From: Robert Baer
X-Mailer: Mozilla 4.75 [en] (Win98; U)
Subject: Re: Music - Do numbers matter?
Date: Fri, 03 Jan 2003 00:58:16 GMT
NNTP-Posting-Date: Thu, 02 Jan 2003 16:58:16 PST
Organization: EarthLink Inc. -- http://www.EarthLink.net
James Llort wrote:
> I'm after a new sound system for my studio. I have very sensitive ears
> so sound quality is of the upmost importance. I mostly enjoy classical
> mustic, and I can hear the nasty 'stepping' on CD players as the
> wonderful sound is horribly broken up into chunks with the digital
> scheme. I'm guessing this is because my old system uses a duel "1 bit"
> ADC. This brings me to my first question: How many bits can a human
> ear detect? I would have thought an 8 bit sound card from my old 286
> would offer better sound quality, but it is rather nasty I'm afraid.
> My 32 bit sound card isn't much chop either - I want a sound card to
> hook up to my new system to listen to my MP3's. I have looked but been
> unable to locate 128 bit soundcards - I guess companies like Creative
> only care out the 'leaded ear majority'. Does anyone know of an
> excellant sound card without digital stepping? I'd prefer an optical
> output of some sort to prevent phasing errors.
> Now, to the meat of my question: What is the best sound system money
> can buy(for under 5 grand)? I need your advice. This is for a small
> room, roughly 15 by 15 feet. My problem is that I am 'super-size' and
> find it very difficult to leave my house. No flames please, it's
> genetic. But I degress...
> Thanks to the internet, I have collected much technical information
> from a varity of sound systems. Do all these numbers mean anything to
> the ear, or is it all meaningless. Would I be better off with 40 WRMS
> or 500WPMPO (whatever that means). I have been looking at quality
> brands like Sony , Phillips and AudioPro. My problem is that because
> of my weight I cannot listen to the systems in question - I'll have to
> buy based only on audiophile magizine reviews and numbers. Which one
> should I trust more? The "golden ears" who have listened to the music,
> or the numbers?
> This is not meant as a troll, I am genuinly interested.
> Dr James Llort,
> Department of Music
> Purdue University
> 1061 Freehafer Hall (FREH)
> West Lafayette, IN 47907-1061
The human ear can detect phase variations to at least 100Khz.
And for waveforms to be accurately rendered from a digitized source,
one needs sampling at least 10X of the top end, to have a reasonable
approximation of the transients and high frequency components.
This means 1Mhz minimum sampling, which *NO* commercial outfit would
even !think! of doing.