Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site unc.unc.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!whuxl!whuxlm!akgua!mcnc!unc!gibson
From: gibson@unc.UUCP (Bill Gibson)
Newsgroups: net.micro.amiga
Subject: Re: Amiga sound, checks
Message-ID: <741@unc.unc.UUCP>
Date: Sat, 14-Dec-85 17:22:15 EST
Article-I.D.: unc.741
Posted: Sat Dec 14 17:22:15 1985
Date-Received: Mon, 16-Dec-85 03:48:21 EST
References: <11217@ucbvax.BERKELEY.EDU> <918@ecsvax.UUCP>
Reply-To: gibson@unc.UUCP (Bill Gibson)
Organization: CS Dept, U. of N. Carolina, Chapel Hill
Lines: 23
Summary: 

>> assume it's an 8-bit DAC.  Considering that a 5-bit DAC is plenty good 
>                                                ^^^^^^^^^^^^^^^^^^^^^^^^
>> enough for voice and most sound synthesis, an 8-bit DAC should be out of
>> this world.
>> 					-Matt
>    If you said 15 bits, I certainly would agree.  On the last point, 
>an 8-bit DAC should be out of this word for musical sound synthesis,
>but we have to specify which direction :-).
>--henry schaffer

Total agreement. 8 bits is *minimal* sound quality for synthesis; 8 bits
of sampled sound gives you a 48 dB signal-to-noise ratio. That's about
as good as a bad tape recording.

Not to mention that the Amiga's output contains only frequencies below 
5 kHz, an *extremely* low limit. I suppose that this is the result of 
an engineering tradeoff so that the sound hardware can use
a standard video rate for writing out samples; the machine apparently
wasn't targeted at the music marketplace. That's too bad, since 
a higher sampling rate could have been used easily, with the major
drawback that more memory would be used per time unit.

Bill Gibson