Conference hydra::amiga_v1

    
    I posted .0 not to start a flame war over the "Amiga killer" or
    the TT, but rather, to start a discussion on what features they
    have that the Amiga could be enhanced with.
    
    What is this 8bit PCM stereo sound?  How would it compare to the
    Amiga stereo?
    
    Expanding the palette from 512 to 4096 seems like a good idea to
    me, but limiting it to the same number of color registers (16,4,2)
    seems like they are targeting the Apple IIGS (Sam also mentions
    the "A" company later on in a reference to expensive upgrade
    boards).  Hmmm... how does the 8bit PCM compare to the Ensoniqs
    stereo in the IIGS?
    
    I kind of expected that they would try for a palette greater
    that 4096, maybe something like VGA's 256,000 or TI's 16mil.
    Guess that would have raised the price too much.
    
    I didn't see any mention of a native multitasking OS, did see a mention
    on usenet that they are working on UNIX for the TT, but then
    so is CBM for their 68030 board.
    
    -Dave

    
    Re: The STE
    
    I don't think this machine qualifies as an Amiga killer.  They've
    enlarged the color palette to what is now an acceptable range for a
    home computer.  The sound chip sounds like the long-awaited 'Amy' chip
    that was promised four years ago.  Even with these improvements, I
    don't see it as offering more value or usefullness than an Amiga 500,
    given the Amiga's edge in sophisticated built-in multitasking system
    software.
    
    As far as targeting the Apple IIGS, the current ST already IS far
    superior to the GS (IMHO) in price, speed, display quality, software
    base, aesthetics, etc.  In order to steal sales from Apple, they'll
    have to advertise, advertise, advertise, advertise.  Atari has a large
    image problem in the US (probably even worse than Commodore's) and will
    have to work very hard at building a distribution and support network. 
    The ST has always been a nice system for the money, and it's pained me
    to see a pile of shit like the IIGS (again, IMHO) get all the press.
    
    Re: The TT
    
    I think this may be a case of too little, too late.  Any new machines
    from Atari or Commodore that would compete with the Mac IIcx or the
    386/VGA machines needs to go above and beyond what is considered the
    lowest common denominator; ie, 4096 colors can't cut it in a VGA world
    of 256,000 colors.  16 MHz is ho-hum with 33 MHz 386 systems shipping
    in droves, and '486 systems ready to go out the door.
    
    Price was not mentioned, but it might make a nice university Unix box.
    
    Ed.
    
                                                                        

    RE: < Note 2888.2 by LEDS::ACCIARDI >

      I agree with you Ed, especially on the TT end of things, 16Mhz is just
    ho hum these days.  I would also like to point out that these were just
    announcements, you can't buy any of this yet, and who knows when you will
    be able to.  I point out that atari is well known for announcing things
    and never delivering.  You might actually see the TT, but I wouldn't
    count on buying it expecting atari to release u**x for it.  In my opinion
    it will never show.  It has taken 4 years to get an upgrade to fix a
    couple bugs in the ST O/S.  How are they ever gonna support an O/S that is
    much more sophisticated than TOS.    

    Bill Powers

                                                                        

    
    re: .2
    
    according to the claims I've seen on Usenet, Atari is going to make
    a big push in the US this fall.  And the technology used isn't rocket
    science, so it is possible they could ship this fall.
    
    I'd love to see them go after the IIGS, but I suspect the target
    for the 1040STE will be the Amiga 500/1meg.  Remember the old battles
    vs. the ST?  That didn't have stereo or a palette of 4096.  Ever
    try to explain the value of a multitasking OS to a first time computer
    user?
    
    That TT will give the A2000 serious competition if it is priced
    cheap.  Running a single tasking OS on a 68030 w/2Meg is almost
    a sin.  But then again, think about how many 80386 systems with
    2Meg are still running MS-DOS.
    
    re: the u**** word
    Atari has already got a sort of unix for the ST line, something
    called IRIDUS (or something spelled close to that).
    But the rumors so far sound like Atari is working on their own
    unix.  That reminds me, Commodore seems to be taking a long time
    to get their unix out the door.  The 68020 boards have been shipping
    for a while now.... Doesn't sound like it is easy to port unix and
    also add support for custom hardware.
    
    -Dave

    
    I thought you could just 'buy' Unix from AT&T or whomever is holding the
    license these days??
    
    Ed.

	Yes you can buy UNIX from ATT, however, if the machine you want it for
is not supported by ATT, it does you no good without a lot of programming.  This
is why many companies sell variants of UNIX.

    WHen is the amiga 3000 coming??? It will blow them all away!! right??
    
    mat
    

    Re: .7
    
    If the new CBM management is any good, the next Amiga with a new
    motherboard design (32 bit) will not see the streets until late
    1990. Why? Very simple, they can't possibly design such a thingie
    including a new 32 bit custom chip set in less time without including
    some major gotchas, like a revision A vs. B evolution...
    An Amiga 2000 with a 68030 card like GVP's or the A2630 is if it
    comes to processing power completly sufficient to blow any Apple
    or Atari machine of the near future out of the water.
    
    But since CBM management usually sucks dead gerbils thru a dirty
    garden hose, be prepared for a completely messed up A3000 or A3500.
    
    Regards,
    
    <CB>

    
    Just out of curiosity, how difficult would it be to increase the Amiga
    color palette in a compatible kind of way?  I assume all those nifty
    palette tools built into application software wouldn't work properly...
    
    For that matter, can Atari expect major incompatibilities with an
    enlarged palette?
    
    Ed.
    

    Re: .9
    
    If done properly (there's been a discussion of this on UseNet recently)
    an old palette tool wouldn't be sufficient, but still functional.
    
    Ya can bet that old Atari software will face the same or worse 
    difficulties.
    
    Regards,
    <CB>

    
    Maybe that's a good argument for a standard palette requestor included
    in the system software, like the rumored file requestor.
    
    Ed.

RE: .8

>    If the new CBM management is any good, the next Amiga with a new
>    motherboard design (32 bit) will not see the streets until late
>    1990. Why? Very simple, they can't possibly design such a thingie
>    including a new 32 bit custom chip set in less time

Actually, I think that "late 1990" is impossible if they only began
during the management shakeup.  However, there have been hints that
they began the process a while back.

>    But since CBM management usually sucks dead gerbils thru a dirty
>    garden hose, be prepared for a completely messed up A3000 or A3500.

I've been afraid of a 68030 machine with interfaces to the current
custom ship set.  That qualifies as messed up to me.  (A new 32 bit
chip set with 16 million colors, 256 color registers, support for 
1200 by 1200 pixels seems about right, though.)

    re: futures
    
    why a new chip set?  How about evolving the OS to support graphics
    boards so that it isn't tied so tightly to one chip set design.
    By the late 1990s the Amiga will be dead unless they upgrade the
    graphics long before that (I regard the Enhanced Chip Set as a
    bandaid, not a fix for this problem).
    
    That 32 bit custom chip set could be put on an A2000 graphics board.
    (32 bit wide access to it's own memory, 16 bit wide access to the
    A2000 memory.)
    
    -Dave
    
    -Dave

Re: .13

>    why a new chip set?  How about evolving the OS to support graphics
>    boards so that it isn't tied so tightly to one chip set design.

The problem is that the Amiga OS and Amiga programs have been written
with the promise that certain graphics operations must be available.
There's a reason why Mac and Clone owners don't grab the screen
drag bar  and pull it down to reveal the screen underneath: the hardware
doesn't support a screen underneath.

Imagine explaining that some program doesn't work because the fancy
blitter miniterm stuff used by the application isn't supported by
your graphics board.

The Amiga people built a machine with some fancy graphics processors
and then designed their system to make full use of them.  This tied
the Amiga to a specific graphics architecture just as choosing a
68000 tied them to Motorola for CPU design.

    I  thought it was a fact of life in the computer industry that sooner
    or later any particular system etc would become obsoleted. Just look at
    our own product range over the last few years.
    
    On the other hand, there are a number of smaller systems (like the MAcs
    and PCs etc) that have been around a number of years and will be around for
    sometime yet. I am hoping the Amiga will fall into that category. IF
    you want the latest and greatest in advances graphics or otherwise I
    suspect it will require system replacement. I bought an Apple II+ back
    in '81 and eventually 'upgraded' to the Amiga.
    
    Regards
    	Mike
                                    

    Re: .13, .14
    
    Yeah, as stated in .14, an Amiga without a set of custom chips that
    supports all the features that we have today, won't be an Amiga
    anymore...
    If CBM doesn't fail to develop new version of that stuff (or
    supplementary chips for use with "common" graphic chips) the Amiga
    will be able to keep it's competitive edge WHILST remaining compatible.
    Anything else will be a disaster....
    
    Regards,
    
    <CB>

    
    
    Hi y'all,
    
    	It would seem that given that this machine is a multitasking
    and possibly multi-user machine wouldn't it be neat if you had a
    plug in board which would provide the graphics output, keyboard
    handling and task handling for another user?  This way one hardware
    platform could handle multiple users.  Each card would have it's
    own graphics memory and chips as well as keyboard handler chips,
    and possibly it's own processor.
    
    	I would think that multi user home machines would be nice.
    
    Harry
    

    Re .17:
    
>   	It would seem that given that this machine is a multitasking
    
    Since when?  Did TOS 1.4 make some great leap forward that I missed?
    
>   ... wouldn't it be neat if you had a plug in board which would provide
>   the graphics output, keyboard handling and task handling for another
>   user?
    
    By the time you do all that, it strikes me as more reasonable to
    network a bunch of small (possibly heterogeneous) systems around a
    file server.

    Re: .18
    
    Uh, Bill, I believe Harry is refering to an non-existent A3000 which
    was discussed later in this note.
    
    I would like to see both options, a nice, fast and well supported
    network AND the ability to plug more CPUs into the main machine
    (a la transputer...)
    
    Regards,
    
    <CB>

    re: tied to hardware
    
    Why assume only the Amiga chipset makes pull down screens
    possible?  The reason you see it so much on the Amiga is because
    they put in pull down screens as part of the OS.  Imagine how
    often you would see it if every application had to implement it.
    
    Of course, they could put it in the Amiga because they had hardware
    to support doing it with low system overhead.  But that was due
    to having a display list available, all they needed to do was 
    to manipulate the list.  Display lists aren't new, my Atari 400
    has one, other systems use raster interrupts on a per scan line
    basis to achieve a pseudo-display list.
    
    The point I'm trying to make is that the mapping of system graphics
    calls is a function of the OS.  If the hardware support doesn't
    exist, it is possible to emulate some particular behavior in software.
    Or by making that behavior do nothing (heard rumors that the some
    of the new graphics modes in the Enhanced Chip set do not allow
    pull down screens).  Each application shouldn't have to worry about
    what graphics board or chipset is attached.
    
    -Dave
    

Re: .20

>(heard rumors that the some of the new graphics modes in the Enhanced
>Chip set do not allow pull down screens).

I haven't heard any rumors to that effect.  However, Commodore has stated
that the Viking and Hedley monitors (1008 by 800 resolution) do not allow
pull down screens.  Since those monitors include a frame buffer that
pastes several (6?) normal Amiga screens together to display as one large
screen, I can understand the restriction.

>    The point I'm trying to make is that the mapping of system graphics
>    calls is a function of the OS.  If the hardware support doesn't
>    exist, it is possible to emulate some particular behavior in software.
>    Or by making that behavior do nothing ...

The problem is such a graphics board may support 60% of the current Amiga
graphics perfectly, and support 20% of the remaining Amiga graphics poorly
(slowly or with small differences), and fail to support the remaining 20%
of Amiga graphic capabilities at all.

Sounds like the machine is 60% to 80% of an Amiga to me.

Consider the following: suppose the next Amiga didn't support HAM mode
at all.  How many applications would now be useless?  How many picture
files would need conversion to a different image format?

>   Each application shouldn't have to worry about
>    what graphics board or chipset is attached.

I agree.  However, there are two ways of doing that: hardware compatibility
or software compatibility.  Commodore has already painted itself into
a corner by promising hardware register-level compatibility.

Commodore did get one major benefit from standardizing the hardware:
all Amigas support all modes (except for extra-halfbrite), and program
developers don't have to worry about "I can't use HAM because only 40%
of all Amiga graphics boards support it."

My point is this:  Even without the promise of register-level compatibility,
Commodore has provided a certain number of "weird" features: copper lists,
HAM mode, extra-halfbrite mode, etc.  I suspect that the requirement of
emulating all of these features well enough that the resulting machine
could be called an Amiga is very difficult, verging on impossible.

I certainly wouldn't mind being proved wrong, and seeing third party
graphics boards produced for the Amiga.  It's just that I am skeptical
that it would work as well for the Amiga as it has for clones and Macs:
Clones and Macs never promised much in the way of unusual graphics.

Personally, I think that Commodore took on a big job by going with 
proprietary graphics chips.  Is Commodore up to providing the next
several generations of graphics chips for the Amiga?  I believe that
they will be able to do a 32 bit version suitable for the next generation;
I don't know if they'll be able to do the generation after that.

    I remember reading about the Apple IIGS in byte and how Apple managed
    to get a very high level of compatiblity even with the fairly major
    change in architecture.
    
    Admittedly the Amiga is a much more complex system to start with but
    still...
    
    Regards
    	Mike

    re: .21
    
    The Viking or A2024 monitors generate the picture from 4 640x400
    frames...
    
    Regards,
    
    <CB>

    re:.22
    
    Computer Shopper has an article on the new genlock board for the
    ][GS, says that the board has the custom graphics chips on it (same
    as the ones in the GS), and wonders if IIe owners could somehow
    access those chips to get GS graphic modes.
    
    I wonder if CBM could do the same thing... there was talk a while
    ago about chaining together multiple custom chips to get more color
    registers, etc.
    
    re:.21 & .23
    So CBM has already declared that there is a graphic mode that doesn't
    support 100% of all the standard functions... What happens when
    a naive user assumes he can pull down the 1000x800 screen?
    Does it ignore him, give him a "nope, can't do that" message, or
    what?
    
    -Dave

Re: .24

>    So CBM has already declared that there is a graphic mode that doesn't
>    support 100% of all the standard functions... What happens when
>    a naive user assumes he can pull down the 1000x800 screen?
>    Does it ignore him, give him a "nope, can't do that" message, or
>    what?

I assume that the user is just ignored: that's how I'd implement it.

Of course, this isn't a compatibility problem as new graphics modes
don't need to provide all the capabilities of old graphics modes, just
so long as the old graphics modes are still available.

The Hedley does support the old graphics modes.  It will even function as
a FlickerFixer when doing interlace.

Re .23

>    The Viking or A2024 monitors generate the picture from 4 640x400
>    frames...

It not quite so simple.  As I remember it, it supports different screen
rates (4 to 6) out of the Amiga, as to trade off screen update time against 
memory bandwidth.  There is a update the screen 15 times a second
versus 10 times a second mode (the video refresh is much higher, of
course).

As I remember, the 15 times a second mode uses 100% of the chip memory
DMA bandwidth.

    Re: .26
    
    Yup, you're absolutly right Randy. However it's much easier to figure
    a 1280x800 screen generated from four 640x400 frames than the wierdo
    other refresh rate the Hedley supports (I believe it was 12.5 Hz).
    While it's true that the faster refresh eats all that the Amy has
    to give, it's no more strain than a 640x400x4 hires interlaced screen.
    The effects when crossing a frame border can be (depending on the
    refresh rate) very, uh, interesting.. ;-)
    
    Regards,
    
    <CB>

    The reason that GS is soo compatible is cause It almost has a apple2
    built in!!
    
    matt
    

    re .-several
    Look at the all the baggage the IBM VGA has to carry along to be
    backwards compatible with the MDA, CGA, and EGA.
    Some clone VGA's even added backwards compatibility for Hercules
    graphics. 
    
    The IBM VGA claims to be BIOS and register level backwards compatible
    so the technology of how to support wildly different graphics standards
    on the same board seems to be fairly well understood.
    
    dennis