🚀 go-pugleaf

> Xref: cbmvax net.micro.6809:182 net.micro:1278 net.micro.atari16:1481 net.micro.amiga:2051
> 
> It's interesting to see how the new Coco III fits into the personal computer 
> marketing niche...
> 
> The Coco III has 128 to 512K, graphics nearly as good as the ST or Amiga...

Really? 640x400 pixels in 16 colors out of a palette of 4096?  Bit blitter?
Video coprocessor?  Well, then, how about comparisons to the Atari.  Lets see,
the Atair gives you 640x400 monochrome.  Nope, can't do that.  640x200 in
4 colors out of a palette of 512.  I think the quoted resolution was 640x192
with 4 colors out of a palette of 64.  Not bad, but if they're really going to
the added expense of an RGB analog (over RGBI digital) monitor, why settle 
for just 64 colors (2 bits each of R, G, and B).  Also, the 6809, while one
of the nicest 8 bitters, is going to be hard pressed speedwise to do graphics 
at this resolution anywhere near the speed of the Atari, much less the Amiga.
Of course, in this price range, you're directly competing with the C128 and
(if anyone's actually buying them) the Atari 130.  The graphics, is true as 
stated, are better in some ways than either of these.  The Atari supports
either 128 or 256 colors (can't recall offhand), but its limited to a display 
of around 40 characters.  The C128 has 16 colors each screen (it can support
both the 40 and 80 column screens simultaneously).  The 40 column screen has
a lower graphic resolution, but supports hardware sprites.  The 80 column
screen can do 80x25 text, 80x50 text (interlaced), and a 640x200 two color
bit-map.  Its also possible to exchange a few horizontal lines for a limited
color bit-map, and the display can actually be set up for around 82 characters
across.  The 80 column display has some features (block moves and copies) that
free up the processor, but there are other things about it that make it kind
of slow for many things too.  Overall, it sounds like the CoCo-III has an edge
over the C128 in the display department.

> Tandy's line now seems to consist of a series of MS-DOS
> machines for the adult/business types at the high end,
> with the low end still handled by an 8-bit micro,
> the Coco's 6809.  Some people may be disappointed that
> the Coco III is not a 68000 machine, but Tandy may be wise
> to stay out of the King Kong -vs- Godzilla war between
> Atari ST and Amiga -- maybe neither combatant will live
> to inherit the ruins of Tokyo.  Tandy has straddled the
> 68K home market by landing salvos on both sides of it.

That's silly.  The 68000 machines, any 68000 machine, is much more powerful
than currently existing MS-DOS machines.  The 6809 is in many ways a more
advanced processor than the 8088, and at your 1.7MHz or whatever the CoCo-III
is running faster memory cycles than a standard or XT PClone.  If Motorola
had the forsight to add on-chip banking like in the 8088, the 6809 would
have had a good chance of catching on as THE 8/16 bit chip.

> ...possibly with the memory mapping and protection that the
> 68K pc's lack.

Has TRS announced an MMU chip for the CoCo-III?  Last I heard, none of the 
6809 suppliers (Motorola, Hitachi, etc.) had.

> ... the 68K machines (which MS-DOS users would call "toys" too).

And all UNIX users call MS-DOS a toy.  Much of this comes from the knowledge
that it truly is, since it runs on computers which (at least until the AT)
couldn't dream of supporting UNIX.  Of course, all of the 68000 machines 
will be able to run UNIX quite nicely, along with OS-9 68K, the version of
OS-9 for grownups.  Show an MS-DOS user an Amiga running a Lotus clone using
four megabytes of RAM for its spreadsheet, while at the same time a database,
EMACS, and downloading E-Mail in the background, and he'll quickly change
his tune.  What's a "memory-resident" utility, anyway.  Come on!  MS-DOS is
just an enhanced CP/M; even the 6809 versions of OS-9 are a more sophisticated
and powerful solution to the operating system question.

> Commodore and Atari have tried to narrow this gap, with 128K editions of 
> their home computers, but in terms of improvements over their 64K 
> predecessors, the Coco III is a quantum leap compared to the other 128K 
> 8-bitters. Of course any of these machines should have blown the Apple II
> away years ago, but some things defy explanation.

The Apple II was successful for exactly the same reasons that the IBM PC was
successful (1) Open Architecture, and (2) the company (Apple or IBM, take your
pick) managed to get a reputation for having a powerful "REAL" computer, despite
opinions to the contrary.  This is called MARKETING.  CBM, TRS, and most others 
haven't quite caught on to the same degree yet.

> Sure, the Coco III could use even more keyboard improvement
> (you STILL need CTRL to get the C-language symbols, but at
> least there's real CTRL key *in the ight place*), 512K tops
> may seem limiting (but 6809 code is more compact than 68K binary,
> I'll bet)

Your're right on the code compactness.  68000 op-codes alone are 16 bits long,
though of course, running the same bus speed, they're loaded just as fast
as a 6809 instruction (8MHz 68000 == 2MHz 6809 for bus speed comparisons).
Of course, if your're doing 32 bit arithmetic or many other things, the 
more advanced 68000 instruction set may produce a tighter solution (like a
block move of any amount of memory in 2 instructions, etc.  The 6809, at the
very least, will have to stop to switch banks).  Much of the size of the
code will depend on whether a compiler or an assembler is used and if a 
compiler, how efficient it is.

> and the RS232 is probably still bit-banger software.

The C64 (and C128 because of it) use the same kind of deal for RS-232.

> But against Commodore 128s and Atari 130XEs, Coco III is heavy
> competition indeed.  (If you believe that $220 tag; seems too
> good to be true -- I'd have guessed $400-$600 for a Tandy product).

Maybe reasonable competition for the C128, heavy competition for the Atari
130.  The C128 still have a number of advantages.  Not the least of which
is the 6,000,000+ C64s that have been sold already (they estimate that about
600,000 C128s have been sold, though this estimate isn't that recent).  The
C128's 8502 at 2.04MHz will do some things faster than the 1.7MHz 6809 in
the CoCo-III.  The 6809 is a better processor, however, and doing some things,
like memory moves, multiplys, etc. it'll be faster despite the slower bus.
The C128's disk drive is more expensive, but it stores around 350K of data,
and can read MS-DOS and most CP/M disks.  The CoCo, as mentioned here, is
internally expandable to 512K, the C128 externally expandable for a total
memory of 640K.  The CoCo's directly addressable memory will be more 
useful, for general applications, than the C128's DMA addresses expansion
memory.  Though much of this usefulness depends on how the CoCo III banks
its extra memory.  The C128 has a very advanced mechanism for this via the
MMU chip, the Atari's method (selecting 1 of 5 banks in one 16K segment) is
weaker.  

The C128 has one undeniable advantage, and that's the bulk of software 
available for it.  There are over 1500 different titles of Educational 
software alone available for the C64, probably around 4000 games, lots of
productivity packages, etc.  Nearly every major consumer oriented software 
house writes a version of every product they sell for the C64.  In C128
mode we already have several professional quality word processors, data
base and spreadsheet programs, etc.  CP/M, while certainly inferior to OS-9,
was supported by most of the major business oriented companies that were
around before MS-DOS, and, as such, there's lots of business and public 
domain stuff around for the C128.  The CoCo-III will certainly be able to
take advantage of the OS-9 software around, but there's much less of that
available, and it may be awhile before OS-9 level 2 comes around to let you 
use any extra memory.  

As for someone looking for power, the ST looks like a good deal.  It offers
a reasonable amount of hardware power, easily comparable to a MAC.  The
fact that it has an operating system far inferior to OS-9 will be overlooked
by many, and will actually be a moot point as OS-9 68K is released, which
will certainly blow away 6809 OS-9.

> -- 
> Mike J Knudsen  __   ...ihnp4!ihwpt!knudsen
>               / NO \
> Bell Labs    / BABY \   (312)-979-4132 (work)
>  (AT & T)   /ON BOARD\
>             \GO AHEAD/    BORED SAILORS
> IH 6D-319    \ & HIT/   go BOARDSAILING.
> x4132         \ ME /
>                 --
-- 
/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
Dave Haynie    {caip,ihnp4,allegra,seismo}!cbmvax!daveh

	"I don't feel safe in this world no more, 
	 I don't want to die in a nuclear war,
	 I want to sail away to a distant shore
	 And live like an ape man."
				-The Kinks

	These opinions are my own, though for a small fee they be yours too.
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/