.1 kHz source.

Steve

No, that is absolutely not true. You only need it greater than 2x, hence the 44.1 although it could have been 44 or 48, not sure why 44.1.

 

[cpp]

You only need it greater than 2x, hence the 44.1 although it could have been 44 or 48, not sure why 44.1

Interesting how 44.1 was established

http://www.cs.columbia.edu/~hgs/audio/44.1.html

http://www.indiana.edu/~emusic/etext/digital_audio/chapter5_rate.shtml

 

[mtrycrafts]

Interesting how 44.1 was established

http://www.cs.columbia.edu/~hgs/audio/44.1.html

Chapter Five: Principles of Digital Audio

How interesting, thanks.

 

[audiofox]

No, that is absolutely not true. You only need it greater than 2x, hence the 44.1 although it could have been 44 or 48, not sure why 44.1.

The 2X factor is the referred to as the Nyquist rate, a key design rule for digital signal processing system:

Nyquist rate - Wikipedia, the free encyclopedia

Note that this is typically used as the minimum required rate, but many digital communications systems exceed that rate if other factors affect the signal quality or particular digital modulation formats are used, so it is one of several variables that are traded against each other when optimizing a digital comm or signal processing system.

 

[avnetguy]

No, that is absolutely not true. You only need it greater than 2x, hence the 44.1 although it could have been 44 or 48, not sure why 44.1.

Remember I'm only talking about digital reconstruction, not about what can actually be heard by human ears. When sampling an input signal with the frequency at half the sampling rate the best achieved would be a triangle wave (samples taken at peaks) and the worst would be no waveform (samples taken at zero crossing), this depends on how your sample clock matched up with the incoming signal. Now as the input signal frequency decreases you are guaranteed to have one of the two points being non-zero and the frequency component is therefore retained. Now all input frequencies from the sampling rate / 2 (22050) to sampling rate / 3 (14700) will be represented by only two points placed "somewhere" on the slopes of the sine wave. So my question here is will those two digital points provide an accurate representation of the amplitude of that input sine wave?

Steve

 

[avnetguy]

Exactly, but my point in all of this is how will we know that lower bitrates and word lengths are sufficient if hardware, in general, is the limiting technology.

DJ

I don't think you'll have to worry about it as the higher standards are already established, both sampling rate (192 or 96 kHz) and resolution (24 bit) are in place for all AVRs that to can use DTS-HD Master Audio.

Of course you'll have to wait at least a few years before the 10TB ipod's and such come out, then the music industry will magically recognize the need for higher rates and of course charge higher prices.

Steve

 

[cpp]

Of course you'll have to wait at least a few years before the 10TB ipod's and such come out, then the music industry will magically recognize the need for higher rates and of course charge higher prices

Of course by then you will be older, your hearing will fade ( trust me your hearing fades with time) and the price of these cool tricks will be sky high and by then you will not hear the difference between 16/44.1 and 24/96

 

[PENG]

So my question here is will those two digital points provide an accurate representation of the amplitude of that input sine wave?

Steve

Theoretically yes if you know the frequency, but I think you know that already, so what is the hidden question?

 

[avnetguy]

Theoretically yes if you know the frequency, but I think you know that already, so what is the hidden question?

Now my answer to that would be no, I really don't see how you could consistantly capture the actual amplitude with only two points per cycle. In my mind you'd see the frequency with a varied amplitude as the two points move along the sine wave on each following cycle.

Going back to my previous statement, if I now have 4 samples per cycle on the sine wave I'd end up with a pretty good approximation of the amplitude regardless of their position on the sine wave.

So no hidden questions, it just looks to me like the representation of input frequencies from the sampling rate / 2 down to sampling rate / 4 will be far from a good approximation. So is my thinking process off here?

Steve

P.S. Gotta take the dog to the park now but maybe there is an easy way to prove/disprove this? How about testing with a lower sampling rate (4 kHz) and an input sine wave signal with a frequency that falls more into our normal hearing range (1000-2000Hz).

 

[avnetguy]

So here is a quick visual example that shows the difference in the actual amplitude captured.

Both sides show sine waves (top to bottom) at 3900, 3500, 3000, 2500 Hz.
The left side is sampled at 16000 and the right side 8000.

So visually it is pretty obvious at the sample rate of 8000 the amplitude of the signal is not represented very well as your input signal approaches 4000.

Steve

 

[PENG]

So visually it is pretty obvious at the sample rate of 8000 the amplitude of the signal is not represented very well as your input signal approaches 4000.

Steve

Can you link us to your source? Just looking at them I don't know what those graphs represent. They don't look like sine waves, but they can be resolved into sine waves of fundamental frequencies and harmonics. All I can say is that as long as you don't sample at zero crossing as you mentiioned before, two points in a cycle should allow you to reconstruct the original sine wave in theory. In practice there are many factors that influence how difficult it is to reconstruct the original signal accurately, but it is theoretically possible and we know it is happening otherwise we won't have CDs.

Take a look of the formula here and you should be able to see that it can done given that you have two values within one cylcle and the frequency is known.

Sine wave - Wikipedia, the free encyclopedia