Fantlist 9th Apr to 19th Apr 97

Date: Wed, 9 Apr 1997 20:41:22 +0100
From: Rob Probin <rob@zedworld.demon.co.uk>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re(5): New way to program?

>On Tue, 8 Apr 1997, Rob Probin wrote:
>
>> I understand Allan's comments - but I also think that it's a real bitch
>> having to rewrite major chunks of operating system code.
>
>I don't know where rewriting chunks of the OS came from, but this wouldn't
>have to be done anyway; at least no more rewriting that has to be done to
>port code between platforms today....

What I meant was '...rewrite major chunks of operating system calling
code'. I tend to try to program in real (i.e. not small) programs with the
OS calls obstracted as much as possible. Ok, this is application specific
(since editors, for instance are a LOT closer to the OS, than say,
compilers). But, as you say, code for multile OS's is done a lot already,
and if I understand you, you are saying 'make your port code standard
across different projects', i.e. modularise the system specific code
(either with macros, or with code) and then reuse in future projects.

> I honestly think you guys are making this more difficult than it really
>is, >or has to be....What I'm talking about is; at its worst, the same
>thing that's
>done now involving porting; and at its best, the best thing to come along
>since
>sliced bread.

:-) We just like to think we are really clever to be able to solve such
problems. :-)

To be honest, the best system 'independance code' is probably developed by
a group. How about starting an interface right here?

Date: Wed, 9 Apr 1997 18:02:20 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re:(6) New way to program?

On Wed, 9 Apr 1997, Rob Probin wrote:

> if I understand you, you are saying 'make your port code standard
> across different projects', i.e. modularise the system specific code
> (either with macros, or with code) and then reuse in future projects.

Originally I never thought about making the independence code standard,
but that is a really good idea.

> To be honest, the best system 'independance code' is probably developed by
> a group. How about starting an interface right here?

I'd like to start work on something like that. The hardest part will be
in identifying which parts should be separated from the main code: do you
make windows code separate, math functions, what about stuff like
WaitNextEvent()... One of the solutions that I came up with, or at least
a start of a solution, is to heavily modularize your programs. From my
point of view this would take a little rethinking about how to
actually write a program. Even though programs may be split between
windows code, and menus, dialogs, video, stuff like that, you would have
to go even further--object orientated program seems to be the way to go.

You guys might think I'm going out on a limb, but... I've been thinking
about writing programs so that they are modeled after the human body. The
human body, or any advanced organism, is rather complex and yet
everything works--you don't see any humans malfunctioning do you. Like
organisms, programs are going to evolve over time to become more complex,
like working on different OSs, and yet still function. I haven't really
sat down to actually think about all the different aspects that would go
into comprising a program, but I'm pretty sure that looking at the
different aspects of the human body and modeling programs after it would
be a good idea.

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Sat, 12 Apr 1997 18:07:38 -0400
From: Stuart Ball <lightsoft@compuserve.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: re(7) New way to program

Hi chaps,
I've just caught up with what happened.
I got dragged away for a few days (looking for German in Peterborough - weird)
but Rob filled me in via phone (thanks Rob!).

Tim>In the real world standards are more or less a bad thing, imagine how
> drab looking a school is that has uniforms instead of letting the kids
> wear what they want. But in the technology world standards are a good
> thing. Back when railroads were first being introduced each company
> decided on their own railway implementation: spacing of the track, number
> of rivets, etc...

That's a good counter. It does not explain though why we have at least two (if
not three) OS's competing for market share. My argument goes that it is
because _nobody_ wants just one choice. You're argument can be given credence
thus: Most everybody has the technology to be able to talk to a TCP/IP network
because it is a sensible standard that get's the job done. In the area of
GUI's and user friendliness however, judging what is good and bad (and
therefore should become a standard?) is subjective and thus unquantifiable via
any rational, logical or measurable means.

And then there are alternatives to standard railway tracks; monorail, linear
induction etc any of which may prove "better" in the future. Rail is a mature
industry. Computing is not. Indeed, some would say Computing has just got it's
head out; the shoulders are a little problematical.

Allan>Although it would be a neat thing if the same code worked on PC's as
>well as Macs, I must say it's a prospect I don't look forward to at all.
>Part of owning a Mac (or any other computer other than a PC) is usually
>about being different.

Here here!

-----------------

If I may be permitted to summarise (and comment on) what I think was the
outcome re. abstracting away from OSes:

1. You can abstract via compile time switches, viz
if MacOS
include this
if windowz
include this
etc.
This everybody does to some extent, as was pointed out we do for various
versions of Fantasm, Eddie etc. No run time speed penalty. The whole reason we
have C preprocessors is to handle this type of compile time reasoning.
However, you do have to write at least some of the program at least twice (and
thats just to support two platforms). re: Rob's "but I also think that it's a
real bitch having to rewrite major chunks of operating system code" 9/4/97.

2. OS interface boxes that are rewritten for a specific target. Your program
talks to the box, the box talks to the OS and back again. A lot of programs
use this method. Slight run time speed penalty, more so if more translation is
performed. Thus a program written on a Mac for both windows and Mac would run
slightly faster on Mac than Windows as most likely more translation would be
required for Windows . My prefered option.

3. Write in a "cross platform" language - i.e. p-coded Pascal, Java, BASIC
etc.
Some run time speed penalty. Language is never flexible enough. Host access
limited. End results can appear amatuerish. However, processors are becoming
more powerful, but without knowledge of the target platform, you can't take
advantage of any "special" attributes of the target. We all know that a Win
prog ported to Mac that looks like a win prog does badly and vice versa? With
these languages as soon as you peek or poke outside (if allowed) the run time,
you loose portability (invalid and obvious agreed).

I can come to no clear conclusion - it's an individual decision as to the
approach to take.

Even C is touted as being a cross platform beast, when in reality it is not.
OK, it is IF you stick to standard C libraries, I/O and maths and you know the
difference between little and big endian and are willing to design in a way
that handles it. If you draw windows, access the screen, play a sound, check
the mouse, optimise around the FPU, or a million and one other necessary
things, then it isn't portable.

The only _truly_ portable program AFAIK is a flow chart scribbled on a piece
of note paper.
Run time speed is less that satisfactory however. (This of course assumes the
person running said program understands the language of the flow chart - oh
dear).

------------------

Tim>>you don't see any humans malfunctioning do you.

Kev> naaa. that's to easy to make a crack at.

ROTFL. Killer.

Stu.

From: Stuart Ball <lightsoft@compuserve.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Lightsoft WWW pages 2nd server now running.

We are pleased to announce that our second WWW server is up and running -
Lightsoft(2).

The address of the mirror site is:
http://www.telemacus.it/lightsoft/

Easy access can be gained from the main page of the www.tau.it/lightsoft
pages.

Hopefully this will provide speedier access during peak times.

Many thanks to Mario Garofano, Chairman & CEO of Telemacus Internet
Services Srl.

We have rechecked all links on the pages on both sites during this phase.
If you do find a bad link on either site, please let us know.
Thank you.

Lightsoft 13 Apr. 97.

Date: Sun, 13 Apr 1997 22:16:56 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: Lightsoft WWW pages 2nd server now running.

On Sun, 13 Apr 1997, Stuart Ball wrote:

> Many thanks to Mario Garofano, Chairman & CEO of Telemacus Internet
> Services Srl.

Hey, this same guy wanted to give me space on his server but he never
wrote back to finalize everything. Hmm, well at least I know he was
serious about it and not joking.

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Mon, 14 Apr 1997 12:33:34 -0800
From: Kevin Avila <kevin@cache-computing.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Cursor Focusing

For almost month now I have been using Tim's cool mousigator program
(http://www.winthrop.edu/~humphret/mousigator.html) The yesterday I came
across the same thing called Sloop, only there charging for it a
http://www.quadratic.com/Sloop/. The Sloop's description says that Cursor
Focusing (switching windows without click the button) has been avalivble on
other OS's for years, so I fired up my old X-Windows server and sure enough
there is that option built-in. So congrats go out to Tim for beating
everyone else to the MacOS, I thought you were just lazy. :)
----------------------------------------
Kevin Avila http://www.cache-computing.com
kevin@cache-computing.com

Date: Mon, 14 Apr 1997 18:22:19 -0400
From: Stuart Ball <lightsoft@compuserve.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Cursor Focusing

Kevin wrote:
>For almostmonth now I have been using Tim's cool mousigator program
>(http://www.winthrop.edu/~humphret/mousigator.html) The yesterday I came
>across the same thing called Sloop, only there charging for it a
>http://www.quadratic.com/Sloop/. The Sloop's description says that Cursor
>Focusing (switching windows without click the button) has been avalivble on
>other OS's for years, so I fired up my old X-Windows server and sure enough
>there is that option built-in. So congrats go out to Tim for beating
>everyone else to the MacOS, I thought you were just lazy. :)

I've used Tim's a fair amount. It's cool. Why don't you undercut them Tim?

When I read Quadratic's PR I must admit to thinking "Funny, I've been using
something similar for about a month."

I noticed that Quadratics doesn't allow typing in BG windows, and I think
Tim is working on that?

Stu.

Date: Mon, 14 Apr 1997 23:56:16 -0500
From: support@quadratic.com
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: Cursor Focusing

Hi folks,
This is Graham (the author of Sloop -- yes, I'm on this list). I had no
idea about Mousigator but I'll definitely check it out now.
Since releasing Sloop I've also found out that there was a MacHack '96
entry called "Macintosh Feable Window Manager" that does the typing in BG
windows. (A pretty neat hack that just patches FrontWindow). The URL is
something like <ftp://ftp.machack.com/Hacks96/MFWM.sit.bin>. So if Tim's
working on BG window typing he might want to check that out.
I know my press release was a bit overblown! (Quadratic Software is just
me -- one person).
I'm not a big company (have been barely living on my shareware income so
far) so I hope this "undercutting" thing isn't advocated under the idea
that I'm some big corporation! (or Bill Gates wannabe, etc.)
(Remember me Stu? I reported the A2 register conflict with Kaleidoscope
that resulted in Power Fantasm 4.1.8.)

Anyway, I just wanted you all to know I'm not *that* evil of a guy!

Regards,
Graham

>Kevin wrote:
>>For almostmonth now I have been using Tim's cool mousigator program
>>(http://www.winthrop.edu/~humphret/mousigator.html) The yesterday I came
>>across the same thing called Sloop, only there charging for it a
>>http://www.quadratic.com/Sloop/. The Sloop's description says that Cursor
>>Focusing (switching windows without click the button) has been avalivble
>on
>>other OS's for years, so I fired up my old X-Windows server and sure
>enough
>>there is that option built-in. So congrats go out to Tim for beating
>>everyone else to the MacOS, I thought you were just lazy. :)
>
>I've used Tim's a fair amount. It's cool. Why don't you undercut them Tim?
>
>When I read Quadratic's PR I must admit to thinking "Funny, I've been using
>something similar for about a month."
>
>I noticed that Quadratics doesn't allow typing in BG windows, and I think
>Tim is working on that?
>
>Stu.
>

Date: Tue, 15 Apr 1997 00:19:19 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: Cursor Focusing

On Mon, 14 Apr 1997, Kevin Avila wrote:

> For almostmonth now I have been using Tim's cool mousigator program
> (http://www.winthrop.edu/~humphret/mousigator.html)

Hey, thanks for the plug!

> the same thing called Sloop, only there charging for it a
> http://www.quadratic.com/Sloop/. The Sloop's description says that Cursor
> Focusing (switching windows without click the button) has been avalivble on
> other OS's for years, so I fired up my old X-Windows server and sure enough
> there is that option built-in.

I tried out Sloop too, maybe because I'm biased toward my own program but
I think mine's a little better in that you can see what window is going
to be activated; it also works somewhat with floating windows, in the
future it will work totally with them.

> So congrats go out to Tim for beating
> everyone else to the MacOS, I thought you were just lazy. :)
Actually, I was just lazy and tired of clicking around:) Seriously, it is
on UNIX, and when I couldn't find the same thing for the Mac, I simply
made it. Of course, I'm going to go further than just porting over an
idea <how boring>, I'm going to add a few little things to it here and there
to make it a bit better and spice it up--although I won't speculate since
I don't want to get people's hopes up and then inadvertantly drop them.

Stu said:
> I've used Tim's a fair amount. It's cool. Why don't you undercut them Tim?

I definately won't go any higher than what they're charging, and yet
Mousigator is going to have more stuff; even still I think $20 is a tad much
for just window focusing.

> When I read Quadratic's PR I must admit to thinking "Funny, I've been using
> something similar for about a month."

Actually, to be fair, in the ReadMe it said that they had a public beta
out in February, or some time around there, so they did have a little
head start on me. But it is true that you've been using it for a month
now, despite their claim:)

> I noticed that Quadratics doesn't allow typing in BG windows, and I think
> Tim is working on that?

Yeah, I'm working on it, or rather I have it planned; I'm still working
on Budding Animator too. As I said above I don't want to speculate too
much--but I guess a relatively small listserv like this is a good place
to speculate, how about programming abstraction:)

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Tue, 15 Apr 1997 00:57:29 -0400
From: Stuart Ball <lightsoft@compuserve.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: Cursor focusing...

Oh, THAT Graham :-) Bugger me.
----
So I'm sitting here burning my nipples with a Marlboro.
Ouch! Ouch!
------------------------------
Hiya Graham,
Fancy seeing you here - small world!
No offence taken I hope. Blame it all on Kev - works for me <BG>.

Still, typing without bringing a wind into the FG a la X (fvwm et al) would be
nice, and I'm sure somebody somewhere must be doing it?

> I'm not a big company (have been barely living on my shareware income so
>far)
One does not "live" off shareware income, one starves and hopes and codes :-)

Stu.

Date: Tue, 15 Apr 1997 01:41:03 -0500
From: support@quadratic.com
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: Cursor focusing...

Hi Stu,
No offense taken at all :) -- I checked out Tim's pages after my last
message and I realized what was going on. That would certainly suck to be
working on a project and then have someone else release something that does
basically the same thing.
So, Tim, I don't know what to say -- it certainly wasn't intentional! If
you ever want tips on problems I ran into while writing Sloop or anything
else like that, let me know.
A couple days ago I heard a rumor that cursor-focusing is already a
preference in alpha versions of Tempo (MacOS 8). So maybe we'll *both* be
had by Apple in a few months!
Anyway, I've got to go to sleep now. (It's almost 2 am here).

Bye,
Graham

p.s. I haven't yet compiled and installed MFWM (it's a public source code
distribution in <ftp://ftp.machack.com/Hacks96/>) but it seems to have the
rudimentary basics of typing into a window without switching to it.

Date: Mon, 14 Apr 1997 22:37:00 -0700
From: Kevin Avila <kevin@cache-computing.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: Cursor focusing...

>No offence taken I hope. Blame it all on Kev - works for me <BG>.

ACTION: Kev posts a "Kick Me!" sign on himself. :-)

----------------------------------------
Kevin Avila http://www.cache-computing.com
kevin@cache-computing.com

Date: Tue, 15 Apr 1997 04:48:37 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re:(5) Cursor focusing...

On Tue, 15 Apr 1997 support@quadratic.com wrote:
> So, Tim, I don't know what to say -- it certainly wasn't intentional! If
> you ever want tips on problems I ran into while writing Sloop or anything
> else like that, let me know.

If you were to hear what I'm going to say to you without vocal inflections
then you might think that it would be a challenge; it's not, just take
it for what it is...
If I were you, Graham, I would continue working on Sloop, that is if you
already planned to. Mousigator isn't just going to focus on
cursor-focusing. It's going to be a navigational tool, hence the name
"Mousigator"--cursor-focusing is just the first part of it. Even if
our programs do the same thing, they may offer different options and do
the same thing in different ways; in short, don't kill Sloop or
otherwise give up just because I'll be doing the same thing.

> A couple days ago I heard a rumor that cursor-focusing is already a
> preference in alpha versions of Tempo (MacOS 8). So maybe we'll *both* be
> had by Apple in a few months!

This is probably true; but can I share a little something with you. I
still program in 68K assembly even though PPC is the future, so why do I
do it? Two reasons, 1) Apple has made it feasible to continue to program
in the old ways by providing an 68K emulator on PowerMacs, so my programs
work on all Macs even if they don't have the speed of a PPC, and 2) I
still see people using old Macs, old as in SEs. Sometimes they still have
them because they can't afford to upgrade--I didn't get my Mac until 7
years after my Apple II! Imagine if they still made software for it, I
wouldn't have *needed* to upgrade. For the people that are content to use
their computers to do word processing or whatever, there really isn't a
reason to upgrade--how fast a processor do you need to blink the cursor.
Even though they're still using old computers they still want the latest
software. Wouldn't it be wonderful if a new program came out that worked
with their old computer, and at the same time worked on modern computers...

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Tue, 15 Apr 1997 05:48:01 -0400
From: Stuart Ball <lightsoft@compuserve.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re(6):Cursor focusing

Tim wrote:
<snip>
>This is probably true; but can I share a little something with you. I
>still program in 68K assembly even though PPC is the future
<snip>

Hi Tim,
You may find this interesting.
I've just completed some tests on a "not a toy" C compiler outputting
native code - not our's I hasten to add. Maximum optimisation for speed
naturally.

I wrote a largish component for Anvil, both in C and 68k assembler (Build
V3 as it happens).
The 68k version runs two to three times the speed of the native compiled
version and the 68k version is running under emulation on this machine.

Why?
The compiler was outputting so much rubbish as to really make things
sluggish, specially on OS calls and accessing structure members.

Thus I'd say you aren't loosing a lot by writing in 68k - indeed you may be
gaining quite a bit, spec when you consider that most of the emulators on
Macs these days are dynamic recompilation emu's, so after the first run,
you're effectively native anyway.

IMO Apple's Emu is the smartest move I've seen any company ever make.

Stu.

Date: Tue, 15 Apr 97 09:51:55 -0700
From: James Hague <jhague@dadgum.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re:(5) Cursor focusing...

>This is probably true; but can I share a little something with you. I
>still program in 68K assembly even though PPC is the future, so why do I
>do it?

I've been surprised that more programs haven't been written in 68K asm.
PPC asm is fun an all that, but (1) Extra bits of speed are worth more on
slower machines, and (2) Compilers seem to do a much crappier job of
spitting out 68K code than PPC compilers. (That's not to say PPC
compilers are great, mind you.)

Have any Mac games been written entirely in 68K asm? The only one I can
think of is "Scruffy II."

>I still see people using old Macs, old as in SEs.

That's why it's very strange that Apple isn't supporting NuBus machines
for Rhapsody. I mean, come on, who's gonna chuck the machine they bought
in 1995? The typical Apple fan will still be using it in 2005.

James

Date: Tue, 15 Apr 1997 14:59:15 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re:(7) Cursor focusing

On Tue, 15 Apr 1997, Stuart Ball wrote:

> I wrote a largish component for Anvil, both in C and 68k assembler (Build
> V3 as it happens).
> The 68k version runs two to three times the speed of the native compiled
> version and the 68k version is running under emulation on this machine.
>
> Why?
> The compiler was outputting so much rubbish as to really make things
> sluggish, specially on OS calls and accessing structure members.

I did a little test on the computers here at school. I made two programs
that did exactly the same thing, print out a prompt and read in a
number. The assembly version was about 48K, while the C++ compiled
version was 358K! Granted, these are UNIX systems where the C libraries
have to *very* general, but it serves to illustrate why I love assembly:
in assembly you just have the code that you want, not the code that you
will need just-in-case. It takes longer to write programs in assembly:
but programs are made only once and used many times, so I'm willing to
put up with spending a week writing a function in assembly that could be
done in a day in C++; and I'm sure the end users won't mind either. I
honestly think programming has left its roots, and is getting more
commercial, "Lets just output code really quick, who cares about the
quality." That's the kind of attitude I perceive nowadays about
programming, and I don't think that's right.

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Tue, 15 Apr 1997 12:55:06 -0700
From: Kevin Avila <kevin@cache-computing.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re:(7) Cursor focusing

>I did a little test on the computers here at school. I made two programs
>that did exactly the same thing, print out a prompt and read in a
>number. The assembly version was about 48K, while the C++ compiled
>version was 358K! Granted, these are UNIX systems where the C libraries
>have to *very* general, but it serves to illustrate why I love assembly:

hehe. You must be using like gcc 1.0. I just did the same thing on my UNIX
box with gcc 2.7.4 and the assembly was 234 bytes and the C version was 7K.
Still a huge differance. Or perhaps yours was mutch larger then mine
becaused I used the XCOFF birnary format, and by defaulty on UNIX the
format is ELF. For a comparason of compile size read what I wrote at
http://www.cache-computing.com/Ontario.

----------------------------------------
Kevin Avila http://www.cache-computing.com
kevin@cache-computing.com

Date: Tue, 15 Apr 1997 22:07:02 +0200
From: Allan Lund Jensen <maccoder@post6.tele.dk>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: (7) Cursor focusing

Figaro Tea wrote:
> I did a little test on the computers here at school. I made two programs
> that did exactly the same thing, print out a prompt and read in a
> number. The assembly version was about 48K, while the C++ compiled
> version was 358K! Granted, these are UNIX systems where the C libraries
> have to *very* general, but it serves to illustrate why I love assembly:
> in assembly you just have the code that you want, not the code that you
> will need just-in-case. It takes longer to write programs in assembly:
> but programs are made only once and used many times, so I'm willing to
> put up with spending a week writing a function in assembly that could be
> done in a day in C++; and I'm sure the end users won't mind either. I
> honestly think programming has left its roots, and is getting more
> commercial, "Lets just output code really quick, who cares about the
> quality." That's the kind of attitude I perceive nowadays about
> programming, and I don't think that's right.
>

I always feel like a fool when I try to justify using assembler to my
friends :(.
Basically assembler is getting a bad reputation by the press. I remember
reading in "The Mac" an article about programming. Assembler was
dismissed as "something with a lot of numbers". Assembler is made to
look like it doesn't make any sense. IMHO, assembler is the language
which makes most sense. Not that it's easy or anything, but compared to
C++, it makes a whole lot more sense (I'm still struggling with figuring
out just how classes work in C++ ;) ).
I can certainly see the advantage in using C++ or other languages (you
can calculate cos and sinus-tables with them ;), but I think that most
people are too quick to dismiss assembler.
I some ways Mac-programmers are among the worst when it comes to this. I
have yet to find some information on the net re. assembler for Mac(other
than Lightsoft and Motorola, that is). PC-users still see some use in
assembler, for texturemapping for example, and writes endless pages on
'how to optimize the innerloop'.

I recently downloaded the source for Steve Israelsons (I think that's
the right name) 3d-engine. You might remember him as the guy behind
Pfhorte, the Marathon-editor. To my surprise there didn't seem to be a
line of assembler in it. Although I have no idea how good code
Codewarrior produces these days, it would surprise me if he couldn't
make his engine a bit faster with a few lines of assembler. Am I wrong
in this? Are all the Mac-games really pure C? If they are, shouldn't it
be possible to make a blindingly fast game in PPC-assembler?
Phew! Had to get that off my chest...

.... erm, one question. How did we get from Mouseigator to the benefits
of assembler? Ahh, the joy of the Fantasm listserver.

Allan

Date: Tue, 15 Apr 1997 16:58:56 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: (8) Cursor focusing

On Tue, 15 Apr 1997, Allan Lund Jensen wrote:

> I can certainly see the advantage in using C++ or other languages (you
> can calculate cos and sinus-tables with them ;), but I think that most
> people are too quick to dismiss assembler.

Different languages have their strengths: Fortran is for scientific work,
COBOL for database, C++ is a general language. Assembly, however, is the
ultimate language, you can do any and everything in it. There are no
explicit data structures like classes, structs, int, floats, chars...
You can interpret the data however you want: in Budding Animator I print
the window number by storing the number as ASCII and I add to it as if it
were an int; sure you wouldn't have to do this in C++ but the fact that I
can do it in assembly is why it's so great. Assembly is more *advanced*
than the other languages because it's so *simple*. Take a look at the
much-hyped PowerPC and how advanced it is, in reality it's simpler than
the 68K, but this gives it enough flexibility to do things that the 68K
simply can't do--it's kinda like the way solids can only pass through
certain substances; but fluids, simple as they are, can pass through
virtually anything; and then take a look at gases... Basically, simpler
is better, because it enables you to do more.

> Am I wrong
> in this? Are all the Mac-games really pure C? If they are, shouldn't it
> be possible to make a blindingly fast game in PPC-assembler?

I don't know how a lot of Mac games are written, but since there really
hasn't been a standalone assembler since Fantasm, I would think that most
of them are in C.

James Hague posted a message once saying that he programmed for the Super
Nintendo--I'm in awe of you James:) Anyway, he said that a lot of the
games were written in assembly, which is why on a 65816 at 3 MHz, or
whatever processor it ran on, the games ran awesome. Yet on my RISC
PowerPC 601 at 60MHz the game Skittles, which is similar to a Super
Nintendo game "Yoshi's Cookie" chokes when the action gets intense.
Skittles is written in C++ and has a few features; Yoshi's Cookie was
probably written in assembly and has a lot of features. Of course,
Yoshi's Cookie performs better on a lesser processor than Skittles does
on a greater processor. This flat out sucks! If you take a look at James'
Bumbler, then you'd see how a computer game should perform; all this talk
that you need a 2000+MHz processor to run the latest games is an excuse
by programmers to be sloppy and lazy. Not that I'm totally downing
high-level languages, I just hate that they get all the respect as the
standard languages. Lightsoft, you guys are a godsend for making Fantasm.

> Phew! Had to get that off my chest...

Yeah, me too!

> .... erm, one question. How did we get from Mouseigator to the benefits
> of assembler? Ahh, the joy of the Fantasm listserver.

I couldn't have put it better myself. It's kind of neat that we can
talk about anything, even if we don't know why we're talking about it!

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Tue, 15 Apr 1997 17:09:40 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re:(7) Cursor focusing

On Tue, 15 Apr 1997, Kevin Avila wrote:
> hehe. You must be using like gcc 1.0. I just did the same thing on my UNIX
> box with gcc 2.7.4 and the assembly was 234 bytes and the C version was 7K.
> Still a huge differance. Or perhaps yours was mutch larger then mine
> becaused I used the XCOFF birnary format, and by defaulty on UNIX the
> format is ELF. For a comparason of compile size read what I wrote at
> http://www.cache-computing.com/Ontario.

I don't know what version of gcc we have, but knowing our school it's
probably the early beta! Either way, the relative size differeces are
more or less the same in our programs, which proves how much more efficient
assembly is over C++.

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Tue, 15 Apr 1997 15:47:47 -0700
From: Kevin Avila <kevin@cache-computing.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: (7) Cursor focusing

>I recently downloaded the source for Steve Israelsons (I think that's
>the right name) 3d-engine. You might remember him as the guy behind
>Pfhorte, the Marathon-editor. To my surprise there didn't seem to be a
>line of assembler in it. Although I have no idea how good code
>Codewarrior produces these days, it would surprise me if he couldn't
>make his engine a bit faster with a few lines of assembler. Am I wrong
>in this? Are all the Mac-games really pure C? If they are, shouldn't it
>be possible to make a blindingly fast game in PPC-assembler?
>Phew! Had to get that off my chest...
>.... erm, one question. How did we get from Mouseigator to the benefits
>of assembler? Ahh, the joy of the Fantasm listserver.

I spoke with him about writing his engine in assembly, he would love to,
but just does not have the time. However, I will be writting it in
assembly. I have taken over development of Pfhorte and now have about 33%
of it done in assembly. It runs mutch faster.

----------------------------------------
Kevin Avila http://www.cache-computing.com
kevin@cache-computing.com

Date: Wed, 16 Apr 1997 01:26:13 +0200
From: Allan Lund Jensen <maccoder@post6.tele.dk>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: (7) Cursor focusing

> I spoke with him about writing his engine in assembly, he would love to,
> but just does not have the time. However, I will be writting it in
> assembly. I have taken over development of Pfhorte and now have about 33%
> of it done in assembly. It runs mutch faster.
>
Well, there you go. I look forward to see the end result. And, hey,
since you'll have a 3d-engine up and running, you should go for broke
and make a game. I know I'd buy it ...

(BTW did anyone notice my weird and scary psychic powers in the last
mail. Let me try again: I sure wish someone would send me one of those
300 Mhz 603 processors ..... I'll post my adress later ;) )

Allan

Date: Tue, 15 Apr 1997 20:33:34 -0400
From: Stuart Ball <lightsoft@compuserve.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Whatever...:-)

Morning chaps and chapesses :-)
Re: game written in 68k assembler...
Well, we have a 3d system in 68k, written originally for the ST. About
three years ago we ported it to the Mac (that's why Fantasm was originally
written).

It really hammers along as it was writtn to run fast on an 8 MHz 68000. Add
in the 020 addressing modes, faster Mac processors/emulators and it flies.

Rewrite for PPC and it'd be scary.

One day....

Were we to "do" a game, I would not think of C for the core. I would think
of C for the shell however. Oh hang on, we do have a game - Eclipse. It'll
get finished one day :-).

One of the reasons Fant5 and Fanta_C are taking so bloody long is because
we need a most flexible system that allows mixing high level and low level
with ease. Should be fun :-)

The joy of assembly to me is:
a). You can write how you like. One minute something is being used as a
"char", the next it's something else.
b). You can develop your own style, system and methods to suit your needs.
c). It provides the ultimate in "tweakability". You can write sloppy (and
fast) initially, then really optimise it over and over until happy.
d). Nothing will ever run faster as long as the programmer had taken the
time to at least learn the rudimentaries.
etc.

And, for the record, some people may be under the impression that we are
rather wrapped up in moving our focus from assembler to C.
Rubbish. What we want is the best assembler, best C compiler and best
system for any other language you care to mention, invent or drag up from
the past.

Granted, you won't get proportional fonts in one of our editors, as it just
slows things down. You'll get Monaco bitmap as always.
(I can see that coming back to haunt me sometime in the future :-))

Stu.
Time for coffee.

Date: Tue, 15 Apr 97 20:22:42 -0700
From: James Hague <jhague@dadgum.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: (8) Cursor focusing

Figaro Tea wrote:
>Anyway, he said that a lot of the
>games were written in assembly, which is why on a 65816 at 3 MHz, or
>whatever processor it ran on, the games ran awesome.

99% of the games were written in assembly, 'cause the 65816 is a hard
processor for compiler writers to deal with. Having only one general
purpose register might be the problem :-)
The SNES was chock full of flashy graphics hardware too, of course.

Throughout the 16-bit game system era (still kicking, really), folks in
the newgroups kept up the dumb ol' "assembly is outdated, use high-level
languages" battle. And at the same time hundreds of complex games were
being written entirely in assembly for the Super Nintendo, Genesis, Game
Boy, Lynx, and Game Gear. Kinda throws that theory...

The best way to react to the "you're crazy for using assembly!" folks is
to listen to their arguments, smile, nod a lot, say "well, yes" every so
often, and then go on doing what you were doing before. It used to be
the assembly people doing the arguing, but for some reason the C and C++
crowds have a real problem with assembly coders these days.

James

Date: Wed, 16 Apr 1997 06:04:48 -0400
From: Stuart Ball <lightsoft@compuserve.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: News from the Tower 16 Apr '97

Hi Chaps and Chapesses,
1. Demos?
2. BB's?
3. Archives?

1. I've just been sorting out all my floppies and I found the old Lightsoft
'95 demo.
Amazingly it runs just fine on my PowerMac which blew me away. Just a few
equalisation problems.

Anyway, it got me thinking, "why not have a demo compo?".

If I remember right, it took about a week to write that old demo, and most
of that time was shape designing in Resedit. Coding was 18 hours. Music was
about an hour with an old tape deck, and it was written with Fant2 so it'd
be faster in 4/5.

If there's any interest, we'll post the '95 demo and source to the WWW
pages and organise some dates/rules etc. I'll see if I can get Lightsoft to
cough up a wee prize to boot.

If it's a "goer" then I want Allan's permission to rip his native fire code
and Brollo's to use the parallel add (I'll use it anyway, but it's polite
to ask) :-)

Looking at the summer time I'd guess as we really need to get Anvil running
first.

Stu,
Deep in a linker in deepest Grantham on a beautiful day.

---------------------------

2. We've been kindly offered a First Class BB via tcp/ip offering all the
usual FC features - chat, sigs, file areas, organised feedback meetings
etc. Sounds great but it's admin work for Lightsoft :-)
Thus we would like your feedback - would you use it?
It'd be for registered Fant users only and access would be free.

----------------------------

3. It's been suggested we should archive the mailing list and place the
archives on the WWW pages. Again feedback required as again it's more admin
work.

These suggestions are for your benefit, not ours, so no input and nothing
happens.
Feedback to Fant list please.
Thanks for your time.
Lightsoft 11021604

Date: Wed, 16 Apr 1997 16:09:13 -0500
From: See Neng Foo <snf@mail.utexas.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: News from the Tower 16 Apr '97

Why are you mailing every hour? Don't spam....
sn

Moon@bbs.telemacus.it wrote:
>
> Questo i un messaggio automatico di conferma alla mail che mi hai
> appena inviato, sfortunatamente non sarr reperibile fino a Venerdl
> 18.4.97, se necessario i possibile contattarmi allo 0338 7721207.
>
> Mario
> This message sent from Telemacus BBS, Milano, Italy.

Date: Wed, 16 Apr 1997 14:26:45 -0700
From: Kevin Avila <kevin@cache-computing.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: News from the Tower 16 Apr '97

>Hey, good idea, count me in!
>Making the whole thing in a week seems a little harsh, though. I=92d first
>have to hook my coffeemaker up to an IV-drop.
>Demos are usually good fun, but there are very few on the Mac (I=92ve only
>found one place on the net which has a few, and they appear to be very
>old).
>Could be good PR for using assembler, as well?
>So come on, guys. It=92s only a week of your life, and it=92s all for the
>good of =93the greater purpose=94.
>(If I=92m the only one who thinks this is a good idea, will I win the =91we=e
>prize=92 by default? I=92m kinda partial to fast sports-cars, just so you
>know ;) )

I'm in. I'm hooking up the coffee maker and the concert speakers for some
intense MOD's :)

----------------------------------------
Kevin Avila http://www.cache-computing.com
kevin@cache-computing.com

Date: Wed, 16 Apr 1997 22:25:32 +0100
From: Rob Probin <rob@zedworld.demon.co.uk>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: The Joy of Assembler [was Re: Cursor focusing]

>99% of the games were written in assembly, 'cause the 65816 is a hard
>processor for compiler writers to deal with. Having only one general
>purpose register might be the problem :-)
>The SNES was chock full of flashy graphics hardware too, of course.
>
To my knowledge 99% of games for all 8-bits, and 16-bits (inc. Atari and Amiga)
were assembler. The only exception worth any note I can think of was
Frontier (Elite 2), which if we believe the old guru himself, 'life was too
short for', pretty amazing if true.

In my experience, the key to writing fast code in C (or C++, or any
high-level language for that matter) is to know (a) what your compiler
outputs for each situation and tailor to suit (hence I think static code
motion optimisation is a waste of space), (b) know where you can ignore
that speed implications (sometimes not as easy to spot), and (c) changing
your high-level code so that it outputs better target code. (d) when it
comes to it - code the core sections in assembler. I'm sure this was Doom's
approach.

Hence the reason why generally high-level language programmers output so
much rubbish...(i.e. you have to be an assembler programmer to do it...)

>Throughout the 16-bit game system era (still kicking, really), folks in
>the newgroups kept up the dumb ol' "assembly is outdated, use high-level
>languages" battle. And at the same time hundreds of complex games were
>being written entirely in assembly for the Super Nintendo, Genesis, Game
>Boy, Lynx, and Game Gear. Kinda throws that theory...
>

They often have the resources to throw at a project to make it fast.
In the end, all the reasons for high level languages (reliablility,
maintainablility, etc, etc) boil down to one thing: FASTER CODE
DEVELOPMENT. Thats all. And when execution speed is not an issue, business
takes the easier, cost effective option. (What they often miss is the
actual *quality* of the program is not improved by doing things faster...)

>The best way to react to the "you're crazy for using assembly!" folks is
>to listen to their arguments, smile, nod a lot, say "well, yes" every so
>often, and then go on doing what you were doing before. It used to be
>the assembly people doing the arguing, but for some reason the C and C++
>crowds have a real problem with assembly coders these days.

Yeah, I know. Its that old "anything they don't know they're scared of and
want to prove to everyone its not an option".... Somewhat reminds me about
the PC vs. Mac arguement:-

MacUser: Macs are better than PC's
PCuser: But hardly anyone buys the Mac, so it can't be better.
MacUser: That doesn't mean anything - really the Mac is better.
PCuser: No the Mac is s**t.

This summaries 99.9999% of the PC vs. Mac debate. The only dent in the PC
users armour is the fact that Mr PCuser has never used a Mac (and
definately not for the required 3 months to know a system), but Mr MacUser
has used a PC for more than that...

Back onto compilers ... who said "some compilers output an efficiency of
110% on RISC processors compared to hand-written code" ??? What I want to
know is what efficiency ... whose code, and who are they trying to fool??
(and where can I get this compiler!)

Oops, I know there are some big compiler writers who read this list - any
comments guys???

>by programmers to be sloppy and lazy. Not that I'm totally downing
>high-level languages, I just hate that they get all the respect as the
>standard languages. Lightsoft, you guys are a godsend for making Fantasm.

We couldn't believe there wasn't a hand-written assembler development tool!
There is nothing _quite_ like getting 'to the metal'.

Rob

Date: Wed, 16 Apr 1997 17:36:15 -0500
From: Andy Clayton <bali@ccwf.cc.utexas.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: News from the Tower 16 Apr '97

>> Questo =C8 un messaggio automatico di conferma alla mail che mi hai
>> appena inviato, sfortunatamente non sar=DA reperibile fino a Venerd=CF
>> 18.4.97, se necessario =C8 possibile contattarmi allo 0338 7721207.
>>
>> This message sent from Telemacus BBS, Milano, Italy.
>
>I'm getting really tired of getting this same message a dozen times per
>hour: turn off your emailer or something--or at least translate what you
>said into english so the rest of us, or at least me, can read it.

Hm, my best guess would be that it is an automatic "I'm away, but your mail
has been received" message. It looks like whoever it is won't be back
until this Friday! Ug, that may mean we'll be getting 10 msgs * 24 hrs * 3
days =3D a lot of these helpful greetings from Italy. Maybe someone on this
list from Italy or somewhere close should call that number and tell him/her
that their auto-reply is screwed up. Or maybe contact the BBS operators?

Or else, can someone with administrative access to this list temporarily
block incoming messages from that address?

Andy Clayton

Date: Wed, 16 Apr 1997 14:56:26 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: News from the Tower 16 Apr '97

On Wed, 16 Apr 1997, Stuart Ball wrote:
> 2. We've been kindly offered a First Class BB via tcp/ip offering all the
> usual FC features - chat, sigs, file areas, organised feedback meetings
> etc. Sounds great but it's admin work for Lightsoft :-)
> Thus we would like your feedback - would you use it?
> It'd be for registered Fant users only and access would be free.
>
> ----------------------------
>
> 3. It's been suggested we should archive the mailing list and place the
> archives on the WWW pages. Again feedback required as again it's more admin
> work.

I was thinking about this for awhile myself, I think you should go ahead
with both ideas: the FirstClass BBS, and archiving the postings. I would
go with archiving if you only went with one idea--that way people could
see everything I've written:)

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Wed, 16 Apr 1997 21:42:10 +0200
From: Allan Lund Jensen <maccoder@post6.tele.dk>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: News from the Tower 16 Apr '97

> If there's any interest, we'll post the '95 demo and source to the WWW
> pages and organise some dates/rules etc. I'll see if I can get Lightsoft to
> cough up a wee prize to boot.

Hey, good idea, count me in!
Making the whole thing in a week seems a little harsh, though. Id first
have to hook my coffeemaker up to an IV-drop.
Demos are usually good fun, but there are very few on the Mac (Ive only
found one place on the net which has a few, and they appear to be very
old).
Could be good PR for using assembler, as well?
So come on, guys. Its only a week of your life, and its all for the
good of the greater purpose.
(If Im the only one who thinks this is a good idea, will I win the wee
prize by default? Im kinda partial to fast sports-cars, just so you
know ;) )

> If it's a "goer" then I want Allan's permission to rip his native fire code
> and Brollo's to use the parallel add (I'll use it anyway, but it's polite
> to ask) :-)

Oh, alright then. Just remember to include my name in 10-feet high
flaming letters while a deep american voice growls what a guy ;)

---------------------------

> 2. We've been kindly offered a First Class BB via tcp/ip offering all the
> usual FC features - chat, sigs, file areas, organised feedback meetings
> etc. Sounds great but it's admin work for Lightsoft :-)
> Thus we would like your feedback - would you use it?
> It'd be for registered Fant users only and access would be free.

Although I have no idea what the above is (well, I assume a BB is a
bulletin board, but the FC puzzles me) , the word free caught my eye
and made me an instant believer. If it means any long-distance calls,
Im still very much for it, but my bank-account wont be :(

----------------------------

> 3. It's been suggested we should archive the mailing list and place the
> archives on the WWW pages. Again feedback required as again it's more admin
> work.

Yep, thats neat too. Maybe I can get a chance to figure out what A new
way of programming was *really* about.

Tim wrote:
> --that way people could see everything I've written:)
Yeah, that would be nice ;)

On the whole it appears Im in favour of everything. More work to web
admin in particular. Spare time? Pah! Overrated in my opinion...

Allan

Date: Thu, 17 Apr 1997 06:15:40 +0200
From: Moon@bbs.telemacus.it
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: News from the Tower 16 Apr '97

<x-fontsize><param>9</param>Questo =E9 un messaggio automatico di conferma =
alla mail che mi hai appena inviato, sfortunatamente non sar=F2 reperibile =
fino a Venerd=EC 18.4.97, se necessario =E9 possibile contattarmi allo 0338=
7721207.

Mario</x-fontsize><x-fontsize><param>10</param>
This message sent from Telemacus BBS, Milano, Italy.
</x-fontsize>

(167 similar messages deleted)

Date: Wed, 16 Apr 1997 18:30:35 -0500
From: "robert c. wagner" <rcw@texas.net>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: The moon is orbiting itself

Is it possible to unsubscribe Moon from the list-server?

What we're seeing is obviously an automated response which looks
something like:
"I've got your message, but I won't be able to responed until Friday the
18th"

So every message posted to the list-server, including this one, will be
auto-replyed over and over and over and over again!!!!

robert

rcw@texas.net

Date: Wed, 16 Apr 1997 21:38:49 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Please Spare Us!

Please Lightsoft, spare us poor humble peasants from that email storm
coming from the Moon@bbs.telemacus.it! Canst thou find it in thine heart
to blocketh messages that spew from that wretched place--at least until
Friday gets here. Thy obedient peasants can barely takest it! We
promise we'll cutest down more trees and mine morest of that precious
substance, gold--at least until Diablo comes out. Please, sparest us!

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Thu, 17 Apr 1997 09:05:22 -0400
From: Stuart Ball <lightsoft@compuserve.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Mail loop stopped.

It looks as if we managed to break the loop about four hours ago.
It appears it was nobody's fault.

The list-server echoes mail back to the sender.
In this case this was causing an auto reply from the sender, which was
being echoed back, etc.

Anyway, should it happen again, we now know hope to stop it fairly
sharpish.
Nightmare!

Hopefully all should now be back to normal.
If anybody posted any mail last night (or yesterday depending on where
you're sitting) it may well be worth reposting it as I should think most
people either disconnected from the list or just hit "Delete". I did :-)

Sorry we couldn't stop it any sooner, first we saw of it was at 05:30 gmt.
Rgds,
Stu.
(I'll also post this message on the web pages for the benefit of those that
unsubscribed).

Date: Thu, 17 Apr 1997 10:06:34 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: Mail loop stopped.

On Thu, 17 Apr 1997, Stuart Ball wrote:
> It looks as if we managed to break the loop about four hours ago.

Oh bless you my liege! Unfortunately we won't be able to harvest any
resources as promised.....uh.....ummmm, it was those nasty Peons!
..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Thu, 17 Apr 1997 12:20:47 -0400
From: Stuart Ball <lightsoft@compuserve.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: BB/Compo's etc.
Tim wrote:
>Oh bless you my liege! Unfortunately we won't be able to harvest any
>resources as promised.....uh.....ummmm, it was those nasty Peons!
Stu thinks that whatever it is Tim is taking, he'd like to try some :-)

Allan wrote:
>This is a reposting of the message I send last night, in case you were
>too busy erasing greetings from sunny Italy ...
Well, live and learn as they say...
>>Demo...
>Hey, good idea, count me in!
I'd say we need at least five entrants to make it worth while. I've checked
and I guess we can put up 2-300 bucks prize money.

>Making the whole thing in a week seems a little harsh, though. Id first
>have to hook my coffeemaker up to an IV-drop.
Seriously, it doesn't take that long - the idea is the hardest part. You
deff. need a good vision of what you want before you start, and then you
find out how to do stupid things that really are only useful in a demo...
For example, in the 95 demo towards the end you'll see the stars suddenly
become like very fast masers (or whatever) shooting out of the corners of
the screen. This was an accident resulting from not checking the carry bit
and with a little bit of math overdriving that was the result. As you say,
great fun.
I think it'd be good if one of the rules stated that the winning entrant
had to publish the source for all to drool over in the interests of
education.

>Demos are usually good fun, but there are very few on the Mac (Ive only
>found one place on the net which has a few, and they appear to be very
>old).

Yes, agreed. I found that site (and promptly lost it again) a while back.
It had the Screaming Cabala demo on it for one...
>Oh, alright then. Just remember to include my name in 10-feet high
>flaming letters while a deep american voice growls what a guy ;)

Umm, most people have sort of smaller than 10 foot monitors, but the "what
a guy" sample we can rip from Red Dwarf no probs :-)
>Although I have no idea what the above is (well, I assume a BB is a
>bulletin board, but the FC puzzles me) , the word free caught my eye
>and made me an instant believer. If it means any long-distance calls,
>Im still very much for it, but my bank-account wont be :(
As I understand it it's accessed via tcp/ip and PPP so it's just your local
isp thing.
FC is First Class. BB was indeed bulletin board.
I'm sure Kev or Mario (when he re-subscribes :-)) can give more explicit
details on FC BB's.

>>List archives
>Yep, thats neat too. Maybe I can get a chance to figure out what A new
>way of programming was *really* about.
Please, don't start Tim off again <BG>. I enjoyed it thou...

>On the whole it appears Im in favour of everything. More work to web
>admin in particular. Spare time? Pah! Overrated in my opinion...
Spare time? This is me 12 hours in and I still have a lot of work to do.
Still, life's too short to sleep.
So, to sum up:
That's 1 for the demo comp and 1 for the archives.

BTW, was there anything about high level/low level I missed last night?
Anybody willing to come out with "There's nothing I can't do in C as fast
as you can in assembler"?

Stu in a "not connected at all with Lightsoft, hard at work on a C
compiler" role.

Date: Thu, 17 Apr 1997 09:39:52 -0700
From: Kevin Avila <kevin@cache-computing.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: BB/Compo's etc.

>I think it'd be good if one of the rules stated that the winning entrant
>had to publish the source for all to drool over in the interests of
>education.

uh-no. You mean I have to comment it and use complex, extended menemonics
so it looks like I know what i'm doing? oh no. Stu, can I post OpenOS as a
demo? It's not done yet, but I tink it will be the only demo of a compleate
operating system. :)

>FC is First Class. BB was indeed bulletin board.
>I'm sure Kev or Mario (when he re-subscribes :-)) can give more explicit
>details on FC BB's.

I like FC. :) I tink it will be an exelant place to post files, and chat.
It's easy toi use (interface is just like a mac) I would recomend you
download it from www.softarc.com and just play with it. It's great. (hows
that for a plug?)

Also check out Tim's mousie thing, James's bumble bee game (waaaaay cool),
anyone else have software they want me to plug? Also, anyone want to
sign-up with Exel communications as the LD carrier? Just had to ask for a
friend. :)

----------------------------------------
Kevin Avila http://www.cache-computing.com
kevin@cache-computing.com

Date: Thu, 17 Apr 97 09:40:09 -0700
From: James Hague <jhague@dadgum.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: The Joy of Assembler [was Re: Cursor focusing]

>Back onto compilers ... who said "some compilers output an efficiency of
>110% on RISC processors compared to hand-written code" ??? What I want to
>know is what efficiency ... whose code, and who are they trying to fool??
>(and where can I get this compiler!)

It wasn't me.

But I've used at least one compiler (GNU C for the SH2) where I couldn't
beat it much of the time. I'd rewrite a routine in assembly and it would
be slower than the C. Not that I'm an SH2 expert, mind you.

Cool optimizations don't usually come from rewriting little routines in
assembly though, but from rewriting big, huge chunks of code. Dodging
the subroutine entry/exit code on the PPC can be a big deal.

Re the comment Rob made about static code motion being a waste
Right on! I always thought compiler writers went out of their way to try
to improve horrendous code. Why bother?

James

Date: Thu, 17 Apr 1997 22:10:57 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Computer Problem

I know I should probably post this elsewhere, but I'll try here first:

In the lab that I work in there is a computer that won't start up and
asks for a floppy disk instead of booting off the hard drive. I don't
have access to the CD so I can't insert that to make things work better.

I read somewhere that holding down a certain keyboard sequence makes the
Mac start at SCSI ID 0 and work its way from there until it finds a
startup disk; anybody know the keyboard combo? I doubt too much can be
done unless I can get ahold of the CD for it, but I figured I'd at least
try asking.

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Thu, 17 Apr 1997 23:26:19 -0700
From: Kevin Avila <kevin@cache-computing.com>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: Computer Problem

>I know I should probably post this elsewhere, but I'll try here first:
>
>In the lab that I work in there is a computer that won't start up and
>asks for a floppy disk instead of booting off the hard drive. I don't
>have access to the CD so I can't insert that to make things work better.
>
>I read somewhere that holding down a certain keyboard sequence makes the
>Mac start at SCSI ID 0 and work its way from there until it finds a
>startup disk; anybody know the keyboard combo? I doubt too much can be
>done unless I can get ahold of the CD for it, but I figured I'd at least
>try asking.

I think it is cmd-opt-right shit pr ctrl-opt-right shit. etc.

----------------------------------------
Kevin Avila http://www.cache-computing.com
kevin@cache-computing.com

Date: Fri, 18 Apr 1997 08:22:28 -0600
From: hamilton@airnet.net (Chris Hamilton)
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re: Computer Problem

>I read somewhere that holding down a certain keyboard sequence makes the
>Mac start at SCSI ID 0 and work its way from there until it finds a

cmnd-option-shift-delete at startup (though it actually starts at SCSI ID 5
and goes backwards)

Alternately, create a boot floppy. You'll need 'System', 'Finder', and all
system enablers for that type of computer. You should be able to find the
files on a computer with the same processor and motherboard, and there
should also be a 'Utilities' floppy from Apple somewhere with the manuals.

***

FYI Macs look for startup disks in the following order:

Internal floppy drive
Second internal floppy drive
External floppy drive
Startup device selected in Startup Disk control panel (stored in PRAM)
Internal hard disk (bypassed if key combination is pressed)
External SCSI devices (beginning with highest SCSI number)
Internal hard disk (in case it wasn't warmed up yet the first time)
Src: Macintosh User's Guide for Macintosh Performa computers, p.90.

---
Chris Hamilton -- hamilton@airnet.net
http://www.airnet.net/hamilton/
http://members.tripod.com/~GeoffryN/

Date: Sat, 19 Apr 1997 15:02:18 -0400 (EDT)
From: Figaro Tea <humphret@lurch.winthrop.edu>
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re:(2) Computer Problem

On Fri, 18 Apr 1997, Chris Hamilton wrote:
> Alternately, create a boot floppy. You'll need 'System', 'Finder', and all
> system enablers for that type of computer. You should be able to find the
> files on a computer with the same processor and motherboard, and there
> should also be a 'Utilities' floppy from Apple somewhere with the manuals.

I wish I could create a boot floopy, but with the Finder and the System
Enabler there isn't 1.4 MB left for the System on the floopy! I honestly
hope that ZIP drives become standard, floopy disks just can't cut it anymore.

Is there a way to trim down the System to get just the stuff you need;
i.e. what files are essential for bootup?

..._Tim_...
--=[Live your life in the present, history will make itself in the future.]=--
http://www.winthrop.edu/~humphret
zzhumphreyt@winthrop.edu

Date: Sat, 19 Apr 1997 23:02:45 -0600
From: hamilton@airnet.net (Chris Hamilton)
To: Multiple recipients of <list-fantasm@mail.tau.it>
Subject: Re:(2) Computer Problem

>On Fri, 18 Apr 1997, Chris Hamilton wrote:
>
>> Alternately, create a boot floppy. You'll need 'System', 'Finder', and all
>> system enablers for that type of computer. You should be able to find the
>> files on a computer with the same processor and motherboard, and there
>> should also be a 'Utilities' floppy from Apple somewhere with the manuals.
>
>I wish I could create a boot floopy, but with the Finder and the System
>Enabler there isn't 1.4 MB left for the System on the floopy! I honestly
>hope that ZIP drives become standard, floopy disks just can't cut it anymore.
>
>Is there a way to trim down the System to get just the stuff you need;
>i.e. what files are essential for bootup?

Get rid of all of the sounds and keyboard layouts by dragging them to the
trash on the desktop. Remove the following resources:

(Do not remove the keyboard layouts from within ResEdit. The standard US
keyboard layout is invisible from the desktop. You certainly don't want to
remove this, and you may do so inadvertantly in ResEdit.)

alis
atlk
dtn
fwst
icl4
icl8 (You'll only have b/w icons, but it's only temporary and
ics4 is worth the >75 kb you'll save on the floppy)
icl8
lmgr
ltlk
mean
movp
and all uneccessary PICT resources.

Note that I'm using MacOS 7.1 on a 68LC040. Some of these may, and
probably will be different. Perhaps someone else can make a suggestion if
you'll tell what kind of machine and OS version you have.

Also, some of the System Enablers aren't 100% neccessary for the minimum
operation of the computer. Somewhere on the Internet I've seen a list of
the different enablers and what each does. Maybe at Apple? Take a look at
that and get rid of anything that you can possibly do without.

If you can find the 'Utilities' disk that came with the machine, it already
has the minimum startup files.

---
Chris Hamilton -- hamilton@airnet.net
http://www.airnet.net/hamilton/
http://members.tripod.com/~GeoffryN/

BACK