Mindset said:
Ok so let me ask you this, why does high resolution = higher frequencies?
Try thinking of it this way: Frequency - measured in cycles per second - is nothing more that how fast or how rapidly changes happen to the waveform. A 20kHz signal goes through a complete cycle 20,000 times per second.
Lets say that there's a deformation in that waveform that happens on the cycle's initial upswing, the kind of deformation that only a "higher resolution" sample rate could pick up on. As the initial upswing part of a complete waveform is in it's entirety only 1/4th of the entire waveform, that upswig takes place in 1/4th the time of the full wave cycle, meaning the entire upswing takes place in only 1/80,000th of a second.
This means that any such "high resolution" deformation of that wave is occuring at a "resolution" of 80,000 cycles per second, which is the equivalent of saying it is happening at a frequency of 80kHz.
When talking in the samples per second realm, "resolution" simply means slicing into shorter periods of time, which means the ability to respond to higher frequency events. Resolution = frequency response.
Mindset said:
So what you guys are saying is that 1hz, or 1 cycle per second = 1 frequency?? And a sample rate of 1hz = 1 frequency
Not exactly, though you're starting to get on the right track. According to proven mathematical theories such as the Nyquest Theorem and others, in order to accurately reproduce any given frequency, one needs to sample at twice that frequency. So theoretically, based upon just that, in order to reproduce up to 20kHz, one needs to sample at 40kHz minimum. There are some other factors that come in with stuff called aliasing and slew rates of frequency filters and such that I won't go into in detail here, but suffice it for now to say that those factors dictate that teh sample rate go just a bit higher than double the desired frequency. Long story short, the engineers ended up at 44.1kHz as the ideal minimum sample rate to handle up to 20kHz "resolution".
Mindset said:
192khz, it's more lively, than 44.1khz.
As far as your question as to why higher sample rates do sometimes sound better, that is a subject of hot debate amongest even the "experts". There are many factors involved there (including mundane logistical stuff like actual hardware quality), but there are those that believe that humans can "pick up on" some frequencies higher than 20kHz, even if they cann't seem to consciously hear them. The elusive "air" that higher sample rates can allegedly provide (though whether your loudspeakers can actually reproduce them is another question altogether and kind of mucks up the arguement a bit
).
But even those who subscribe to the supersonic "air" theory will tell you - well, the vast majority of them anyway - that there is a limit even to those supersonic frequencies that usually falls well below 40kHz. Meaning that even those frequencies would fall below the capabilities of an 88.2k sample rate. 192k sampling (~90kHz frequency response) is simply overkill, even there.
And this isn't even bringing in factors such as loudspeaker limitations, the fact that the recording chain from microphone to pre-converter isn't designed to accurately - if at all - capture those high frequencies anyway, or the fact that for any given converter, it's a crapshoot of engineering design as to which sample rate selection it actually performs at best. Brand A converter may have excellent 48k sampling but it's 96k stage simply doesn't perform as well because of a design thing. It happens. Not to mention the difference in quality between brands; I'll take an Apogee at 44.1 over a Presonus at 96 any day of the week.
G.
But, just go listen to Joshua Judges Ruth by Lyle Lovett - it was recorded on late 80's/early nineties converters at 16 bit 44.1 (recorded straight to digital) - fabulous sounding. I did record at 48 to keep the "air' a bit tighter, but I got tired of having to keep switching my converters back and forth........
