On Friday, June 13, 1997 1:18 PM, Random Junk [SMTP:jsd@gamespot.com]
wrote:
} the whole idea of a 44khz RECORD (vinyl anyway) is to laugh... how
} many people's turntables even go close to 16khz? (that would be a
} 32k
} record, of course). how many "average human" ears are even good
} enough to hear above 16khz any more? if you go to loud clubs a lot,
} chances are your cutoff is even lower than that.
Sorry to drag this out further, but now I've got a question:
What is the maximum number of samples ("bumps" I suppose) that can be
contained on a record (vinyl) per second? 45rpm would allow the most
samples if a single song were contained on one side of a record, right?
So what's the limit for vinyl?
On Fri, 13 Jun 1997, Christopher Fahey wrote:
} Chris Fahey Sez:
} I just thought of something. If the human ear is supposed to be so very
} weak that CD audio far exceeds it's ability to detect, then how come so
} many folks can hear the difference between vinyl and CD? Usually they
} are probably liars, just like people who think they know something
} about wine.
I don't know jack about wine, but I know a little about CDs. :)
It's got to do with the sampling rate, the distortion, and the
quantization. CDs are created using a 44.1KHz sampling frequency -- i.e.
44,100 samples per second. This number is set by what's known as the
"Nyquist Criterion", which states that the sampling frequency should be
two times or greater than the highest frequency component in your signal.
The signal to be sampled here is audio meant for human ears, which
theoretically can detect frequencies up to 20KHz. Doing the math here
gives a minimum required sampling frequency of 40KHz. If you drop below
this sampling frequency you will introduce what is known as "aliasing", which
causes distortion because the signal isn't being properly sampled. I'm
not exactly sure, but I think the sampling frequency used is 44.1KHz to
allow for error correction, so there is extra information if needed.
Then there is the matter of quantization. From what I understand,
16 bits/sample is common for CDs. Since each bit can take on a value of
one or zero, a sample is quantized to one of 65,536 different amplitude
levels (i.e. 2 raised to the 16th power). This number of possible
amplitudes for each sample seems more than adequate to me, but I'd like to
hear the DVDs with 24 bits/sample giving about *16.78 million* possible
amplitude levels just to make sure. :) Also the DVDs are supposed to have
96KHz (according to g.), which is 4.8 times higher than the highest
frequency detectable by humans. What's the point in this for audio?
For video, maybe, but it seems like overkill in the sample rate for audio
to me. Anyone?
Anyway, I know about digital, but vinyl and record players are a
bit more mysterious to me. It seems like the analogous thing to
quantization in CDs would be handled by the height of each bump in vinyl,
is this correct? If so, how accurate is this process? It just seems that
with surface noise, possible warpage, worn out needles, etc. CDs are bound
to sound better to most people than vinyl, despite the whole technical
aspect, which I imagine is in favor of CDs as well.
} But here's an idea: Although our *ears* cannot detect the difference in
} the sound form the speaker, maybe the *environment* itself reacts
} differently to the sound. The little vibrations which the sound causes
} in the room, the way the room acoustically alters what we hear, may
} significantly differ from vinyl to CD. These differences are
} accentuated by the vibrating objects in the room/environment, so that
} they then become detectible to the anal retentive audiophile ear.
}
} Imagine a dish on a shelf which vibrates one way when a 44khz cd plays
} a tone and another way when a record plays the tone. The sound of it's
} vibration becomes part of the listener's experience of the sound.
}
} Am I on drugs?
Hmm. If this somehow plays a part in things with regards to
the way we hear audio from vinyl and CDs, I'd imagine it's a *very* small
component in comparison to all the other elements going on in the way the
music was recorded onto the medium and how it's played back, but it's an
interesting idea I've never thought about before...
| E r i c | [mail] franse@engr.arizona.edu |
| F r a n s | [web]
http://engr.arizona.edu/~franse |
"Make the events occur that you want to occur" - The Black Dog