Impulse Tracker 2.01

6 views
Skip to first unread message

ri...@ccwf.cc.utexas.edu

unread,
May 2, 1996, 3:00:00 AM5/2/96
to

Lissen to me an' lissen good.

Impulse Tracker 2 is now the ultimate tracker made.
Coders--go get off your butts and support the new format.

If you don't like the IT format, expect to find yourself wondering where
the 20th century went in approximately 3.66 years. This is the
shit--this is the fucking shit. Give the coder a butt-load of credit for
listening to what we musicians really wanted and finally giving us a
tracker that does EVERYTHING.

There is nothing left to discuss...please send flames to /dev/null.


--
Jimmy Rimmer Rimbo / Lucid http://ccwf.cc.utexas.edu/~rimbo/
* "If you hide your ignorance, nobody will hit you and you'll never learn." *
* [from Fahrenheit 451] *
"There's something noble about procrastinating now instead of later." --J.C.

Trixter / Hornet

unread,
May 7, 1996, 3:00:00 AM5/7/96
to

In article <4m9lmm$s...@piglet.cc.utexas.edu>,

<ri...@ccwf.cc.utexas.edu> wrote:
>Lissen to me an' lissen good.
>
>Impulse Tracker 2 is now the ultimate tracker made.
>Coders--go get off your butts and support the new format.

I don't know any one coder who's willing to support New Note
Actions--which, by the way, actually *reduce* the quality of your
song if you've got a GUS. Otherwise, it's S3M meets XM in terms
of support.

New Note Actions will kill that format amongst coders and GUS owners.
What the heck was he thinking?
--
Jim Leonard (Trixter / Hornet) Email: tri...@mcs.com
*THE* PC Demo WWW page: http://www.cdrom.com/pub/demos/hornet/html/demos.html
The 8086 Compo is OVER! Results at http://www.cdrom.com/pub/demos/hornet/8086
Make A Computer easy enough for a fool to use, and only fools will use it!

Andrew G Sega

unread,
May 7, 1996, 3:00:00 AM5/7/96
to

trixter (tri...@mcs.com) wrote:
: New Note Actions will kill that format amongst coders and GUS owners.

: What the heck was he thinking?

well the thing is, if you are smart and don't use the NNA silliness, it ends up
being THE perfect tracker for anyone who is used to st3. if you like the
ft2 gui, then stick with that, but for us who need the speed and macros,
use impulse. and come on, like you couldn't just code an .it player which
ignored NNA triggers :>

cheers
andy

# necros / five musicians
# nec...@fm.org
# as...@ic.sunysb.edu

Paul Furber

unread,
May 7, 1996, 3:00:00 AM5/7/96
to

: I don't know any one coder who's willing to support New Note

: Actions--which, by the way, actually *reduce* the quality of your
: song if you've got a GUS. Otherwise, it's S3M meets XM in terms
: of support.

You're right about the reduction in quality there Jim, but most of the time
the new notes that are still playing (if you set continue=on in your
instrument list - fades and loops are a different story) in the background
are fairly
soft. Of corz if it's a 32-channel song which uses NNAs then the quality
loss can be pretty noticeable (on a GUS that is...)

: New Note Actions will kill that format amongst coders and GUS owners.
: What the heck was he thinking?

I don't know so much about this. Being both a coder and a GUS owner I am
actually so excited about this format that I want to get off my lazy butt
and write a player for the .IT format. Main reason: all the useability
and simplicity of ST32 combined with 16-bit samples on the GUS.
Yes it is a problem that NNAs gobble channels. Eg: Not for kids by Skaven
when converted to an IT takes 28 channels! (Using my interpretation for
the insts of course). This is originally a 4-channel MOD.

I think there are plenty of new features other than NNAs which justify
writing a GUS player for ITs - one of them being I've written some
songs which
allowed me to work with the music and the instruments instead of having
to work around the tracker.
Beta testers anyone? Mail me at the address below. I'll be writing it in
Watcom C and TASM and will be supporting GUS ONLY (unless any kind soul
wants to help by writing an SB driver). It will probably take about three
months given my various personal time restrictions. And yes, I will
definitely release the source........

****************************************************************
* Paul Furber { pa...@is.co.za } *
* News Editor ITWeb SA *
* Mad and Hopeless demo coder
****************************************************************


ri...@ccwf.cc.utexas.edu

unread,
May 7, 1996, 3:00:00 AM5/7/96
to

In article <4mnv9n$g...@Mars.mcs.com>, Trixter / Hornet <tri...@MCS.COM> wrote:
>In article <4m9lmm$s...@piglet.cc.utexas.edu>,
> <ri...@ccwf.cc.utexas.edu> wrote:
>New Note Actions will kill that format amongst coders and GUS owners.

Two words--"Software Mixing." Now that the world has Pentiums and
they can afford to have better mixing than the GUS hardware can provide
(eg--I never listen to DOPE.MOD unless I'm using CP or something and the
Max 48kHz codec).

>What the heck was he thinking?

My Ensoniq allows you to program things to do New Note Actions. Why
can't the demoscene, too? He was thinking, hey, this is a feature that
everybody in the world except for mod-writers gets to use, why not now?
Plus, he was able to code them, wasn't he? So what's everyone else's excuse?

Speaking as a musician, new note actions are the best thing to ever
happen to the MOD scene--we aren't dealing with the Amiga's 4-voice
hardware or 22kHz 8-bit SoundBastards any more. We've got cards and
machines fast enough to do the work to make this new format work. The
only thing that held me back initially was that IT didn't support 16-bit
samples or samples longer than 64k. Well, now it does!

So, if IT's author can program new note actions, why can't the rest of
you? Is he really that much better than all the rest of you?

Mr.P / Powersource

unread,
May 8, 1996, 3:00:00 AM5/8/96
to

Msg.Text("7 May 1996 16:32:05 -0500","ri...@ccwf.cc.utexas.edu");

>In article <4mnv9n$g...@Mars.mcs.com>, Trixter / Hornet <tri...@MCS.COM> wrote:
>>In article <4m9lmm$s...@piglet.cc.utexas.edu>,
>> <ri...@ccwf.cc.utexas.edu> wrote:
>>New Note Actions will kill that format amongst coders and GUS owners.
>Two words--"Software Mixing." Now that the world has Pentiums and
>they can afford to have better mixing than the GUS hardware can provide
>(eg--I never listen to DOPE.MOD unless I'm using CP or something and the
>Max 48kHz codec).

Experience in this newsgroup should tell you that the minute you
mention "Software Mixing", people will WHINE ABOUT THAT 5% OF CPU TIME
THAT THEY HAVE JUST LOST!

Sorry, but I'm bitter. :)

>>What the heck was he thinking?
>My Ensoniq allows you to program things to do New Note Actions. Why
>can't the demoscene, too? He was thinking, hey, this is a feature that
>everybody in the world except for mod-writers gets to use, why not now?
>Plus, he was able to code them, wasn't he? So what's everyone else's excuse?

There are several.

A) Laziness
B) I can't figure it out.
C) S3M and XM work for my group's musicians.
D) I HATE SOFTWARE MIXING! SB DIE DIE DIE YAH YAH HUHUH! ;)
E) I hate this guy.
F) What?
G) Could you repeat that?
H) Stop picking on me!

>Speaking as a musician, new note actions are the best thing to ever
>happen to the MOD scene--we aren't dealing with the Amiga's 4-voice
>hardware or 22kHz 8-bit SoundBastards any more. We've got cards and
>machines fast enough to do the work to make this new format work. The
>only thing that held me back initially was that IT didn't support 16-bit
>samples or samples longer than 64k. Well, now it does!

Again, people are not willing to do software mixing anymore. Well,
they don't want to write their own routines. A pity, since the same
routine is used basically in a mixer as it is in a texture mapping
routine or something.

>So, if IT's author can program new note actions, why can't the rest of
>you? Is he really that much better than all the rest of you?

I dunno about these things, but if Trixter says something, I'll
believe it. :)

- destroyed at request & boredom -

Sam

unread,
May 8, 1996, 3:00:00 AM5/8/96
to

On Wed, 8 May 1996, Mr.P / Powersource wrote:

> Experience in this newsgroup should tell you that the minute you
> mention "Software Mixing", people will WHINE ABOUT THAT 5% OF CPU TIME
> THAT THEY HAVE JUST LOST!

32 channels at 44 kHz, properly interpolated, takes more than 5% cpu,
doesn't it?

Don't forget GUS PnP lets you do that (more than 14 chans). *And* the
docs for that are already available...

Sam (who thinks XM is a good format and we should stick with the damn thing.)

>> homepage * http://www.dur.ac.uk/~d405ua/ * fiction, art, links and...
>> s0ftware f0rge programs * entertainment : MIDI utilities * Win 3.1/95
>> Latest software * SysEx Solution 95 * data store / librarian / editor


Trixter / Hornet

unread,
May 8, 1996, 3:00:00 AM5/8/96
to

In article <4mofgl$d...@piglet.cc.utexas.edu>,

<ri...@ccwf.cc.utexas.edu> wrote:
>In article <4mnv9n$g...@Mars.mcs.com>, Trixter / Hornet <tri...@MCS.COM> wrote:
>>In article <4m9lmm$s...@piglet.cc.utexas.edu>,
>> <ri...@ccwf.cc.utexas.edu> wrote:
>>New Note Actions will kill that format amongst coders and GUS owners.
>
>Two words--"Software Mixing." Now that the world has Pentiums and
>they can afford to have better mixing than the GUS hardware can provide
>(eg--I never listen to DOPE.MOD unless I'm using CP or something and the
>Max 48kHz codec).

One word: "Wasteful". New Note Action songs could easily hit 64
or more channels if you're NewNoting the chords--which Pentium Pro
are you going to use to mix that one to get the full quality you
desire? Got a couple lying around?

>My Ensoniq allows you to program things to do New Note Actions. Why

Of course it can, it's supposed to--it's a MIDI device. It's
hardware. Hardware is *supposed* to do this. But it convienently
drops notes when it hits its 24 or 32 channel limit, so it can
afford to.

>Plus, he was able to code them, wasn't he? So what's everyone else's excuse?

It was his vision--let *him* support them. Even better, he can
release the player code. I don't want my demos to require a
Pentium. (*Designing* for a Pentium is a different story...)

>Speaking as a musician, new note actions are the best thing to ever
>happen to the MOD scene--we aren't dealing with the Amiga's 4-voice

.
.


.
>So, if IT's author can program new note actions, why can't the rest of
>you? Is he really that much better than all the rest of you?

Of course not. The point is that you're on your soapbox in the
wrong forum--your sentiments would be received much better in a
.mod group or something. Democoders pride themselves on doing more
with less; IT NNAs go completely against that. For a musician,
I'm sure that IT is fantastic, but I'm sure most democoders aren't
interested.

I hope you won't hold my opinions against me. Well, okay, you
can--just don't ambuse me with a baseball bat at NAID. :-)

Trixter / Hornet

unread,
May 8, 1996, 3:00:00 AM5/8/96
to

In article <319018b1...@news.innovplace.saskatoon.sk.ca>,

Mr.P / Powersource <m...@spartan.pei.edu> wrote:
>
>Experience in this newsgroup should tell you that the minute you
>mention "Software Mixing", people will WHINE ABOUT THAT 5% OF CPU TIME
>THAT THEY HAVE JUST LOST!

We're not talking about 5% here. How much CPU time do you think it
would take to mix 64 channels?

>Again, people are not willing to do software mixing anymore. Well,
>they don't want to write their own routines. A pity, since the same
>routine is used basically in a mixer as it is in a texture mapping
>routine or something.

The mixing itself is easy. It's the DMA/hardware/IRQ programming,
interpreting different formats, etc. that is annoying.

Mr.P / Powersource

unread,
May 8, 1996, 3:00:00 AM5/8/96
to

Msg.Text("7 May 1996 18:13:19 GMT","as...@engws12.ic.sunysb.edu
(Andrew G Sega)");

>trixter (tri...@mcs.com) wrote:
>: New Note Actions will kill that format amongst coders and GUS owners.
>: What the heck was he thinking?
>well the thing is, if you are smart and don't use the NNA silliness, it ends up
>being THE perfect tracker for anyone who is used to st3. if you like the
>ft2 gui, then stick with that, but for us who need the speed and macros,
>use impulse. and come on, like you couldn't just code an .it player which
>ignored NNA triggers :>

Yeah, but then you'd say it was shit. :)

Mr.P / Powersource

unread,
May 8, 1996, 3:00:00 AM5/8/96
to

Msg.Text("Wed, 8 May 1996 10:09:59 +0100","Sam
<Samuel....@durham.ac.uk>");

>On Wed, 8 May 1996, Mr.P / Powersource wrote:
>> Experience in this newsgroup should tell you that the minute you
>> mention "Software Mixing", people will WHINE ABOUT THAT 5% OF CPU TIME
>> THAT THEY HAVE JUST LOST!
>32 channels at 44 kHz, properly interpolated, takes more than 5% cpu,
>doesn't it?

You don't use all 32 channels at the same time, do you?

Timed, my linear interpolation integer only mixer will eat about 20%
CPU time on this P133. This is for 32 channels at 44kHz. [Sample
resolution doesn't count for much.]
If I moved it to use the FPU, it could be reduced to maybe 5%-10%.

>Don't forget GUS PnP lets you do that (more than 14 chans). *And* the
>docs for that are already available...

You're just bitter about me attacking the GUS PnP, right? :)

I'm not paying $300 for the damn thing.

Joshua C Shepard

unread,
May 8, 1996, 3:00:00 AM5/8/96
to

On 8 May 1996, Trixter / Hornet wrote:

rimbo wrote:
> >Speaking as a musician, new note actions are the best thing to ever
> >happen to the MOD scene--we aren't dealing with the Amiga's 4-voice
> .
> .
> .
> >So, if IT's author can program new note actions, why can't the rest of
> >you? Is he really that much better than all the rest of you?
>
> Of course not. The point is that you're on your soapbox in the
> wrong forum--your sentiments would be received much better in a
> .mod group or something. Democoders pride themselves on doing more
> with less; IT NNAs go completely against that. For a musician,
> I'm sure that IT is fantastic, but I'm sure most democoders aren't
> interested.

Hmm. Last time I checked this newsgroup was comp.sys.ibm.pc.demos... I
don't see how that makes it a demo-coders newsgroup and not a
demo-musicians as well... :)

Personally, as a coder and a musician, I don't use the NNAs in my music. I
also don't see why they should kill the format. Some ppl might choose to
support them while others may not. Big deal, IT is still a great tracker
for those of us who are used to ST3 (or who hate mice).


- josh / winghead

__________________________________________________________________________

Joshua C. Shepard -- wing...@cyberspace.com -- apex /\ danaan --
==========================================================================
home page: www.cyberspace.com/winghead (perpetually under construction)
__________________________________________________________________________

Andrew G Sega

unread,
May 9, 1996, 3:00:00 AM5/9/96
to

first of all
IT is fantastic that is very true
but please don't assume that NNAs are godsends to musicians

if i know my song is going to be played on a GUS, or even
something with an 8-bit output, it's going to sound horrible with NNA's.

if the song is being played on something perhaps similar to an sb16 which
is being software mixed, you still have the problem of creating huge
polyphonic songs which can use over a hundred channels, and the problem with
that is, of course, that your tracking is a) very ineffcient, since
most of the channels can't even be heard amidst the clangor, or b) your
song takes us 85% rastertime on a P6, which is just plain disgusting

perhaps when the day comes where we can have 128 channels and FFT reverb
and dynamic oscillators simulated through software, then we can use NNA's,
but for now, i don't really see the point. having control over your music is
one of the things which makes tracking so different from MIDI
implementations.

- necros / five musicians
nec...@fm.org

ri...@ccwf.cc.utexas.edu

unread,
May 9, 1996, 3:00:00 AM5/9/96
to

In article <4mrgks$m...@abel.cc.sunysb.edu>,

Andrew G Sega <as...@engws12.ic.sunysb.edu> wrote:
>first of all
>IT is fantastic that is very true
>but please don't assume that NNAs are godsends to musicians

Godsends? No. Very nice, yes. No, the REAL godsend is that I can now
have both instrument AND channel panning.

>if i know my song is going to be played on a GUS, or even
>something with an 8-bit output, it's going to sound horrible with NNA's.

Not true...the demo song with IT2.01 sounded beautiful on my GUS.

...

>- necros / five musicians

See the bit about voice-stealing. It DOESN'T HAVE TO BE AS HARD AS
EVERYONE'S MAKING It...

And what's this bit about having lots of control? NNAs by no means make
one have to give up any control...

Sam

unread,
May 9, 1996, 3:00:00 AM5/9/96
to

On 8 May 1996 ri...@ccwf.cc.utexas.edu wrote:

> Yes. It takes about 7%.
>
> big FUCKING deal

okay, okay... *grin*. I just got told by some guy in some other newsgroup
that it wasn't possible to emulate a gus on a p90 (and still sound as
good), but I think he was talking out of his backside. (Tommy someone,
author of a program that converts midi -> wav, non-realtime)

> >Don't forget GUS PnP lets you do that (more than 14 chans). *And* the
> >docs for that are already available...
>

> So, what's the holdup?

I think Cubic already supports 32 channels on gus PnP. I'm sure I
heard one of the players does, but perhaps whoever told me that was wrong.

> >Sam (who thinks XM is a good format and we should stick with the damn thing.)
>

> It's good. But it's not great. (Ever try to do channel panning in FT2?
> IT lets you CHOOSE whether you want instrument or channel panning,
> depending on what you're doing...)

Yep, sorry, didn't quite say what I meant there - I think starting with
the xm format and improving on that would be better than yet another new
one; after xm is only just getting reasonably amount of players, so..

Just that it's easier to add features to an existing format than to start
a new one.

Sam

Mr.P / Powersource

unread,
May 9, 1996, 3:00:00 AM5/9/96
to

Msg.Text("8 May 1996 13:39:16 -0500","tri...@MCS.COM (Trixter /
Hornet)");
>In article <319018b1...@news.innovplace.saskatoon.sk.ca>,

>Mr.P / Powersource <m...@spartan.pei.edu> wrote:
>>Experience in this newsgroup should tell you that the minute you
>>mention "Software Mixing", people will WHINE ABOUT THAT 5% OF CPU TIME
>>THAT THEY HAVE JUST LOST!
>We're not talking about 5% here. How much CPU time do you think it
>would take to mix 64 channels?

Are you going to use 64 channels?

Come on. The only 32 channel song I ever heard was written by one of
my friends, and he used 23 channels for an ECHO.

>>Again, people are not willing to do software mixing anymore. Well,
>>they don't want to write their own routines. A pity, since the same
>>routine is used basically in a mixer as it is in a texture mapping
>>routine or something.
>The mixing itself is easy. It's the DMA/hardware/IRQ programming,
>interpreting different formats, etc. that is annoying.

Interpreting different formats: tough it, you gotta read them even for
the GUS.
DMA/IRQ: GUS timers/downloading/uploading/whateverloading/freeloading?
Hardware: OK, so maybe this bit is a little hard to expound on, but
the basic output command for the SB 16 is to send two bytes to the
DSP. Information on that can be easily found, or if you want, I can
post my entire SB interface unit.

Mr.P / Powersource

unread,
May 9, 1996, 3:00:00 AM5/9/96
to

Msg.Text("8 May 1996 18:53:05 -0500","ri...@ccwf.cc.utexas.edu");


>In article <4mqpok$2...@Mars.mcs.com>, Trixter / Hornet <tri...@MCS.COM> wrote:
>>The mixing itself is easy. It's the DMA/hardware/IRQ programming,
>>interpreting different formats, etc. that is annoying.

>*slap* @#$&@*$(*!!!!! ****ALERT: THE FOLLOWING TEXT HAS BEEN AUTODELETED FOR THE C.D.A.
>****
>****
>****
>What, do I have to go out and do this myself??!?!?!?
>KICKING and FUCKING SCREAMING will the coders of the demoscene be dragged
>into the 21st century...
>Virtualizing channels is NOT difficult, folks. MIDI programmers do it
>every day. And if you're having real difficulties interpreting multiple
>formats, go grab MikMod's sources and look at how MikMak did it. (This is
>the right way.)

How did he do it? Multiple players for multiple loaders? That is the
most correct way IMO. Conversion between formats gets to be a pain in
the ass.

Oh, and the CDA can censor that too.

Scott Stuart Buchanan

unread,
May 9, 1996, 3:00:00 AM5/9/96
to

Andrew G Sega (as...@engws12.ic.sunysb.edu) wrote:
:

song takes us 85% rastertime on a P6, which is just plain disgusting

I know an example of this. A friend of mine has allready written a song that
(without difficulty) can't be played on many p5-100's and below, using SB16's
The GUS has absolutely no hope at even pretending to play the song.
The reason? This .IT, at times, gets up to around 128 channels! Being a
486dx2-66 user, with a 512k GUS, I'll NEVER have any hope of playing this
particular song on my system, and neither will a LOT of other people.

So, it's allready being done, and no doubt it will continue, and, I'm sure
the music is wonderful, BUT, how the fuck, is a demo-coder meant to write
a demo, using a player, that takes nearly 100% CPU time on a p5-133, AND have
any effects (other than maybe a text-mode scrolly) working at any decent
speed?

-Goblin/Xtatic!-


Trixter / Hornet

unread,
May 9, 1996, 3:00:00 AM5/9/96
to

In article <31923844...@news.innovplace.saskatoon.sk.ca>,

Mr.P / Powersource <m...@spartan.pei.edu> wrote:
>>We're not talking about 5% here. How much CPU time do you think it
>>would take to mix 64 channels?
>
>Are you going to use 64 channels?
>
>Come on. The only 32 channel song I ever heard was written by one of
>my friends, and he used 23 channels for an ECHO.

I don't think you understand: If you use New Note Actions a lot,
new notes will be played while old notes haven't ended yet. Even
the sample "8" or "10" channel song (I can't remember how many it
is right now) goes up to 23 or 24 channels using NNAs. Imagine
what a 16-channel song will do!

>DSP. Information on that can be easily found, or if you want, I can
>post my entire SB interface unit.

I'd rather you write an .IT player and prove me wrong.

yrm...@sara.cc.utu.fi

unread,
May 9, 1996, 3:00:00 AM5/9/96
to

In article <Dr3tE...@rci.ripco.com>, domi...@ripco.com (Michael Chisari) writes:

> Trixter / Hornet (tri...@MCS.COM) wrote:
> : New Note Actions will kill that format amongst coders and GUS owners.
> : What the heck was he thinking?
>
> very simple... he developed impulse tracker with sb16 users in mind.
^^^^
So what things can SB16 do that a standard GUS can't?

- play OPLwhatever FM-crap
- record 16bit samples

And which of these does IT use???

IOW: GUS can (oh yes!) play 16-bit samples, and
it is not necessary to use its ability to mix
channels. Just play two samples, left&right.
-
Let's add some steam to this discussion:

What are you arguing about somebody making a
new complex music file format? If you are
a coder, no panic... Unless you are a
developer of a common standalone mod player,
there is absolutely no need to code a
player for this impulsive format.

And if you are a musician, why would you
demand anyone to code a player for this
thing? So you are making music for a demo,
and there is no other way of making music
than IT with NNA? :)
If I asked someone (else than me) to make
music for my demo, I would require it to be
a 6 channel MOD with bpm rate 175, take it
or leave it... This is not disneyland. ;)

--
yzi/fit, yrm...@sara.cc.utu.fi

Niklas Beisert

unread,
May 9, 1996, 3:00:00 AM5/9/96
to

Hi Andrew,

>: New Note Actions will kill that format amongst coders and GUS owners.
>: What the heck was he thinking?
>

>well the thing is, if you are smart and don't use the NNA silliness, it ends
>up being THE perfect tracker for anyone who is used to st3. if you like the
>ft2 gui, then stick with that, but for us who need the speed and macros,
>use impulse. and come on, like you couldn't just code an .it player which
>ignored NNA triggers :>

then why don't you save it in XM format? everyone will be happy.
or save it in .AMS... or .MDL... those are all nice formats, and they
surely have all the effects .IT has, except for some tracker coder's
stupid ideas.

Is there anyone that would like to code an .IT loader for cp???

oh... and then the 64 channels... HAHAHAHAHAHAHHAHAHAHAHHAHAH!!


cu
Niklas / pascal / cubic team

-> pas...@nightmare.harz.de, Niklas Beisert@2:2437/301.44

Quantum Porcupine

unread,
May 9, 1996, 3:00:00 AM5/9/96
to

tri...@MCS.COM (Trixter / Hornet) wrote:

>In article <4mofgl$d...@piglet.cc.utexas.edu>,
> <ri...@ccwf.cc.utexas.edu> wrote:
>>In article <4mnv9n$g...@Mars.mcs.com>, Trixter / Hornet <tri...@MCS.COM> wrote:
>>>In article <4m9lmm$s...@piglet.cc.utexas.edu>,
>>> <ri...@ccwf.cc.utexas.edu> wrote:

>>>New Note Actions will kill that format amongst coders and GUS owners.
>>

>>Two words--"Software Mixing." Now that the world has Pentiums and
>>they can afford to have better mixing than the GUS hardware can provide
>>(eg--I never listen to DOPE.MOD unless I'm using CP or something and the
>>Max 48kHz codec).

>One word: "Wasteful". New Note Action songs could easily hit 64
>or more channels if you're NewNoting the chords--which Pentium Pro
>are you going to use to mix that one to get the full quality you
>desire? Got a couple lying around?

actually... pulse told me himself that he only has a DX4/100 and he
can get 112 channels.

also, if you read ittech.doc it tells you the channel virtualization
scheme. it drops the quietest channels first.

>>My Ensoniq allows you to program things to do New Note Actions. Why

>Of course it can, it's supposed to--it's a MIDI device. It's
>hardware. Hardware is *supposed* to do this. But it convienently
>drops notes when it hits its 24 or 32 channel limit, so it can
>afford to.

again, read ittech.doc.

>>Plus, he was able to code them, wasn't he? So what's everyone else's excuse?

>It was his vision--let *him* support them. Even better, he can
>release the player code. I don't want my demos to require a
>Pentium. (*Designing* for a Pentium is a different story...)

agian, he only has a DX4/100 and he can get 112 (or was it 119)
channels...

>>So, if IT's author can program new note actions, why can't the rest of
>>you? Is he really that much better than all the rest of you?

>Of course not. The point is that you're on your soapbox in the
>wrong forum--your sentiments would be received much better in a
>.mod group or something. Democoders pride themselves on doing more
>with less; IT NNAs go completely against that. For a musician,
>I'm sure that IT is fantastic, but I'm sure most democoders aren't
>interested.

as both a coder and a musician, i must say that i love NNAs. :)

>I hope you won't hold my opinions against me. Well, okay, you
>can--just don't ambuse me with a baseball bat at NAID. :-)

lucky for you i'm gonna be in france during NAID :)


---
quantum porcupine, coder, musician | that which is, is not
and porcupine. mailto:jsh...@nmsu.edu | that which can, can not
http://infinity.beve.blacksburg.va.us/~porcpine | that which does, does not


Mr.P / Powersource

unread,
May 10, 1996, 3:00:00 AM5/10/96
to

Msg.Text("9 May 1996 21:24:39 GMT","s219...@cse.unsw.edu.au (Scott
Stuart Buchanan)");

>Andrew G Sega (as...@engws12.ic.sunysb.edu) wrote:
> song takes us 85% rastertime on a P6, which is just plain disgusting
>I know an example of this. A friend of mine has allready written a song that
>(without difficulty) can't be played on many p5-100's and below, using SB16's
>The GUS has absolutely no hope at even pretending to play the song.
>The reason? This .IT, at times, gets up to around 128 channels! Being a
>486dx2-66 user, with a 512k GUS, I'll NEVER have any hope of playing this
>particular song on my system, and neither will a LOT of other people.

4xGUS?

Hey, that's what somebody would suggest.

>So, it's allready being done, and no doubt it will continue, and, I'm sure
>the music is wonderful, BUT, how the fuck, is a demo-coder meant to write
>a demo, using a player, that takes nearly 100% CPU time on a p5-133, AND have
>any effects (other than maybe a text-mode scrolly) working at any decent
>speed?

I dunno. There's the problem. Hey, can I have the module? I've a
P5-133. :)

ri...@ccwf.cc.utexas.edu

unread,
May 10, 1996, 3:00:00 AM5/10/96
to

In article <319238c9...@news.innovplace.saskatoon.sk.ca>,

Mr.P / Powersource <m...@spartan.pei.edu> wrote:
>How did he do it? Multiple players for multiple loaders? That is the
>most correct way IMO. Conversion between formats gets to be a pain in
>the ass.

No, he has his own format, which can do anything any other format does.
When the loaders load in a song, they convert it to the format. The
format uses different effects for things which need to be implemented
differently--i.e., I forget the specifics, but if S3Ms need effect X
implemented differently than Protracker MODs implement effect X, then
there are two effects---X_s3m and X_pt_mod. Something like that.

I'll be getting much more familiar with MikMod this Summer, let me tell
you...

ri...@ccwf.cc.utexas.edu

unread,
May 10, 1996, 3:00:00 AM5/10/96
to

In article <4mtnqn$6...@mirv.unsw.edu.au>,

Scott Stuart Buchanan <s219...@cse.unsw.edu.au> wrote:
>Andrew G Sega (as...@engws12.ic.sunysb.edu) wrote:
> song takes us 85% rastertime on a P6, which is just plain disgusting
>I know an example of this. A friend of mine has allready written a song that
...

>The reason? This .IT, at times, gets up to around 128 channels! Being a

That has nothing to do with NNAs...that's just bad songwriting on his
part. Anyone can write a song that uses over 128 channels on an IT. But
it's baaaaaaaaaad.

>So, it's allready being done, and no doubt it will continue, and, I'm sure
>the music is wonderful, BUT, how the fuck, is a demo-coder meant to write
>a demo, using a player, that takes nearly 100% CPU time on a p5-133, AND have

You don't USE a player that takes up 100% CPU time. You use a player
that sets a limit on how many Virtual Channels are available. Sure,
you'll have voice-stealing, but the point of NNAs is not lost that way.

NNAs are not effects in the same way that A08 is an effect; they're
effects in the same way reverb is. Once a note ends, it's OVER...NNAs
are just a different way of treating it. So what if a note that's
supposed to fade out isn't quite 100% finished when its channel is
needed? Nobody will notice, and nobody will care.

Scott Stuart Buchanan

unread,
May 10, 1996, 3:00:00 AM5/10/96
to

Mr.P / Powersource (m...@spartan.pei.edu) wrote:
: > song takes us 85% rastertime on a P6, which is just plain disgusting
: >I know an example of this. A friend of mine has allready written a song that
: >(without difficulty) can't be played on many p5-100's and below, using SB16's

: >The GUS has absolutely no hope at even pretending to play the song.
: >The reason? This .IT, at times, gets up to around 128 channels! Being a
: >486dx2-66 user, with a 512k GUS, I'll NEVER have any hope of playing this

: >particular song on my system, and neither will a LOT of other people.

: 4xGUS?

: Hey, that's what somebody would suggest.

Right, and you can pay for it :P~
Someone mentioned the song included with Impulse tracker... Here's
what I think :) With the song included with Impulse Tracker v1.06,
Jeffrey Lim SAYS that quite a powerful computer is needed to play the song
(a 16 channel .IT), because it can get up to 44 channels, so unless you
have a GUS, you'll need some good fast system. IF you have a GUS, you
miss out on those other 12 channels. I haven't yet got ITv2.01, but I
can't imagine that with these NNA things, its going to make things any
less processor intensive, so, to listen to these songs, we're going to
need either some really FAST system, OR a GUS, and miss out on half the
song. Now, since this is comp.sys.ibm.pc.demos, I assume we're talking
about demos, and demo-music, i.e. music to be used in demos. Now, if one
needs a super-fast system JUST to play .IT music, then, demo's supporting
the new .IT format will simply not work on lower-end machines, and on
higher end machines, there won't be enough CPU power left over to do
anything on the screen. OR, if one uses a GUS, well, you could just miss
out on half the song, which kinda defeats the purpose of the extra channels.

erk, after all that babbling, I'll try and conclude wtf I'm trying to say.

- The new .IT format may be a godsend for musicians, but, it is obviously
impractical for demos, and this IS a newsgroup regarding DEMOS.

: >So, it's allready being done, and no doubt it will continue, and, I'm sure


: >the music is wonderful, BUT, how the fuck, is a demo-coder meant to write
: >a demo, using a player, that takes nearly 100% CPU time on a p5-133, AND have

: >any effects (other than maybe a text-mode scrolly) working at any decent
: >speed?

: I dunno. There's the problem. Hey, can I have the module? I've a
: P5-133. :)

Actually, I don't have it (even if I did have it, I couldn't play it, so
there's not much point me having it), so, no :)

cya l8r
-Goblin/Xtatic-


ri...@ccwf.cc.utexas.edu

unread,
May 10, 1996, 3:00:00 AM5/10/96
to

In article <4mv29s$d...@mirv.unsw.edu.au>,

Scott Stuart Buchanan <s219...@cse.unsw.edu.au> wrote:
>have a GUS, you'll need some good fast system. IF you have a GUS, you
>miss out on those other 12 channels. I haven't yet got ITv2.01, but I

This is the fallacy! Right here! You don't "Miss Out" on anything (I'm
assuming the tracker knows what he was doing, right?). A couple of
voices get stolen (12 out of 44 is NOTHING, believe me), but go listen to
the song on both soundcards. Make a recording if you want. Then tell me
if you hear THAT much difference (you WON'T).

Niklas Beisert

unread,
May 10, 1996, 3:00:00 AM5/10/96
to

Hi Andrew,

asega # engws12.ic.sunysb.edu@2:2437/120.999 wrote:

>perhaps when the day comes where we can have 128 channels and FFT reverb
>and dynamic oscillators simulated through software, then we can use NNA's,
>but for now, i don't really see the point. having control over your music is
>one of the things which makes tracking so different from MIDI
>implementations.

yepp... either use MIDI or don't use NNAs.

Who needs a NNA? if you hit a new note, why don't you then use a new
channel? you have 64 of them... (...)

how to produce a chord on a channel? hitting 3 notes at the same time?
this does not work.
what happens to old notes? will effects affect them?
then the loss of a logical<->physical channel correspondence.

i simply cannot put this into my player unless i recode it.
and if i ignore the NNAs, several guys will complain.

Scott Stuart Buchanan

unread,
May 11, 1996, 3:00:00 AM5/11/96
to

ri...@ccwf.cc.utexas.edu wrote:
: In article <4mtnqn$6...@mirv.unsw.edu.au>,

: >So, it's allready being done, and no doubt it will continue, and, I'm sure
: >the music is wonderful, BUT, how the fuck, is a demo-coder meant to write
: >a demo, using a player, that takes nearly 100% CPU time on a p5-133, AND have

: You don't USE a player that takes up 100% CPU time. You use a player

: that sets a limit on how many Virtual Channels are available. Sure,
: you'll have voice-stealing, but the point of NNAs is not lost that way.

Ok, if you limit the number of Virtual Channels, you CAN loose "vital"
parts of the music. I'm pretty sure that with a certain MIDI file I have,
when I play it on my GUS, it, at times, exceeds the 32 channel GUS limit,
and I lose (noticeably) bits of the music, and instruments cut-out that
shouldn't. So, tell me, why isn't this going to happen with .IT's as well?

: NNAs are not effects in the same way that A08 is an effect; they're

: effects in the same way reverb is. Once a note ends, it's OVER...NNAs
: are just a different way of treating it. So what if a note that's
: supposed to fade out isn't quite 100% finished when its channel is
: needed? Nobody will notice, and nobody will care.

If nobody will notice, and nobody will care, then why use NNA's at all?
If you intend to cut-off the notes before they are "100% finished", then
why use NNA's?

cya l8r
-Goblin/Xtatic!-
s219...@cse.unsw.EDU.AU


Scott Stuart Buchanan

unread,
May 11, 1996, 3:00:00 AM5/11/96
to

ri...@ccwf.cc.utexas.edu wrote:
: In article <4mv29s$d...@mirv.unsw.edu.au>,
: >miss out on those other 12 channels. I haven't yet got ITv2.01, but I

: This is the fallacy! Right here! You don't "Miss Out" on anything (I'm
: assuming the tracker knows what he was doing, right?). A couple of
: voices get stolen (12 out of 44 is NOTHING, believe me), but go listen to
: the song on both soundcards. Make a recording if you want. Then tell me
: if you hear THAT much difference (you WON'T).

Unfortunately I don't have an SB16 to compare with, I'd have to go to my
friends house, listen, then come back home, and by that time, I think I'd
have forgotten what it sounded like on the SB16.

Ok. Lets say, you compose a 16 channel song, with every channel using
"surround" panning. Unless I'm mistaken, that allready takes 32 channels.
Say, now, you have NNA's going on all of those channels, its 64 channels.
So, either all the NNA's are wasted on a GUS, OR you miss some of the
surround channels, which mean the song plays on the left hand side only,
etc... either way, it seems very easy to make a song that will play
noticeably wrongly on a GUS.

Take a 24 channel song, and, well, don't expect it to play properly on a
GUS. I wouldn't expect to play on an SB16, unless you had about a p166.

Don't get me wrong, I'm sure IT is great, I know that my groups musician
loves it, and lots of other musicians love it, and, most importantly its
an Australian product, and everyone knows Australian stuff is best ;)
BUT, as a demo music composer, I don't think it is god's gift.

cya l8r
-Goblin/Xtatic!-
s219...@cse.unsw.EDU.AU


Alex Statix

unread,
May 11, 1996, 3:00:00 AM5/11/96
to

To add to this little debate... (flames/comments welcome <grin>)

what's the problem with having NNA's implemented in a tracker/player? You
can for example set them all to cut, except maybe for one or two samples
where you want it to ring on a bit... then your actual channel count is never
going to rise more than 1 or 2 above the "normal" level. It's convenient
for musicians (in the same way that volume envelopes are)... they could
do it all (pretty much) with commands and loads of channels, but why
bother when you're doing such a simple task?

as to quality... if you set a 32 channel limit (eg GUS, or just proc.
time for software mixing) you can do as the hardware does, (and IT does)
and drop the quiet notes. That's what professional synths do...

as to coding... as trixter said (? I think), implementing NNAs and virtual
channels is not hard at all... I just put it into my player routines, which
now have (non bugged) AWE32 support, gus support, blah blah blah. And it
wasn't difficult... hats off to pulse blokey for implementing them in
the first place.

another variation on standard trackers is to have MIDI style overlapping
patterns/variable number of channel patterns and so on. Psychic Link have
had this for two years, and it really can help out for some sorts of file
(particularly techno). (take a look at dtrk200a.zip). If anyone would be
interested in beta-testing version 3, this summer, then email me and I'll
stick your name on a list, and post it out when (if) I get it all working.
It does all IT/XM style NNAs,instruments etc, dtrk style patterns, and runs
in a FT style interface (Ie, not another st3 clone. Not that I want to
alienate people who've grown up on st3, its a fast interface, but then
so is any you know well. And I don't know it at all)

arg! how long have I been babbling? too much... cut.
---------------------------------------------------------------------------
Statix (aka Alex Evans) Psychic Link - UK demo team
email: sta...@sv.span.com WWW: http://crystal.clare.cam.ac.uk/~sah32

ri...@ccwf.cc.utexas.edu

unread,
May 11, 1996, 3:00:00 AM5/11/96
to

In article <4n1vm2$8...@mirv.unsw.edu.au>,

Scott Stuart Buchanan <s219...@cse.unsw.edu.au> wrote:
>ri...@ccwf.cc.utexas.edu wrote:
>: This is the fallacy! Right here! You don't "Miss Out" on anything (I'm
>: assuming the tracker knows what he was doing, right?). A couple of
>Ok. Lets say, you compose a 16 channel song, with every channel using
>"surround" panning. Unless I'm mistaken, that allready takes 32 channels.
>Say, now, you have NNA's going on all of those channels, its 64 channels.

duh duh duh...LISTEN to the ABSURDITY of the case you're coming up with
here! WHAT KIND OF AN IDIOT TRACKER SENDS 16 OF 16 CHANNELS TO THE
SURROUND CHANNEL? I know a lot of trackers, but I don't know any stupid
enough to try this. You're just coming up with an example for the sake
of countering my argument; you're not coming up with a realistic example.

Certainly, any tracker who feels malicious can come up with some stupid
idea like the one you just gave to spite people...but most trackers I
know, if they're going to put effort into something, they'll put effort
into making good samples, good riffs, good chord progressions, good
rhythm sections, good melodies, good use of effects--NOT in using as many
channels as humanly possible.

Okay, not including PeriSoft, that's true. ;) (Just kidding, Peri.)

Trixter / Hornet

unread,
May 11, 1996, 3:00:00 AM5/11/96
to

In article <3190fe7...@news.innovplace.saskatoon.sk.ca>,

Mr.P / Powersource <m...@spartan.pei.edu> wrote:
>Msg.Text("Wed, 8 May 1996 10:09:59 +0100","Sam
><Samuel....@durham.ac.uk>");
>>On Wed, 8 May 1996, Mr.P / Powersource wrote:
>>> Experience in this newsgroup should tell you that the minute you
>>> mention "Software Mixing", people will WHINE ABOUT THAT 5% OF CPU TIME
>>> THAT THEY HAVE JUST LOST!
>>32 channels at 44 kHz, properly interpolated, takes more than 5% cpu,
>>doesn't it?
>
>You don't use all 32 channels at the same time, do you?

Argh... No, the composer doesn't use all 32 channels at the same
time, but NNA's do.

Do you understand what New Note Actions do?

ri...@ccwf.cc.utexas.edu

unread,
May 11, 1996, 3:00:00 AM5/11/96
to

A note on CPU processor time...

DOPE.MOD, a 28-channel module, using 44.1kHz interpolated ("iDO" in
Mod4Win) uses ...anywhere from 6-28%. Right now, it says 16%, but as
I've written this it has varied widely.

Whoa...waitaminit...it just went up to 35%.

Still...that's why we use GUS, right?

:)

Mr.P / Powersource

unread,
May 11, 1996, 3:00:00 AM5/11/96
to

Msg.Text("Thu, 09 May 96 10:32:00
+0200","niklas....@p44.nightmare.harz.de (Niklas Beisert)");

>Hi Andrew,
>>: New Note Actions will kill that format amongst coders and GUS owners.
>>: What the heck was he thinking?
>>well the thing is, if you are smart and don't use the NNA silliness, it ends
>>up being THE perfect tracker for anyone who is used to st3. if you like the
>>ft2 gui, then stick with that, but for us who need the speed and macros,
>>use impulse. and come on, like you couldn't just code an .it player which
>>ignored NNA triggers :>
>then why don't you save it in XM format? everyone will be happy.
>or save it in .AMS... or .MDL... those are all nice formats, and they
>surely have all the effects .IT has, except for some tracker coder's
>stupid ideas.

You're bitter, right?

>Is there anyone that would like to code an .IT loader for cp???
>oh... and then the 64 channels... HAHAHAHAHAHAHHAHAHAHAHHAHAH!!

Yup, you're bitter.

Right. We used to do things not because they were easy, but because
they were hard. Now we have other people filling up the gaps and not
willing to take the challenge.

We had GUS and 32 channels and now people don't want to move forward
because of the fact that IT is going to force them to abandon them.
So, instead of merely ignoring it, we f