Havemose's NG Amiga illusions (Mathias happens to be right this time)
Bennu Oing
> And - how do you justify calling OS5 a successor to AmigaOS after having
> dropped all plans of including a high-level OS 3.x compatibility layer?
> After all, UAE and the Amiga PCI card can be used under other operating
> systems as well.
For once, I couldn't agree more. I suspect their new "software"
emphasis is because they lost the AGA schematics, and haven't got adequate
manpower to either track them down, reverse-engineer them, use the many
printed references to create new (but register compatible) hardware, or
just use address shadowing. They need to get more original Amiga people
on board- and not as piecemeal "consultants".
You won't find many Amiga supporters bigger than me. When people like me
become openly critical of Amiga, it's time to get scared. And I say that
Havemose's stated approach is suicide for the Amiga. Many of his
solutions appear to be taking the easy way out, intead of doing the right
thing (or doing things the right way). His castration of 3.x API's has no
useful purpose. You can add new routines without gutting current ones. He
has yet to explain this approach to anybody's satisfaction.
The biggest verification I've seen that Havemose is the wrong man for the
job, is his failure to confirm or deny (AFAIK) that he was responsible for
the laughable blit routines in 3.0. Instead of dong one blit for each
plane, his approved approach was to one blit per line per plane. The
overhead was obscene. Put an A4000 into Super HiRes at 256 colors and you
can SEE it moving windows line by line instead of plane by plane. It's
slow as hell. What kind of person would approve of that kind of
programming? If a person doesn't know how wrong that is-- despite
blitting being very simple, and can't SEE how slow and inefficient it is,
then IMHO, he has no business heading up the OS team. We'd be better off
hiring one of those kids from Europe who write demos. :-)
If Jim Collas doesn't reign Havemose in, his Amiga NG is going to be
dismissed by the community he claims to respect. Users aren't going to
buy a machine if they have to run a kind of UAE on to use the library of
thousands of existing programs. They can do that right now on a PPC or
Intel-x86.
The appearence is one of `hey we've got 800,000 Amiga zealots who'll act
as unpaid salespeople for our new box...stick an "Amiga" logo on it and
they'll buy anything'. Sorry but that's how it looks. If it hasn't got
sprites, playfields, hardware collision detection, and a Raster-Interupt
Copper, and compatibility with existing 3.x routines, then it's not an
Amiga. That's the baseline reference to be an Amiga. They can add all
they want, but don't mess with the parts that work.
Dion Wooning
>
> On Sun, 22 Nov 1998, Mathias Ortmann wrote in reply to "Allan Havemose":
>
> > And - how do you justify calling OS5 a successor to AmigaOS after having
> > dropped all plans of including a high-level OS 3.x compatibility layer?
This is debatable. Does the mere fact that a comparibility layer exist
in
itself make the enveloping OS a succesor? Think of ixemul.library - does
this
make AmigaOS a successor to a UNIX? However, I can see what you are
getting
at. To me, and I suspect a large group of other people, AOS5 could
justifyably
be called a successor to AOS3.x if it retains enough of the flavour of
the
current OS, without shying away from doing things differently where they
are
needed. I admit this is not very specific, and it may well be that AOS5
will
end up a fantastic OS, but with too little of the original Amiga way of
doing things to feel like a true successor.
> For once, I couldn't agree more. I suspect their new "software"
> emphasis is because they lost the AGA schematics, and haven't got adequate
> manpower to either track them down, reverse-engineer them, use the many
> printed references to create new (but register compatible) hardware, or
> just use address shadowing.
Then again, there are people like Bennu that still think AGA was Great
and
prefer devoting lots of time and money to a design that will irrevocably
be held back by such a legacy, to doing a fresh design in which
everything
learned from the Amiga - both in positive and negative sense - could be
used.
> You won't find many Amiga supporters bigger than me.
Yes you will. But not all are so backwards looking as you, it is to be
hoped.
> When people like me become openly critical of Amiga, it's time to get
> scared.
ooooooooh *scared* :))
> Havemose's [...] castration of 3.x API's has no
> useful purpose. You can add new routines without gutting current ones.
Sure you can - and you end up with a bloated OS held back by obvious
shortfalls because you need to support all mistakes made in the past. As
much as I like AmigaOS, there are some things that definately could use
redesign. Now that we are moving to new hardware, why not take advantage
of this and redisign the OS to be clean, lean and modern again (as this
is why we originally liked the Amiga, right) ?
> The biggest verification I've seen that Havemose is the wrong man for the
> job, is his failure to confirm or deny (AFAIK) that he was responsible for
> the laughable blit routines in 3.0. [...]
I don't know about this. Luckily, even if it were so, AOS5 is not going
to
be his pet project; instead a team of diverse professionals will work on
it, and it is quite likely that very little code will actually come from
Havemose's hand.
> If Jim Collas doesn't reign Havemose in, his Amiga NG is going to be
> dismissed by the community he claims to respect. Users aren't going to
> buy a machine if they have to run a kind of UAE on to use the library of
> thousands of existing programs. They can do that right now on a PPC or
> Intel-x86.
Indeed, and UAE is shaping up more and more. It is just that the host
OSes
you have to run AUE under suck in their own ways, at least to many
Amigans.
> The appearence is one of `hey we've got 800,000 Amiga zealots who'll act
> as unpaid salespeople for our new box...stick an "Amiga" logo on it and
> they'll buy anything'. Sorry but that's how it looks.
Well you could read that into their intentions if you feel you are a
Amiga
zealot (even though they have repeatedly stated that the original Amiga
owners are not their main concern). Instead, I will decide for myself
whether or not I think they have come up with something interesting, and
I suggest you simply do the same.
> If it hasn't got
> sprites, playfields, hardware collision detection, and a Raster-Interupt
> Copper, and compatibility with existing 3.x routines, then it's not an
> Amiga. That's the baseline reference to be an Amiga. They can add all
> they want, but don't mess with the parts that work.
Look, if you need AmigaOS3/68k/AGA compatibility, the best thing to do
would be to take UAE, adapt it for AOS5 (provided it is good), and take
it
further by integrating it into its host environment as good as you can.
Among others:
- allow launching Amiga apps into one or more UAE environments directly
from shells/desktop of the host OS, without having to explicitly tell
it is an Amiga app or having to create appropriate icons manually;
- allow UAE windows to open directly on the host OS desktop, without
forcing them all into the same single window on the desktop;
- allow clipboard sharing (cut/paste etc) between host OS apps and UAE;
- write specialised drivers to allow UAE to drive the available
hardware
more efficiently (mainly gfx and audio subsystems: CGfx/Picasso/AHI).
Work has already started on some of these parts (e.g. Picasso drivers).
But
please don't wish for AOS5 to be held back by too much legacy issues.
---
Dion Wooning
Matt 'Neko' Sealey
Bennu Oing <bo...@waste.org> wrote in message
> For once, I couldn't agree more. I suspect their new "software"
> emphasis is because they lost the AGA schematics, and haven't got adequate
> manpower to either track them down, reverse-engineer them, use the many
> printed references to create new (but register compatible) hardware, or
> just use address shadowing. They need to get more original Amiga people
> on board- and not as piecemeal "consultants".
Where is your proof that they lost the schematics for AGA? This is another
"Siamese said they`re dropping the Classics" rumour.
Who told you? Mathias? Ha! Tim Rue???? Ralph Schmidt????????
No doubt it was no-one working near or for Amiga, or anyone who has
even a smidgen of respect for them.
> The biggest verification I've seen that Havemose is the wrong man for the
> job, is his failure to confirm or deny (AFAIK) that he was responsible for
> the laughable blit routines in 3.0. Instead of dong one blit for each
> plane, his approved approach was to one blit per line per plane. The
> overhead was obscene. Put an A4000 into Super HiRes at 256 colors and you
> can SEE it moving windows line by line instead of plane by plane. It's
> slow as hell. What kind of person would approve of that kind of
> programming?
The kind of person who had seen the new AAA chipset working? The kind
of person who knew that that method would be inifinitely faster on RTG
hardware? The kind of person who was planning on the future?
> If a person doesn't know how wrong that is-- despite
> blitting being very simple, and can't SEE how slow and inefficient it is,
> then IMHO, he has no business heading up the OS team. We'd be better off
> hiring one of those kids from Europe who write demos. :-)
I have an idea, Bennu. Why don`t you completely re-write graphics.library
and
prove him wrong? You seem to know so much better than him.
Hey, why don`t you start a new convergance company and design an (alleged)
revolutionary OS?
Can`t? Won`t? Shut the fuck up then.
> If Jim Collas doesn't reign Havemose in, his Amiga NG is going to be
> dismissed by the community he claims to respect. Users aren't going to
> buy a machine if they have to run a kind of UAE on to use the library of
> thousands of existing programs. They can do that right now on a PPC or
> Intel-x86.
You`re right. People didn`t buy an Intel to play Amiga games. They bought
it to do different things. UAE is a bonus. You don`t really think they`d
stake
the entire future of the company/OS/system on the fact that it emulates an
A1200?
And besides, it`s not as if an A1200 is expensive even more brand new. Hell,
you can get an A4000 for next to nothing. If people are that desperate to
use legacy apps, they can use the Amigas they had BEFORE they bought the NG.
Or use the emulator.
> The appearence is one of `hey we've got 800,000 Amiga zealots who'll act
> as unpaid salespeople for our new box...stick an "Amiga" logo on it and
> they'll buy anything'. Sorry but that's how it looks.
That`s probably how it is ;)
> If it hasn't got sprites, playfields, hardware collision detection, and a
Raster-
> Interupt Copper, and compatibility with existing 3.x routines, then it's
not an
> Amiga. That's the baseline reference to be an Amiga. They can add all
> they want, but don't mess with the parts that work.
Sprites, playfields, hardware collison detection, raster interrupt coppers
and
compatibility with existing routines is what Classic does.
You buy an NG and it`ll do NG stuff.
You keep your Classic to do Classic stuff. Or just use that PC/Mac/Unix box
you`ve no doubt got.
--
Matt Sealey mw...@le.ac.uk
Distributed Systems Support
Computer Centre
University of Leicester
Bill Hoggett
>Sprites, playfields, hardware collison detection, raster interrupt coppers
>and compatibility with existing routines is what Classic does.
Absolutely. There are already plenty of ongoing efforts to make the Classic
line faster, although not the aspects Bennu holds so dear. No one cares
about them that much to actually try improving them.
If the NG Amiga doesn't drop all these primitive ideas altogether, I won't
be buying one and neither will many people who want to see a really powerful
and lasting machine. Alarmingly for Bennu and his supporters, we also tend
to be the people who spend most money on our machines, so who is going to
buy this white elephant he's advocating?
The old technology is DEAD Bennu. Live with it. If that makes you believe
the Amiga is also DEAD then so be it.
Bill
Clockmeister
Bennu Oing <bo...@waste.org> wrote in article
<Pine.LNX.3.96.990505...@waste.org>...
> On Sun, 22 Nov 1998, Mathias Ortmann wrote in reply to "Allan Havemose":
>
> > And - how do you justify calling OS5 a successor to AmigaOS after
having
> > dropped all plans of including a high-level OS 3.x compatibility layer?
> > After all, UAE and the Amiga PCI card can be used under other operating
> > systems as well.
>
> For once, I couldn't agree more. I suspect their new "software"
> emphasis is because they lost the AGA schematics, and haven't got
adequate
> manpower to either track them down, reverse-engineer them, use the many
> printed references to create new (but register compatible) hardware, or
> just use address shadowing. They need to get more original Amiga people
> on board- and not as piecemeal "consultants".
AGA isn't an issue anymore, it's a dead horse. Why use an old dead chipset
as the basis for a new machine?. I'm hoping for something a little better
then what AGA is capable off, and what AAA would have been capable off.
>
> You won't find many Amiga supporters bigger than me. When people like
me
> become openly critical of Amiga, it's time to get scared.
There is a fine line between being a supporter and a fanatic. You certainly
seem to fit the latter.
And I say that
> Havemose's stated approach is suicide for the Amiga. Many of his
> solutions appear to be taking the easy way out, intead of doing the right
> thing (or doing things the right way). His castration of 3.x API's has
no
> useful purpose. You can add new routines without gutting current ones.
He
> has yet to explain this approach to anybody's satisfaction.
Something a little more cutting edge then that simplistic solution is
needed to bring the OS up to date.
> The biggest verification I've seen that Havemose is the wrong man for
the
> job, is his failure to confirm or deny (AFAIK) that he was responsible
for
> the laughable blit routines in 3.0. Instead of dong one blit for each
> plane, his approved approach was to one blit per line per plane. The
> overhead was obscene. Put an A4000 into Super HiRes at 256 colors and
you
> can SEE it moving windows line by line instead of plane by plane. It's
> slow as hell. What kind of person would approve of that kind of
> programming? If a person doesn't know how wrong that is-- despite
> blitting being very simple, and can't SEE how slow and inefficient it is,
> then IMHO, he has no business heading up the OS team. We'd be better off
> hiring one of those kids from Europe who write demos. :-)
Banging the hardware directly eh?. Why bother with an OS at all.
> If Jim Collas doesn't reign Havemose in, his Amiga NG is going to be
> dismissed by the community he claims to respect. Users aren't going to
> buy a machine if they have to run a kind of UAE on to use the library of
> thousands of existing programs. They can do that right now on a PPC or
> Intel-x86.
Why buy a new Amiga just so that you can run old software?. I'm looking
forward to a machine which is far more usefull then a modern PC, and with
software that sets it apart from anything else on the market. If it runs
the old stuff through some form of emulation, who cares, it runs doesn't
it?.
> The appearence is one of `hey we've got 800,000 Amiga zealots who'll act
> as unpaid salespeople for our new box...stick an "Amiga" logo on it and
> they'll buy anything'. Sorry but that's how it looks. If it hasn't got
> sprites, playfields, hardware collision detection, and a Raster-Interupt
> Copper, and compatibility with existing 3.x routines, then it's not an
> Amiga. That's the baseline reference to be an Amiga. They can add all
> they want, but don't mess with the parts that work.
Have you used a modern PC?. All that stuff was great in the eighties, but
it's obsolete now. (Still a marvel of modern electronic engineering though,
especially considering the timeframe...)
To me, an Amiga is a computer that fits into the same mould as that of the
A1000 when released, that is, unique capabilities, excellent performance,
and revolutionary rather then evolutionary design.
If the new Amiga fits that mould, and they pin the Amiga badge on it, it is
indeed worthy to be called an Amiga.
But that's just my opinion...
Regards,
Clockmeister.
>
>
Ralph Schmidt
>
>Bennu Oing <bo...@waste.org> wrote in message
>
>> For once, I couldn't agree more. I suspect their new "software"
>> emphasis is because they lost the AGA schematics, and haven't got
adequate
>> manpower to either track them down, reverse-engineer them, use the
many
>> printed references to create new (but register compatible)
hardware, or
>> just use address shadowing. They need to get more original Amiga
people
>> on board- and not as piecemeal "consultants".
>
another
>"Siamese said they`re dropping the Classics" rumour.
>
>Who told you? Mathias? Ha! Tim Rue???? Ralph Schmidt????????
>
>No doubt it was no-one working near or for Amiga, or anyone who has
>even a smidgen of respect for them.
>
Could you please keep me out of this thread...i have nothing to do
with AInc, with this AA demo/amigaos fanatic starting this...and not
with the list of people you name here.
It seems threads only work on .misc if they contain certain keywords
and one must be "Ralph Schmidt" to make it a selfrunner.
--
Ralph Schmidt,la...@popmail.owl.de(private),NextMail welcome
Kirk Strauser
>AGA isn't an issue anymore, it's a dead horse. Why use an old dead chipset
>as the basis for a new machine?. I'm hoping for something a little better
>then what AGA is capable off, and what AAA would have been capable off.
Furthermore, it's not even the definitive Amiga graphics standard (not
that there *is* one). I have a hopped-up 3000 (see sig) that has ECS
(although I can't verify, because I haven't seen an ECS screen in months),
and only run RTG-friendly software.
I sincerely hope noone's advocating the Windows "ideal" of 100% backward
compatibility for *everything*, including specific workarounds for buggy
software.
--
Kirk Strauser Member // http://members.dialnet.net/teknique/
Team AMIGA \X/ http://csc.smsu.edu/~strauser/RA.html
New page! See http://csc.smsu.edu/~strauser/honeypot.html for system info
John Millington
: For once, I couldn't agree more. I suspect their new "software"
: manpower to either track them down, reverse-engineer them, use the many
: printed references to create new (but register compatible) hardware, or
: just use address shadowing.
Heh. Or maybe they're not bothering with the Amiga chepset because they
judge it to be _worthless_. The conditions under which the chipset was
invented (memory was expensive) are no longer applicable. HAM is obsolete
because 24-bit frame buffer memory is cheap. Sprites are semi-obsolete
(though admitted still cool and elegant) since blitters are so damn fast
and cheap. The list goes on and on. Most of the "features" of the Amiga
chipset were actually very _very_ clever hacks for getting around the
technological and economic limitations of 1982-1985. Those limitations
are gone. It's time for a new platform that makes the most of 1999 tech
to achieve performance comparable to 2007 tech.
: You won't find many Amiga supporters bigger than me. When people like me
: become openly critical of Amiga, it's time to get scared. And I say that
: Havemose's stated approach is suicide for the Amiga.
Different people like the Amiga for different reasons. Just about any change
in any direction is likely to offend someone. I would be scared if some Amiga
supporters did _not_ openly criticize Amiga.
[lots of snippage]
: If Jim Collas doesn't reign Havemose in, his Amiga NG is going to be
: dismissed by the community he claims to respect. Users aren't going to
: buy a machine if they have to run a kind of UAE on to use the library of
: thousands of existing programs. They can do that right now on a PPC or
: Intel-x86.
True, but what is the point in buying a machine that is only useful when
emulating another? I would like a new machine that has a modern _native_
OS in addition to Amiga Classic emulation. Whether something like that
is currently available for PPC/x86 or not, is subject to debate (i.e.
depends on how much you like BeOS).
: The appearence is one of `hey we've got 800,000 Amiga zealots who'll act
: as unpaid salespeople for our new box...stick an "Amiga" logo on it and
: they'll buy anything'. Sorry but that's how it looks. If it hasn't got
: sprites, playfields, hardware collision detection, and a Raster-Interupt
: Copper, and compatibility with existing 3.x routines, then it's not an
: Amiga. That's the baseline reference to be an Amiga. They can add all
: they want, but don't mess with the parts that work.
I agree. It won't be an Amiga, except in a very legaleese "we own the
trademark" sense of the word. And yes, I think they are using that
trademark to get some out-of-the-box name recognition and psychological
association.
But that doesn't matter. Havemose isn't _trying_ to make a better Amiga.
He's trying to make something new that is _better_ than an Amiga.
You see the old Amiga userbase as simply being supporters of just the Amiga
specifically. Other people think the Amiga userbase (or at least _most_
of that userbase) consists of people who will support cool computers _in_
_general_, and that the Amiga Classic simply happens to be the only computer
to date that has been worthy enough to merit their attention. If the second
point of view turns out to be correct, then AI's plans may be good.
(I say "may" because their excellence (and even their competence) is still
unproven.)
Yog-Sothoth Neblod Zin,
John Millington
Soy Copper
> > > And - how do you justify calling OS5 a successor to AmigaOS after having
> > > dropped all plans of including a high-level OS 3.x compatibility layer?
>
> This is debatable. Does the mere fact that a comparibility layer exist
> in
> itself make the enveloping OS a succesor?
It might help. If it doesn't have routines that existing
programs can call (we're assuming a 68K emulator is in place just like on
the PPC Macs) and rely on, then it's not Amiga OS... it's a scam.
> Think of ixemul.library - does
> this
> make AmigaOS a successor to a UNIX?
No, but that's your argument. it's apples and acorns.
> > For once, I couldn't agree more. I suspect their new "software"
> > emphasis is because they lost the AGA schematics, and haven't got adequate
> > manpower to either track them down, reverse-engineer them, use the many
> > printed references to create new (but register compatible) hardware, or
> > just use address shadowing.
>
> Then again, there are people like Bennu that still think AGA was Great
It still is. The thing it needs most is a faster clock rate, pixel
decoupling, 24-bit chunky. Not real hard to do if you know anything about
engineering these kinds of circuits. Throw a good texture mapping engine
on the die and you have a good product. Why not go that way? The ugly
truth they won't tell you: they lost the schematics.
> and
> prefer devoting lots of time and money to a design that will irrevocably
> be held back by such a legacy,
You're one of those "compatibility is bad" dullards are you? Here's some
news: that's true sometimes, sometimes not. Compatibility with bad stuff
is bad. The Amiga was flexible enough that compatibility would be a plus,
and not all that hard. If they hadn't lost the AGA schematics.
BTW, you note we've already waited for years, so waiting one more and
doing it right is OK with me. Despite your mantra, you can make huge
leaps while staying compatible. Look at Cyrix, AMD, etc. They created
new, efficient circuits to execute old crappy x86 instructions. And they
didn't just copy Intel's schematics. And if you have a good design
already, implementing compatibility is usually that much easier.
> to doing a fresh design in which
> everything
> learned from the Amiga - both in positive and negative sense - could be
> used.
Please see my "False Dichotomy" thread-- and then take some EE courses.
And oh yeah, get a read mail editor. you're making a mess.
> > You won't find many Amiga supporters bigger than me.
>
> Yes you will. But not all are so backwards looking as you, it is to be
> hoped.
You're a fanatic. An ignorant one at that. Amiga wants to sell you a
bill of goods, and you just suck it right up, excuses and all. Your
experience with x86 makes you think compatibility is bad. It can be if
you want compatibility with quano hardware. But the Amiga isn't guano.
You can maintain the same resources, even many of the same register
locations, with very little effort. If they're too craven to do that,
wimp out and use shadowing. Of course, if you were an engineer, you'd
know this. This is how advances are almost always made. Haven't you ever
wondered how top notch Wintel video boxes can still do CGA, EGA, etc?
> > Havemose's [...] castration of 3.x API's has no
> > useful purpose. You can add new routines without gutting current ones.
>
> Sure you can
You fanatics will walk lock step with any marketroid spewtum that comes
out.
> - and you end up with a bloated OS
1) AmigaOS doubled from 256K to 512K under Havemose's watch.
2) It got slower overall.
3) AOS 5 is already starting at 4MB-- so much for your gripes about bloat.
It's already pretty big. We don't see you griping abot THAT.
> held back by obvious
> shortfalls because you need to support all mistakes made in the past.
You're obviously not a programmer either. Retargetability will both
cause bloat and emeliorate many of your concerns. Other compatibility
issues are simple: don't fix what's not broken. The existing AmigaOS is
still relatively lean, and works. Havemose's vision has it grow by 800%
AND lose compatibility. You and he must be smoking from the same bag. I
mean, a whole emulator is smaller than that for chrissake!
> As
> much as I like AmigaOS, there are some things that definately could use
> redesign.
Agreed. A whole list of things.
>Now that we are moving to new hardware, why not take advantage
> of this and redisign the OS to be clean, lean and modern again
Got news for you. The days of AmigaOS Assembly are over. Clean and lean
mean assem. They went to C under the impression that it's "more portable"
(apprently having never used cross-assemblers). Granted Assem is best,
but we got C. You compile it and you're done (sometimes). You act like
the Amiga is an 8088 DOS box. It was fully 32 bit and modern from
inception. When you do something right you need not scrap it (unless
there are other motives). GW/Amiga is in a rush due to market forces,
hence their current approach. Schindler spun our wheels for so long, that
other platforms may squash GW's ability to get out of the PC commodity
game. So they're rushing it, just like IBM did in 1980.
Let me ream you another way. QNX is a derivative of a 30 year old OS.
But you fanatics who repeat the official propaganda sheets aren't bitching
about that. You're hypocritical... or don't know what you're talking
about. Which is it?
>
> > The biggest verification I've seen that Havemose is the wrong man for the
> > job, is his failure to confirm or deny (AFAIK) that he was responsible for
> > the laughable blit routines in 3.0. [...]
>
> I don't know about this.
He approved them, if he didn't himself write them. Start up an AGA box
and have a laugh.
> Luckily, even if it were so, AOS5 is not going
> to
> be his pet project;
It already is: it adds some improvements but i's gotten fatter and loses
compatibilty. Same old Havemose.
> instead a team of diverse professionals
Many of whom have no business touching the Amiga since they have no
experience with it.
> will work on
> it, and it is quite likely that very little code will actually come from
> Havemose's hand.
Which is probably how he let those absurd blit routines into 3.0.
> Indeed, and UAE is shaping up more and more. It is just that the host
> OSes
> you have to run AUE under suck in their own ways, at least to many
> Amigans.
So our choice is: Run UAE now on a fast, cheap machine OR
Run UAE on an Amiga (guffaw) and wait for the vapor to solidify.
> > The appearence is one of `hey we've got 800,000 Amiga zealots who'll act
> > as unpaid salespeople for our new box...stick an "Amiga" logo on it and
> > they'll buy anything'. Sorry but that's how it looks.
>
> Well you could read that into their intentions if you feel you are a
> Amiga
> zealot
OK, I'm a zealot, you're a schill.
> (even though they have repeatedly stated that the original Amiga
> owners are not their main concern).
Which makes you wonder why Ted Waitt said legacy apps are essential.
>Instead, I will decide for myself
Too late. you're already havemosed.
> whether or not I think they have come up with something interesting
I hope they do. We're just saying we want the baseline hardware resources
and register shadowing to help with compatibility.
> > If it hasn't got
> > sprites, playfields, hardware collision detection, and a Raster-Interupt
> > Copper, and compatibility with existing 3.x routines, then it's not an
> > Amiga. That's the baseline reference to be an Amiga. They can add all
> > they want, but don't mess with the parts that work.
>
> Look, if you need AmigaOS3/68k/AGA compatibility, the best thing to do
> would be to take UAE,
UAE is a crock. Software emulation is a crock. Period. A real Amiga
doesn't need UAE.
> adapt it for AOS5 (provided it is good), and take
> it
> further by integrating it into its host environment as good as you can.
That would mean a lot of bloat-- at least 512K worth. Which is what you
claim you oppose. You really haven't a clue, have you?
> - allow clipboard sharing (cut/paste etc) between host OS apps and UAE;
You do realize the insanity you've bought into don't you?
Alex Calara
Clock> There is a fine line between being a supporter and a fanatic.
Clock> You certainly seem to fit the latter.
"A fanatic is one who can't shut up, and won't change the subject."
--
------------------------------------------------------------------------
Posted with Amiga NewsRog
------------------------------------------------------------------------
Matt 'Neko' Sealey
> >
> >No doubt it was no-one working near or for Amiga, or anyone who has
> >even a smidgen of respect for them.
>
> Could you please keep me out of this thread...i have nothing to do
> with AInc, with this AA demo/amigaos fanatic starting this...and not
> with the list of people you name here.
>
> It seems threads only work on .misc if they contain certain keywords
> and one must be "Ralph Schmidt" to make it a selfrunner.
Sorry. I just used you as an example of people who consistently flame Amiga.
Won`t happen again. But you can see my point, I can tell ;)
Christian Brandt
> > Then again, there are people like Bennu that still think AGA was Great
>
> It still is. The thing it needs most is a faster clock rate, pixel
> decoupling, 24-bit chunky. Not real hard to do if you know anything about
> engineering these kinds of circuits. Throw a good texture mapping engine
> on the die and you have a good product. Why not go that way? The ugly
> truth they won't tell you: they lost the schematics.
earth to apollo11: Continue on PLANED cruise to moon, travelling to the
andormeda-galaxy is NOT an option, especially after you nearly exploded.
First, such capabilities wouldn`t even impress my hamster, and second,
because every $3 ATI-Chipset feature MORE and third, you can buy those
stuff in unlimited quantities.
these are the adventures of the starship amiga, which boldly stays,
where everyone has already been. they haltet five years ago in their
mission to ignore everyting out of sight.
> > and
> > prefer devoting lots of time and money to a design that will irrevocably
> > be held back by such a legacy,
>
> You're one of those "compatibility is bad" dullards are you?
Compability isn`t bad, its just sometimes too expensive and a waste of resources.
> Here's some
> news: that's true sometimes, sometimes not.
Things would be a piece of cake if AmigaOS would be opensourced like
GEM, DR/NW/OpenDos, several unices, hurd and so on. But now I just have
to conclude: No OpenSource for Amiga within 2000 will drag things just
deeper into the dirt.
Look, things have turned into mere ruins for over five years, theres
nothing to earn with classic amigas anymore and so isn`t anything to
earn with an emulation or compatible hardware. Oh, yes, I also want
them, but I also want peace on earth and naked women dancing on my
table, but I don`t get em.
With one difference: an emulation is easy to do, has allreay been done,
and is acceptable stable and fast, in most times faster and better than
the originals. Only a minority of all amigas is faster than a
PentiumII-400 running UAE, which well outperformes my Amiga3000 and
features FAST graphics up to 1600x1200 in 24Bit.
Ok, and now imagine how AmigaOS rocks in native i386/ppc/sparc-code...
Maybe the best thing would be to wait another year for gateway go
chapter11 and then buy the whole amiga-stuff-trademarks-andsoon for
$10.000 and then but the damnd thing under GPL ;-)
> Compatibility with bad stuff
> is bad. The Amiga was flexible enough that compatibility would be a plus,
> and not all that hard. If they hadn't lost the AGA schematics.
AGA, concerning to Dave, Mitch and several others, was a piece of
clutchworking. Guess my PC isn`t nearly as bloathed by compability-hacks
as an Amiga4000...
> BTW, you note we've already waited for years, so waiting one more and
> doing it right is OK with me.
While waiting for the TNG-Amiga I am using Linux and Windows on x86. My
good old 486-system has been running since 1994, when amiga blew up, and
cost me just a nikel and a dime. Meanwhile its an
486dx4-160/64MB-RAM/15GB-Harddisks running linux and it costs me less
than my first Amiga2000 or my Amiga3000. Btw, quake1-timedemo does
18fps, uae is running at full Amiga2000-Speed.
> Despite your mantra, you can make huge
> leaps while staying compatible. Look at Cyrix, AMD, etc. They created
> new, efficient circuits to execute old crappy x86 instructions. And they
> didn't just copy Intel's schematics. And if you have a good design
> already, implementing compatibility is usually that much easier.
Cyrix got bankrupt because they only sold 50 millions CPUs and would
have to sell 100 millions to get revenues. So what...
> UAE is a crock. Software emulation is a crock. Period. A real Amiga
> doesn't need UAE.
SWIV and SilvwormIV run perfectly. So does Spot, Trapdoor and Finalword.
What else?
> That would mean a lot of bloat-- at least 512K worth. Which is what you
> claim you oppose. You really haven't a clue, have you?
Linux is bloathed, Windows is bloathed, my 486 is overdosed much too
much... but hey, things still run blindingly fast and cost nothing (or
at least I only paid $40 for upgrading CPU and RAM to 486dx4-160/64MB)
--
Christian Brandt
I.S.A.R Netzwerke Dienstleistungs GmbH
Arabellastr.15 - 81925 Muenchen
http://www.isar.net/ - Telefon +49-89-92091-127
Mika Yrjola
> On Wed, 5 May 1999, Dion Warwick wrote:
> > Then again, there are people like Bennu that still think AGA was Great
>
> It still is. The thing it needs most is a faster clock rate, pixel
> decoupling, 24-bit chunky. Not real hard to do if you know anything about
> engineering these kinds of circuits. Throw a good texture mapping engine
> on the die and you have a good product. Why not go that way? The ugly
> truth they won't tell you: they lost the schematics.
Oh god. And improve the design that it catches the speed difference of
a few hundred times - class, all this alone, while there are tens of
competitors who don't have to start almost decade from the
past. Realistic? Not, IMO.
> You're a fanatic. An ignorant one at that. Amiga wants to sell you a
> bill of goods, and you just suck it right up, excuses and all. Your
> experience with x86 makes you think compatibility is bad. It can be if
> you want compatibility with quano hardware. But the Amiga isn't guano.
> You can maintain the same resources, even many of the same register
> locations, with very little effort. If they're too craven to do that,
> wimp out and use shadowing. Of course, if you were an engineer, you'd
> know this. This is how advances are almost always made. Haven't you ever
> wondered how top notch Wintel video boxes can still do CGA, EGA, etc?
Haven't you ever wondered how much more *developement resources* there
have been during this decade? And now you except a marvellous
superhero-laden team to surpass it all in a year or so, doing work
tens or hundred times more efficiently, while being tied to making
compatible hardware?
A solution in software ("this access tries to start blitting") and
converting it to approriate actions using whatever hardware there
might be would take far less developement resources, and be ready far
faster, IMO. And as the whole hardware wouldn't be emulated, like UAE,
it would probably work with acceptable speed anyway.
> 1) AmigaOS doubled from 256K to 512K under Havemose's watch.
> 2) It got slower overall.
How much? I really didn't ever notice any major difference...
> 3) AOS 5 is already starting at 4MB-- so much for your gripes about bloat.
> It's already pretty big. We don't see you griping abot THAT.
Sigh. And has far more features to do. Or do you want just 1:1 clone
of AmigaOS 3.x ? Add networking, RTG and RTA and see if you can keep
it using 512 kilobytes.
> You're obviously not a programmer either. Retargetability will both
> cause bloat and emeliorate many of your concerns. Other compatibility
> issues are simple: don't fix what's not broken. The existing AmigaOS is
> still relatively lean, and works. Havemose's vision has it grow by 800%
> AND lose compatibility. You and he must be smoking from the same bag. I
> mean, a whole emulator is smaller than that for chrissake!
Who wants 1:1 replica without networking etc?
> Let me ream you another way. QNX is a derivative of a 30 year old OS.
This takes things very, very far.
> But you fanatics who repeat the official propaganda sheets aren't bitching
> about that. You're hypocritical... or don't know what you're talking
> about. Which is it?
And you are the only objective person on universe?
> UAE is a crock. Software emulation is a crock. Period. A real Amiga
> doesn't need UAE.
Your definition is the only right definition?
--
/-------------------------------------------------------------------------\
I Fantasy, Sci-fi, Computers, Marillion, Oldfield, Vangelis, Clannad, Irc I
I Odd Experiences, Worms, Tuna, Synths. See http://www.lut.fi/~myrjola/ I
\-------------------------------------------------------------------------/
Remote
[Snip]
What a bunch of crap!?!?!
Why don't you take a hike to alt.religion.amiga and stay there????
Fanatic zealots like you are the worse thing that can happen to any
platform!