Will advancements kill the demo scene?
ri...@minnie.cc.utexas.edu
new doodads and technological advancements which may kill the demo
scene for IBM machines as we know it, or at the very least push it
into obscurity.
(1) Out-of-order execution.
Newer processors, such as AMD's K5, pick out the order they want to
process code, so heavily optimized compilers or heavily tweaked
Assembly, as processors get more powerful this way, will no longer be
needed.
(2) WinG.
Two facts: (1) Microsoft rules the world whether we like it or not,
and (2) they aren't supporting DOS after next year, whether we like it
or not. They've come out with a new graphics engine which is still
slower than DOS and full-screen page-flipping, but is adequate to
handle animation under Windoze.
(3) GUSMax problems.
As great of a card as it is, I've heard of many people who simply
cannot get one to work in their machine. Apparently, it conflicts
with just about any other soundcard, and even some SCSI adapters.
This could very well end Gravis' chances in the soundcard market.
Okay, there's the little pieces of almost-factual information which I
know. So, what I want to hear is,
about (1): If chips with out-of-order execution become commonplace,
will Assembly Language coding die out?
about (2): Has anyone had the chance to work with WinG betas, and
what did you think of it? Also, what will happen to DOS once the
Almighty Bill has stopped supporting it? Will it struggle to survive
for a few years later, only to fizzle into obscurity...such as the
Apple II series, or will it gain a life of its own, where the founding
company will no longer be in control of its own product, such as the
whole IBM architecture (remember Micro Channel?) or the whole "New
Coke" fiasco?
about (3): I think we've heard enough about this already, and anybody
who even thinks about responding to this one deserves to be
core-dumped liberally. But, if anyone has something LEGITIMATE TO SAY
ABOUT HOW THIS EFFECTS THE DEMO SCENE AND NOT JUST ANOTHER DAMNED "WHY
CAN'T YOU WRITE FOR MY SOUNDCARD" POST, even then think thrice before
responding. Who knows? There might actually be someone with
something intelligent to say about this yet.
Well, just trying to stimulate some speculation. Have fun responding
to this.
--
Jimmy Rimmer, Rimbo / Lucid ri...@ccwf.cc.utexas.edu
...because the eagle will still fly in the rain.
Zolotukhin, Alex
>From: ri...@minnie.cc.utexas.edu
>Subject: Will advancements kill the demo scene?
>Date: 28 Nov 1994 09:47:33 -0600
>Okay, the title isn't entirely accurate, but I've heard about a few
>new doodads and technological advancements which may kill the demo
>scene for IBM machines as we know it, or at the very least push it
>into obscurity.
>
>Two facts: (1) Microsoft rules the world whether we like it or not,
>and (2) they aren't supporting DOS after next year, whether we like it
>or not. They've come out with a new graphics engine which is still
>slower than DOS and full-screen page-flipping, but is adequate to
>handle animation under Windoze.
>
>what did you think of it?
Well, I've been working with it for about a month now, and it seems pretty
nice. You can do things like communicating directly with the video card (
which Windows as is won't allow you), fast blt's. I think you can use your
fast DOS graphics routines in assembly in Windows now (with some changes, of
course), although I haven't tried it yet. Animations are pretty fast. WinG
comes with some demo programs, and one of them is rotating cube. When I run
it, the rotation and the color change was smooth and perfect.
Of course for time being DOS is faster in graphics than Windows, even
if you use WinG's libraries, but Microsoft is working on it. Just look at
DOOM for Windows. ID made it using Microsoft's WinG, and it runs FASTER than
the DOS version. That's a good step forward :-).
>Also, what will happen to DOS once the
>Almighty Bill has stopped supporting it? Will it struggle to survive
>for a few years later, only to fizzle into obscurity...such as the
>Apple II series, or will it gain a life of its own, where the founding
>company will no longer be in control of its own product, such as the
>whole IBM architecture (remember Micro Channel?) or the whole "New
>Coke" fiasco?
That I don't know. I have been coding for Windows for a long time now, and I
like it a lot, but I sure would like to have good old friend DOS around too.
I hope it survives :-(.
Bye,
+======================================================+
| Alexey Zolotukhin / ale...@hauk.hsr.no / (aka) Nexus.|
| Student of Stavanger College in Norway. |
| -----------------------------------------------------|
| Lovings to usual internet friends: |
| Master Bo, Fat Brother, Blue Crystal, and evrybody |
| else I know on Internet :-) |
+======================================================+
Chris Hargrove
>about (2): Has anyone had the chance to work with WinG betas, and
>what did you think of it? Also, what will happen to DOS once the
>Almighty Bill has stopped supporting it? Will it struggle to survive
>for a few years later, only to fizzle into obscurity...such as the
I doubt DOS will die completely, if at all. That is, at least until another
OS comes out that gives you so much control with very little overhead.
Coders want to be able to control their machines. Without an OS like
DOS, they can't accomplish this, which would piss off a LOT of
developers. Even Microsoft wouldn't be able to take the wrath of
millions of power users should they kill DOS, without giving us a
replacement with the same performance ability. WinG, for all its
gains, just won't cut it.
Kiwidog / Terraformer / TGD
R. Eijkelhof
>Okay, the title isn't entirely accurate, but I've heard about a few
>new doodads and technological advancements which may kill the demo
>scene for IBM machines as we know it, or at the very least push it
>into obscurity.
No way. The demo scene is a totally different world than all the
microsoft business windows crap.
>(1) Out-of-order execution.
>Newer processors, such as AMD's K5, pick out the order they want to
>process code, so heavily optimized compilers or heavily tweaked
>Assembly, as processors get more powerful this way, will no longer be
>needed.
asm will always be the fastest language around. So whatever some
windows freak can do in any windows coding language, it can be
done faster in asm.
>(2) WinG.
>Two facts: (1) Microsoft rules the world whether we like it or not,
>and (2) they aren't supporting DOS after next year, whether we like it
>or not.
(1) True (2) So what. Demo coders DO support dos next year.
In fact I haven't had much support from microsoft the last
ten years anyway.
> They've come out with a new graphics engine which is still
>slower than DOS and full-screen page-flipping, but is adequate to
>handle animation under Windoze.
demos are coded in dos. this new engine is slower than dos if
it comes to memory xs. So: dos still rulez
>(3) GUSMax problems.
>As great of a card as it is, I've heard of many people who simply
>cannot get one to work in their machine. Apparently, it conflicts
>with just about any other soundcard, and even some SCSI adapters.
>This could very well end Gravis' chances in the soundcard market.
Well there's still the regular GUS which dominates the demo scene
for 99%. And whatever, Gravis ain't lamers so they will fix this
for sure.
>about (1): If chips with out-of-order execution become commonplace,
>will Assembly Language coding die out?
Nope.
>about (2): Has anyone had the chance to work with WinG betas, and
>what did you think of it?
Even WITHOUT working with it I think it sux.
> Also, what will happen to DOS once the
>Almighty Bill has stopped supporting it?
It still rulez for demo coding
> Will it struggle to survive
>for a few years later, only to fizzle into obscurity...such as the
>Apple II series, or will it gain a life of its own, where the founding
>company will no longer be in control of its own product, such as the
>whole IBM architecture (remember Micro Channel?) or the whole "New
>Coke" fiasco?
yep... dos won't die
>about (3): I think we've heard enough about this already, and anybody
>who even thinks about responding to this one deserves to be
>core-dumped liberally. But, if anyone has something LEGITIMATE TO SAY
>ABOUT HOW THIS EFFECTS THE DEMO SCENE AND NOT JUST ANOTHER DAMNED "WHY
>CAN'T YOU WRITE FOR MY SOUNDCARD" POST, even then think thrice before
>responding. Who knows? There might actually be someone with
>something intelligent to say about this yet.
Everybody supports your soundcard as long as you have a GUS.
>Well, just trying to stimulate some speculation. Have fun responding
>to this.
Sure did!
- Red Ice / Genuine
Chris Pimlott
ZA> In article <3bcu2l$5...@minnie.cc.utexas.edu> ri...@minnie.cc.utexas.edu
ZA> writes:
ZA> >From: ri...@minnie.cc.utexas.edu
ZA> >Subject: Will advancements kill the demo scene?
ZA> >
ZA> >about (2): Has anyone had the chance to work with WinG betas, and
ZA> >what did you think of it?
ZA> Of course for time being DOS is faster in graphics than Windows, even
ZA> if you use WinG's libraries, but Microsoft is working on it. Just look at
ZA> DOOM for Windows. ID made it using Microsoft's WinG, and it runs FASTER
ZA> than the DOS version. That's a good step forward :-).
Oh really? That's big news to me... At least on my 486, it's a helluva lot
slower in Windows than it is in DOS.
/\ < "Just because you're paranoid
Chris Pimlott -' \ .- piml...@shadowso.com \ Don't mean they're not after
\/ > you..." -- Nirvana
Mats A Knip
> Newer processors, such as AMD's K5, pick out the order they want to
> process code, so heavily optimized compilers or heavily tweaked
> needed.
> about (1): If chips with out-of-order execution become commonplace,
> will Assembly Language coding die out?
The way I see it, out-of-order execution is no reason to give up
assembly programming (or optimizing compilers). There's a lot more to
optimizing than ordering the instructions in the most optimal
way. Assembly coding is as much about rewriting a routine that
originally required 15 instructions into a 5 instruction routine, and
no matter how efficiently OOO-CPUs execute the instructions 5
instructions will still execute in shorter time than 15 (or
20). OOO-CPUs will probably close up the gap between high-level
languages and pure assembly, but real coders don't code in assembly
just for speed anyway. They do it because it's more fun :-)
(at least on non-Intel CPUs...)
--
------------------------------------------------------------------------------
Mats Knip, mk...@snakemail.hut.fi
Dick Verweij
>>(3) GUSMax problems.
>>As great of a card as it is, I've heard of many people who simply
>>cannot get one to work in their machine. Apparently, it conflicts
>>with just about any other soundcard, and even some SCSI adapters.
>>This could very well end Gravis' chances in the soundcard market.
BULLL!
I have spoken to many people owning GUSMAXes, and never had any problems
with the card. There are always some lamers with hardware problems, and not
ONLY with this card.. CHECK OUT the newsgroups, and see that other sound-
boards are having even more trouble with IRQs, DMA, etc...
BTW: GUS is one the best cards around for its price, and it can do
about everything. READ: 'WHY GUS IS GREAT FAQ'.
--
DickyDick
email: hpve...@cs.ruu.nl
ONLY A DICK MAKES IT POSSIBLE
Tomi Aarnio
> new doodads and technological advancements which may kill the demo
> scene for IBM machines as we know it, or at the very least push it
The demo scene will not die, it'll only change like it's
already been doing for the past ten years.
: (1) Out-of-order execution.
> Newer processors, such as AMD's K5, pick out the order they want to
> process code, so heavily optimized compilers or heavily tweaked
I'm not familiar with the concept, but I really don't think
that "out-of-order execution" would make it unnecessary to
optimize your code by hand. After all, they said that the
Pentium would make Assembly language optimizing useless with
its dual pipelines etc... True, everything that you had learnt
from all those clock cycle tables for the earlier cpus is
useless, but optimizing by hand most certainly is not. In
fact it's even more interesting (and rewarding) for the Pentium
than it was for the earlier 86's. Hmm, how about optimizing for
Dual-Pentium machines?
Anyway. Even if you could "kill" the assembly language, you
would not kill the demo scene. The scene will adapt to changes.
> (2) they aren't supporting DOS after next year, whether we like it
> or not. They've come out with a new graphics engine which is still
Actually demo coders haven't much been supporting the DOS for
the past year either... Quite the opposite - considerable
efforts have been made to bypass the DOS and its limits. :-)
Most demos are coded in flat protected mode, DOS only serves
as the program loader. It'll be left behind. So what's next?
Most likely the same platform as the game industry will choose;
games and demos have similar interests. Eg. real-time graphics
is a must for decent demos and games - stupid solitaire games
and feeble screen savers not counted as games and demos, of
course. Win3.x will not do for either. OS/2 maybe?
> This could very well end Gravis' chances in the soundcard market.
You make it sound like Gravis Inc. were the demo scene. As
great support as they've given us -- the GUS is still just
a soundcard. At the moment it's THE soundcard, yes, but in
just two years it might very well be a memory. The demo
scene has seen entire architectures go down (Atari ST, C64,
soon the Amiga?), but still it's only been growing as whole.
Perhaps the IBM PC demo scene will vanish some day, but who
cares, there'll always be other - better - machines to play
with. Although the PC/ST/Amiga demo freaks have not always
been very friendly towards each other, we're still one big
family (what a cheesy way to put it). :-)
> Well, just trying to stimulate some speculation.
Congratulations for writing the best csipd article for at
least a month. :-) IMHO of course.
-- dev/emf
Ben Cooley
WinG... what a joke.. just another Microsoft ploy to get those unruly coders
to do it the Big M way. Who wants to write a demo to play in a window? what's
the point anyway?
o Hardware coding in Windows
Anyhow.. why does having Windows as your operating system limit you from
getting down to the hardware of the machine. Amiga demo coders have been
doing this for a long time. Though a lot of amiga demo's must be booted from
a floppy, most (some??) will actually run from the Amiga shell and seem to
take over the machine just fine. The same can be easily done in Windows, just
not quite as easily as DOS. I write games in Windows, and I don't have any
trouble moving the stuff I don't want out of the way. Ever hear of Death()
and Ressurection()? (PageLock() is also rather useful).
o DOS dying
DOS, or something very close to it, will still be the underlying OS underneath
Win95. It'll be around for at least another 2..3 years. Plus, if NT takes
over, you can still add ring 0 drivers to it to overcome any nasty system
limitations that might bug you. (Same goes for Win 3.1 with VXD's).
Plus, I don't see what point there is in worrying about DOS vs. Windows when
they both will do the same stuff. So what if DOS dies... just means you hack
past Windows limits the same way you do DOS.
o 3D hardware
Didn't see the beginning of your thread so I didn't notice whether you
mentioned 3D hardware or not. Sure demo's that compete by having the fastest
3D texture mapping, shading, etc. might be hurt by 3D hardware, but coders
will just go on to do fancier 3D effects and shows when freed from doing the
grunt work of polygon drawing. I expect to see stuff rivaling Hollywood (in
real time) in the not too distant future. In fact, demo's are going that way
right now, faster than any comercial mm products or games in a lot of ways.
Also, VESA is creating a low level hardware standard for accessing the
capablities of 3D video cards.. complete with a software stand-in when the
capablities aren't there. That might fit the down to the the metal
requirement for demos.. if it doesn't there's always hard coding to the most
popular cards.
o GUS
Hey, the GUS was the first RAM based card first. No other cards have
RAM (well.. except the MAUI and the AWE..), and demo's must have the RAM to
play back custom instruments. As long as the rest of the sound card bozo's
keep ignoring RAM, the GUS will live on.
o Demo's in general
Demo's are programmer art. As long as there's somthing difficult and tricky
to do that will impress the hell out of someone else, and there's the hardware
to do it on, there will be demo's. The subjects and focus may change, but I
think that tricky audio/video/3D programming will always be popular.
jsomers
>new doodads and technological advancements which may kill the demo
>scene for IBM machines as we know it, or at the very least push it
>into obscurity.
>
>(2) WinG.
>Two facts: (1) Microsoft rules the world whether we like it or not,
>and (2) they aren't supporting DOS after next year, whether we like it
>or not.
This just in....
According to the latest issue of PC Week (12/5/94) Microsoft will release
an upgrade to DOS. Here's why...
1) To appease vertical-market (niche) software companies, who are concerned
that their software won't run under Win 95.
2) Microsoft doesn't want to abandon the DOS market to a competitor. IBM
is coming out with a 32 bit PC-DOS 7.0 in February.
3) Contrary to what Microsoft says publicly, it is easily technically to
uncouple MS-DOS 7.0 from Win 95. So if users just want DOS, it would be
simple to have them buy it from Microsoft so Microsoft can get their bucks.
4) "The Justice Department's consent decress specifies that Microsoft is
not permitted to 'tie' separate operating systems and graphical environments
together in a way that would prevent other vendors' products from working
with Windows. 'Microsoft could get in trouble if they try to coerce vendors
who want just DOS to also take Windows or if users were unable to run
Windows [95] on top of things other than MS-DOS 7.0' "
The article says Microsoft is publicly denying that MS-DOS 7.0 will exist,
but is privately telling some companies that, in fact, DOS 7.0 will be
available separately. Of course, there's no guarantee this will happen,
but PC Week is usually right on the money.
And now back to the demo scene....(the one computing area not dominated
by Bill Gates).
jeff s.
jso...@id.wing.net
ri...@huey.cc.utexas.edu
[chainsaw applied to info about MS-DOS 7.0, which was kinda nice to hear]
>And now back to the demo scene....(the one computing area not dominated
>by Bill Gates).
Unless, of course, you count the way Verses shook up Asm '94 with the
hilarious warped-Bill routines...
Sorry, dev, I couldn't resist... :)
Samuel Marshall
|> >(2) WinG.
|> >Two facts: (1) Microsoft rules the world whether we like it or not,
|> >and (2) they aren't supporting DOS after next year, whether we like it
|> >or not.
|>
|> (1) True (2) So what. Demo coders DO support dos next year.
|> In fact I haven't had much support from microsoft the last
|> ten years anyway.
Who needs DOS anyway? Demos don't need to use any of its `features', except for
file access. The new ones use extenders for memory alloc etc. Demos could load
from boot sector or a simple `DOS' (Demo Operating System) disk like this:
1) You have a `demo operating system' boot disk.
2) When you get a new demo, after copying it onto your hard disk, you register it
onto the boot disk (using software on the boot disk).
This copies the demo's fileposition details onto the disk, and the demo then
appears on the disk's startup menu.
3) You boot from the disk (straight into protected mode, no drivers QEMM etc
loaded) and select the demo from the menu.
4) The demo is loaded into RAM as usual, with a command-line argument to its file
position information so it can use BIOS to do file reads.
There you go. If the worst comes to the worst, who needs DOS anyway? [I'm not
saying this system is sensible, but it's *possible* ]. Saves you all the hassle
with drivers, setups etc.
Of course if the demo fitted on a floppy it would be even simpler, as could just
boot from the boot sector, but many demos are 2 big for this and I don't
think pc users want to get into disk swapping now...
It would be easier to just keep a copy of DOS around, but then who cares...
[I just realised this would be incompatible with smartdrv etc. Oh dear. Well
demos will just have to go on the uncompressed drive then.]
--
Sam
--> Home Page <-- ** Go on, try it! ** --> http://www.dur.ac.uk/~d405ua <--
Teemu Valtonen
> Who needs DOS anyway? Demos don't need to use any of its `features', except
> for file access. The new ones use extenders for memory alloc etc. Demos
> could load from boot sector or a simple `DOS' (Demo Operating System) disk
The problem is that no-one would want to code demos under
such a primitive OS, even if one could install it on a
hard disk.
> Of course if the demo fitted on a floppy it would be even simpler, as could
> just boot from the boot sector, but many demos are 2 big for this and I
You can't be serious...
-- saracen /emf
Eric Rasmussen
: with the card. There are always some lamers with hardware problems, and not
: ONLY with this card.. CHECK OUT the newsgroups, and see that other sound-
: boards are having even more trouble with IRQs, DMA, etc...
: BTW: GUS is one the best cards around for its price, and it can do
: about everything. READ: 'WHY GUS IS GREAT FAQ'.
Look at the majority of people who are having these problems... They are
usually a bunch of lame-asses who don't know an IRQ from a DMA from a
CPU, and so on... if only people knew more about their computers, they
wouldn't be having these problems... :)
--Mussen
Eric Rasmussen
University of Nevada, Reno
ras...@scs.unr.edu
"Well, no demo group yet, but working on it! :)"
ri...@huey.cc.utexas.edu
Eric Rasmussen <ras...@scs.unr.edu> wrote:
>Look at the majority of people who are having these problems... They are
>usually a bunch of lame-asses who don't know an IRQ from a DMA from a
>CPU, and so on... if only people knew more about their computers, they
>wouldn't be having these problems... :)
I just thought you'd like to know this.
I most certainly know the difference between a DMA and an IRQ. I have
worked professionally doing hardware technical support for years. On
my own computer I have successfully installed four soundcards (up to
three at once), DRAM, SIMMs, a new hard drive, a mathco, and I've
reversed the connections on the floppy drives so that the 3.5" is
drive A:.
I am having problems with my MAX not because I'm a lame-ass who can't
tell an IRQ from a DMA, but because the MAX doesn't play correctly
things my old GUS could...and not for lack of memory, but because the
thing is BROKEN. Something wrong WITH THE CARD is causing the
problem.
The majority of problems come from people who have SCSI cards on a
second sound card that they don't want to give up (e.g., for better SB
support for games). The MAX simply WILL NOT co-exist with another
sound card. The only reason my MAX co-exists is because my LAPC-1 is
a very non-controversial card (leave 330h and IRQ 5 alone, and it's
happy), and even then, as I said before, the card ITSELF is faulty.
Don't go judging people before you've been in their shoes.
Jani Mattsson
> [...] and I've reversed the connections on the floppy drives so that the 3.5" is
> drive A:.
Woaah.. and this was the top point of your career?->
(Sorry, couldn't resist :))
> The MAX simply WILL NOT co-exist with another sound card.
You are talking here about the specific Max you have, which
might be faulty, at least my GUS Max works perfectly with
my SB Pro and NE2000-compatible ethernet card.