Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!rutgers!ucsd!ucbvax!hplabs!pyramid!cbmvax!jesup
From: jesup@cbmvax.UUCP (Randell Jesup)
Newsgroups: comp.arch
Subject: Re: Self-modifying code
Message-ID: <4370@cbmvax.UUCP>
Date: 29 Jul 88 21:44:58 GMT
References: <1988Jul26.024039.28579@utzoo.uucp> <4333@cbmvax.UUCP> <3057@tekig5.PEN.TEK.COM>
Reply-To: jesup@cbmvax.UUCP (Randell Jesup)
Organization: Commodore Technology, West Chester, PA
Lines: 53

In article <3057@tekig5.PEN.TEK.COM> wayneck@tekig5.PEN.TEK.COM (Wayne Knapp) writes:
>Yes but there is a valid reason.  The Amiga is only (at least the workbench)
>updating a 320x200x2bit plane display.  The sun screen is much, much larger,
>(maybe 1000 x 800?).   It is a whole lot easier to move around 16k or so
>than up to say 100k on the sun.

	Wrong: 640x200x2bit or 640x400x2bit (for workbench, more bitplanes
are allowed for "custom" screen.)  Also, When I'm using suntools, I tend to
use 80-column wide windows, 30 or so lines high, 1 bitplane deep.  Overall,
about the same amount of display memory used for the window I'm scrolling
in each case.

>This just like comparing the Amiga to the Mac II.  The Amiga isn't that
>speedy if you are using HAM 640x400, which is 192k of memory.  Then 
>the Mac II also slows down in its 640x400x8 mode which is 256k of memory.
>But I would having a 256 color palette is much nicer than HAM in my opinion. 

	HAM only operates in lores (320x{200,400}) in the current chips.
The biggest amount of memory for a screen/window is 128K (640x400x4bits).
The Mac-II is not a reasonable comparison to the current Amigas (look at
price, when they were designed, processor, etc.)  The interesting thing
is that in windowing and screen update/scrolling speed the 68000 8Mhz
16-bit $600 Amiga can do so well compared to the 68020 16Mhz 32-bit Mac-II
that costs several thousand, which also uses VRAMs (a win due to bus
contention), which didn't exist when the current amiga was designed.

>Things like blitters are nice, but I would much rather have a better CPU.
>Now how about a 68000 like CPU with say 128 data reg. and 128 address reg..
>That is somethin I really could get into using.  Things like blitters are
>a dime a dozen, I want some real power!

	Ok, where can we buy this CPU for less than $20?  In fact, where
can one buy this CPU at all?  Warning: task switch time on that will
be atrocious, and instructions will have to double in size or so to hold
those register addresses.

	Leading edge CPU design is an expensive thing compared to building
a blitter.  The last major CPU MOS (Commodore's silicon arm) designed was
the 6502/6510.  They concentrate on other things now, like the custom
chips for the C64 and now the Amiga.

>P.S. The Amiga was pretty hot three years ago, but I just think that in 
>general it is no longer that wonderful.  Time for a major upgrade? 

	The new non-interlaced chips are more or less public knowlege,
though most of the details aren't.

	I can say that we certainly aren't standing still; Gould announced a
few months ago in Germany that we are working on the Amiga 3000, based on the
68030.

-- 
Randell Jesup, Commodore Engineering {uunet|rutgers|allegra}!cbmvax!jesup