Attn Commodore: You are making a Big Mistake (Hardware Manual)
Adam Costello
hardware. It was documented, so I could program it, experiment with
it, and play with it.
One of the main reasons I chose to buy an Amiga was that I was
impressed by the hardware. I knew that there was much less useful
software for the Amiga, but that's not what turns me on.
For people who want their computers to be productive and don't want to
have to think too hard, there's Apple and IBM. For people who want to
enjoy their computers, and especially for people who want to enjoy
programming their computers, there's Amiga.
If Commodore refuses to release a hardware manual for future Amigas,
they will be betraying their most loyal customers, the very customers
for whom IBM and Apple were never even an option. Commodore should
compete by striving to be different from its competitors, not by trying
to mimic them. Compatibility has always been one of the strengths of
IBM and Apple. Capability has always been the strength of Amiga. The
Amiga can do things other computers simply cannot do, but only if
programmers know how the machine works.
If the OS is documented, programmers can choose to use it for
compatibility. If the hardware is documented, programmers can choose
to use it for capability. If both are documented, the programmer has a
choice. It's a trade-off that is best evaluated by the programmer on a
per-program basis, not once-and-for-all by Commodore.
If Commodore fails to document the hardware, lots of Commodore's paying
customers will suffer. They may even desert.
And if none of the above is convincing, consider this: hackers will
figure out much of the hardware anyway, but not all of it, which is an
even more dangerous situation than the one Commodore is afraid of. If
Commodore does document the hardware, at least hardware programmers
have a chance to know what they're doing, and Commodore can make some
money from books sales.
It's not too late for a change-of-mind!
AMC
Ewout Kramer
>One of the things I loved most about my old C-64 was the custom
>hardware. It was documented, so I could program it, experiment with
>it, and play with it.
>One of the main reasons I chose to buy an Amiga was that I was
>impressed by the hardware. I knew that there was much less useful
>software for the Amiga, but that's not what turns me on.
This is exactly what is the problem right now. Look, a MAJOR part of the
Amiga owners bought their Amiga to be the successor to the C64.
Hacking registers, getting the max out of the hardware etc. etc.
To them the Amiga should be a game machine. That's it.
However, it should be clear that those days are over. Commodore has
to look at the future. If Commodore wants to avoid being known as
"game machine company nr. 1" forever , they'll have to enhance the OS
to meet the needs for advanced system layers like RTG.
>For people who want their computers to be productive and don't want to
>have to think too hard, there's Apple and IBM. For people who want to
>enjoy their computers, and especially for people who want to enjoy
>programming their computers, there's Amiga.
"Enjoy programming" stands for hitting the hardware, playing games,
use optimized gfx routines etc. etc. To be fair, I did that too and
I really enjoyed it. However you can't blame Commodore for dropping support
for that. If they don't, Commodore will be faced with Super-Nintendos,
NeoGeos etc. Clearly NOT the market Commodore wants to be in. They make
computers, not game machines. Plainly said, Commodore has to drop the
home-computer like machines.
>If Commodore refuses to release a hardware manual for future Amigas,
>they will be betraying their most loyal customers, the very customers
>for whom IBM and Apple were never even an option. Commodore should
This is the big dilemma. What should Commodore do ? Trying to keep the
hacking clients happy will doom them to make computers adapted to
game-programming, however trying to get in the market of IBM's and MAC's
will surely betray about 80% of the Amiga owners, so that's not the
solution too.
>compete by striving to be different from its competitors, not by trying
>to mimic them. Compatibility has always been one of the strengths of
>IBM and Apple. Capability has always been the strength of Amiga. The
>Amiga can do things other computers simply cannot do, but only if
>programmers know how the machine works.
Not for too long. PC's are getting faster and faster at doing graphics.
To stay in the race, Commodore has to produce new GFX cards, faster machines
etc. However, when we keep doing access at HW level we'll get at the same
rotten position PC's have been in for years: You'll have to support
all $%^#%^$#%^# graphics cards available. Do you really think game
companies will do that? The Amiga is not such a big market for them anyway
so wasting time on adapting games for multiple cards will be out of
the question.
>If the OS is documented, programmers can choose to use it for
>compatibility. If the hardware is documented, programmers can choose
>to use it for capability. If both are documented, the programmer has a
>choice. It's a trade-off that is best evaluated by the programmer on a
>per-program basis, not once-and-for-all by Commodore.
>If Commodore fails to document the hardware, lots of Commodore's paying
>customers will suffer. They may even desert.
No. I foresee chaos. When Commodore doesn't do something for the
C64-style users, other manufaturers will release faster graphics boards,
thus creating even more confusion and incompatibilities.
Commodore will go on doing 3.0, 4.0 etc, advancing the OS even more
and the HW-ers will stay as a HUGE subgroup, producing faster and better
games. In fact, you'll have the Amiga I and Amiga II.
Commodore has noticed you can't go on expanding the C64-like type of machine,
only meant to be used as a toy ("Look I can do 1000 32x32 bobs in less that
2 scanlines!").
I love the OS, and as long as it's fast enough to let me play NetHack,
I'm satisfied :-). I would never have upgraded my 2-floppy A500 otherwise if
I just wanted to play games.
>And if none of the above is convincing, consider this: hackers will
>figure out much of the hardware anyway, but not all of it, which is an
>even more dangerous situation than the one Commodore is afraid of. If
Hackers will, game companies won't. What this means is that unlike games,
not all demos will work on newer systems, or maybe on none of them.
I don't think anyone at Commodore cares about that.
Regards,
Ewout
ANTHONY FRANCIS PRESTON
will be created that will peek and poke at the new chips until most
of the neat functions(some that maybe even C= had not thought of)
can be documented(even if uncorrectly) for all to know... It will
become the Grimoree of Graphics...
Lothar Fritsch
: Amiga can do things other computers simply cannot do, but only if
: programmers know how the machine works.
lots of games don't work on my A2000/A2630/1MB Chip because some stupid
kids like self-modifying code and assume the computer to have 512k Chip
and 512k fast ram. Sad for all those people having newer machines than
the 1.3 Versions of the OS, isn't it?
: to use it for capability. If both are documented, the programmer has a
: choice. It's a trade-off that is best evaluated by the programmer on a
: per-program basis, not once-and-for-all by Commodore.
As long as I have to pay for a program, i want it to run on my Amigas.
I can't imagine a company taking back a program when I say it doesn't run
on my Amiga. I haven't seen that there's a label "not for accelerated Amigas"
or "for KS1.3" or "for AGA" only.
: If Commodore fails to document the hardware, lots of Commodore's paying
: customers will suffer. They may even desert.
You mean those people still hacking out demos on an A1000? Do you think
they all go A1200 or A4000 because of the documentation?
: have a chance to know what they're doing, and Commodore can make some
: money from books sales.
This is a true point.
Yours,
Lothar
--
Lothar Fritsch, Student der Universitaet des Saarlandes, Fachb. Informatik
e-mail: fri...@fsinfo.cs.uni-sb.de phone: (ger) 06821-72562
'Keep Scotland clean - run over a tourist!'
(Autoaufkleber auf PKW in Fort William, Schottland)
Gregory M. Stelmack
and third party companies are releasing graphics boards like crazy. Software
written to be published has GOT to start following the OS, or it simply won't
work on many machines. Look at the fits we went through finding games that
would work under 2.04 / 68030, and that is only now getting straightened out.
If you want to publish commercial software, you shouldn't be going to the
hardware any more (at least past ECS), and Commodore is providing a
sufficiently optimized OS to reduce the need for direct hardware banging.
If you don't want to publish, but just tinker, then you'll probably be able
to figure enough out to do what you want, and don't need to know the rest.
Just don't try to sell it to me...
I for one support Commodore's decision NOT to release a new Hardware manual.
--
-- Greg Stelmack (stel...@eggo.csee.usf.edu)
-- FullTime Grad Student, PartTime Amiga Salesman, PartTime Amiga Developer
-- Author of: Spades, Pro Port Analyzer Plus, more to come...
-- DISCLAIMER: The opinions reflected here are mine and mine alone.
Marc Schaefer
>Except that Commodore has already stated they have more chipsets in the pipe,
>and third party companies are releasing graphics boards like crazy. Software
>written to be published has GOT to start following the OS, or it simply won't
>work on many machines. Look at the fits we went through finding games that
>would work under 2.04 / 68030, and that is only now getting straightened out.
Agreeing. Better put more time in making Libraries, Devices, and Style Guide
books & examples than documenting hardware.
Robert Wille
|>
|> This problem can be solved very easily. In the U.S., where the coders
|> want to run their c-coded spreadsheets in 24 bit color - relase NO
|> hardware ref. In Europe, very there is still some fun involved in
|> programming, release it!
|>
|> I do feel a bit sorry for the people trying to port Linux, however,
|> but then again, C= is probably to proud of their Unix efforts so they
|> won't make it easy for other people to make a freely distributable version.
|>
|> --
|> | Stig A. Olsen | st...@ifi.uio.no | "Ja til EF" |
|> | Never hit a man with glasses. Hit him with a baseball bat. |
If you've been following the thread 'New hardware reference guide?' you
would know that someone (I forgot who) from C= stated that people working
on OS's would be able to hardware docs. Just because there isn't a
hardware reference guide doesn't mean that documentation on the hardware
doesn't exist. According to the previously mentioned article, people who
have a *legitimate* reason (creating the best-ever demo doesn't qualify
as a legitimate reason) for needing docs on the hardware will be able
to get them. I don't know if the Linux group will be able to get
documentation or not, but I hope so and it isn't too unreasonable that
they will.
--
Robert
There is evidence that the brain is like a computer. If that's true,
then there really aren't any stupid people. Just people running DOS.
JARDAR SUNDE OLSEN
>a...@wucs1.wustl.edu (Adam Costello) writes:
[stuff deleted]
>>If Commodore refuses to release a hardware manual for future Amigas,
>>they will be betraying their most loyal customers, the very customers
>>for whom IBM and Apple were never even an option. Commodore should
>
>This is the big dilemma. What should Commodore do ? Trying to keep the
>hacking clients happy will doom them to make computers adapted to
>game-programming, however trying to get in the market of IBM's and MAC's
>will surely betray about 80% of the Amiga owners, so that's not the
>solution too.
Why? "the market of IBM's and MAC's" is the very same market of the likes of
the A4000: The professional market. Whether it is MM og dtp or whatever.
>>compete by striving to be different from its competitors, not by trying
>>to mimic them. Compatibility has always been one of the strengths of
>>IBM and Apple. Capability has always been the strength of Amiga. The
>>Amiga can do things other computers simply cannot do, but only if
>>programmers know how the machine works.
[an answer for this deleted]
The only way to be able to develop the OS any considerably lenght further is
by ensuring that people re following the rules of compatibility.
>>And if none of the above is convincing, consider this: hackers will
>>figure out much of the hardware anyway, but not all of it, which is an
>>even more dangerous situation than the one Commodore is afraid of. If
>
>Hackers will, game companies won't. What this means is that unlike games,
>not all demos will work on newer systems, or maybe on none of them.
>I don't think anyone at Commodore cares about that.
And they shouldn't!! Nobody will spend $, $ and more $ on a computer just to
be looking at demos. If you buy an A4000 you probably want to do some
serious work on it. Meaning that if you have spent $1000 on some piece of
software you would like to have it work on the next edition of the OS too.
>Regards,
>
>Ewout
Jardar Sunde Olsen - An Amiga amigo forever! | Disclaimer: These
/// Also known as: | are your opinions!
/// jard...@dhhalden.no | Your opinions!!
\\\///
\/// "If I go insane, please don't put your wires in my brain"
- Pink Floyd
Stig Arne Olsen
Stig Arne Olsen
In article <1992Nov18....@beaver.cs.washington.edu>, wi...@june.cs.washington.edu (Robert Wille) writes:
> In article <1992Nov18.1...@ifi.uio.no>, st...@ifi.uio.no (Stig Arne Olsen) writes:
> If you've been following the thread 'New hardware reference guide?' you
> would know that someone (I forgot who) from C= stated that people working
> on OS's would be able to hardware docs. Just because there isn't a
Well, so they think it's better that SOME get the hardware ref., but not
all? That SUCKS!
> hardware reference guide doesn't mean that documentation on the hardware
> doesn't exist. According to the previously mentioned article, people who
I never said that the documentation did not exist, the whole discussion
here is whether it should be made available to the public or not.
I imagine it's pretty hard to design new chips without documenting how
they work :-)
> have a *legitimate* reason (creating the best-ever demo doesn't qualify
> as a legitimate reason) for needing docs on the hardware will be able
> to get them. I don't know if the Linux group will be able to get
> documentation or not, but I hope so and it isn't too unreasonable that
> they will.
I doubt it. Linux is not a commercial product, so it wouldn't surprise me
a bit if C= wouldn't release it to them. I also remember one time that
someone wanted to include some CLI-commands in their public-domain work,
this was something _everybody_ already had, and would only make it more
easy to install the utility (forgive me for not remembering the exact
program),but Commodore refused. I have no reason to think they have
become more easy to deal with now.
> --
> Robert
John Bickers
> I for one support Commodore's decision NOT to release a new Hardware manual.
There's a lot of useful stuff in the current hardware manual that
has nothing to do with graphics or audio. What about info on the
serial port hardware, for people who want to make their own
devices? Info on the CIAs, for the same reason? Etc.
> -- Greg Stelmack (stel...@eggo.csee.usf.edu)
--
*** John Bickers, TAP. jbic...@templar.actrix.gen.nz ***
*** "Radioactivity - It's in the air, for you and me" - Kraftwerk ***
Dale Larson
>Agreeing. Better put more time in making Libraries, Devices, and Style Guide
>books & examples than documenting hardware.
Well, *I* would like to have nice documentation :-)
Actually, there is a lot of internal engineering documentation, but you do have
a point because it would take work to integrate into existing documentation to
create a new hardware manual.
--
Dale Larson, Software Engineer | Anyone who believes that what I say or do
| represents the policies or procedures of
| Commodore has more lawyers than they have
dla...@commodore.com | common sense.
David Benn
Inquisition - in which thousands of volumes of books brim-full of wonderful
and sometimes dangerous knowledge, were hidden in a tower by the head of a
monastry owing to his belief that the books contained "forbidden
knowledge".
Sound at all familiar?
It's true that many people don't but their Amigas for "serious" purposes
(whatever they are), and even if they do, don't you think that people
should be given SOME credit for self-discipline? Failing that, what's so
darn catastrophic about playing around at the hardware level sometimes?
Sometimes I think that the attitudes surrounding the maintainance of Amiga
standards border on the paranoid. YES, let's have standards and stick to
them, but for cryin' out loud, don't prevent the distribution of knowledge
in order to achieve this end.
--
db...@leven.appcomp.utas.edu.au
David Benn - Applied Computing, University of Tasmania at Launceston.
Everything should be as simple as possible, but not simpler.
(Albert Einstein)
Brian Jackson - Amiga Networking
>In article <1992Nov18....@beaver.cs.washington.edu>, wi...@june.cs.washington.edu (Robert Wille) writes:
>> on OS's would be able to hardware docs. Just because there isn't a
>Well, so they think it's better that SOME get the hardware ref., but not
>all? That SUCKS!
Hey. Commodore makes the chips, Commodore makes the machines,
Commodore writes the docs, Commodore pays the engineers and Commodore
decides what does and does not get released to the public.
We did it your way up til now and it created FAR more problems for
engineering, users and developers than it was worth. So, this time,
we're doing it differently. If, in your view, that "SUCKS!" then so
be it. The fact is that "demo coders", while creating 'neato' things
to look at, do little or nothing for sales and they create an entire
generation of Amiga programmers that have failed to learn how to do it
right. The result is application software (you know, the stuff that
people pay MONEY for and expect to work on new machines) that is chock
full of silly, pointless coding errors and, often, _slower_ code than
if it had been coded properly from the start.
>> hardware reference guide doesn't mean that documentation on the hardware
>> doesn't exist. According to the previously mentioned article, people who
>
>I never said that the documentation did not exist, the whole discussion
>here is whether it should be made available to the public or not.
>I imagine it's pretty hard to design new chips without documenting how
>they work :-)
No. A discussion is where people _discuss_ a topic. What we have here
is people _demanding_ something and then complaining like children
when they are told "no".
>> have a *legitimate* reason (creating the best-ever demo doesn't qualify
>> as a legitimate reason) for needing docs on the hardware will be able
>> to get them. I don't know if the Linux group will be able to get
>> documentation or not, but I hope so and it isn't too unreasonable that
>> they will.
>I doubt it. Linux is not a commercial product, so it wouldn't surprise me
>a bit if C= wouldn't release it to them. I also remember one time that
>someone wanted to include some CLI-commands in their public-domain work,
>this was something _everybody_ already had, and would only make it more
>easy to install the utility (forgive me for not remembering the exact
>program),but Commodore refused. I have no reason to think they have
>become more easy to deal with now.
Please don't spread such bogus mythology. Anyone with an application that
requires any/all of the Workbench files on their distribution can
acquire a Workbench license for a _small_ fee and have at it. Matt
Dillon's DICE (registered version) comes with this. No big deal.
There is nothing whatsoever "hard to deal with" with that.
>> Robert
+ + + + + + + + + + + + + + + + + + + + +
Brian Jackson Amiga Networking Group, Commodore-Amiga Inc.
b...@cbmvax.commodore.com
{uunet|rutgers}!cbmvax!bj or netwo...@cbmvax.commodore.com
uva uvam vivendo varia fit
Bjoern Reese
> Hey. Commodore makes the chips, Commodore makes the machines,
> Commodore writes the docs, Commodore pays the engineers and Commodore
> decides what does and does not get released to the public.
Hail almighty Commodore! (Do you support dictatorship too?)
Just remember that Commodore is absolutely NOTHING without the public.
> We did it your way up til now and it created FAR more problems for
> engineering, users and developers than it was worth. So, this time,
So the first (and completely incomprehensive) version of the hardware
reference guide was you way of supporting "our way" ? Thanks for nothing.
Well, pick up an issue of this book, which was the only material from C=
that we had back then, and you'll find that C= also is guilty of this
problem.
> we're doing it differently. If, in your view, that "SUCKS!" then so
> be it. The fact is that "demo coders", while creating 'neato' things
> to look at, do little or nothing for sales and they create an entire
> generation of Amiga programmers that have failed to learn how to do it
Most people I know bought their Amiga because a) the games, b) the
demos, and c) the possibility of learning how to create one (game/demo)
themselves. It is my strong belief that Amiga only survived the initial
sale problems (I remember that sales were low i US, and that only the
sales in Europe - especially in Germany - rescued the Amiga) because
of kids buying a programmable game machine. If you wanted applications
back then, you bought a PC or a MAC. Now, has that "little or nothing"
to do with the sales?
BTW, do you actually KNOW any demo coders personally?
> right. The result is application software (you know, the stuff that
> people pay MONEY for and expect to work on new machines) that is chock
How much MONEY do you pay for demos? If it's anything above the prize of
a disk, you've been fooled.
> full of silly, pointless coding errors and, often, _slower_ code than
> if it had been coded properly from the start.
Most applications are written in C, or another highlevel language, not
in assembly language, like almost all demos. I doubt that the problems
above erupt from the demo scene the way you want us to believe. I think
many people have their first programming experiences on the Amiga, and
beginners _do_ make mistakes, whether they do demos or applications.
Just don't blame demo coders for _all_ the mistakes done by application
programmers.
Which leads me to another question. What is worst: a demo coder with
docs on how to bang the hardware properly, or a demo coder without
these docs?
--
Bjoern Reese | Email: bre...@imada.ou.dk
Odense University, Denmark | Voice: +45 65 932 182 (private)
Simon Lee
>I think you missed my point earlier that a lot of people buy machines
>because of the demos. This is something that is very unique for the Amiga.
>So, I think your point about not doing anything for sales is wrong.
>How many A500 vs A3000 has there been sold? If you don't release any
>new machines between the A1200 and A4000, I think you will see that the
>same thing will happen again, the A1200 will be the clear winner.
>I don't know if you guys at Commodore have even seen programs like ProTracker,
>that have become almost a standard in the demo world. Now there even exists
>mod-players for ms-dos machines!
I for one would not buy a computer because of a neat little demo. I've seen
many on the PC's, but I still wouldn't buy a PC. I bought one because it had
a multi-tasking system, where I could do more than one thing at once. I hate
to wait for processes to finish before being able to do the next thing.
>and demos. Demos are not supposed to last a lifetime. When you release new
>chip sets, people just adjust after a while. It's not so critical as you
>seem to think. If you worry about getting the best possible hardware, let
>me worry about my programs working.
Your programs, if you ever decide to sell it, would make many a frustrated
customers. Sure, you may not upgrade your computer, but those of us who do
upgrade are stuck with code made only for the past years model. Not very good
for the computer's reputation.
>So, I was right. They have to pay a fee (how much is this small
>fee, anyway?). If a program requires kickstart 2.0, it's not very likely
>that the ones who will use it do not have wb 2.0 too. And this makes it
>it impossible to do for public-domain/freeware programmers.
I'm glad they charge. It will attract the "serious" programmers to write for
the Amiga. Why would it be impossible for PD/Freeware programmers to do
(what?) They can get 2.0 specific libraries from their compiler manufacturer.
They usually come with documentation for it also. Maybe no hardware banging
code, but that's unnecessary since you can do what you want through the 2.0
library calls. Why hardware hack? People will realize that if something is
too slow, they should upgrade the processor. You might say why should the
processor be upgraded since that's more expensive rather than optimize the
code? Well if that's the case, the Amiga's as good as dead because it would
be stuck in yesteryear with old technology. The lowest Mac has an 020, the
majority of PC users have 386's. The majority of Amiga owners still have
68000's. People complain the Amiga is falling behind. That kind of hardware
banging isn't gonna help the Amiga catch up.
--
* Simon Lee * Microscopy and Imaging Resource *
* si...@ivem.ucsd.edu * Intermediate Voltage Electron Microscopy *
* su...@ucsd.edu * UC San Diego, Dept. of Neuroscience *
-------------------------------------------------------------------------
Joel Edward Swan
>In article <37...@cbmvax.commodore.com>, b...@cbmvax.commodore.com (Brian Jackso
n
> - Amiga Networking) writes:
>> In article <1992Nov19.1...@ifi.uio.no> st...@ifi.uio.no (Stig Arne O
l
>sen) writes:
>> Hey. Commodore makes the chips, Commodore makes the machines,
>> Commodore writes the docs, Commodore pays the engineers and Commodore
>> decides what does and does not get released to the public.
>
So what?
>
>> We did it your way up til now and it created FAR more problems for
>> engineering, users and developers than it was worth. So, this time,
>> we're doing it differently. If, in your view, that "SUCKS!" then so
>> be it. The fact is that "demo coders", while creating 'neato' things
>> to look at, do little or nothing for sales and they create an entire
>> generation of Amiga programmers that have failed to learn how to do it
>> right. The result is application software (you know, the stuff that
>> people pay MONEY for and expect to work on new machines) that is chock
>> full of silly, pointless coding errors and, often, _slower_ code than
>> if it had been coded properly from the start.
>
>Ok, if you didn't like my use choice of words, I'm sorry, no offence
>intended.
>
>I think you missed my point earlier that a lot of people buy machines
>because of the demos. This is something that is very unique for the Amiga.
All I can say to this is BULL. I've been with the Amiga since 1985. I know
0 (that's zero) people who dished out between $500US and $3000Us to buy an
Amiga for some demos. I DO know of dozens who have bought Amigas because
they run software that either makes money for them or can entertain for more
than 5 minutes.
>So, I think your point about not doing anything for sales is wrong.
And 99.99% of Amiga owners think your lame reasons are wrong.
>How many A500 vs A3000 has there been sold? If you don't release any
>new machines between the A1200 and A4000, I think you will see that the
>same thing will happen again, the A1200 will be the clear winner.
>I don't know if you guys at Commodore have even seen programs like ProTracker,
>that have become almost a standard in the demo world. Now there even exists
>mod-players for ms-dos machines!
Does protracker break all of the OS, kill multitasking and go entirely to
the hardware? No? Wow! So why are you using it as an illustration? It
doesn't belong here.
>I wouldn't blame you (C=) if you redesigned the whole chip-set so that all
>hardware coded stuff broke. It's not your problem if my code doesn't work!
Ah, but how wrong you are. When programs break, users get hacked off. This
gives the machine a bad name and can sink sales. Remember, you wouldn't be
the only one going directly to the hardware. Unfortunately, "serious"
applications would do the same because it's a bad habit.
>And you must realize that there is a difference between coding applications
>and demos. Demos are not supposed to last a lifetime. When you release new
Only problem is, application developers would also write broken code.
Unfortunately it can only be everyone or no one. That's the unfortunate
CRUX of the matter.
>chip sets, people just adjust after a while. It's not so critical as you
>seem to think. If you worry about getting the best possible hardware, let
>me worry about my programs working.
And CBM must also "worry" about their image. They don't need to be seen as
a crash prone machine with software that breaks too many rules.
>> + + + + + + + + + + + + + + + + + + + + +
>> Brian Jackson Amiga Networking Group, Commodore-Amiga Inc.
>> b...@cbmvax.commodore.com
>> {uunet|rutgers}!cbmvax!bj or netwo...@cbmvax.commodore.com
>> uva uvam vivendo varia fit
>
>| Stig A. Olsen | st...@ifi.uio.no | "Ja til EF" |
>| Never hit a man with glasses. Hit him with a baseball bat. |
-Joel
-------
========================================================================
/ Joel E. Swan....Pres./...Media Specialties, Ltd., Oak Forest, IL.USA /
/ & Senior Producer..../...Moody Broadcasting Network, Chicago, IL. USA /
/ Portal ID: joeles..../...joe...@cup.portal.com /
========================================================================
Gregory R. Block
>We all know that, but who buys the machines? Commodore?
Does it matter? 98% of the people who buy the machines aren't
programmers. That 2% of programming types is going to have a rather
large % being registered developers under one program or another.
Those are the people who are going to WANT to do it the way C=
says--because they see more of the future than those 98%, or even the
part of the 2% that aren't developers.
Whether you want a HRM for good intentions or bad, C= producing one
enables people to keep hitting the hardware in non-future-compatible
ways. They no longer want to see that happen. Not putting out a HRM
is the best way to do it.
Listen, it's simple. Read through the RKM-Libraries. Read how
some of those library functions are written. They say to do certain
things, or not to do certain things.
People constantly break them.
Who the hell is going to listen to a few warnings that say not to hit
the hardware in a book that lists all of the hardware to hit for you?
Not enough people to make a difference. So, looking at it from the
perspective of DIG, putting out a HRM HURTS C= MORE THAN IT HELPS
PROGRAMMERS.
See?
>I think you missed my point earlier that a lot of people buy machines
>because of the demos. This is something that is very unique for the Amiga.
>So, I think your point about not doing anything for sales is wrong.
>How many A500 vs A3000 has there been sold? If you don't release any
>new machines between the A1200 and A4000, I think you will see that the
>same thing will happen again, the A1200 will be the clear winner.
>I don't know if you guys at Commodore have even seen programs like ProTracker,
>that have become almost a standard in the demo world. Now there even exists
>mod-players for ms-dos machines!
Mods will continue to work under AGA, and they don't hit graphics
hardware. There's no reason ProTracker can't use OS routines for its
graphics, is there?
What would I like to see? I'd like to see Chris Green come out with a
demo to prove that it can be done. He's obviously got extensive
knowledge of real-time texture mapping. I say let's see it. ;)
He said that once upon a time, he was a bit of a hardware-hitter
himself. Well, show us how an old pro does it then! Come out with a
demo (doesn't even have to be source-released, y'know!) and knock our
socks off. Besides, it'll give those corporate buggers a real
pissing, make a lot of happy user groups, and a ton of happy dealers
showing off lightning-fast real-time graphics response on Amiga sales
floors.
>I wouldn't blame you (C=) if you redesigned the whole chip-set so that all
>hardware coded stuff broke. It's not your problem if my code doesn't work!
You're wrong: It's their problem. That should be obvious. How ever
much I happen to like the guys, it should be dead obvious that these
idiots take the bad programming practices of others seriously--they
put all of those bugfixes into 2.04 and even more of them into 2.1.
It's NOT meant as a slam against C= or the engineers, SW or HW. It's
just that if it's not obvious enough that C= takes the blame for buggy
programs, and that C= puts huge amounts of manpower into letting those
buggy programs get away with what they do, then I don't know how else
to show it.
>And you must realize that there is a difference between coding applications
>and demos. Demos are not supposed to last a lifetime. When you release new
>chip sets, people just adjust after a while. It's not so critical as you
>seem to think. If you worry about getting the best possible hardware, let
>me worry about my programs working.
Democoders of past become applicatoins programmers of now. I think
Chris Green might agree. And with those democoders comes the crappy
programming practices that most demos utilize. Banging hardware
becomes so instinctive that to do anything else takes more work than
it needs to due to total unfamiliarity. Things that were natural to
do with hardware become unnatural to do with the OS, which seems slow
and sluggish in comparison--or should I say DID, because that isn't
true any longer.
>Get real. I guess you would say that a discussion is a debate where
>all ends up having your opinion.
You don't seem to get it. C= didn't say "Well, guys, we're thinking
about not putting out a HRM. What's your opinion?". They said "There
shall be no HRM, mortal. Cope."
Users demanding from C= that they do something that impinges, or in
this case, devastates future plans (as it would do to any DIG plans,
or for that matter, future chipset design) is different from holding a
discussion, isn't it? I sure think so.
>So, I was right. They have to pay a fee (how much is this small
>fee, anyway?). If a program requires kickstart 2.0, it's not very likely
A small fee. Seriously. Why do you think they charge 20 bucks for
includes/autodocs? It's to retain their copyrights. It's also to
ensure quality.
>that the ones who will use it do not have wb 2.0 too. And this makes it
>it impossible to do for public-domain/freeware programmers.
It's really meant for large applications, I believe. To create
bootable disks, so to put. There are still lots of people, I imagine,
because I personally wouldn't know, that reboot their machine when
they want to run some program by using a bootable disk.
Greg
--
(: (: (: (: Have you overdosed on smileys today? Why NOT!?! :) :) :) :)
(: "You said the world was magic, I was wide eyed and laughing." :)
(: -The Indigo Girls, "Joking" :)
(: (: (: (: (: (: (: (: (: (: (: (: (:) :) :) :) :) :) :) :) :) :) :) :)
Gregory R. Block
>Hail almighty Commodore! (Do you support dictatorship too?)
>Just remember that Commodore is absolutely NOTHING without the public.
"The Public" and "Demo-coders" are two different things.
>So the first (and completely incomprehensive) version of the hardware
>reference guide was you way of supporting "our way" ? Thanks for nothing.
>Well, pick up an issue of this book, which was the only material from C=
>that we had back then, and you'll find that C= also is guilty of this
>problem.
C= matured. They came out with later reference manuals, and then the
2.0 version of the Hardware Reference Manual with ECS.
The democoders didn't.
>Most people I know bought their Amiga because a) the games, b) the
>demos, and c) the possibility of learning how to create one (game/demo)
>themselves. It is my strong belief that Amiga only survived the initial
>sale problems (I remember that sales were low i US, and that only the
>sales in Europe - especially in Germany - rescued the Amiga) because
>of kids buying a programmable game machine. If you wanted applications
>back then, you bought a PC or a MAC. Now, has that "little or nothing"
>to do with the sales?
A) the games, will still exist. There are companies now that have
pledged support for AGA. I do believe Psygnosis was one of them.
Just so you know, using the OS does NOT mean you have to use
intuition. Graphics.library can do lots of fast things.
>BTW, do you actually KNOW any demo coders personally?
Yes. And boy, are they pissed. But they're getting over it. They're
not USED to using the OS for very much, and it's a big change.
>How much MONEY do you pay for demos? If it's anything above the prize of
>a disk, you've been fooled.
Does democoding breed poor coding habits? You tell me. It's
something C= has to consider.
>Most applications are written in C, or another highlevel language, not
>in assembly language, like almost all demos. I doubt that the problems
>above erupt from the demo scene the way you want us to believe. I think
>many people have their first programming experiences on the Amiga, and
>beginners _do_ make mistakes, whether they do demos or applications.
>Just don't blame demo coders for _all_ the mistakes done by application
>programmers.
Obviously not. But good demo coding does not good application coding
make. Unfortunately, very often, they're two totally different
things. Good demo coding makes good hardware-banging games. But
there's no reason that you can't make games using the OS. Nobody has
said they'd be as fast--they won't. There is something in there. But
it's NOT as slow as some would claim.
>Which leads me to another question. What is worst: a demo coder with
>docs on how to bang the hardware properly, or a demo coder without
>these docs?
A demo coder without, of course, because he'll try to hack, slash, and
bootleg his way into the chipset.
You don't seem to understand. C= is making a hell of a statement by
not releasing a HRM. They're not saying "you're on your own", they're
saying "Don't do it any more."
Big difference, dontcha think?
Borge Noest
>Does it matter? 98% of the people who buy the machines aren't
>programmers. That 2% of programming types is going to have a rather
>large % being registered developers under one program or another.
[I think C= not releasing the AGA HWRM is the way to go, but]
I now fully understand that we live in different worlds.
Every single Amiga owner I know is programming, be it AMOS, C, or
assembler.
It must be some kind of cultural thing. People('kids') _are_ playing games,
but it's not like you either play or program; programming is another form
of having fun.
Sounds like different worlds to me.
--
|/// bor...@stud.cs.uit.no (Boerge Noest) | Amiga B2000 \\\|
|// Box 218, 9001 Tromsoe, Norway | Remember to :-) when needed \\|
|/ The worlds northernmost university | Life is worth living. \|
Erik Lundevall
>In article <37...@cbmvax.commodore.com> b...@cbmvax.commodore.com (Brian Jackson
>> to look at, do little or nothing for sales and they create an entire
>Most people I know bought their Amiga because a) the games, b) the
>demos, and c) the possibility of learning how to create one (game/demo)
>themselves. It is my strong belief that Amiga only survived the initial
Yes, that is probably true, at least for part a (the games) combined with
what kind of computer their friends used. There are probably a number
of people who bought the Amiga just for making demos etc, but compared
to the rest they are not many.
>> right. The result is application software (you know, the stuff that
>> people pay MONEY for and expect to work on new machines) that is chock
>How much MONEY do you pay for demos? If it's anything above the prize of
>a disk, you've been fooled.
And? You have missed the point. People pay money for software and expect
it to work properly when upgrading to a newer model. Obviously this has
not been fulfilled completely, since there has been compability problems
with some software when a hardware/software upgrade has been made.
One of these problem areas is caused by people banging on the hardware
directly. Now, some of these hardbangers are pure demo coders, others
are not. It does not matter, because you can't separate them in practice.
You have to look at it in a broader perspective than just the demos you
might want to write.
>> full of silly, pointless coding errors and, often, _slower_ code than
>> if it had been coded properly from the start.
>Most applications are written in C, or another highlevel language, not
>in assembly language, like almost all demos. I doubt that the problems
>above erupt from the demo scene the way you want us to believe. I think
The point here was most likely that some people have constructed their
own ways of doing things, in spite of existing support in the OS. This
has not always been more effective. Which language it was written in is
of secondary importance.
>Which leads me to another question. What is worst: a demo coder with
>docs on how to bang the hardware properly, or a demo coder without
>these docs?
Wrong question. The point is that you should not bang the hardware.
Commodore does not keep the AGA hardware info to themselves because
they want you to bang the hardware without docs. They don't want you
to do it at all.
--
-Erik Lundevall lu...@nada.kth.se
lu...@adder.adsp.sub.org
cbmehq!cbmswe!adder!lu...@cbmvax.commodore.com
Your fault. Core dumped.
Paul Kolenbrander
Hmm. Not totally true. It kinda depends on what country your in. Would it
surprize you that in the Netherlands when you want to become a registered
developer, you need to
1. Already have a product on the market
2. Be registered at the Chamber of Commerce
3. Be prepared to pay US$ 500.- a year...
Sound familiar? Yes it IS the Commercial Developer program! And it is
also the ONLY one they are offering!!! How about us Freeware/Shareware
programmers? I can meet the first two requirements, but US$ 500,- a
year? No way. I don't mind having to pay C= a reasonable sum for the
'privilige' of being registered, and getting the info I need, but $500
is way out of my league... :-(
> Brian Jackson Amiga Networking Group, Commodore-Amiga Inc.
> b...@cbmvax.commodore.com
CYa, Paul
--
Paul Kolenbrander \ InterNet: boi...@myamy.hacktic.nl
Turfveldenstraat 37 \ Fido: 2:284/114.3 Paul Kolenbrander
NL-5632 XH EINDHOVEN | - - - - - - - - - - - - - - - - - -
Voice: +31-40-415752 | Timezone:GMT+1 | Fax: +31-40-426446
pod...@qut.edu.au
>>I think you missed my point earlier that a lot of people buy machines
>>because of the demos. This is something that is very unique for the Amiga.
>
> All I can say to this is BULL. I've been with the Amiga since 1985. I know
> 0 (that's zero) people who dished out between $500US and $3000Us to buy an
> Amiga for some demos. I DO know of dozens who have bought Amigas because
> they run software that either makes money for them or can entertain for more
> than 5 minutes.
>
I know dozens of people here in Australia who bought the Amiga specifically
for the games AND DEMO scene. I know only one or two that have Amigas (a3000)
who bought it mainly for other purposes - mainly graphics and ray-tracing.
Even these people would be very unhappy if the games/demo scene died off.
James.
Paul Kienitz
to this:
Supporting applications programmers properly may be an inconvenience to demo
coders, but the kind of support you want for demos is DEADLY to the
applications market.
At least half of the problems that have existed with upgrading all amigas have
come from EXACTLY the kinds of things you want from C= to help support your
useless demo coding.
I want a machine that can offer retargetable graphics. A machine that does not
is simply not going to compete with "real" computers. I want a "real"
computer, and if you have your way the Amiga will never be one.
If you want a machine that's optimized for k00l demos, then get a damn Atari.
Don't expect C= to gamble its entire future on sales to demo fans. It's too
small and undependable a market, and it'll never support any "real" computer.
To sell Amigas for the demo fan market would be the biggest WASTE of a fine
hardware & software development to ever happen in the industry. What you're
asking for really does amount to asking C= to GIVE UP the serious app market to
satisfy the game/demo market. It really is one or the other -- trying to have
it both ways, on your terms, will never work. The kind of software that has
come into existence because of trying to do things your way has ALREADY held
the Amiga back and hurt all users of non-entertainment software.
Gregory R. Block
>I now fully understand that we live in different worlds.
Umm, last time I checked, my internet connection, my body, and my
astrally-projected self were all on earth. ;)
>Every single Amiga owner I know is programming, be it AMOS, C, or
>assembler.
>It must be some kind of cultural thing. People('kids') _are_ playing games,
>but it's not like you either play or program; programming is another form
>of having fun.
>
>Sounds like different worlds to me.
How dare you make me feel old. I just turned 20, I've ceased being a
teeny-bopper as of Nov. 20th. :)
However, all of my friends are programmers. Well, not all, but a very
large portion of. But you've got to remember that it doesn't
represent the entire amiga community--Many people's lives are spent
doing zero programming. Many more than those who do.
Remember that just because your friends are cool, it doesn't mean the
rest of the world is. After all, Intel still exists, right? ;)
Liam Greenwood
> Which leads me to another question. What is worst: a demo coder with
> docs on how to bang the hardware properly, or a demo coder without
> these docs?
hardware sing and dance without docs.
--
Liam Greenwood ------ li...@durie.amigans.gen.nz ------ Wanganui, N.Z.
Don't tell my Mother I'm a programmer,
she thinks I'm a piano player in a brothel
Bjoern Reese
Lundevall) writes:
I have erased much of your posting as I happen to agree with it. :-o
But there are a few places where you seem to have missed my points.
> In <1992Nov22....@imada.ou.dk> bre...@monet.imada.ou.dk (Bjoern
Reese) writes:
> >In article <37...@cbmvax.commodore.com> b...@cbmvax.commodore.com (Brian
Jackson
> >> right. The result is application software (you know, the stuff that
> >> people pay MONEY for and expect to work on new machines) that is chock
> >How much MONEY do you pay for demos? If it's anything above the prize of
> >a disk, you've been fooled.
> And? You have missed the point. People pay money for software and expect
> it to work properly when upgrading to a newer model. Obviously this has
Well, my point was that you don't pay anything for demos, and hence you
can't expect them to work on future machines; it would be nice if they
did, but you have no right to expect it.
> You have to look at it in a broader perspective than just the demos you
> might want to write.
Don't worry about me. I've stopped democoding some two years ago. I am
advocating the release of a new hardware guide partly because I am
opposed restrictions and limitations... sort of the cyberpunkish
information-wants-to-be-free attitude.
> >Which leads me to another question. What is worst: a demo coder with
> >docs on how to bang the hardware properly, or a demo coder without
> >these docs?
>
> Wrong question. The point is that you should not bang the hardware.
> Commodore does not keep the AGA hardware info to themselves because
> they want you to bang the hardware without docs. They don't want you
> to do it at all.
Demo-kids don't care! They want to do it their way. If C= doesn't
release a new hardware guide, the challenge and the glory becomes
much bigger for them. Creating a demo without an docs will be
considered ultra-cool. All of you don't seem to understand that
simply denying to release any docs doesn't stop them at all. I am
afraid of the damage they will cause if they don't have any (or
only unofficial) docs.
All rights reserved to say "I told you so." :) :)
Vlod Kalicun
: release a new hardware guide, the challenge and the glory becomes
: much bigger for them. Creating a demo without an docs will be
: considered ultra-cool. All of you don't seem to understand that
: simply denying to release any docs doesn't stop them at all. I am
: afraid of the damage they will cause if they don't have any (or
: only unofficial) docs.
:
: All rights reserved to say "I told you so." :) :)
:
: --
:
: Bjoern Reese | Email: bre...@imada.ou.dk
: Odense University, Denmark | Voice: +45 65 932 182 (private)
The thing is that if C= release it, they are sort of saying that 'if you
want hack the hardware - here are the docs'. Now some applications
are also written this way, so when you upgrade to a new machine, who do C=
get complaints about? the users who have hardware-hacked code.
So in order to get maximium compability (much I think most users prefer when
upgrading) they insist you don't hardware-hack.(By not releasing the docs).
Then if your software doesn't work then C= can say 'well we told you not to do
it that way etc..".
Personaly I would prefer to have software that will wont be redundant the next
time I upgrade to a new machine.
Just my opinion..
Vlod
Jesse Michael
wrote:
> I wouldn't blame you (C=) if you redesigned the whole chip-set so that
all
> hardware coded stuff broke. It's not your problem if my code doesn't
work!
> And you must realize that there is a difference between coding
applications
> and demos. Demos are not supposed to last a lifetime. When you release
new
> chip sets, people just adjust after a while. It's not so critical as you
> seem to think. If you worry about getting the best possible hardware, let
> me worry about my programs working.
I agree with this completely.
I would be more than happy if C= completely redesigned the chipset as long
as they released the documentation for it.
I would just modify my code so that it would work correctly on the new
chipset. It's not such a big deal.
-- Via DLG Pro v0.985b
Hamish Macdonald
>>>>> In message <Jesse_Mic...@amwbbs.rain.com>,
>>>>> Jesse_...@amwbbs.rain.com (Jesse Michael) wrote:
Jesse> I would be more than happy if C= completely redesigned the
Jesse> chipset as long as they released the documentation for it.
Jesse> I would just modify my code so that it would work correctly on
Jesse> the new chipset. It's not such a big deal.
Talk about missing the whole point....
Will you also provide free upgrades to all users of your code at your
cost within 1 day of the new hardware being available?
If you stay within the Amiga OS guidelines, you won't even have to
worry about that.
Erik Lundevall
>Well, my point was that you don't pay anything for demos, and hence you
>can't expect them to work on future machines; it would be nice if they
>did, but you have no right to expect it.
But then it does not matter if any docs are released or not.
>Don't worry about me. I've stopped democoding some two years ago. I am
>advocating the release of a new hardware guide partly because I am
>opposed restrictions and limitations... sort of the cyberpunkish
>information-wants-to-be-free attitude.
There is information and there is information. The effects of available
information have to be considered in different aspects.
>> Wrong question. The point is that you should not bang the hardware.
>> Commodore does not keep the AGA hardware info to themselves because
>> they want you to bang the hardware without docs. They don't want you
>> to do it at all.
>Demo-kids don't care! They want to do it their way. If C= doesn't
>release a new hardware guide, the challenge and the glory becomes
>much bigger for them. Creating a demo without an docs will be
Yes, I am quite aware of that. So they would actually be more happy without
the docs...
>considered ultra-cool. All of you don't seem to understand that
>simply denying to release any docs doesn't stop them at all. I am
>afraid of the damage they will cause if they don't have any (or
>only unofficial) docs.
Of course it does not stop everyone. I know people are working on unofficial
docs. But some will definitely think twice before banging the hardware. Maybe
not the demo guys, but those who are into producing commercial software.
And that's the point.
>Bjoern Reese | Email: bre...@imada.ou.dk
>Odense University, Denmark | Voice: +45 65 932 182 (private)
Joel Edward Swan
i
>tes:
>>>I think you missed my point earlier that a lot of people buy machines
>>>because of the demos. This is something that is very unique for the Amiga.
>>
>> All I can say to this is BULL. I've been with the Amiga since 1985. I know
>> 0 (that's zero) people who dished out between $500US and $3000Us to buy an
>> Amiga for some demos. I DO know of dozens who have bought Amigas because
>> they run software that either makes money for them or can entertain for more
>> than 5 minutes.
>>
>BULL.
>
>I know dozens of people here in Australia who bought the Amiga specifically
>for the games AND DEMO scene. I know only one or two that have Amigas (a3000)
Interesting how you put that "games" in there when I said "demos." Now, ask
how many of them will buy games that die if they try to add another floppy
drive, or some memory, or an accelerator, or..... Demos can come and go,
they are free. Once you start shelling out money, it's a different story.
The point is to keep any program from breaking - it's those programs that
get and keep customers. Did those "dozens" of people spend all that hard
earn money JUST to watch demos? No. They wanted to play games too. I'll
bet they also get interested in WPs, etc.
>who bought it mainly for other purposes - mainly graphics and ray-tracing.
>Even these people would be very unhappy if the games/demo scene died off.
>
>James.
>
As I would be unhappy. But, if I had to choose between the two I would not
pick the hardware hacking demos (this of course assumes that you can't write
a good demo without going to the hardware.)
Besides, the point is really moot. They won't be released - period.
Thorne Kontos
I think that not releasing a hardware reference manual is possibly
the worst mistake C= can make. How many other firms release a chip
without releasing the specifications? It doesn't seem to me that
they would be in business very long. 8^)
Greg Stelmack's thinking is short-sighted, much like C='s
marketing staff, not to mention most of the other staff. The engineers
are great, but how they can work in such a repressive environment
amazes me constantly. John Bicker's comments are right on the money!
People use the hardware reference manual for other things than
just the graphics. So what we have is now is a black box (AGA chip set)
that people just out of curiosity will try to figure out. And they will
succeed, and the information will slowly disseminate like it did before.
At least if they were to release a hardware reference manual for the
new AGA-equipped machines, they could make a few extra $$$$ in the
process.
Of course I have said this before, and have gone as far as stating
this is one of a number of reasons I have not trotted down to the local
dealer and forked over the cash for the A4000. I simply consider buying
a product that lacks documentation a product I don't want to have.
Jeff Dickson
>
> I think that not releasing a hardware reference manual is possibly
>the worst mistake C= can make. How many other firms release a chip
>without releasing the specifications? It doesn't seem to me that
>they would be in business very long.
A chip verses a whole implementation? Puhleaze! If Commodore sold
the chip seperately (which I doubt) - then they'd have to.
> Greg Stelmack's thinking is short-sighted, much like C='s
>marketing staff, not to mention most of the other staff. The engineers
>are great, but how they can work in such a repressive environment
>amazes me constantly. John Bicker's comments are right on the money!
Hiding the internals of the AGA chipset as well as the rest of
the stuff is necessary for DIG (device independant graphics).
> People use the hardware reference manual for other things than
>just the graphics. So what we have is now is a black box (AGA chip set)
>that people just out of curiosity will try to figure out. And they will
>succeed, and the information will slowly disseminate like it did before.
>At least if they were to release a hardware reference manual for the
>new AGA-equipped machines, they could make a few extra $$$$ in the
>process.
No plausible way to bang the hardware directly means Commodore would
not be tied to ensuring backward compatibility hardwarewise.
> Of course I have said this before, and have gone as far as stating
>this is one of a number of reasons I have not trotted down to the local
>dealer and forked over the cash for the A4000. I simply consider buying
>a product that lacks documentation a product I don't want to have.
Yeah, but you do own an Amiga. If you really abide by that statement
then you wouldn't have bought an Amiga in the first place!
Jeff
Department of Computing Science, QUT
> The thing is that if C= release it, they are sort of saying that 'if you
> want hack the hardware - here are the docs'. Now some applications
I don't see how you make this assumption. The developers know damn well
that Commodore doesn't want them to hack the hardware. So it shouldn't
make any difference wether or not they release docs.
> are also written this way, so when you upgrade to a new machine, who do C=
> get complaints about? the users who have hardware-hacked code.
>
> So in order to get maximium compability (much I think most users prefer when
> upgrading) they insist you don't hardware-hack.(By not releasing the docs).
>
So we kill the demo and game scene to increase compatibility ???
> Then if your software doesn't work then C= can say 'well we told you not to do
> it that way etc..".
>
They can say that anyway, whether they release docs or not.
> Personaly I would prefer to have software that will wont be redundant the next
> time I upgrade to a new machine.
>
> Just my opinion..
>
> Vlod
>
James
Department of Computing Science, QUT
>>I know dozens of people here in Australia who bought the Amiga specifically
>>for the games AND DEMO scene. I know only one or two that have Amigas (a3000)
> ^^^^^
> Interesting how you put that "games" in there when I said "demos." Now, ask
> how many of them will buy games that die if they try to add another floppy
> drive, or some memory, or an accelerator, or..... Demos can come and go,
> they are free. Once you start shelling out money, it's a different story.
You can't write good action games without banging the hardware. So any
arguing about demos implicitly includes arguing about action games.
Don't try and tell me you can do fast action games through the OS coz you
can't as anybody who has written one will tell you.
There is no reason for a game to die when another floppy is added, or some
memory is added just because the hardware has been hit.
Most games buyers are not about to upgrade to an accelerator so they don't
care if the damn thing will work on a different machine. All that is
necessary is writing on the package that tells what machines the game
will run on. That's what all the IBlames have been doing successfully for
many years.
> The point is to keep any program from breaking - it's those programs that
> get and keep customers. Did those "dozens" of people spend all that hard
> earn money JUST to watch demos? No. They wanted to play games too. I'll
> bet they also get interested in WPs, etc.
Please stop bring Application programs into this argument. I still don't
see that they are relevant. No application programmer would hack the
hardware nowadays so its just wasting time talking about it.
Huge amounts of people buy the amiga coz the games are as good as the
consoles, but there are more and there are demos and application programs.
If the machine no-longer had brilliant games - which it won't without
banging the hardware - then these people wouldn't give a damn about
buying an amiga or a pc. Most end-users don't even know what a computer
really is and certainly aren't going to realise that the Amiga has a
far superior operating system to the competition.
>
> As I would be unhappy. But, if I had to choose between the two I would not
> pick the hardware hacking demos (this of course assumes that you can't write
> a good demo without going to the hardware.)
As I said above, by picking that your also picking the fast arcade action
games, so your in a minority group.
>
> Besides, the point is really moot. They won't be released - period.
>
> -Joel
There's no such thing as a certainty, though I'm inclined to agree with you.
Still, the argument raises some interesting issues from time to time.
I'm much more inclined to attempt some games using the OS than I was before
it started. However, it's a pity the arcade game scene has to die out.
James
Paul Kienitz
> release a new hardware guide, the challenge and the glory becomes
> much bigger for them.
programs.
Hamish Macdonald
>>>>> In message <1992Nov27.1...@qut.edu.au>,
>>>>> pod...@qut.edu.au (Department of Computing Science, QUT) wrote:
Department> You can't write good action games without banging the
Department> hardware. So any arguing about demos implicitly includes
Department> arguing about action games. Don't try and tell me you can
Department> do fast action games through the OS coz you can't as
Department> anybody who has written one will tell you.
You people still don't get it....
With AGA and an A1200 or A4000, how do *you* or these other "game
authors" (read twits) *KNOW* that it the OS isn't fast enough without
banging the hardware.
By all accounts, the OS IS fast enough.
We're not talking ECS here.
Josef Karthauser
The point about writing games in Machine Code and "Banging the hardware" is that
the programmer has TOTAL control over the environment. There is a programming
phrase that you can't _trust_ any piece of code that wasn't written by you, and a
lot of "hackers" still subscribe to this idea. If a bug occurs you can guarantee
that it is in your code, but using the OS, which is totally unnecessary for a
SINGLE task, you can't prove that the bug is your fault. I've had really obscure
bugs which occured and weren't my fault at all!
I'm not saying that you can't or shouldn't write games that use the OS, but part
of the FUN of writing games in machine code is to see just how far you can push
the machine, try pushing the machine in C using OS calls.
Cheers,
Joe
Hamish Macdonald
>>>>> In message <1992Nov2...@cogs.sussex.ac.uk>,
>>>>> jo...@cogs.sussex.ac.uk (Josef Karthauser) wrote:
Josef> I'm not saying that you can't or shouldn't write games that use
Josef> the OS, but part of the FUN of writing games in machine code is
Josef> to see just how far you can push the machine, try pushing the
Josef> machine in C using OS calls.
So fine, do it, just do two things:
1) Don't expect C= to tell you how to use the hardware without the OS.
2) Make sure that people who get this game *know* that it is likely to
break on new machines.
Dave Proctor
>In article <70...@cup.portal.com>, joe...@cup.portal.com (Joel Edward Swan) writes:
>> [lots of stuff deleted]
>>
>> Besides, the point is really moot. They won't be released - period.
>>
>> -Joel
>There's no such thing as a certainty, though I'm inclined to agree with you.
>Still, the argument raises some interesting issues from time to time.
>I'm much more inclined to attempt some games using the OS than I was before
>it started. However, it's a pity the arcade game scene has to die out.
>James
Ok, just my 2 pennies worth.
a) If *all* games used the OS we wouldn't and couldn't have games like ZOOL etc
What would happen ? My guess, consoles like the SEGA etc would monopolise the
games market. Companies would not produce any games on the computers, as no one
in their right mind would buy a computer to buy games which were, in comparison
to the consoles, complete crap. Results : Amiga becomes a totally business machine.
Ok, it would be used for graphics as well - how many teenagers buy a computer for
graphics ??? Not many. Certainly not enuff for companies to write software for it.
b) If commodore aren't going to release a hardware ref, what would the legal position
be if, for instance, I published (on nn) a set of registers, and their function ?
- Dave.
Bjoern Reese
True, but games evolve, and at some point in time the games will
have to do some things which are too slow, going the OS way (please
remember that the gameconsoles also evolve.) At some point it
will be necessary to bang the hardware to keep up speed (and yes,
I know that at that point there will be even faster machines, but
the base machine will be the "old" AGA machines.) Point is, that
OS routines are written as all-purpose routines. Even if some of
the routines are optimized by "the guys who know" (ie. the
software engineers at Commodore) they still contain minor overhead.
To call an OS routine you need a JSR; the OS then does a JMP and
at the end of the routine a RTS. Three small instructions which
are executed in no time, but it all adds up if you have to put
up, say, 5000 dots per frame (hence calling the proper OS routine
5000 times.)
Tom R Krotchko
My experience with the 4000 doesn't bear this out. The chance of
a game working (roughly 50-50 from what I've seen) is not dependant
on any one factor other than the mood of the programmer when he
coded the stuff:
Ferinstance:
Monkey Island 2: Doesn't Work
Anything Accolade: Doesn't Work
Superbase 4: Doesn't Work.
Yet, Legends of Valour, a fast scrolling game does work. And doesn't
blow away the OS.
The "blame" is the competitive pressure of the market: most game
makers could care less about the OS, they're trying to make money.
If that means cutting a few corners, so be it. Its hard to argue with
that line of reasoning except to point out that not all game makers
do this
To...@cup.portal.com
Tom Krotchko
Osma Ahvenlampi
(.....)
>the base machine will be the "old" AGA machines.) Point is, that
>OS routines are written as all-purpose routines. Even if some of
>the routines are optimized by "the guys who know" (ie. the
>software engineers at Commodore) they still contain minor overhead.
>To call an OS routine you need a JSR; the OS then does a JMP and
>at the end of the routine a RTS. Three small instructions which
>are executed in no time, but it all adds up if you have to put
>up, say, 5000 dots per frame (hence calling the proper OS routine
>5000 times.)
Yeah, sure, like jsr+jmp+rts would mean anything when you've got to calculate
the bit position... And you'd probably have your own routine (jsr+rts) to do
it anyway...
Whatever, this shows something: A simple dot plotting has NO use for hardware.
NOONE has said it's illegal to write directly to the bitmap! If you need to
know the hardware registers for something like this, I don't want to see
your code nearer than 6 meters from my Amiga.
--
Osma Ahvenlampi - oahv...@snakemail.hut.fi * Workstation power for micro-
All my opinions are not necessarily really mine * computer price: Amiga := FUN
Denny Lee Atkin
>
>My experience with the 4000 doesn't bear this out. The chance of
>a game working (roughly 50-50 from what I've seen) is not dependant
>on any one factor other than the mood of the programmer when he
>coded the stuff:
>
>Ferinstance:
>
>Monkey Island 2: Doesn't Work
>Anything Accolade: Doesn't Work
Those will work fine if you boot in ECS mode. Running in ECS mode
with caches off fixes a TON of games that appear broken with the 4000.
-------------------------------------------------------------------------------
Denny Atkin | den...@cup.portal.com | "No matter where
Editor | Portal, GEnie, and BIX: dennya | you go, there you
COMPUTE's Amiga Resource | CIS: 75500,3602 | are." - B.Banzai
---------------=====<<<<< YES! COMPUTE _DOES_ COVER THE AMIGA! >>>>>=====------
Olaf Barthel
> In a message dated Sun 22 Nov 92 09:30, Pa...@terapin.com (paul Kienitz)
> wrote:
>
> > Give me just ONE reason not to use LoadView instead of poking the
> bitplane
> > pointers by hand. And don't try any lame answers like "you can't do
> > interleaved bitplanes through the OS" -- I've done interleaved bitplanes
> on
> > draggable intuition screens under 1.3 with no difficulty at all.
>
> Sufficiently fast double or triple buffering. There's one reason.
No longer a problem with Kickstart 3.0.
> -Jesse
--
Olaf Barthel | Internet: ol...@sourcery.mxm.sub.org
Brabeckstrasse 35 | o.ba...@a-link-h.comlink.de
D-3000 Hannover 71 | cbmvax.commodore.com!cbmehq!sourcery!olsen
MXM, ECG127 | Z-Netz: O.BARTHEL@A-LINK-H
-------------------------------------------------------------------------
Ceci n'est pas une signature.
Gregory M. Stelmack
|> It's true that many people don't but their Amigas for "serious" purposes
|> (whatever they are), and even if they do, don't you think that people
|> should be given SOME credit for self-discipline? Failing that, what's so
|> darn catastrophic about playing around at the hardware level sometimes?
It's that it's not just sometimes. There is an abundance of commercial software,
much of it games, that bangs the hardware, and when the user upgrades to a new
machine that software won't work. We shouldn't have to have "3000 / 2.04"
compatability lists, but we do. My feeling is that Commodore gave everyone
a chance, and they blew it, so now they are doing it themselves. Programmers
have not exercised restraint, so Commodore had to take action.
If Commodore were so intent on censorship and restricting knowleded, would
there be RKMs, or a Developer program? They have chosen to shut off one small
section that will be rapidly changing in the future in order to promote
good software and ease later transitions, and again I applaud them for it.
--
-- Greg Stelmack (INTERNET: stel...@eggo.csee.usf.edu BIX: gstelmack)
-- FullTime Grad Student, PartTime Amiga Salesman, PartTime Amiga Developer
-- Author of: Spades, Pro Port Analyzer Plus, more to come...
-- DISCLAIMER: The opinions reflected here are mine and mine alone.
Bjoern Reese
> If Commodore were so intent on censorship and restricting knowleded, would
> there be RKMs, or a Developer program? They have chosen to shut off one small
> section that will be rapidly changing in the future in order to promote
> good software and ease later transitions, and again I applaud them for it.
Isn't that the same as censorship? To me it is. It's not total censorship,
but it's censorship nonetheless.
--
Bjoern Reese
> In article <1992Nov28....@imada.ou.dk> bre...@imada.ou.dk (Bjoern
Reese) writes:
> (.....)
> >the base machine will be the "old" AGA machines.) Point is, that
> >OS routines are written as all-purpose routines. Even if some of
> >the routines are optimized by "the guys who know" (ie. the
> >software engineers at Commodore) they still contain minor overhead.
> >To call an OS routine you need a JSR; the OS then does a JMP and
> >at the end of the routine a RTS. Three small instructions which
> >are executed in no time, but it all adds up if you have to put
> >up, say, 5000 dots per frame (hence calling the proper OS routine
> >5000 times.)
>
> Yeah, sure, like jsr+jmp+rts would mean anything when you've got to calculate
I could calculate a lot of bit positions in the time that has been wasted
jumping back and forth (Remember to multiply the jumping time by 5000.)
> the bit position... And you'd probably have your own routine (jsr+rts) to do
> it anyway...
No, and that's one of the important ideas of optimizing - eliminating
overhead. I'll only have a loop which contains all the code (seen through
the optimizer-glasses, doing a JSR/BSR+RTS myself would be plain stupid.)
> Whatever, this shows something: A simple dot plotting has NO use for
hardware.
Ok, bad example. Let's say 5000 blitter lines instead. You know, people
have invented correct methods for doing blitter lines which are faster
than those of the OS.
> NOONE has said it's illegal to write directly to the bitmap! If you need to
> know the hardware registers for something like this, I don't want to see
> your code nearer than 6 meters from my Amiga.
Watch out! My code has a range of 10 meters :) :)
> --
> Osma Ahvenlampi - oahv...@snakemail.hut.fi * Workstation power for
micro-
> All my opinions are not necessarily really mine * computer price: Amiga :=
FUN
--
Michael van Elst
>Ok, bad example. Let's say 5000 blitter lines instead. You know, people
>have invented correct methods for doing blitter lines which are faster
>than those of the OS.
The question is _how much_ faster.. A speedup of 10% might be good for
a competition ("I can draw 10 lines more than you..") but otherwise you
can forget such an 'improvement'.
Regards,
--
Michael van Elst
UUCP: universe!local-cluster!milky-way!sol!earth!uunet!unido!mpirbn!p554mve
Internet: p55...@mpirbn.mpifr-bonn.mpg.de
"A potential Snark may lurk in every tree."
Bjoern Reese
> In <1992Nov29....@imada.ou.dk> bre...@monet.imada.ou.dk (Bjoern
Reese) writes:
> >Ok, bad example. Let's say 5000 blitter lines instead. You know, people
> >have invented correct methods for doing blitter lines which are faster
> >than those of the OS.
>
> The question is _how much_ faster.. A speedup of 10% might be good for
> a competition ("I can draw 10 lines more than you..") but otherwise you
> can forget such an 'improvement'.
A 10% speedup might be the difference between a playable and an unplayable
game (same design, different speeds.) We must remember that games evolve
all the time. Sufficient speed today, doesn't necessarily mean sufficient
speed tomorrow. A 2D Rainbow Island may become a 3D vector Rainbow Island.
Amiga games are up against consoles with dedicated hardware (where
hardware banging is allowed :)
Loren James Rittle
>In article <1992Nov25.0...@ariel.ec.usf.edu>
>stel...@wraith.csee.usf.edu (Gregory M. Stelmack) writes:
>>If Commodore were so intent on censorship and restricting knowleded, would
>>there be RKMs, or a Developer program? They have chosen to shut off one small
>>
>>good software and ease later transitions, and again I applaud them for it.
>Isn't that the same as censorship? To me it is. It's not total censorship,
>but it's censorship nonetheless.
Last time I checked the above was called a ``trade secret'', not
censorship... :-)
As always, a company that releases a product of this nature (i.e.
not food, not hazardous and not used in critical places) can fail
to disclose any set of information on the details of a product.
Regards,
Loren
--
Loren J. Rittle (rit...@comm.mot.com)
Systems Technology Research (IL02/2240)
Motorola, Inc.
(708)576-7794
Michael T Pins
->>You can't write good action games without banging the hardware.
->
->My experience with the 4000 doesn't bear this out. The chance of
->a game working (roughly 50-50 from what I've seen) is not dependant
->on any one factor other than the mood of the programmer when he
->coded the stuff:
->
->Ferinstance:
->
->Monkey Island 2: Doesn't Work
->Anything Accolade: Doesn't Work
>Those will work fine if you boot in ECS mode. Running in ECS mode
>with caches off fixes a TON of games that appear broken with the 4000.
Let's make sure our terminology is straight here....
Running in ECS mode *doesn't* fix a ton of games that appear broken,
it allows a ton of games that *are* broken to still run on the A4000.
--
*****************************************************************************
* Michael Pins | Internet: ami...@isca.uiowa.edu *
* ISCA's Amiga Librarian | #include <std.disclaimer> *
*****************************************************************************
Mattias Myrberg
>A demo coder _with_ docs. A real coder will work out how to make the
>hardware sing and dance without docs.
Yes ! Indeed he will. I don't understand why we have this silly
discussion anyway.In the end when demo-coders have found out how
AGA works, all these persons that want NO HREF to be released will
be there drooling and wish that they could do all those things.
As a matter of fact this is a question of wether the artist should use
the best brushes and the high-quality paint, or ask a monkey to do the painting.
The freedom of the artist, see ?
--------- ---------
Mattias Myrberg Uppsala Univ.
David Benn
Point taken.
--
db...@leven.appcomp.utas.edu.au
David Benn - Applied Computing, University of Tasmania at Launceston.
Everything should be as simple as possible, but not simpler.
(Albert Einstein)
Mr. Subliminal
the quality of public domain programs for the Amiga, giving the edge to
software companies who I would assume would be able to get such information.
This may be bad for the pd side of the amiga but it may help get more software
companies interested in developing for the amiga if they don't have to compete
with pd programs that are basically free in most cases. From a business
viewpoint it's probably a good idea, in any case application programs that are
public domain should use the operating system calls.
Byron...
___________________________________________________________________________
| /// "Byron Montgomerie" Internet: by...@odie.cs.mun.ca /// |
| \\\/// "Does anybody here remember Vera Lynn? \\\/// |
| \/// How she said that we would meet again, \/// |
| Amiga some sunny day..." -- Pink Floyd -- Amiga |
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Russell McOrmond
In a previous article, bre...@imada.ou.dk (Bjoern Reese) says:
>have to do some things which are too slow, going the OS way (please
>remember that the gameconsoles also evolve.) At some point it
Yes, and you also have to realize that a majority of the game console
authors go through the operating system on these machines. There may not
be a 'DOS' or something like that, but there is a very large amount of
hardware independance which allow for expansion without having to always
have new cartridges. It encourages upgrades to the latest and greatest
console, something that the Amiga Game authors don't seem to want to
allow.
A few real stupid statements have been said in this thread. One is that
allowing for direct hardware hitting is going to allow the Amiga to
'stay above the game consols'. This could be the furthest statement from
the truth. Once you limit a game to only running on a SINGLE Amiga
('This Game runs on an A500 with ECS', 'This game only runs on an A1200...')
then there is no advantage to using the Amiga at all as you have to have
a separate machine for your 'Game Playing' anyways. So, people will opt
for the game consols where the more popular games are available.
The second statement that has been made is that the 'Average MS-DOS
machine is more different than each other than the Average Amiga'. These
people don't seem to see past the Video and Audio of the computers and
realize that it is only those two aspects of the Amiga that are
hardware specific. They also don't seem to realize that as far
as hardware independance is concerned that the Amiga is FAR superior to an
MS-DOS machine in ALL respects. Even the Video and Audio are only
hardware dependant at the lowest level. One can write an
application that does basic screen/console operations without large
amounts of graphics and can do that in a TOTALLY hardware independant
manner. When it comes to things like add on serial ports and parallel ports,
I find it amusing that these people don't seem to recognize the power
of the .device interface and recognize that something as standardized is
not in use on the PC (Yes, Fossil exists, but not EVERY application uses
it as EVERY serial application on the Amiga uses the .device interface).
Oh, well ;-)
--
Russell McOrmond, Ottawa Ontario, Canada | Standard Disclaimer applies.
Freenet: aa...@freenet.carleton.ca (Faster) | Current WELMAT 'keeper of
Home: r...@Atronx.OCUnix.On.Ca, 1:163/109 | sources.
Gregory M. Stelmack
>In article <70...@cup.portal.com>, joe...@cup.portal.com (Joel Edward Swan) writes:
>>>because of the demos. This is something that is very unique for the Amiga.
>>
>> All I can say to this is BULL. I've been with the Amiga since 1985. I know
>> 0 (that's zero) people who dished out between $500US and $3000Us to buy an
>> Amiga for some demos. I DO know of dozens who have bought Amigas because
>> they run software that either makes money for them or can entertain for more
>> than 5 minutes.
>>
>BULL.
>
>I know dozens of people here in Australia who bought the Amiga specifically
>for the games AND DEMO scene. I know only one or two that have Amigas (a3000)
>Even these people would be very unhappy if the games/demo scene died off.
>
>James.
Overseas, maybe. Here in the US, it is very small. The kind of demos that sell
machines here (at least in the store I work for) are fast VistaPro anims
(run a VANIM off the harddrive on a 4000 -- impressive!), seeing PageStream
or Final Copy work, watching ADPro (and now ImageFX) work, and seeing the
Toaster in action. Occasionally, people buy just for games. Never sold a one
based on a EuroDemo, and most who buy for games constantly complain about
harddrive installability / not working on 3000s (2.04 and accelerators),
exactly the kinds of things releasing AGA docs will perpetuate.
Those in the hardware hacking games / demo area still have the ECS to tinker
with, but let's let the serious users move on with AGA and what comes after...
Paul Kolenbrander
> >You can't write good action games without banging the hardware.
>
> My experience with the 4000 doesn't bear this out. The chance of
> a game working (roughly 50-50 from what I've seen) is not dependant
> on any one factor other than the mood of the programmer when he
> coded the stuff:
>
> Ferinstance:
>
'Scuse me? Guess what's running in the background right now? :-) Yep
Superbase 4, version 1.0 1st April 1991... I've not tested the forms
editor, but I can still run applications I created on my A2500. Pray
tell me what happens at your end and what version you have...
CYa, Paul
--
Paul Kolenbrander \ InterNet: boi...@myamy.hacktic.nl
Turfveldenstraat 37 \ Fido: 2:284/114.3 Paul Kolenbrander
NL-5632 XH EINDHOVEN | - - - - - - - - - - - - - - - - - -
Voice: +31-40-415752 | Timezone:GMT+1 | Fax: +31-40-426446
Tomas Arvidsson
>In article 19739 Liam Greenwood writes:
>> A demo coder _with_ docs. A real coder will work out how to make the
>> hardware sing and dance without docs.
> Yes ! Indeed he will. I don't understand why we have this silly
> discussion anyway.In the end when demo-coders have found out how
> AGA works, all these persons that want NO HREF to be released will
> be there drooling and wish that they could do all those things.
Will they? And why will they not be able to do the same things?
All that's needed is documented together with the rest of the OS.
> As a matter of fact this is a question of wether the artist should
> use the best brushes and the high-quality paint, or ask a monkey to
> do the painting.
No, it is a question of weather you know what you are doing or not.
Most people banging at the hardware don't really know what they are
doing or what will happen when someone try to run their code on a
different machine configuration.
> The freedom of the artist, see ?
Yeah, and graffitti is also art as long as it's in an art gallery.
>Mattias Myrberg Uppsala Univ.
/Tomas
--
Tomas Arvidson ///
d91...@csd.uu.se, ///
tom...@nada.kth.se, tom...@sprawl.adsp.sub.org \\\///
Expressed opinions are mine unless stated otherwise \XX/
--
Tomas Arvidson ///
d91...@csd.uu.se, ///
tom...@nada.kth.se, tom...@sprawl.adsp.sub.org \\\///
Expressed opinions are mine unless stated otherwise \XX/
Vlod Kalicun
: 2) Make sure that people who get this game *know* that it is likely to
: break on new machines.
Yes! and give their maoney back if it fails to comply..
Vlod
monty@sagpd1
>
>Most people I know bought their Amiga because a) the games, b) the
>demos, and c) the possibility of learning how to create one (game/demo)
>themselves. It is my strong belief that Amiga only survived the initial
>sale problems (I remember that sales were low i US, and that only the
>sales in Europe - especially in Germany - rescued the Amiga) because
>of kids buying a programmable game machine. If you wanted applications
>back then, you bought a PC or a MAC. Now, has that "little or nothing"
>to do with the sales?
>
Most people DID NOT buy an Amiga, because of the games. The game image is
probably the second largest reason that the Amiga has not sold in the US
(lack of CBM Marketing/support in 1986 was the first reason).
Ask any serious IBMer or MACer and all you here is "Oh you mean that game
machine". If the fanatic demo makers would spend their time making demos
of productive software, ie CAD, wordprocessing, multimedia presentations,
and then produce the real application package, then maybe, just maybe, you
would see US sales soar.
Multimedia and desktop video are nice niches, but they will never support
the Amiga on there own, at least not on the scale we would all like to see.
It's always been my belief that games were the downfall of the Amiga in
the US. So you might say the the game makers actually sunk the Amiga
and didn't save it at all.
IMHO of course,
Monty Saine
Tom R Krotchko
>
>'Scuse me? Guess what's running in the background right now? :-) Yep
>Superbase 4, version 1.0 1st April 1991... I've not tested the forms
>editor, but I can still run applications I created on my A2500. Pray
>tell me what happens at your end and what version you have...
Calling up Oxxi reveals that its not a Superbase problem as such; it
happens to "certain machines". They don't know what causes it, nor
do they know when they'll have a fix.
Here's what happens:
You start Superbase 4, the disk spins for a while, and then...nothing.
The forms editor starts, a requester pops up saying "Not Enough Memory",
and then software failure requester.
I haven't done anything serious like yanking the GVP Series II card, YET.
But GVP is sending me new ROMS free because of a known incompatibility
problem with the 4000. I asked tech support to describe the known problems,
his only response was "its not worth getting into". I thought it was, but
he didn't seem eager to talk, and I was paying for the call, so I didn't
pursue it.
>CYa, Paul
Probably not. But if you're in the area, you're welcome to stop by
for a drink and some cheer. We are approaching the Christmas season
after all.
To...@cup.portal.com
Tom Krotchko
Tom R Krotchko
>
>'Scuse me? Guess what's running in the background right now? :-) Yep
>Superbase 4, version 1.0 1st April 1991... I've not tested the forms
>editor, but I can still run applications I created on my A2500. Pray
>tell me what happens at your end and what version you have...
Calling up Oxxi reveals that its not a Superbase problem as such; it
Josef Karthauser
Who says that C= won't tell us how to use the hardware? Their track record up to
date has been pretty good. If you follow their specifications for what to do and
what not to do there will be NO problems with newer versions of the firmware. The
problem is that most people have to wait until a hardware manual is published to
know what to do, but if you are an official developer you get told by C= as
things happen! If C= didn't want us to play with the hardware why do they tell us
how to?
Also if you play by the rules it WILL work on new machines!!! Look at the list of
do's and dont's in the front of the Hardware Guide.. Looks pretty specific to me!
Cheers,
Joe!
Josef Karthauser
|> In article <Jesse_Mic...@amwbbs.rain.com> Jesse_...@amwbbs.rain.com (Jesse Michael) writes:
|> > In a message dated Sun 22 Nov 92 09:30, Pa...@terapin.com (paul Kienitz)
|> > wrote:
|> >
|> > > Give me just ONE reason not to use LoadView instead of poking the
|> > bitplane
|> > > pointers by hand. And don't try any lame answers like "you can't do
|> > > interleaved bitplanes through the OS" -- I've done interleaved bitplanes
|> > on
|> > > draggable intuition screens under 1.3 with no difficulty at all.
|> >
|> > Sufficiently fast double or triple buffering. There's one reason.
|>
|> No longer a problem with Kickstart 3.0
Fine.. So I can write a game using OS calls and it will only work on KS3.0!!
Fine, except that it will have NO commercial value as only a handful of people
will be able to use it. If it's written to "bang" the hardware and written
following C= guidelines then there is no reason why it shouldn't work on ANY
amiga, unless of course C= change the hardware completely.. But then it wouldn't
be an amiga would it?
Cheers,
Joe
Stig A. Olsen
The nice thing about the amiga (up to this point) was that we
had the best of two worlds. We could have smashing good demos/games
AND we could have nice productivity software running in a
multitasking environment.
Some coders like to code in one environment, some in the other. I
think that the hardware reference manual should be released, so
that us wanting to use the hardware directly should be allowed to
do it. That way, we can still have the best of both worlds.
If you don't like demos/games don't run them. Some people say they
can make games that are fast using the OS. Maybe, but you can never
match the quality and speed of hardware coded games. There is little
use in releasing new hardware if we are restricted to using it
below its full potential.
--
| Stig A. Olsen | st...@ifi.uio.no | "Ja til EF" |
| Never hit a man with glasses. Hit him with a baseball bat. |
Author of "SnackMan" - 100 % hardware coded, works on A500-A3000
PAL, source included, available at a site near you.
Marius Mortensen
> following C= guidelines then there is no reason why it shouldn't work on ANY
> amiga, unless of course C= change the hardware completely.. But then it wouldn't
> be an amiga would it?
Yes, it would! This is the exact point where out views differ - It's the OS that
*makes* the Amiga such a nice machine, not the HW. And RTG will most certainly
teach us this.
Marius Mortensen
Chris Green
>You can't write good action games without banging the hardware. So any
>arguing about demos implicitly includes arguing about action games.
>Don't try and tell me you can do fast action games through the OS coz you
>can't as anybody who has written one will tell you.
Which games have you written? I did FS2, Jet, 3d code for Ultima UnderWorld,
3d & 2d code for Car & Driver, and graphics code for a sega game. If I was
redoing (FS2, Jet) or porting (the others) to an AGA machine right now, they'd
be running in intuition screens with full multi-tasking. And I'd also
put in the appropriate (simple) fallbacks do that they would work even
on non-amiga chips. While developing those games, I'd have enforcer
in my startup-sequence (as I do now), and MungWall. They would be
standard loadfiles with the data files in standard dos format (as
all pc games are), most likely with "look up a word in the manual" protection.
They'd work on all amigas, and they'd work on future OS versions without
any of my compatriots here having to put special hacks in for them, and without
the user having to run some "degrader" type of program. I'd spend more time
on gameplay and art then worrying about what register was set to what, and if
it could have been 3% faster if I'd taken over the machine. And in the end,
I'd make more money than you, because the time spent on gameplay would pay off in
continued sales, which would be possible because the game would work on both
current and future machines. Also, the game would get better with future
machines, because it would get faster and faster with better amigas. And,
my publisher would be happier about doing more amiga games, because there
wouldn't be millions of support calls about the game not working on
certain configurations. And if some new whiz bang chipset came out
with higher resolutions that were worth supporting, I'd have it working in
those quickly modes by changing constants and redoing art, while you'd have
to re-write your graphics routines (again).
--
*-------------------------------------------*---------------------------*
|Chris Green - Graphics Software Engineer - chr...@commodore.COM f
| Commodore-Amiga - uunet!cbmvax!chrisg n
|My opinions are my own, and do not - icantforgettheimpression o
|necessarily represent those of my employer.- youmadeyouleftaholeinthe r
|"A screaming comes across the sky..." - backofmyhead d
*-------------------------------------------*---------------------------*
Chris Green
>have invented correct methods for doing blitter lines which are faster
>than those of the OS.
>
lea my_5000LineTable(An),Am
move.w #5000-1,d4
1$: movem.w (Am)+,d0/d1/d2/d3
bsr My_Amazing_Line_Routine_Which_Pays_Attention_to_BytesPerRow_
And_Calls_WaitBlit
dbra d4,1$
jsr _LVODisownBlitter(a6)
To make this work under RTG, call GetBitMapAttr(bm,BMA_FLAGS).
If BMF_STANDARD is set, go right to the screen. If not, allocate a temporary
bitmap of the proper size, do your 5000 lines to it, and after all rendering
is complete, call BltBitMap to copy the temp to the real bitmap.
Jeff Dickson
>
>The nice thing about the amiga (up to this point) was that we
>had the best of two worlds. We could have smashing good demos/games
>AND we could have nice productivity software running in a
>multitasking environment.
>
>Some coders like to code in one environment, some in the other. I
>think that the hardware reference manual should be released, so
>that us wanting to use the hardware directly should be allowed to
>do it. That way, we can still have the best of both worlds.
When Commodore upgrades the system they have to consider the
installed base (hardware/software). Commodore is not and/or the Amiga's
popularity vast enough that they can afford to upgrade the hardware with
no regard to existing applications. Us users wouldn't be too happy either!
This further complicates the upgrade process since Commodore needs to main-
tain somekind of backward compatibility (if possible) with
programs that in this case bang the hardware directly.
If everyone used the O.S., then Commodore wouldn't have to lose any
sleep over this issue. But, that's not the way it has been. By not releasing
a hardware manual, assuring backward compatibility does not have to be a
concern in future upgrades.
Hiding the hardware is necessary for the Amiga's longetivity.
Jeff
Mike Meyer
> In article <olsen...@sourcery.mxm.sub.org>, ol...@sourcery.mxm.sub.org (Olaf Barthel) writes:
> |> In article <Jesse_Mic...@amwbbs.rain.com> Jesse_...@amwbbs.rain.com (Jesse Michael) writes:
> |> > In a message dated Sun 22 Nov 92 09:30, Pa...@terapin.com (paul Kienitz)
> |> > wrote:
> |> >
> |> > > Give me just ONE reason not to use LoadView instead of poking the
> |> > bitplane
> |> > > pointers by hand. And don't try any lame answers like "you can't do
> |> > > interleaved bitplanes through the OS" -- I've done interleaved bitplanes
> |> > on
> |> > > draggable intuition screens under 1.3 with no difficulty at all.
> |> >
> |> > Sufficiently fast double or triple buffering. There's one reason.
> |>
> |> No longer a problem with Kickstart 3.0
>
> Fine.. So I can write a game using OS calls and it will only work on KS3.0!!
> Fine, except that it will have NO commercial value as only a handful of people
> will be able to use it.
Hmm - so why are games programmers complaing about not being able to
hit the AGA hardware directly? Obviously, AGA games have no commercial
value, as only a handful of people will be able to use them. As a
matter of fact, the set of people with AGA is *smaller* than the set
with OS 3.0, so there games that use 3.0 should be *more* valuable
than games that hit the AGA hardware directly.
> If it's written to "bang" the hardware and written
> following C= guidelines then there is no reason why it shouldn't work on ANY
> amiga, unless of course C= change the hardware completely.. But then it wouldn't
> be an amiga would it?
Only if your definition of Amiga is "the hottest cheap graphics
harware available in 1985". Some of us would like graphis hardware
that was spec'ed and designed in the 90s. Getting that requires being
able to change the hardware completely, which means they have to get
people to quit bashing the hardware.
<mike
ANTHONY FRANCIS PRESTON
|3d & 2d code for Car & Driver, and graphics code for a sega game. If I was
|redoing (FS2, Jet) or porting (the others) to an AGA machine right now, they'd
|be running in intuition screens with full multi-tasking. And I'd also
|put in the appropriate (simple) fallbacks do that they would work even
|on non-amiga chips. While developing those games, I'd have enforcer
|in my startup-sequence (as I do now), and MungWall. They would be
|standard loadfiles with the data files in standard dos format (as
|all pc games are), most likely with "look up a word in the manual" protection.
| They'd work on all amigas, and they'd work on future OS versions without
3.0 to the Amiga?
Marc Heuler
>Yeah, sure, like jsr+jmp+rts would mean anything when you've got to calculate
>it anyway...
If you are to call it 5000 times per frame you would rather 'hardcode' it,
with 5000 times the routine at once, maybe using 50KB of memory, but it's
blowing fast. You save loops, subroutine calls, returns, counting
registers, conditions.. Everything is hard-wired while assembling, not
while executing. You need such methods for an up-to-date 'euro-demo'.
>Whatever, this shows something: A simple dot plotting has NO use for hardware.
>NOONE has said it's illegal to write directly to the bitmap! If you need to
>know the hardware registers for something like this, I don't want to see
>your code nearer than 6 meters from my Amiga.
Not yet, but what about RTG which will (hopefully) soon be implemented to
the OS?
--
_____________ _____________________________________ _______________________
| Marc Heuler | UUCP: ma...@aargh.adsp.sub.org | Manche Rechner kosten |
| Wuerzburg | ZNet: marc%aargh....@aworld.zer | ein Vermoegen, manche |
| Germany | Tel.: [+49]-(0)931-53295 | nur den Verstand.. |
~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~
Per Christian deg}rd
In article <37...@cbmvax.commodore.com>, chr...@cbmvax.commodore.com (Chris Green) writes:
> In article <1992Nov27.1...@qut.edu.au> pod...@qut.edu.au (Department of Computing Science, QUT) writes:
> >You can't write good action games without banging the hardware. So any
> >arguing about demos implicitly includes arguing about action games.
> >Don't try and tell me you can do fast action games through the OS coz you
> >can't as anybody who has written one will tell you.
>
> Which games have you written? I did FS2, Jet, 3d code for Ultima UnderWorld,
> 3d & 2d code for Car & Driver, and graphics code for a sega game.
These are hardly the types of games he was talking about. How about a multi-layer
parallax scrolling shoot-em-up?
> If I was
> redoing (FS2, Jet) or porting (the others) to an AGA machine right now, they'd
> be running in intuition screens with full multi-tasking. And I'd also
> put in the appropriate (simple) fallbacks do that they would work even
> on non-amiga chips. While developing those games, I'd have enforcer
> in my startup-sequence (as I do now), and MungWall. They would be
> standard loadfiles with the data files in standard dos format (as
> all pc games are), most likely with "look up a word in the manual" protection.
Yes, YOU would probably do that, because you work for Commodore, and this how they
want it done. Do they want the kind of games we see for Windows? SLOW, simple, and
ultimately BORING stuff. I certainly don't.
> They'd work on all amigas, and they'd work on future OS versions without
> any of my compatriots here having to put special hacks in for them, and without
> the user having to run some "degrader" type of program. I'd spend more time
> on gameplay and art then worrying about what register was set to what, and if
> it could have been 3% faster if I'd taken over the machine. And in the end,
> I'd make more money than you, because the time spent on gameplay would pay off in
> continued sales, which would be possible because the game would work on both
> current and future machines. Also, the game would get better with future
> machines, because it would get faster and faster with better amigas. And,
> my publisher would be happier about doing more amiga games, because there
> wouldn't be millions of support calls about the game not working on
> certain configurations. And if some new whiz bang chipset came out
> with higher resolutions that were worth supporting, I'd have it working in
> those quickly modes by changing constants and redoing art, while you'd have
> to re-write your graphics routines (again).
NO! This is simply WRONG!!! And let me tell you why:
* Even if you DO use libraries, you're can never be 100% certain that it will be
working on future machines.
* You'd spend MORE time writing a multitasking game than writing a 'harware' one.
You'd have to make sure it in all the situations you might end up in using
Intuition and the system, you'd have to make sure you don't mess up any other
program that might be running etc etc etc. It's a lot of work!
* 3%??? This is absolutely CRAP! I'd say 50% if the programmer knows anything about
coding. Using AreaFill (or something) in a flight sim would slow it down far into
boredom city.
* No you would NOT make more money! Who would buy a slow game that they knew would be
at least twice as fast if the programmer had accessed the hardware direcely?
None of the A500/A2000 owners anyway, and they dominate the market. Maybe your game
would run at an acceptable rate on A3000, but how many people owns it? And any other
sim they'd got would be faster or more detailed. And why bother with multitasking
anyway. It't great for utilities, but if the game can return to Workbench, most (all)
players would be satisfied. Accessing hardware directly and TEMPORARILY shutting down
the system, you can guarantee a steady frame update (which lots of arcade games
require) and you can read have you own interrupts that you KNOW will execute every
fifth of a secound and you'd have a SAFE environment. In the multitasking world, you
can't be sure of anything. What's good about having fast reflexes if the game can't
catch up?
Phew! But don't get me wrong. I've got NOTHING against Intuition or libraries at all!
To tell the truth, I think it surpasses anything I've seen on other computers. But
good as it is, it's simply not suited for fast arcade games with the processor speed
of the current Amiga park. Believe me, I know what I'm talking about. I've made
'banging the hardware' stuff that works on any Amiga I've tested it on and I've also
done my fair share of safe library coding. I know how the Amiga works and I've seen
the differences in speed. Last time my 'hardware' game crashed, I found the bug
quicklt using a debugger. Last time my system friendly utility crashed, it was because
of a bug in asl.library, which I could do NOTHING about, so I had to use ReqToolsLib
instead. So if you think as a games programmer, and not as a Commodore emnplyee,
Chris, you'd want the new hardware reference guide too. I know the rest of us do.
Commodre have made some mistakes in the past, what with the Plus/4, C16 A600 without
numeric keyboard (compatibility???), but this may be their worst. And final?
Per Christian Odegard (pe...@ifi.uio.no)
Josef Karthauser
Hmmm excuse me.... But am I missing something?
Are you saying that without out the Hardware that the Amiga has it would still be
just as good.. as long as it had Amiga Dos?
Is an ST with Amiga Dos an Amiga?
Joe.
Josef Karthauser
> In article <olsen...@sourcery.mxm.sub.org>, ol...@sourcery.mxm.sub.org (Olaf Barthel) writes:
- hit the AGA hardware directly? Obviously, AGA games have no commercial
- value, as only a handful of people will be able to use them. As a
- matter of fact, the set of people with AGA is *smaller* than the set
- with OS 3.0, so there games that use 3.0 should be *more* valuable
- than games that hit the AGA hardware directly.
- > If it's written to "bang" the hardware and written
- > following C= guidelines then there is no reason why it shouldn't work on ANY
- > amiga, unless of course C= change the hardware completely.. But then it wouldn't
- > be an amiga would it?
- Only if your definition of Amiga is "the hottest cheap graphics
- harware available in 1985". Some of us would like graphis hardware
- that was spec'ed and designed in the 90s. Getting that requires being
- able to change the hardware completely, which means they have to get
- people to quit bashing the hardware.
-- <mike
I understand this need... And don't get me wrong I love the Amiga O/S, but we
MUST not fall into the PC trap... You know the one.... People begin to write
software that becomes lazier and lazier because the machine's run faster and
faster, and before you know it your programs are 40 Megs a piece and won't run
fast enough on a 386. How many of you have seen Ultima 7 on the PC. It runs at an
ample speed on a 33Mhz 486, but if too slow on a 386. It's a crime! By writing it
efficiently it should be fine on a 286! When I buy my Amiga 4000 I want the speed
and power that makes the Amiga such a good machine. I don't want it to become too
slow because people cant write fast, efficient programs. I dont want to have to
say "it will be okay when I get my 68060!)
Long live the amiga
Joe
-~=Zaphod=~-
> Which games have you written? I did FS2, Jet, 3d code for Ultima UnderWorld,
>3d & 2d code for Car & Driver, and graphics code for a sega game. If I was
>redoing (FS2, Jet) or porting (the others) to an AGA machine right now, they'd
Well, who can argue with a list of games like that!
Practically house hold names...
>it could have been 3% faster if I'd taken over the machine.
You've been doing something *seriously* wrong if you can only get a 3% increase.
You should be able to get at least 50%.
>I'd make more money than you, because the time spent on gameplay would pay off in
>continued sales, which would be possible because the game would work on both
>current and future machines. Also, the game would get better with future
>machines, because it would get faster and faster with better amigas. And,
>my publisher would be happier about doing more amiga games, because there
>wouldn't be millions of support calls about the game not working on
>certain configurations. And if some new whiz bang chipset came out
>with higher resolutions that were worth supporting, I'd have it working in
>those quickly modes by changing constants and redoing art, while you'd have
>to re-write your graphics routines (again).
FACT:You run an OS friendly version of a game next to a hardware hacked game,
the hardware hacked game would look slicker and play better.
Check out the scroll on SIM CITY for a good laugh at OS based games.
--
*******************************************************************************
* /// A member of S.H.I.T. (Super High Intelligence Team) /// *
* /// Fight, defeat and kill organized laming. /// *
****\\\X//********Steven Haggerty**csg...@uk.ac.cov.cck ok?*****\\\X//*********
Dr Peter Kittel Germany
>3d & 2d code for Car & Driver, and graphics code for a sega game. If I was
>redoing (FS2, Jet) or porting (the others) to an AGA machine right now, they'd
>be running in intuition screens with full multi-tasking.
Well put. Now I have an idea: Just to blame the ignorants, how about
porting just one of these programs (or some relevant, teaching part of one)
in fact and show to everyone what is the way to go? And teach them the
methods they obviously simply don't know?
--
Best regards, Dr. Peter Kittel // E-Mail to \\ Only my personal opinions...
Commodore Frankfurt, Germany \X/ {uunet|pyramid|rutgers}!cbmvax!cbmger!peterk
Wer's nicht kann, soll's bleiben klopfen oder Steine lassen!
Eivind Eklund
>If BMF_STANDARD is set, go right to the screen. If not, allocate a temporary
>bitmap of the proper size, do your 5000 lines to it, and after all rendering
>is complete, call BltBitMap to copy the temp to the real bitmap.
How am I, as of today, going to be able to do that? This function is NOT
documented as of Kickstart 2.04, which is the latest public one. To get
information on this, I would have to register as a developer, of a version
that would cost me about $400,- a year... I think CBM should make
information as of how to write WORKING CODE FOR THIS AND HIGHER VERSIONS
OF THE OS as public domain TODAY. No waiting and BUYING documentation on
this topic; documentation enough to make it possible for me to write 100%
working code should be made available NOW; most people banging ECS are NOT
developers (registered), and need documentation made available if you want
their code to keep working. I have no problem with following guidelines
GIVEN THAT I AM TOLD THE GUIDELINES. All my programs work on all machines
(A4000 excepted, bcoz it was NOT documented earlier about LoadView(0);
WaitTOF(); WaitTOF(); )
My $.02
Eivind Eklund AKA Vishnu CRB
Eivind Eklund
>In article <1992Dec2.1...@ugle.unit.no>, am...@Lise.Unit.NO (Marius Mortensen) writes:
>|>
>|> Yes, it would! This is the exact point where out views differ - It's the OS that
>|> *makes* the Amiga such a nice machine, not the HW. And RTG will most certainly
>|> teach us this.
>
>
>Are you saying that without out the Hardware that the Amiga has it would
>still be just as good.. as long as it had Amiga Dos?
What I think he is trying to say (and which should be pretty clear) is
that with the same or BETTER HW, and running AmigaOS (NOT DOS, This is NOT
a crappy MessyDos machine) it would still be an Amiga.
>Is an ST with Amiga Dos an Amiga?
We're talking about Amiga *OS*, not Amiga *DOS*.
But after mine, and I think, from knowing him, Marius' opinion, it would
NOT be an Amiga due to worse HW. But if you equipped it with the OS and HW
cpable of running the same SW: Yes, for my money it would be an Amiga.
Philippe Berard
>
> It's true that many people don't but their Amigas for "serious" purposes
> (whatever they are), and even if they do, don't you think that people
> should be given SOME credit for self-discipline? Failing that, what's so
> darn catastrophic about playing around at the hardware level sometimes?
>
> Sometimes I think that the attitudes surrounding the maintainance of Amiga
> standards border on the paranoid. YES, let's have standards and stick to
> them, but for cryin' out loud, don't prevent the distribution of knowledge
> in order to achieve this end.
This is what wise men say ;-)
Now, I would like to say that I'm a system programmer (C & Asm) but I
really enjoy my Amiga when playing fast games like Interceptor,
Indianapolis or Gravity Force. If anyone is able to make a game like
Gravity Force with the OS routines, I will agree that no HRM is
needed to make something fast, but it seems impossible... ;-)
-- Philippe
.----------------------------------------------------------------------------.
| Philippe Berard (French Amiga User) | |
| "They hold a cup of wisdom, | UseNet : am...@amipb.gna.org |
| But there is nothing within" (Kate Bush). | |
`----------------------------------------------------------------------------'
Gregory M. Stelmack
>Demo-kids don't care! They want to do it their way. If C= doesn't
>release a new hardware guide, the challenge and the glory becomes
>much bigger for them. Creating a demo without an docs will be
>considered ultra-cool. All of you don't seem to understand that
>simply denying to release any docs doesn't stop them at all. I am
>afraid of the damage they will cause if they don't have any (or
>only unofficial) docs.
Except that they're not doing it right now even with the docs! The way I
see it, they are trying to discourage using the hardware at all by not
supporting. Some will continue to bang the hardware for the challenge, but
having the docs hasn't helped them get it right up until now, so what will
change if an AGA doc is published? Without it, maybe a few less hardware
hacked programs to clutter up the shelves, with it, the same situation we
have now on upgrades.
As an example, take a look at the recently released Road Rash. Why do I
still have to hit a key when a disk is inserted on a game released in
1992? Why won't it install on the harddrive (the main program has everything
hardcoded as df0:, and the loader is apparently powerpacked...)? This
situation is getting ridiculous, so let's start teaching them to do it
RIGHT!
Michael van Elst
>You've been doing something *seriously* wrong if you can only get a 3% increase.
>You should be able to get at least 50%.
Well, if he runs OS conform already at 97% the speed of the hardware he can't
get much improvement. BTW, there's a difference between OS conform programming
and using AreaFill for 5x5 pixel large polygons.
>Check out the scroll on SIM CITY for a good laugh at OS based games.
Or to look at not-so-cute programming techniques :)
Regards,
--
Michael van Elst
UUCP: universe!local-cluster!milky-way!sol!earth!uunet!unido!mpirbn!p554mve
Internet: p55...@mpirbn.mpifr-bonn.mpg.de
"A potential Snark may lurk in every tree."
Jan-Olof Hendig
>In article <1992Nov24....@imada.ou.dk> bre...@monet.imada.ou.dk (Bjoern Reese) writes:
>>Demo-kids don't care! They want to do it their way. If C= doesn't
>>release a new hardware guide, the challenge and the glory becomes
>>much bigger for them. Creating a demo without an docs will be
>>considered ultra-cool. All of you don't seem to understand that
>>simply denying to release any docs doesn't stop them at all. I am
>>afraid of the damage they will cause if they don't have any (or
>>only unofficial) docs.
>
>Except that they're not doing it right now even with the docs! The way I
>see it, they are trying to discourage using the hardware at all by not
>supporting. Some will continue to bang the hardware for the challenge, but
>having the docs hasn't helped them get it right up until now, so what will
>change if an AGA doc is published? Without it, maybe a few less hardware
>hacked programs to clutter up the shelves, with it, the same situation we
>have now on upgrades.
>
>As an example, take a look at the recently released Road Rash. Why do I
>still have to hit a key when a disk is inserted on a game released in
>1992? Why won't it install on the harddrive (the main program has everything
>hardcoded as df0:, and the loader is apparently powerpacked...)? This
>situation is getting ridiculous, so let's start teaching them to do it
>RIGHT!
>
Your gripes against Road Rash are justified, but it has nothing at all to
do with banging the hardware.
>
>--
>-- Greg Stelmack (INTERNET: stel...@eggo.csee.usf.edu BIX: gstelmack)
>-- FullTime Grad Student, PartTime Amiga Salesman, PartTime Amiga Developer
>-- Author of: Spades, Pro Port Analyzer Plus, more to come...
>-- DISCLAIMER: The opinions reflected here are mine and mine alone.
Jan-Olof
Chris Green
>FACT:You run an OS friendly version of a game next to a hardware hacked game,
>the hardware hacked game would look slicker and play better.
>Check out the scroll on SIM CITY for a good laugh at OS based games.
SIM CITY is a golden example. It does operate far more slowly than
most games, and could have been much faster without blowing away the
OS. HOWEVER, SimCity is one of the best selling games of the past 10
years, and continues to sell very well. Why? because all of the effort
went in to making it a playable fun game. Another example of this is
Tetris. Or Lemmings (which could have run in an intuition screen under
3.0 or a view in 2.0).
Hans-Joachim Widmaier
> In article <1992Nov27.1...@qut.edu.au> pod...@qut.edu.au (Department of Computing Science, QUT) writes:
> >You can't write good action games without banging the hardware. So any
> >arguing about demos implicitly includes arguing about action games.
> >Don't try and tell me you can do fast action games through the OS coz you
> >can't as anybody who has written one will tell you.
>
> 3d & 2d code for Car & Driver, and graphics code for a sega game. If I was
> redoing (FS2, Jet) or porting (the others) to an AGA machine right now, they'd
> put in the appropriate (simple) fallbacks do that they would work even
> on non-amiga chips. While developing those games, I'd have enforcer
> in my startup-sequence (as I do now), and MungWall. They would be
> standard loadfiles with the data files in standard dos format (as
> all pc games are), most likely with "look up a word in the manual" protection.
Right said! I'd be glad if the hardware banging and OS avoiding games where
found only with action games. But when I look at Civilization, I get very
angry! What a nice game, what a terrible program! I can't believ they ever tested
it! What could this program be if it was written like an application program!
While US games are somewhat more OS friendly than european, almost all of them
just can't repect the keymap! IMHO, it's easier to get an translated key than
a raw one and look it up yourself.
It's so depressing!
Maybe you should do a good Amiga game to show them that it IS possible.
> |Chris Green - Graphics Software Engineer - chr...@commodore.COM f
jbhr
--
jb...@sofal.adsp.sub.org
Philippe Berard
>
> Hey. Commodore makes the chips, Commodore makes the machines,
> Commodore writes the docs, Commodore pays the engineers and Commodore
> decides what does and does not get released to the public.
But does Commodore buy their machines ?
> We did it your way up til now and it created FAR more problems for
> engineering, users and developers than it was worth. So, this time,
> we're doing it differently. If, in your view, that "SUCKS!" then so
> be it. The fact is that "demo coders", while creating 'neato' things
> to look at, do little or nothing for sales and they create an entire
> generation of Amiga programmers that have failed to learn how to do it
> right. The result is application software (you know, the stuff that
> people pay MONEY for and expect to work on new machines) that is chock
> full of silly, pointless coding errors and, often, _slower_ code than
> if it had been coded properly from the start.
Yes, you're 90% right. Demo coders often do what is forbidden. I get
really tired of this discussion now, but I would like to know how
to program a fast >GAME< (with 3D/copperlists/parallax scrolling, put
your favorite here) with the OS stuff. All the demos (some in the RKMs,
some not) I've seen were really slow, even on a 4000. I hope this will
change, or the Amiga game market will surely be dead in a couple of
months, and it's still what makes the Amiga live (it's easy to tell
when you see CBM results in the US and in Europe). If it doesn't
change, you could add 'Commodore is Hara-kiri' ;-)
> No. A discussion is where people _discuss_ a topic. What we have here
> is people _demanding_ something and then complaining like children
> when they are told "no".
Everyone knows programmers are still children in their head ;-)
We pay the machines, so we have the right to give our opinions. And
please stop talking about demos, and tell us about games. Thanks.
Sincerely,
Peter Cherna
>In article <37...@cbmvax.commodore.com> chr...@cbmvax.commodore.com (Chris Green) writes:
>>
>>Which games have you written? I did FS2, Jet, 3d code for Ultima UnderWorld,
>>3d & 2d code for Car & Driver, and graphics code for a sega game. If I was
>>redoing (FS2, Jet) or porting (the others) to an AGA machine right now, they'd
>>be running in intuition screens with full multi-tasking.
>
>Well put. Now I have an idea: Just to blame the ignorants, how about
>porting just one of these programs (or some relevant, teaching part of one)
>in fact and show to everyone what is the way to go? And teach them the
>methods they obviously simply don't know?
For exactly this reason, Chris will be giving a talk on OS-friendly
games programming at the Orlando DevCon. Expect to see examples of
how to get your display up-and-running under the OS, and how to
do your rendering efficiently, including fast blitting and so on.
A big source of confusion is that "using the OS" really means using
the OS calls to get your display open, and cooperating with the OS
in order to do rendering. A lot of noise has been generated on the
irrelevant subject of the speed of AreaDraw() and bobs/gels, which
although part of the OS, aren't the only legal way to render such things.
>Best regards, Dr. Peter Kittel // E-Mail to \\ Only my personal opinions...
Peter
--
Peter Cherna, User Interface Development Group, Commodore-Amiga, Inc.
{uunet|rutgers}!cbmvax!peter pe...@cbmvax.commodore.com
My opinions do not necessarily represent the opinions of my employer.
"Opinions enlarged to show detail"
Jesse Michael
wrote:
>Why hardware hack? People will realize that if something is
>too slow, they should upgrade the processor.
This is the exact mentality that creates `programs' for clones that require
4 Megs of ram or more, a 80386, and a monster harddrive.
This is also something that the Amiga has successfully avoided.
This is _NOT_ something to want.
>You might say why should the
>processor be upgraded since that's more expensive rather than optimize the
>code? Well if that's the case, the Amiga's as good as dead because it
would
>be stuck in yesteryear with old technology. The lowest Mac has an 020,
the
>majority of PC users have 386's. The majority of Amiga owners still have
>68000's. People complain the Amiga is falling behind. That kind of
hardware
>banging isn't gonna help the Amiga catch up.
The reason the majority of PC users have 386's is because _THEY NEED IT_ to
run most of their software. The situation with Macs is similar.
The best reason to optimize code instead of requiring a faster processor is
that you won't alienate part of your potential market. Some people just
can't afford to upgrade...
I'd _LOVE_ to get a 1200 or a 4000, but I have other more pressing
expenses.
Going directly to the hardware _DOES NOT_ mean that your program will have
incompatibility problems.
-Jesse
-- Via DLG Pro v0.995
--
Haze of Epsilon / Check out Epsilon for the best in NTSC demos! Ok, so if
you compare us to groups like Anarchy and Silents then we're lame, but what
the hell do you expect from us lazy Americans?!? :^)
Jesse Michael
wrote:
>All I can say to this is BULL. I've been with the Amiga since 1985. I
know
>0 (that's zero) people who dished out between $500US and $3000Us to buy an
>Amiga for some demos. I DO know of dozens who have bought Amigas because
>they run software that either makes money for them or can entertain for
more
>than 5 minutes.
I have a friend (yes, he's an American) who bought his Amiga solely so he
could write demos on it. In Europe, you'll find a hell of a lot of people
who did this also.
> And 99.99% of Amiga owners think your lame reasons are wrong.
If only .01% of Amiga owners think that going directly to the hardware is
ok, then *why were there more people than that at The Party last year*?
It was a party only for people in demo groups.
I think a lot of people who wanted to go couldn't. I know I'm one of them.
There are a _LOT_ more demo coders/demo groups than you seem to think.
>Does protracker break all of the OS, kill multitasking and go entirely to
>the hardware? No? Wow! So why are you using it as an illustration? It
>doesn't belong here.
It doesn't kill multitasking, but it _DOES_ go `entirely' to the hardware.
Get a clue.
Stephen Tell
If the new machine had equivilent or better but totaly incompatible hardware,
Amiga Dos X.? (X >= 4?) with RTG, then yes, its an Amiga. Even if the old
hardcoded demos don't work.
>Is an ST with Amiga Dos an Amiga?
Probably not, but depends on the hardware capabilities, not the exact
register bits in the hardware.
>Joe.
We'll really learn when Motorola finally reaches the limit of the 68K
architecture and Commodore has to go RISC. Then the "must be in assembly
for speed folks" all lose a few years rewriting, while the C programs compile
right up. Personally, I'd like to see a DEC Alpha based Amiga - and doesn't
C= already have some sort of relationship with DEC?
(Enough gasoline for the fire for now...)
--
Steve Tell te...@cs.unc.edu H: 919 968 1792 | #5L Estes Park apts
UNC Chapel Hill Computer Science W: 919 962 1845 | Carrboro NC 27510
Engineering is a _lot_ like art: Some circuits are like lyric poems, some
are like army manuals, and some are like The Hitchhiker's Guide to the Galaxy..
Chris Green
> Now, I would like to say that I'm a system programmer (C & Asm) but I
> really enjoy my Amiga when playing fast games like Interceptor,
> Indianapolis or Gravity Force. If anyone is able to make a game like
> Gravity Force with the OS routines, I will agree that no HRM is
> needed to make something fast, but it seems impossible... ;-)
I have never seen Gravity Force. However, both Interceptor and Indianapolis
could be written under the OS for the A1200 and not suffer for it.
--
*-------------------------------------------*---------------------------*
|Chris Green - Graphics Software Engineer - chr...@commodore.COM f
Chris Green
>
>These are hardly the types of games he was talking about. How about a multi-layer
>parallax scrolling shoot-em-up?
>
components of the scroll (usually a small set of scanlines at the bottom
of the screen via blitting, and do the full-screen component with
ScrollVPort). Note that PC games sell 100,000's of copies with no more
hardware assist than a bitmap they can write into, double buffering,
and full screen scroll. The Amiga OS model of the hardware provides far more
than this.
>Yes, YOU would probably do that, because you work for Commodore, and this how they
>want it done. Do they want the kind of games we see for Windows? SLOW, simple, and
>ultimately BORING stuff. I certainly don't.
No, I'd do what would make my task easier and my game better. It would
run as fast as if I took over the system (faster, because I would have less
code to write, and more time to profile & optimize).
>
>* Even if you DO use libraries, you're can never be 100% certain that it will be
> working on future machines.
You can be pretty sure, if you program carefully. And, if I was unlucky
and broke on a future machine, the fix would most likely be only one
thing, and a simpel one at that.
>
>* You'd spend MORE time writing a multitasking game than writing a 'harware' one.
> You'd have to make sure it in all the situations you might end up in using
> Intuition and the system, you'd have to make sure you don't mess up any other
> program that might be running etc etc etc. It's a lot of work!
It's not a lot of work. In fact, running with the system helps
the debugging, because you can use mmu tools and other debug tools to
find bugs that might bite you later, even if you had taken over the system.
For instance, if your program trashes some memory it doesn't own via a wild
pointer, it's just as likely to trash itslef as some other program, but
you might not notice until too late.
>
>* 3%??? This is absolutely CRAP! I'd say 50% if the programmer knows anything about
> coding. Using AreaFill (or something) in a flight sim would slow it down far into
> boredom city.
Who said I'd use AreaFill? I'd go right to the blitter and the display
memory. And, as I've said, this can even work on RTG devices with a simple
check.
>* No you would NOT make more money! Who would buy a slow game that they knew would be
> at least twice as fast if the programmer had accessed the hardware direcely?
ridiculous. I challenge you to show a 50% difference between the
time available to a priority 25 task and a system with the whole OS
shut down. And I'd have more time to optimize my 3d code and create better
objects and art than the person who had to write his own copper and disk
routines.
> None of the A500/A2000 owners anyway, and they dominate the market. Maybe your game
> would run at an acceptable rate on A3000,
Than there's no need for an AGA hardware doc, because anything you use
it for won't work on an A500/A2000 either.
>It't great for utilities, but if the game can return to Workbench, most (all)
> players would be satisfied. Accessing hardware directly and TEMPORARILY shutting down
> the system, you can guarantee a steady frame update (which lots of arcade games
> require) and you can read have you own interrupts that you KNOW will execute every
> fifth of a secound and you'd have a SAFE environment. In the multitasking world, you
> can't be sure of anything. What's good about having fast reflexes if the game can't
> catch up?
>
The 1200 is perfectly capable of delivering interrupts 5 times per second.
Five times per second is, in fact, ridiculously slow. I wouldn't even use an interrupt
for that, I'd use a task. Just for a test, I created a user copper-list
which causes a copper interrupt on a given scanline. In the interrupt routine,
it sets color0 to red. It gets into the interrupt routine at exactly the right time,
and the transition to red is rock-solid. The interrupt signals a high priority task,
which changes the color to green. The length of the red bar is approixmately 30
microseconds long, and varies by about 2.2 microseconds.
Chris Green
>In article <37...@cbmvax.commodore.com> chr...@cbmvax.commodore.com (Chris Green) writes:
>> To make this work under RTG, call GetBitMapAttr(bm,BMA_FLAGS).
> ^^^^^^^^^^^^^
>>If BMF_STANDARD is set, go right to the screen. If not, allocate a temporary
>>bitmap of the proper size, do your 5000 lines to it, and after all rendering
>>is complete, call BltBitMap to copy the temp to the real bitmap.
>
>How am I, as of today, going to be able to do that? This function is NOT
>documented as of Kickstart 2.04, which is the latest public one. To get
>information on this, I would have to register as a developer, of a version
>that would cost me about $400,- a year... I think CBM should make
>information as of how to write WORKING CODE FOR THIS AND HIGHER VERSIONS
>OF THE OS as public domain TODAY. No waiting and BUYING documentation on
>this topic; documentation enough to make it possible for me to write 100%
>working code should be made available NOW; most people banging ECS are NOT
>developers (registered), and need documentation made available if you want
>their code to keep working.
I agree with you, but it's not my decision and would probably give
the lawyers fits.
Here's the doc for GetBitMapAttr:
GetBitMapAttr(bm,attrnum)(a0,d1)
#pragma libcall GBASE gfx_GetBitMapAttr 3c0 1802
#define BMB_STANDARD 3
#define BMF_STANDARD (1l<<BMB_STANDARD)
#define BMB_INTERLEAVED 2
#define BMF_INTERLEAVED (1l<<BMB_INTERLEAVED)
#define BMA_FLAGS 12
Note that an interleaved bitmap will have the INTERLEAVED & STANDARD
flags set. A "Standard" bitmap is defined as a bitmap stored in planar form,
with its PlanePtr's pointing to memory accessible by the blitter.
Fred Mitchell - Product Assurance
>In article <37...@cbmvax.commodore.com> chr...@cbmvax.commodore.com (Chris Green) writes:
>> Which games have you written? I did FS2, Jet, 3d code for Ultima UnderWorld,
>>3d & 2d code for Car & Driver, and graphics code for a sega game. If I was
>>redoing (FS2, Jet) or porting (the others) to an AGA machine right now, they'd
>Well, who can argue with a list of games like that!
>Practically house hold names...
>
>>it could have been 3% faster if I'd taken over the machine.
>
>You've been doing something *seriously* wrong if you can only get a 3% increase.
>You should be able to get at least 50%.
It never ceases to amaze me how some think they know more about the
Amiga than the engineers who designed it and wrote the OS. Chris does
know what he's talking about.
>>I'd make more money than you, because the time spent on gameplay would pay off in
>>continued sales, which would be possible because the game would work on both
>>current and future machines. Also, the game would get better with future
>>machines, because it would get faster and faster with better amigas. And,
>>my publisher would be happier about doing more amiga games, because there
>>wouldn't be millions of support calls about the game not working on
>>certain configurations. And if some new whiz bang chipset came out
>>with higher resolutions that were worth supporting, I'd have it working in
>>those quickly modes by changing constants and redoing art, while you'd have
>>to re-write your graphics routines (again).
>
>FACT:You run an OS friendly version of a game next to a hardware hacked game,
>the hardware hacked game would look slicker and play better.
>Check out the scroll on SIM CITY for a good laugh at OS based games.
"Slickness" is a matter of how much effort you put into it, not of its
OS friendliness. I have seen some really wretched hardware banger
games as well.
I get the distinct feeling that you hardware bangers are afraid of
a minor challenge: Writing an equally slick OS-friendly game, with
all the speed and craftiness of banging on metal. You will have to
do certain things differently, of course, but it can and has been done.
It's not that hard to do, and it will be equally slick. Intuition now
supports double-buffering and dual-playfield modes directly as of
3.0, as does the Graphics subsystem. There are many other enhancements
to the OS to make the life of the programmer eaiser. So why not take
advantage of them and reap all the benefits?
I, myself, like OS-friendly games that I can play during a compile
or a download. I should not have to loose my entire entire machine
just to play a shoot-em-up.
>--
>*******************************************************************************
>* /// A member of S.H.I.T. (Super High Intelligence Team) /// *
>* /// Fight, defeat and kill organized laming. /// *
>****\\\X//********Steven Haggerty**csg...@uk.ac.cov.cck ok?*****\\\X//*********
=== Fred Mitchell (mitc...@cbmvax.UUCP) ===================================
Lorinda, the .. /// Heroic rescues are embarrassing ..
9th wonder of .. \\\ /// if you're not really in trouble. ..
the world! (34) .. \\X// - Cally, Blake's 7 ..
============================================================================
$lI%y he(%ti=l:6f!?HWA(Z?`T*O)oLb4.u3j8h6o`@g)2K vCKTZM8EIg)t:uiVqhr81THL9`M
Alex Topic
tons of information in how to do it properly. So we don't have awful
programs floating around.
It will be depressing if they don't and think some programs will
suffer
if they have to resort to the OS in the area of graphics and sound.
To tell you the truth the only reason that the Amiga attracted me in
the first place was the custom chips, nothing more. Without them alot
of the things on the Amiga would not be possible. I know the OS relies
heavily on them, and helps the OS become efficient in certain areas.
Sooner or later someone will figure them out, and they may not know
everything and in the process lots of head aches will exist for alot of
us. I strongly suggest that an AGA Hardware Reference manual exist.
Anyhow you can see what I'm leaning towards, I love coding in Assembly
and tapping on the hardware. If I used the OS, I would run into limitations
in the future. And I only want to be limited by HARDWARE. Also I just
love the feel of total control of the computer, not partial...
I'm totally against the code you usually see from SEKA or MASTERSEKA
that uses absolute memory, or has alittle for example ORG $10000 at the
beginning of their source. Basically they totally rape the system, and
you have to reset. I believe in friendly takeover, and bring back. This
in effect makes things convient for the user.
One thing I'am worry about is the VBR. Since in some of my code I
write directly to say memory location.. LEVEL3 vector $6c , now if your
CPU has the VBR ability...it could be somewhere else in memory. However
my friend who has an A3000, says the MMU handles this and re-directs it
somehow...hmmm I think he uses ZKICK, or something like that. Not too
sure..
Welp C'ya all later.
-Alex
Stig Arne Olsen
To bring another point into this debate. Someone posted in
?.amiga.games that several companies are now working on AGA
games. I will find it very strange if companies like Psygnosis,
Gremlin Graphics and the likes are going thru the OS. I don`t
think the Beast Series etc. is coded that way.
If they already have games in the works, maybe they`ve already got the
information? This would mean that the only people they don`t want
to hit the hardware is the demo-types/PD programmers etc? If that`s
a fact, Commodore certainly is not doing the right thing.
Another point is that we probably have to wait until the end of
the century (another 7 years?) before the next graphic chip set
comes out (based on experience). Thus the problem this far in
Amiga history have not really been other hardware than the
processor. The problem has been people not using the OS the right
way. If you hit the hardware on your A1000, it probably
works with the 1992 A600 (now that's gonna be a flop, don't you
think?).
If RTG is a goal, can we expect future OS`es to support smooth
scrolling, hardware sprites and individual playfields? Probably not,
and that takes away a lot when it comes to making games/demos.
A question here to C=: When are the ordinary (non HW) reference
manuals for OS3.0 going to be released? As soon as possible? If
you want us to do OS coding, it's nice to have the docs.
Now Commdore: Release the hardware ref, make a new sound chip and
we`re back in action.
Quoting Dijkstra doesn't impress me (hw ref. cons. harmful),
--
| Stig A. Olsen | st...@ifi.uio.no | "Ja til EF" |
| Never hit a man with glasses. Hit him with a baseball bat. |
Osma Ahvenlampi
>most games, and could have been much faster without blowing away the
>OS. HOWEVER, SimCity is one of the best selling games of the past 10
>years, and continues to sell very well. Why? because all of the effort
>went in to making it a playable fun game. Another example of this is
>Tetris. Or Lemmings (which could have run in an intuition screen under
> 3.0 or a view in 2.0).
Yeah, right, Lemmings... OK, it's a great game, I love it, but back when I
had A500 w/ 2 external drives (well, I kinda managed to blow it, bad, if
you're wondering), I noticed that it required that I turn of the DF2:..
Well, It could of course have benn the fact that I had only 1MB RAM, so that
the DF2: took what was needed by the game, but anyway, not very system
friendly I'd say.. Or did you mean that it's simple enough gfx to use OS
and not be slowed down? In that I'd agree...
--
Osma Ahvenlampi - oahv...@snakemail.hut.fi * Workstation power for micro-
All my opinions are not necessarily really mine * computer price: Amiga := FUN
Tomas Arvidsson
>In a message dated Sun 29 Nov 92 17:20, Joe...@cup.portal.com (joel Edward
>wrote:
>I have a friend (yes, he's an American) who bought his Amiga solely so he
>could write demos on it. In Europe, you'll find a hell of a lot of people
>who did this also.
>> And 99.99% of Amiga owners think your lame reasons are wrong.
>If only .01% of Amiga owners think that going directly to the hardware is
>ok, then *why were there more people than that at The Party last year*?
>It was a party only for people in demo groups.
And how many of the people in those groups are programmers and not traders,
musicians, graphic artists, personal friends of some member in the group, etc?
>I think a lot of people who wanted to go couldn't. I know I'm one of them.
>There are a _LOT_ more demo coders/demo groups than you seem to think.
As Erik Lundevall pointed out a couple of days ago; How many are they? 5000?
10000? 0.5% or even 1% of the total Amiga community? In any case, they are
a (small) minority of the Amiga community.
>-Jesse
/Tomas
--
Tomas Arvidson ///
d91...@csd.uu.se, ///
tom...@nada.kth.se, tom...@sprawl.adsp.sub.org \\\///
Expressed opinions are mine unless stated otherwise \XX/
--
Tomas Arvidson ///
d91...@csd.uu.se, ///
tom...@nada.kth.se, tom...@sprawl.adsp.sub.org \\\///
Expressed opinions are mine unless stated otherwise \XX/