Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!cbmvax!daveh
From: daveh@cbmvax.commodore.com (Dave Haynie)
Newsgroups: comp.unix.amiga
Subject: Re: second thoughts on buying a 3000UX
Message-ID: <19755@cbmvax.commodore.com>
Date: 11 Mar 91 21:23:10 GMT
References: <61751@masscomp.westford.ccur.com> <1135@applix.com> <5PR9W6D@xds13.ferranti.com> <1145@applix.com> <13376@hubcap.clemson.edu> <19512@cbmvax.commodore.com>
Reply-To: daveh@cbmvax.commodore.com (Dave Haynie)
Organization: Commodore, West Chester, PA
Lines: 24

In article <19512@cbmvax.commodore.com> jesup@cbmvax.commodore.com (Randell Jesup) writes:

>	Note: I haven't measured these things, I'm going on reports from IBM-
>types about the number of wait-states to access video ram (I've heard numbers
>as high as 12).

Dr. Dobbs had an in-depth article on these VGA toys a few months ago.  They
had PC-AT bus wait-states in the 20-30 range, which would translate to a mean
of 12 XT bus wait-states, to VGA memory.  The other interesting thing about
the VGA register set is that everything is intrinsically 8 bits wide.  So you 
don't get an AT bus version of a VGA card for its bus width, but generally 
because the board runs at AT bus rates (4 cycle 8MHz clock) rather than XT bus
rates (4-5 cycle 4.77MHz clock).  There are a number of non-standard additions
to VGA boards that gives you a more reasonable CPU to memory bandwidth, but
to use such features requires special code; a generic VGA-X wouldn't do much
for you.  This architectural ugliness accounts for the big speedup IBM is
claiming with XGA, even though VGA and XGA use the same generation memory.

>Randell Jesup, Keeper of AmigaDos, Commodore Engineering.

-- 
Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests"
   {uunet|pyramid|rutgers}!cbmvax!daveh      PLINK: hazy     BIX: hazy
	"What works for me might work for you"	-Jimmy Buffett