Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!rutgers!ucla-cs!zen!ucbvax!COGSCI.BERKELEY.EDU!bryce
From: bryce@COGSCI.BERKELEY.EDU (Bryce Nesbitt)
Newsgroups: rec.music.synth,comp.sys.atari.st,comp.sys.amiga
Subject: Re: Amiga midi
Message-ID: <8708220150.AA24029@cogsci.berkeley.edu>
Date: Fri, 21-Aug-87 21:50:07 EDT
Article-I.D.: cogsci.8708220150.AA24029
Posted: Fri Aug 21 21:50:07 1987
Date-Received: Sun, 23-Aug-87 06:35:37 EDT
Sender: daemon@ucbvax.BERKELEY.EDU
Lines: 56
Xref: mnetor rec.music.synth:1327 comp.sys.atari.st:4870 comp.sys.amiga:7636

In article <2489@trwrb.UUCP> sansom@trwrb.UUCP (Richard Sansom) writes:
)In article <.AA07072@cogsci.berkeley.edu> bryce@COGSCI.BERKELEY.EDU writes:
)>In article <628@cup.portal.com> Henry_Burdett_Messenger@cup.portal.com writes:
)>>
)>> These memory systems are *all* compatible with the new blitter chip, and,
)>> unlike the Amiga, all memory on the ST is created equal (no "fast" and
)>> "slow")
)>
)> Oh, come on!!  If you are going to claim a feature "win" of the ST, pick
)> one of the real feature "wins".  Using the Amiga terminology of "fast" and
)> "slow" memory, *all* Atari ST memory would be defined as "slow".
)
)I think you got this one wrong.  The ST's memory is clocked at 16MHz - the
)odd cycles belong to the CPU, and the even cycles belong to the DMA chip (or
)the other way around).  In this way, there is never a time when the CPU is
)idle due to "stolen cycles" (during screen refreshes).  I think that means
)the ST's memory should be defined as "equal" as the fellow stated in his
)original article.

Wrong again.  The situation you describe is exactly the way the Amiga's "slow"
memory works.  *Exactly*.  The memory in the Amiga that runs at twice the
processor speed, and shares alternate/odd cycles with the video is "slow". 

Read the posting... "Using the Amiga terminology of "fast"
			       ^^^^^^^^^^^^^^^^^
and "slow", *all* Atari ST memory would be defined as "slow".  This is true.

The Amiga also can access fast memory.  The memory allocator has a bit for
"fast".  True "fast" memory is *not* shared with the video at all.  Even
if the Amiga is displaying a lo-res screen with 64 colors/line (*twice the ST's
maximum*) _and_ running the blitter to do an area fill, "fast" memory
is 100% avalable for the processor.
If you manage to get an Atari blitter and then do a blit your ST's processor
will hit a slowdown.  Thus it's "slow" by Amiga terminology.
As was said in the part of the article you chopped off, "chip" is
the more technically correct term for what people call "slow" memory on
the Amiga.

BTW: The ST will have "stolen cycles" on instructions that don't hit even video/
processor slots.  Check carefully,  I think that you will find that "exotic"
instructions (like all the branches and branch subroutines :-) will often cause
a wait of two clocks.

Amiga "slow" memory = ST main memory = "chip" memory (= "equal" memory :-)


>>Facts, not flames.
>Ditto.
Di-Ditto.  Please check your "facts".

-----
|\ /|  . Ack! (NAK, EOT, SOH)
{o O} . 
( " )	bryce@cogsci.berkeley.EDU -or- ucbvax!cogsci!bryce
  U	"You are welcome to prove me wrong,
	 but make sure I'm wrong first."