Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

AWEUTIL and NMI

205 views
Skip to first unread message

Jason G Fourier

unread,
Sep 25, 1994, 1:01:11 AM9/25/94
to
Excerpts from mail: 24-Sep-94 Re: AWEUTIL and NMI by Chris Baug...@mv.MV.COM
> I own an Awe32. If you want GM under dos, get something else,
> the aweutil is 45k plus other drivers and isn't very good. No sweat under
> windows. Fortunately the Awe is great in windows or when supported
> directly.
> If you must have GM or MT-32 under Dos, find a card with a
> hardware MIDI interpreter and save yourself some agony. TIE fighter does
> sound great with the AWE, but I am sure it does with any good Wave Table.
> A roland daughter card with a sb16 would be a pretty good combo...

Thanks for the advice, but I'm still convinced that the AWE32 is the
best sound card for me. True, my primary use for the card will be
playing (and developing) games, but the main feature that I'm looking
for is wavetable RAM. This feature gives the sound card so many
possibilities for games, demos, music composition, and even just plain
listening. Not only does it let you choose from a limitless selection
of samples, but it can enhance games that use sound effects by storing
the sound effects as samples. It has 32-note polyphony and the on-board
mixer can mix the samples in stereo at 44KHz without any reduction in
sound quality (...a great plus for MOD's!). By performing the mixing
on-board, it also relieves the CPU of this cumbersome task, allowing
applications to run at a much higher performance (...a great plus for
games!). Besides, I'm looking forward to making my own samples and
using them in music. Now it's true that most of the games that are
already out there (OK, all of them save a few new ones...) can't take
advantage of this feature since they don't support the AWE32 natively.
However, Creative Labs is a big name in the sound card business, and I'm
rather confident in their ability to promote native support for the
card. It's already been out for just a few months and it's gaining
support faster than it's leading competition, the Gravis Ultrasound. If
you just look at the number of games that support Sound Blaster or even
Sound Blaster Pro, it gives you a good indication of the strong hold
that Creative Labs has on the sound card market. This is another reason
why I'm in favor of the AWE32. With it's built-in support for the SB16,
I can use it with all of the hundreds of games that have already come
out with SB support. While I know that the GUS can do SB emulation, I
know from my roommate's experience that it isn't really adequate. These
are the main factors that make me choose the AWE32 over the GUS (a nice
card nonetheless...) and all of the other cards with wavetable RAM.
I've heard that the samples that come in it's ROM are relatively poor,
but this is of no concern to me, since I can have any samples I want in
the RAM. The upgradability is a big plus, too. Besides, it comes
bundled with some pretty nice software.

The bulky drivers are a minus, but I'm willing to muck around with a
memory manager and sacrifice some other drivers to get it working
properly. I've been forced to deal with the shortcomings of DOS for so
long now, anyway. :-) Now I'm not expecting to get good MT-32 emulation
without sysex. This doesn't matter much, since I already own a Roland
CM-32L (a successor of the MT-32). I plan on using the AWE32's MPU-401
MIDI port as it's adapter. The AWE32 comes in handy for general MIDI
support, though, since the Roland doesn't do GM. I've been dismayed at
the recent number of games released without MT-32 support but with GM
support, so GM emulation will be good to have. I've had a Roland/SB
combo for a few years now, so I know what it gives you. That's why I'm
looking for a sound card with wavetable RAM. So if you own an AWE32, my
advice to you is this: Keep what you've got, it's already everything
you'll need (at least until the next line of sound cards come out....
;-).

-Jason Fourier

david alan taylor

unread,
Sep 25, 1994, 1:34:40 AM9/25/94
to
>you just look at the number of games that support Sound Blaster or even
>Sound Blaster Pro, it gives you a good indication of the strong hold
>that Creative Labs has on the sound card market. This is another reason

Well, I could say the same about Adlib. I still see new games that come out
with support for it. But I think Creative is losing their grip.
Game Co.'s are diving in to support some kind of standardization (ala
HMI or AIL) which should allow all cards to be used without emulations.
Seems to me Creative should be very opposed to any standardization but their
own.

GUS RuleZ!

unread,
Sep 25, 1994, 2:33:18 AM9/25/94
to
In article <wiVEEL200...@andrew.cmu.edu>,

>
>Thanks for the advice, but I'm still convinced that the AWE32 is the
>best sound card for me. True, my primary use for the card will be
>playing (and developing) games, but the main feature that I'm looking
>for is wavetable RAM. This feature gives the sound card so many
>possibilities for games, demos, music composition, and even just plain
>listening. Not only does it let you choose from a limitless selection
>of samples, but it can enhance games that use sound effects by storing
>the sound effects as samples. It has 32-note polyphony and the on-board
>mixer can mix the samples in stereo at 44KHz without any reduction in
>sound quality (...a great plus for MOD's!). By performing the mixing
^^^^^^^^^^^^^^^^^^^^^^^

>on-board, it also relieves the CPU of this cumbersome task, allowing
>applications to run at a much higher performance (...a great plus for
>games!).

Yeah, sounds great, but there isn't any MOD player yet that supports the
AWE-32, and I think there won't be any soon. Bye2 CPU resources. :)

AWE-32 does have great features, but currently GUS is still the best for
MOD stuffs.

--
+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+
| W i t y G a n d a | +:::::::::::::::::::::::::::::::::::::::::::+ |
| +::::::::::::::::+ | GUS: The Most Supported Soundcard on Internet |
+=-G=-R=-A=-V=-I=-S=-+=-U=-L=-T=-R=-A=-S=-O=-U=-N=-D=-+=-R=-U=-L=-E=-Z=-+

Cornel Huth

unread,
Sep 25, 1994, 10:19:22 AM9/25/94
to
dta...@cis.ohio-state.edu (david alan taylor) writes:
> Game Co.'s are diving in to support some kind of standardization (ala
> HMI or AIL) which should allow all cards to be used without emulations.
> Seems to me Creative should be very opposed to any standardization but their
> own.

HMI? A standard? AIL might be used by many games, but hardly most.
HMI? I don't know of a single game that uses its SOS (though I'm
not a gamer, but I've yet to hear of any game (or anything) that used
SOS).

I did see ads placed in trade mags about "SOS supports [this...that...
GUS]" when in fact it supported just a few. This was about, oh, last
October (when I saw these ads).

Since few people require support for anything but what they own, all
that really matters to "Yukon Joe" is that his GUS is supported, or
his SB16, or Roland, etc. As for a standard, that doesn't matter
at all. Game makers (or other audio apps) just need to work with
your soundcard -- doesn't matter with what.

All that game makers care about is that they get something that works,
and works without them having to spend a year on R&D. Heck, let MI
do the work and license it. The game makers, after all, make games,
not soundcard toolkits. And if you notice, most sound in games is
only adequate -- just enough to make the visuals work.

Games typically have basic sound needs and that's about all you can
get in AIL, or HMI. If your needs or wants are basic, you can
stop looking -- there's plenty of 'adequate' stuff available.


--
Internet: corne...@LChance.sat.tx.us or cor...@ephsa.sat.tx.us
BBS/Fax: 40th Floor v32b@1(210)684-8065 M-F:5pm-9am/WE:24hrs [-0500]

dar...@onramp.net

unread,
Sep 25, 1994, 4:14:51 PM9/25/94
to

Correction :-),
The AWE32 is able to play at 32 voices only when using ROM.
When using RAM the AWE32 is able to play at 30 voices.
The two lost voices are used (of all things) to refresh and provide timing
to the AWE32's ram.
Why did Creative Labs opt to use the last two voices for Dram support?
God only knows, since the EMU8000 is completely able to support DRAM without
having to use up voices but with the addition of a little external logic.
I guess CL was trying to keep prices down.
Rumor has it that CL is designing new AWE's as we type.
I.E. An AWE32 with SCSI II support and a PCI version.

Benjamin Y. Lee

unread,
Sep 25, 1994, 5:30:10 PM9/25/94
to
In message <364fjt$h...@news.onramp.net> - dar...@onramp.net writes:
>
>Rumor has it that CL is designing new AWE's as we type.
>I.E. An AWE32 with SCSI II support and a PCI version.


PCI version? Do tell. I hope it use my computers main memory/harddrive via
the pci bus instead of having its own ram and rom because then we'll finally
we'll have a card like the GUS but done right. Wish IBM made a pci version of
their mwave but if creative makes a pci awe32 tomorrow I will be first in
line with cash in hand. 8-)


-Ben


dar...@onramp.net

unread,
Sep 27, 1994, 12:00:25 AM9/27/94
to

> PCI version? Do tell. I hope it use my computers main memory/harddrive via
> the pci bus instead of having its own ram and rom because then we'll finally
> we'll have a card like the GUS but done right. Wish IBM made a pci version of
> their mwave but if creative makes a pci awe32 tomorrow I will be first in
> line with cash in hand. 8-)

I won't be first in line until Creative fixes the AWE32 to be able to support
32 (not 30) voices while using AWE32 ram.
Like I said, the EMU8000 is able to do 32 voices with Ram and I'm going to have
my 32 voices ram before I buy the next AWE's.
I get sick of cheap work-arounds.
I would rather pay more for the AWE32 than have it have El Cheapo Fixo's.

Yossi Oren

unread,
Oct 2, 1994, 7:03:04 AM10/2/94
to
In article <wiVEEL200...@andrew.cmu.edu>

Jason G Fourier <jf...@andrew.cmu.edu> writes:

>Excerpts from mail: 24-Sep-94 Re: AWEUTIL and NMI by Chris Baug...@mv.MV.COM
>> I own an Awe32. If you want GM under dos, get something else,
>> the aweutil is 45k plus other drivers and isn't very good. No sweat under
>> windows. Fortunately the Awe is great in windows or when supported
>> directly.
>> If you must have GM or MT-32 under Dos, find a card with a
>> hardware MIDI interpreter and save yourself some agony. TIE fighter does
>> sound great with the AWE, but I am sure it does with any good Wave Table.
>> A roland daughter card with a sb16 would be a pretty good combo...
>
>Thanks for the advice, but I'm still convinced that the AWE32 is the
>best sound card for me. True, my primary use for the card will be
>playing (and developing) games, but the main feature that I'm looking
>for is wavetable RAM. This feature gives the sound card so many
>possibilities for games, demos, music composition, and even just plain
>listening.

I think you've been spoiled by the GUSRAM implementation of the GF1. The sort
of things you can do with the AWERAM are much more limited - for example, you
can not set up an IRQ trigger at a certain point in RAM. This is vital for
effect accelleration and also for fitting more than the amount of RAM you have
on-board (hot-swapping, I think it's called). Access to the RAM is limited
by the programmer since you need to buy an SDK (I'll admit the freeware efforts
are nice, but they're not it).

>Not only does it let you choose from a limitless selection
>of samples,

Does not. You have to find .SBK samples, and there's only one program which
can create them. We GUSsers use 2PAT and get our patches fresh off a Kurzweil
or an EPS.

>but it can enhance games that use sound effects by storing
>the sound effects as samples.

Nope. Sound effects are supported via the SB16ASP part of the AWE. The synth
chip will only be able to play effects if you put them all on-board. This
means there's no effect accelleration - you're just getting around the
problem.

>It has 32-note polyphony and the on-board
>mixer can mix the samples in stereo at 44KHz without any reduction in
>sound quality (...a great plus for MOD's!).

No MOD player for the AWE, bud.

>By performing the mixing
>on-board, it also relieves the CPU of this cumbersome task, allowing
>applications to run at a much higher performance (...a great plus for
>games!).

Creative forces developers to use SB16 effects since they want to keep the
SB16 around as well. NO ACCELLERATION.

>Besides, I'm looking forward to making my own samples and
>using them in music. Now it's true that most of the games that are
>already out there (OK, all of them save a few new ones...) can't take
>advantage of this feature since they don't support the AWE32 natively.

Native support means they took the Creative MIDI emulation in a static-linked
library and emulated directly instead of via a TSR. Still sounds like an
Aria 16se. Nobody can afford the time or money to write something that
actually programs the EMU.

>However, Creative Labs is a big name in the sound card business, and I'm
>rather confident in their ability to promote native support for the
>card.

Eh? Like the SDK, right?

>It's already been out for just a few months and it's gaining
>support faster than it's leading competition, the Gravis Ultrasound.

Thank you, thank you. Given that you can buy 2.5 GUSses for the price of one
AWE that's quite a compliment.
Look, I'll say it again. AWE native is - you take the AWEUTIL.LIB or whatever
and add it to your project, tie a few dongles around and use your SB16 code
for effects. The MIDI interpreter is written by Creative and still takes up
the RAM (this time the GAME has a higher RAM requirement - no big difference)
and you still don't get what I call NATIVE support. You see - supporting the
AWE is easy. NATIVELY, now, nearly nobody's ever done that.

>If you just look at the number of games that support Sound Blaster or even
>Sound Blaster Pro, it gives you a good indication of the strong hold
>that Creative Labs has on the sound card market.

Most of the bundle market is in the hands of the Aztec/Reveal family who sell
Sound Galaxies. The FM synths are dominated by the PAS on the low end (low
prices) and SB16ASP on the high-end. Moving up we have the Orchid and the Logi
SM16W. The GUS is all over the place since you can't fit it in a category.
Support for soundcards is ususally by SOS and AIL now (Lucasarts is still using
their iMuse system and I think Sierra has their own sets too, but that's all),
and these 2 do the GUS like a dream - we're talking 50% speed boosts with SOS!

>This is another reason
>why I'm in favor of the AWE32. With it's built-in support for the SB16,
>I can use it with all of the hundreds of games that have already come
>out with SB support.

When I got a GUS, I couldn't find anybody who'd buy my SB. I left it in at
220 and put the gus on 240. Neener neener neener.

>While I know that the GUS can do SB emulation, I
>know from my roommate's experience that it isn't really adequate. These
>are the main factors that make me choose the AWE32 over the GUS (a nice
>card nonetheless...) and all of the other cards with wavetable RAM.

What cards? Maui?

>I've heard that the samples that come in it's ROM are relatively poor,
>but this is of no concern to me, since I can have any samples I want in
>the RAM.

No you can't. You have 512K and nothing to put in it, with a promise for more.
There's no efficient patch-caching on the AWE since you're supposed to be
using the ROM. This means you buy 8MB simms and don't tell anybody how much
the cost you.

>The upgradability is a big plus, too. Besides, it comes
>bundled with some pretty nice software.

The GUS comes with, what, 8 disks, right? Match that.

Man, this is boring. Look - the AWE doesn't show much promise because it's
a quick glue-circuit job of an EMU ripped out of a keyboard and an SB16ASP.
The synth isn't made for computers and game playing. Creative thinks
developers will come crawling to it, but they've got sound libraries now for
$500 and they don't need an SDK. These libraries give similar results on
an AWE and an SW32 and friends, so people won't buy it. The GUS works
faster on SOS (the only card to do so) so people will get one to get the
most out of games.
The only hope the AWE has is the great mass of CL behind it, but that's not
gonna help for long - look at OS/2... it had IBM behind it. Now AMD is putting
itself behind Advanced Gravis (they make the GF1 now) and AMD is something to
be reckoned with - they're BIG.
Yossi.

Jason G Fourier

unread,
Oct 2, 1994, 8:41:05 PM10/2/94
to
Excerpts from netnews.comp.sys.ibm.pc.soundcard.games: 2-Oct-94 Re:
AWEUTIL and NMI by Yossi Or...@weizmann.weiz
> >Thanks for the advice, but I'm still convinced that the AWE32 is the
> >best sound card for me. True, my primary use for the card will be
> >playing (and developing) games, but the main feature that I'm looking
> >for is wavetable RAM. This feature gives the sound card so many
> >possibilities for games, demos, music composition, and even just plain
> >listening.
> I think you've been spoiled by the GUSRAM implementation of the GF1.
The sort
> of things you can do with the AWERAM are much more limited - for example, you
> can not set up an IRQ trigger at a certain point in RAM. This is vital for
> effect accelleration and also for fitting more than the amount of RAM
you have
> on-board (hot-swapping, I think it's called).

How is it vital? What are you talking about? I don't see how the AWE32
implementation of wavetable RAM is significantly different from the GUS.
I therefore don't see how the GUS could do anything that the AWE32
couldn't do. The reverse is a different matter....

> Access to the RAM is limited
> by the programmer since you need to buy an SDK (I'll admit the
freeware efforts
> are nice, but they're not it).

Who's forcing anyone to use a developer's kit? If I write my own
library to interface directly with the AWE32, how is my access to RAM
going to be limited?

> >Not only does it let you choose from a limitless selection
> >of samples,
> Does not. You have to find .SBK samples, and there's only one program which
> can create them. We GUSsers use 2PAT and get our patches fresh off a
Kurzweil
> or an EPS.

But you can put ANY samples you want into an SBK, especially since you
can convert WAV files into SBK patches. In fact, I could even use GUS
samples if I wanted to by converting the GUS samples into WAV's and the
WAV's into SBK samples. Where's the limitation here? And what's the
difference between the program the AWE32 uses and the program the GUS
uses? You still haven't described anything that the GUS can do that the
AWE32 can't do!

> >but it can enhance games that use sound effects by storing
> >the sound effects as samples.
> Nope. Sound effects are supported via the SB16ASP part of the AWE.
The synth
> chip will only be able to play effects if you put them all on-board. This
> means there's no effect accelleration - you're just getting around the
> problem.

What's to stop me from ditching the SB16 compatibility and storing sound
effects as samples, as I said before? Again, why MUST I use an SDK?
You're not making any sense.

> >It has 32-note polyphony and the on-board
> >mixer can mix the samples in stereo at 44KHz without any reduction in
> >sound quality (...a great plus for MOD's!).
> No MOD player for the AWE, bud.

Fine, you can have your little fun gloating until the first MOD player
that supports the AWE32 comes out. You better gloat as fast as
possble....

> >By performing the mixing
> >on-board, it also relieves the CPU of this cumbersome task, allowing
> >applications to run at a much higher performance (...a great plus for
> >games!).
> Creative forces developers to use SB16 effects since they want to keep the
> SB16 around as well. NO ACCELLERATION.

Once again you're saying there's ABSOLUTELY NO WAY to support the AWE32
natively without paying for Creative Lab's SDK. In reality, all that's
needed is the specs for controlling the EMU chip and you've got native
support.

> >Besides, I'm looking forward to making my own samples and
> >using them in music. Now it's true that most of the games that are
> >already out there (OK, all of them save a few new ones...) can't take
> >advantage of this feature since they don't support the AWE32 natively.
> Native support means they took the Creative MIDI emulation in a static-linked
> library and emulated directly instead of via a TSR. Still sounds like an
> Aria 16se. Nobody can afford the time or money to write something that
> actually programs the EMU.

Why is it so much harder to program the EMU than to program the GUS?
Your whole argument is based on some notion that game developers will
always support the AWE32 through some half-baked MIDI emulation and SB16
compatibility. Developers that are familiar with supporting the GUS
natively will scarcely have trouble doing the same thing with the AWE32.
Native AWE32 support in combination with Creative Labs' backing make
the AWE32 extremely competitive.

> >However, Creative Labs is a big name in the sound card business, and I'm
> >rather confident in their ability to promote native support for the
> >card.
> Eh? Like the SDK, right?

WRONG!

> >It's already been out for just a few months and it's gaining
> >support faster than it's leading competition, the Gravis Ultrasound.
> Thank you, thank you. Given that you can buy 2.5 GUSses for the price of one
> AWE that's quite a compliment.

Maybe you could point out the compliment in that statement. Anyway, you
get what you pay for. :-)

> Look, I'll say it again. AWE native is - you take the AWEUTIL.LIB or
whatever
> and add it to your project, tie a few dongles around and use your SB16 code
> for effects. The MIDI interpreter is written by Creative and still takes up
> the RAM (this time the GAME has a higher RAM requirement - no big
difference)
> and you still don't get what I call NATIVE support.

<Sigh>....
Provided you use the SDK, correct. Your point?

> You see - supporting the
> AWE is easy. NATIVELY, now, nearly nobody's ever done that.

And it's about damn time somebody did! People will support the AWE32
for the same reasons they support the GUS natively.

> >If you just look at the number of games that support Sound Blaster or even
> >Sound Blaster Pro, it gives you a good indication of the strong hold
> >that Creative Labs has on the sound card market.
> Most of the bundle market is in the hands of the Aztec/Reveal family who sell
> Sound Galaxies. The FM synths are dominated by the PAS on the low end (low
> prices) and SB16ASP on the high-end. Moving up we have the Orchid and
the Logi
> SM16W. The GUS is all over the place since you can't fit it in a category.

Huh?
How many games have you played? Do you have any idea how many games
have a Sound Blaster option? Do you have any idea how small GUS support
is in comparison? The PAS and Sound Galaxy get their support through
Sound Blaster compatibility, so what's your point?

> Support for soundcards is ususally by SOS and AIL now (Lucasarts is
still using
> their iMuse system and I think Sierra has their own sets too, but
that's all),
> and these 2 do the GUS like a dream - we're talking 50% speed boosts
with SOS!

The turnout of games that use SOS and AIL hasn't been very high. Even
so, AWE32 support for these packages has just as much potential for
"speed boosts" as does GUS support.

> >This is another reason
> >why I'm in favor of the AWE32. With it's built-in support for the SB16,
> >I can use it with all of the hundreds of games that have already come
> >out with SB support.
> When I got a GUS, I couldn't find anybody who'd buy my SB. I left it in at
> 220 and put the gus on 240. Neener neener neener.

^^^^^^^^^^^^^^^^^^^
Oh, now I see things your way. That makes complete sense! Why didn't
you say that before?

That's exactly the topic of countless discussions on the sound
newsgroups by frustrated GUS owners that can't get both cards to
coexist. This is why built-in SB16 compatibility is such a plus for the
AWE32!

> >While I know that the GUS can do SB emulation, I
> >know from my roommate's experience that it isn't really adequate. These
> >are the main factors that make me choose the AWE32 over the GUS (a nice
> >card nonetheless...) and all of the other cards with wavetable RAM.
> What cards? Maui?

Yes, another nice card without built-in SB compatibility.

> >I've heard that the samples that come in it's ROM are relatively poor,
> >but this is of no concern to me, since I can have any samples I want in
> >the RAM.
> No you can't. You have 512K and nothing to put in it, with a promise
for more.

Wait a second.... The GUS also comes with 512K, and it's expandable to
1MB at MOST. That's quite a bit less than 28MB. Even if you only
support the minimum amount of RAM, you're no better off with the GUS.
For some reason, you still seem to think that you can't put your own
samples in the AWE32. What's the sample RAM for?!!!

> There's no efficient patch-caching on the AWE since you're supposed to be
> using the ROM.

There's no patch caching on the GUS either. And maybe you're supposed
to be using the ROM, but not me.

> This means you buy 8MB simms and don't tell anybody how much
> the cost you.

So just because the extra RAM costs money, the expandability is worthless?

> >The upgradability is a big plus, too. Besides, it comes
> >bundled with some pretty nice software.
> The GUS comes with, what, 8 disks, right? Match that.

And how many of those are full of samples? The AWE32 comes with better
software, just read the box.

> Man, this is boring.

Hey man, you're right, man. So if this is so boring, man, why are you
replying?

> The only hope the AWE has is the great mass of CL behind it, but that's not
> gonna help for long - look at OS/2... it had IBM behind it. Now AMD
is putting
> itself behind Advanced Gravis (they make the GF1 now) and AMD is something to
> be reckoned with - they're BIG.

Sure. IBM is BIG too. But what place do they have in the operating
system market? Look at Windows, it had Microsoft behind it--their main
product is operating systems. Creative Labs' main product is sound
cards. Somehow, I think Creative Labs behind the AWE32 is a better bet.

One last question--what need was there to be so ARROGANT in your reply?
That only suggests insecurity on your behalf. Grow up.

-Jason Fourier

Anssi Saari

unread,
Oct 3, 1994, 5:13:06 AM10/3/94
to
In <4iXpAVm00...@andrew.cmu.edu> Jason G Fourier <jf...@andrew.cmu.edu> writes:

:>Fine, you can have your little fun gloating until the first MOD player


:>that supports the AWE32 comes out. You better gloat as fast as
:>possble....

What, are you making one? Seeing is believeing.

:>Once again you're saying there's ABSOLUTELY NO WAY to support the AWE32


:>natively without paying for Creative Lab's SDK. In reality, all that's
:>needed is the specs for controlling the EMU chip and you've got native
:>support.

Very good. Now, where do you get the specs? EMU doesn't give them out.
So, you're stuck with the SDK.

:>That's exactly the topic of countless discussions on the sound


:>newsgroups by frustrated GUS owners that can't get both cards to
:>coexist.

I've seen it asked countless times whether it is possible or not, but
never a long discussion. It's such a trivial thing.

:>And how many of those are full of samples? The AWE32 comes with better


:>software, just read the box.

Wow, that's how you judge software? By what it says on the box?

Anssi

Anthony Robert Ruggeri

unread,
Oct 3, 1994, 10:40:07 AM10/3/94
to
Having watched this argument for a little while, I have to jump in.

In article <36ohv2$5...@proffa.cc.tut.fi>,


Anssi Saari <s10...@cc.tut.fi> wrote:
>In <4iXpAVm00...@andrew.cmu.edu> Jason G Fourier <jf...@andrew.cmu.edu> writes:
>
>:>Fine, you can have your little fun gloating until the first MOD player
>:>that supports the AWE32 comes out. You better gloat as fast as
>:>possble....
>
>What, are you making one? Seeing is believeing.
>

Hmmm, sounds like the infamous "Patch Maker Heavy" everybody was waiting
for. Seeing is indeed believeing. Maybe nobody really wants to hear
MODs on the AWE.

>:>Once again you're saying there's ABSOLUTELY NO WAY to support the AWE32
>:>natively without paying for Creative Lab's SDK. In reality, all that's
>:>needed is the specs for controlling the EMU chip and you've got native
>:>support.
>
>Very good. Now, where do you get the specs? EMU doesn't give them out.
>So, you're stuck with the SDK.
>

Reverse engineering perhaps? Hacking? This is how impatient but talanted
programmers get low-level information. The first AWE32 patch editor (eskeebay)
got most of the description data for patch parameters from a variety of beta
users who played around with it and discovered them (just an example).

>:>That's exactly the topic of countless discussions on the sound
>:>newsgroups by frustrated GUS owners that can't get both cards to
>:>coexist.
>
>I've seen it asked countless times whether it is possible or not, but
>never a long discussion. It's such a trivial thing.

Even a GUS and an AWE32 will coexist, under both DOS and Windows.

>
>:>And how many of those are full of samples? The AWE32 comes with better
>:>software, just read the box.
>
>Wow, that's how you judge software? By what it says on the box?
>

I think he means, look at what software is supplied with the AWE, because
it blows away what the GUS bundles. The sequencer (Cakewalk Apprentice) is
the best such program bundled with any soundcard, and WaveStudio is one of
the better wave editors. And, right out of the box, I was far more impressed
with the AWE32's software for loading and working with custom patches
than what came with the GUS. Plus, with two extra programs available via
FTP--one from Creative and one from a third party--the AWE32's ability to
create and manipulate custom patches is much, much better than what is
available for the GUS after... what, more than three years or something?
The last time Gravis issued a major update in the functionality of the
Windows drivers was several months ago. Since then we've been waiting for
"real" bank support, to no avail. The GUS is a good card, but for serious
work it's crippled by the software workarounds. And yes, I have tried some
of the shareware patch management programs. They're admirable for what they
accomplish but they don't come close to the ease of use the AWE has.

My $.02.

>Anssi
>


--
Anthony Ruggeri - arug...@phoenix.princeton.edu // E-mu Proteus MPS+
Molecular Biology '95 - Princeton University // Korg DW-8000
Weevils are intriguing but irrelevent. // Gravis Ultrasound
Jurassic Park was cool. Yeah. Huh huh huh. // Sound Blaster AWE32

Craig Lemas

unread,
Oct 4, 1994, 12:12:22 PM10/4/94
to

> > >It has 32-note polyphony and the on-board
> > >mixer can mix the samples in stereo at 44KHz without any reduction in
> > >sound quality (...a great plus for MOD's!).
> > No MOD player for the AWE, bud.
>
> Fine, you can have your little fun gloating until the first MOD player
> that supports the AWE32 comes out. You better gloat as fast as
> possble....

I would like to clear up something for all the AWE32 bashers out there who say
that the GUS is great 'cause it can play MODS without bogging down the system
whereas the AWE32 has to use the digital channel on the card to play MOD files.
We here in the Engineering department at Creative Labs have taken and created
a sample program that enables MOD files to be played through the AWE32 via the
RAM & MIDI channel on the card. While this is just a sample program for
internal use, it proves that it can be done and hopefully will be done in the
future as a real release for AWE32 users.

Craig Lemas
QA Technician, CLI
[cle...@creaf.com]

John Stephens

unread,
Oct 5, 1994, 12:55:25 AM10/5/94
to
Craig Lemas (cle...@creaf.com.) wrote:

: We here in the Engineering department at Creative Labs have taken and created

: a sample program that enables MOD files to be played through the AWE32 via the
: RAM & MIDI channel on the card. While this is just a sample program for
: internal use, it proves that it can be done and hopefully will be done in the
: future as a real release for AWE32 users.

Since you work in the engineering department of Creative Labs, perhaps
you can help me with a Sound Blaster 16 problem. I have used a Sound
Blaster 16 Basic (not Value) and had it connected to an Onkyo receiver
with decent speakers. For some reason, the card produces a loud hissing
sound when playing digital samples below 22 KHz. This isn't background
noise. As the frequency of the sound gets lower, the hiss gets louder.
Wolfenstein 3D uses 8 KHz samples, and sounds aweful. This problem has
nothing to do with 8-bit or 16-bit samples; only the frequency is
important. I enjoyed the quality of 16-bit 44 KHz stereo samples, and was
quite pleased with the small amount of background noise and the good
mixer. But because of this hiss problem, I was unable to keep the card
since it makes DOS games extremely annoying. DOOM, which uses 11 KHz
samples, is another example.

I want to buy a Sound Blaster 16 Basic or MCD and a Turtle Beach Rio, but
only after this hissing problem is fixed, or there is some other
solution. As it stands right now, the Sound Blaster Pro is still better
for games since it doesn't have the hissing problem. It does, however,
have much more background noise.

Right now I have an Advanced Gravis UltraSound 3.73 and a Logitech
SoundMan Wave. The SoundMan Wave has its own set of problems, such as low
MIDI/FM volume relative to other sound sources, low overall output
volume, a continuous background tone that's fairly loud and somewhat
annoying, and less-than-perfect Sound Blaster Pro compatibility. Also,
the MIDI patches don't approach the quality of the Rio, which I could
easily add to almost any Sound Blaster 16 card. The UltraSound is unique,
but I will probably sell it if I can get a Sound Blaster 16 without the
hiss problem and a Turtle Beach Rio.

Any suggestions?

Ari Laakkonen

unread,
Oct 5, 1994, 10:41:48 AM10/5/94
to
In article <36ru6v$g...@clihost.cli.creaf.com> cle...@creaf.com. writes:

>> Fine, you can have your little fun gloating until the first MOD player
>> that supports the AWE32 comes out. You better gloat as fast as
>> possble....
>

> We here in the Engineering department at Creative Labs have taken and created
>a sample program that enables MOD files to be played through the AWE32 via the
>RAM & MIDI channel on the card. While this is just a sample program for
>internal use, it proves that it can be done and hopefully will be done in the
>future as a real release for AWE32 users.

Oh good, now I can stop developing my own mod-player ("esbeekay II")...
I did wonder how many other people were occupied with the same task.
Any idea of a release date?

Ari
--
Ari Laakkonen (a...@vapl.demon.co.uk)

John D. Palmer

unread,
Oct 6, 1994, 9:25:46 AM10/6/94
to
>> > >It has 32-note polyphony and the on-board
>> > >mixer can mix the samples in stereo at 44KHz without any reduction in
>> > >sound quality (...a great plus for MOD's!).
>> > No MOD player for the AWE, bud.
>
>I would like to clear up something for all the AWE32 bashers out there who say
>that the GUS is great 'cause it can play MODS without bogging down the system
>whereas the AWE32 has to use the digital channel on the card to play MOD files.
>We here in the Engineering department at Creative Labs have taken and created
>a sample program that enables MOD files to be played through the AWE32 via the
>RAM & MIDI channel on the card. While this is just a sample program for
>internal use, it proves that it can be done and hopefully will be done in the
>future as a real release for AWE32 users.

While a MOD player that loads the samples into the AWE's ram and
plays them as instruments -- like the GUS -- would be nice, why is it
such a big deal to people??? (GUS and AWE owners)

How much work does everybody do while listening to MODs?! When I
play 'em, I like to listen to 'em and watch the 'scopes or whatever the
current modplayer shows for graphics...

On my old 486DLC-40 the Mod4Win (or similar) showed 17% CPU usage
when playing MODs in Windows... and since it's expired I haven't tested
it on my 486DX-66, but I'd guess around 10-15% CPU usage. This is not a
big percentage to lose to pitch-bend digital samples!

It must be the fact that it's (currently) _something_ that the
GUS can do that the AWE can't do that makes it so "important."

I hate to join the bashing, but once they release the MODplayer
for the AWE then we'll have the choice of adding chorus and reverb to
the MOD's samples... oughtta sound pretty good, eh?
--
John D. Palmer | "I have never let my schooling
jd...@Virginia.edu | interfere with my education."

John Stephens

unread,
Oct 7, 1994, 5:25:44 AM10/7/94
to
John D. Palmer (jd...@watt.seas.Virginia.EDU) wrote:

: On my old 486DLC-40 the Mod4Win (or similar) showed 17% CPU usage

: when playing MODs in Windows... and since it's expired I haven't tested
: it on my 486DX-66, but I'd guess around 10-15% CPU usage. This is not a
: big percentage to lose to pitch-bend digital samples!

What type of MOD were you playing (how many channels) and were you using
16-bit 44 KHz stereo output?

: It must be the fact that it's (currently) _something_ that the

: GUS can do that the AWE can't do that makes it so "important."

Software mixing really degrades sound quality. This is why MODs sound so
much better on an UltraSound than a Sound Blaster 16 or other sound
cards. If a MOD player for the AWE32 that uses independent hardware
channels is released, you will notice a dramatic improvement in sound
quality.

Peter Modin

unread,
Oct 7, 1994, 6:10:08 AM10/7/94
to
jd...@watt.seas.Virginia.EDU (John D. Palmer) writes:
> While a MOD player that loads the samples into the AWE's ram and
>plays them as instruments -- like the GUS -- would be nice, why is it
>such a big deal to people??? (GUS and AWE owners)

> How much work does everybody do while listening to MODs?! When I
>play 'em, I like to listen to 'em and watch the 'scopes or whatever the
>current modplayer shows for graphics...

Well on my ol' Amiga I did from time to time, because of the great quality
on som eof them. And on my Amiga 1200 (that's a 14 MHz '020, not a 40 MHz
486 DX) you hardly noticed it was there, except for taking up memory (and
the sound of course). Since there are no dedicated sound chips on the PC
it makes a lot of sense to make use of whatever dedicated card you have...
Then you can listen to .mod's while getting new ones, and it doesn't even
stop when you move a window (mod4win steals a lot of memory for buffers,
doesn't leave that much for my 4 MB "monster").

The AWE32 (but the GUS would have sufficed :-) is in fact the main reason
I dumped the Amiga.

--
==Peter Modin, Computer Science, Linkoping U.========================
| e91p...@und.ida.liu.se DISCLAIMER : The devil made me do it. |
| mo...@lysator.liu.se RECLAIMER : I talked him into it. |
==4 line signature sucks=====REDISCLAIMER: I got that idea from TV===

Craig Lemas

unread,
Oct 7, 1994, 12:31:05 PM10/7/94
to

> >Bashers? All you have to do is release complete low-level documentation
> >on the thing and we'll support it. Keep that info secret and we'll bash
> >it as long as it remains useless (w/o docs, it's useless). And as with the
> >CSP, that something can be done by the programmers at CL means absolutely
> >nothing to the other programmers that are not privy to this "secret"
> >documentation.
> >If you want to see the AWE32 get "respect", well, you know what to do.
>
> I hope Creative Labs is reading this!
> -Jason Fourier

Yes, we ARE reading these messages. I pass all information on to the head of
the Developer Support department and he does read the info.

As for the information being useless, several "professional" developers have
written drivers for the AWE32 using just the tools that you are condemning as
usless. These developers include Miles Design (AIL), HMI (SOS), Ratcliff &
Associates (MIDPAK), Digital Expressions (DOOM), LucasArts, PC Music
(Psygnosis), & Accolade.

Anway, how many registers do you think there are on the EMU8000 chip? How
about 4096? I'm sure that would be loads of fun to play with.

Jason G Fourier

unread,
Oct 7, 1994, 1:32:07 AM10/7/94
to
Excerpts from netnews.comp.sys.ibm.pc.soundcard.games: 6-Oct-94 Re: MOD
player(s) for the A.. by John D. Pal...@watt.seas
> While a MOD player that loads the samples into the AWE's ram and
> plays them as instruments -- like the GUS -- would be nice, why is it
> such a big deal to people??? (GUS and AWE owners)
>
> How much work does everybody do while listening to MODs?! When I
> play 'em, I like to listen to 'em and watch the 'scopes or whatever the
> current modplayer shows for graphics...
>
> On my old 486DLC-40 the Mod4Win (or similar) showed 17% CPU usage
> when playing MODs in Windows... and since it's expired I haven't tested
> it on my 486DX-66, but I'd guess around 10-15% CPU usage. This is not a
> big percentage to lose to pitch-bend digital samples!
>
> It must be the fact that it's (currently) _something_ that the
> GUS can do that the AWE can't do that makes it so "important."

While it may seem that the desire for a MOD player for the AWE32 is just
a childish race for superiority, this truly isn't the case. When you
were using 17% CPU on your older 486, what sound card was the MOD
playing through? I would expect that it was a mono 8-bit card, but
mixing samples can actually tie up a significant portion of the CPU.
For example, trying to play a standard 4-channel MOD through a Sound
Blaster 16 in stereo at 44KHz can eat up as much as 33% of the CPU
cycles. Playing a MOD through a sound card with wavetable RAM, like the
GUS, will not only yield higher CPU performance, but better sound
quality as well, because the mixing that occurs on the sound card causes
no comparable loss in sound quality. Even if you're using 10% of your
CPU for sound, wouldn't it be so much better to use only a fraction of
1% instead, and devote more of your CPU to cool graphics or whatever
your application might demand? (Remember that games and demos often
play MOD's or equivalent music files.) I think so! Therefore, I think
a MOD player that uses the AWE32's wavetable RAM is sorely needed.

-Jason Fourier

Jason G Fourier

unread,
Oct 7, 1994, 1:55:56 AM10/7/94
to
Excerpts from netnews.comp.sys.ibm.pc.soundcard.tech: 5-Oct-94 Re:
AWEUTIL and NMI by Cornel Hu...@ephsa.sat.tx
>Craig Lemas <cle...@creaf.com.> writes:
>> I would like to clear up something for all the AWE32 bashers out
there who sa
>> that the GUS is great 'cause it can play MODS without bogging down
the system
>> whereas the AWE32 has to use the digital channel on the card to play
MOD file
>> We here in the Engineering department at Creative Labs have taken
and create
>> a sample program that enables MOD files to be played through the
AWE32 via th
>> RAM & MIDI channel on the card. While this is just a sample program for
>> internal use, it proves that it can be done and hopefully will be
done in the
>> future as a real release for AWE32 users.
>> Craig Lemas
>> QA Technician, CLI
>> [cle...@creaf.com]
>

Anssi Saari

unread,
Oct 7, 1994, 10:54:17 AM10/7/94
to
In <Cx96M...@murdoch.acc.Virginia.EDU> jd...@watt.seas.Virginia.EDU (John D. Palmer) writes:

:> How much work does everybody do while listening to MODs?! When I

:>play 'em, I like to listen to 'em and watch the 'scopes or whatever the
:>current modplayer shows for graphics...

I never do any important work on my while listening to mods, I consider
it too risky in both dos and Windows. However, reading and wrinting
news and many other tasks I prefer listening to something, usually
mods. For example, cmod plays in the background as I write this.

:> On my old 486DLC-40 the Mod4Win (or similar) showed 17% CPU usage

:>when playing MODs in Windows... and since it's expired I haven't tested
:>it on my 486DX-66, but I'd guess around 10-15% CPU usage. This is not a
:>big percentage to lose to pitch-bend digital samples!

Mod4Win managed to choke on a 26-channel mtm on my 66, i.e. over
100% CPU usage.

As for bashing, does anyone want to bet CL's player will play at least half
the effects wrong?

Anssi

Chuck Dedman

unread,
Oct 7, 1994, 11:51:22 PM10/7/94
to
Jason G Fourier (jf...@andrew.cmu.edu) wrote:
: Excerpts from netnews.comp.sys.ibm.pc.soundcard.games: 2-Oct-94 Re:

: AWEUTIL and NMI by Yossi Or...@weizmann.weiz
<<<<<<<<< CUT CUT >>>>>>>>>>>>>>>>>>
: > >It's already been out for just a few months and it's gaining

: > >support faster than it's leading competition, the Gravis Ultrasound.
: > Thank you, thank you. Given that you can buy 2.5 GUSses for the price of one
: > AWE that's quite a compliment.

: Maybe you could point out the compliment in that statement. Anyway, you
: get what you pay for. :-)

: > Look, I'll say it again. AWE native is - you take the AWEUTIL.LIB or
: whatever
: > and add it to your project, tie a few dongles around and use your SB16 code
: > for effects. The MIDI interpreter is written by Creative and still takes up
: > the RAM (this time the GAME has a higher RAM requirement - no big
: difference)
: > and you still don't get what I call NATIVE support.

: <Sigh>....
: Provided you use the SDK, correct. Your point?

: > You see - supporting the
: > AWE is easy. NATIVELY, now, nearly nobody's ever done that.

: And it's about damn time somebody did! People will support the AWE32
: for the same reasons they support the GUS natively.

:
: <<<<<<<<<<<< CUT CUT >>>>>>>>>>>>
: -Jason Fourier

Maybe that _IS_ Yoshi's point Jason!!
You <almost> admit that the GUS can do all that the AWE32 does; albiet
in less memory on-board; but the Gus has the native support NOW , it's
working, and you get a heck of a lot more change at the checkout counter.

To me, it doesn't make sense to argue the merits that are GOING to be,
but in the meantime merrily pay over twice as much money.

Chuck

Chuck Dedman

unread,
Oct 8, 1994, 12:14:05 AM10/8/94
to
Craig Lemas (cle...@creaf.com.) wrote:

: > > >It has 32-note polyphony and the on-board

Come On Craig!! That sounds like the same Pipe Dream Promises that CLI
has been giving me since the days of the Game Blaster.

If you guys want to argue a point and say you KNOW it can be done
because it's been done in the lab; then give us PROOF by letting the
users that are paying your salary have the magic carpets.
Have you ever heard of Beta testing??
I got so sick and tired of hearing the same old songs from you folks, I
just chucked ALL my SB stuff and paid for one of the first GUS cards 6
months before they were even sold!
I'll admit, the GUS was and is harder to get tweaked, but once you
optimize the installation, it can't be touched-IMO. Most of that sucess
comes at the expense of a phone call to Brian, Bill, or some of the other
gurus at A.G., but I've NEVER had a problem that one of them didn't fix
by calling me back with some info, or even DownLoading **TEST** programs
that usually work.
Now that I'm into programming the card I have discoverd almost ALL the
others programmers swear by the ease of the GUS to program.

Chuck

Ari Laakkonen

unread,
Oct 7, 1994, 10:31:22 AM10/7/94
to

> I hate to join the bashing, but once they release the MODplayer
>for the AWE then we'll have the choice of adding chorus and reverb to
>the MOD's samples... oughtta sound pretty good, eh?
>--

I'm developing a mod-player for the AWE that will use the AWE ram and
"hardware mixing", and the ability to use LFO's, filters, reverb and chorus
etc. for effects is the reason I was putting one together.

The code to play the samples from ram using midi sequences seems straight-
forward, but not knowing anything much about mod effects I wondered if
anyone out there had any good sources on this? If you do, I would appreciate
it if you could send me a copy.

Rogene Talento

unread,
Oct 8, 1994, 3:21:25 PM10/8/94
to
Ari Laakkonen (a...@vapl.demon.co.uk) wrote:
: In article <36ru6v$g...@clihost.cli.creaf.com> cle...@creaf.com. writes:

: > We here in the Engineering department at Creative Labs have taken and created

: >a sample program that enables MOD files to be played through the AWE32 via the
: >RAM & MIDI channel on the card. While this is just a sample program for
: >internal use, it proves that it can be done and hopefully will be done in the
: >future as a real release for AWE32 users.

: Oh good, now I can stop developing my own mod-player ("esbeekay II")...
: I did wonder how many other people were occupied with the same task.
: Any idea of a release date?

Hey, I wouldn't quit just yet. There can never be too many programs to
choose from. I would love to write a program that would play MODs on the
AWE32 using the RAM and MIDI channel, but I wasn't blessed with any
programming skills. Just write it! Esbeekay rules!

Rogene
--
lab...@teleport.COM Public Access User --- Not affiliated with Teleport
Public Access UNIX and Internet at (503) 220-1016 (2400-14400, N81)

Lord Soth

unread,
Oct 9, 1994, 9:47:59 PM10/9/94
to
-----BEGIN PGP SIGNED MESSAGE-----

In article <373nep$f...@proffa.cc.tut.fi>, s10...@cc.tut.fi (Anssi Saari) wrote:
> :> On my old 486DLC-40 the Mod4Win (or similar) showed 17% CPU usage
> :>when playing MODs in Windows... and since it's expired I haven't tested
> :>it on my 486DX-66, but I'd guess around 10-15% CPU usage. This is not a
> :>big percentage to lose to pitch-bend digital samples!
>
> Mod4Win managed to choke on a 26-channel mtm on my 66, i.e. over
> 100% CPU usage.
>

Really? I've never had any problems with any file, including 26+ channel
S3M's and MTM's.


-----BEGIN PGP SIGNATURE-----
Version: 2.6

iQCVAwUBLpideuaziJ1BeSU1AQE3WwQAuJodC3J2DMJL6Xl2f9XyhFtuDz0SM/62
YZC17NgqYC3j3lKFQ0F47NytjhXgTv6OoywomK92wj33a6OgFag9TCVX80jkf+5q
pEh6p1HPzyOjTePbIRl1td8d+Rtiyg5wjgK9+NKcYW+HIxHcCdRmg3zDEQ84Agxj
joxgabhyRXg=
=NFYo
-----END PGP SIGNATURE-----

David Masten

unread,
Oct 10, 1994, 9:40:13 AM10/10/94
to
John Stephens (jo...@ccnet.com) wrote:
: John D. Palmer (jd...@watt.seas.Virginia.EDU) wrote:

First, sorry this is going to all the damn csip.soundcard groups, haven't
figured out how to change that. Then again, not really up to me to redirect
the thread...

: : On my old 486DLC-40 the Mod4Win (or similar) showed 17% CPU usage

: : when playing MODs in Windows... and since it's expired I haven't tested
: : it on my 486DX-66, but I'd guess around 10-15% CPU usage. This is not a
: : big percentage to lose to pitch-bend digital samples!

: What type of MOD were you playing (how many channels) and were you using
: 16-bit 44 KHz stereo output?

FWIW, I used Mod4Win with a dx2/66 and the Ensoniq card and scored about 15%
with a standard mod with it set at 16-bit 44 KHz stereo. I too generally
wan't doing anything else. In fact, when I tried, it choked. But I could
see others wanting to have them as background music.

: Software mixing really degrades sound quality. This is why MODs sound so

: much better on an UltraSound than a Sound Blaster 16 or other sound
: cards. If a MOD player for the AWE32 that uses independent hardware
: channels is released, you will notice a dramatic improvement in sound
: quality.

Certainly sounded better on my GUS than on an SBPro :-)
But can't honestly say they sounded better than on the Ensoniq. Then again,
I was only using standard mods. Does mod4win play the mods with more
channels?

And can someone clear up another point. What is the bandwidth of the
standard 4-channel mod? Someone said something like 16 kHz, but is
that 16 KHz audio (thus 32 KHz sampling) or 16 K sample, thus 8 K audio?

--
Dave Masten
dma...@spy.org

John Stephens

unread,
Oct 10, 1994, 4:39:35 PM10/10/94
to
David Masten (dma...@spy.org) wrote:

: : Software mixing really degrades sound quality. This is why MODs sound so

: : much better on an UltraSound than a Sound Blaster 16 or other sound
: : cards. If a MOD player for the AWE32 that uses independent hardware
: : channels is released, you will notice a dramatic improvement in sound
: : quality.

: Certainly sounded better on my GUS than on an SBPro :-)
: But can't honestly say they sounded better than on the Ensoniq. Then again,
: I was only using standard mods. Does mod4win play the mods with more
: channels?

Using Mod4Win, the UltraSound won't sound any better than other cards
because Mod4Win just uses the standard Windows Wave device, which has
only two channels: left and right. If you use MODUS, Intertia Player, or
other players that support UltraSound directly, the sound quality will be
substantially better. Also, your stereo equipment has a huge impact on
the difference. In general, the better the stereo, the better the
UltraSound sounds compared to other cards playing MOD files. The AWE32
and Tropez might get native MOD players, though, that should sound really
good.

Saari Anssi

unread,
Oct 11, 1994, 4:37:17 AM10/11/94
to
John Stephens (jo...@ccnet.com) wrote:
: David Masten (dma...@spy.org) wrote:

: Using Mod4Win, the UltraSound won't sound any better than other cards ...

Assuming the other card is 16bit.

: If you use MODUS, Intertia Player, or

: other players that support UltraSound directly, the sound quality will be
: substantially better.

Well, no, as mod4win now supports linear interpolation. As I mentioned
earlier, mod4win choked on my 66 with a 26 channel mtm, but that seemed
to be a momentary glitch as it now takes 'only' 85% of CPU time, 16bit
44kHz, ido and panning.

Anssi

James Richard Webster

unread,
Oct 11, 1994, 5:16:09 AM10/11/94
to
jo...@ccnet.com (John Stephens) writes:

>Software mixing really degrades sound quality. This is why MODs sound so
>much better on an UltraSound than a Sound Blaster 16 or other sound
>cards. If a MOD player for the AWE32 that uses independent hardware
>channels is released, you will notice a dramatic improvement in sound
>quality.

I can only agree! But unfortunately, no-one is going to be writing a MOD-
player for the AWE32 until Creative releases decent low-level info for the card,
I am not the world's greatest coder (by a long shot ;) but anyone can tell you
that the info in the DIP (Developer's Information Package, not a Software
Developer's Kit!) ain't worth jack. God know's the coders on comp.sys.ibm.pc.
demos like to go on about that! WHen Creative follows Gravis lead, and
release the technical info we need to program this card, THEN we'll get MOD
players written.

I mean, how are most people supposed to even read a PostScript or HP Printertalk
document???

L8tr all,
Woody

--
jrwe...@socs.uts.edu.au (James 'Woody' Webster in real life)

[this space for rent]

John Stephens

unread,
Oct 11, 1994, 6:35:57 PM10/11/94
to
Saari Anssi (s10...@korppi.cs.tut.fi) wrote:

: John Stephens (jo...@ccnet.com) wrote:
: : David Masten (dma...@spy.org) wrote:
:
: : Using Mod4Win, the UltraSound won't sound any better than other cards ...
:
: Assuming the other card is 16bit.

Of course. I am assuming a 16-bit sound card is being used.

: : If you use MODUS, Intertia Player, or

: : other players that support UltraSound directly, the sound quality will be
: : substantially better.
:
: Well, no, as mod4win now supports linear interpolation. As I mentioned
: earlier, mod4win choked on my 66 with a 26 channel mtm, but that seemed
: to be a momentary glitch as it now takes 'only' 85% of CPU time, 16bit
: 44kHz, ido and panning.

I have done comparisons with Mod4Win on my UltraSound and SoundMan Wave.
For one thing, the volume is lower in Mod4Win, and if I amplify it, it
distorts. The SoundMan Wave sounds better using Mod4Win than the
UltraSound, but the UltraSound sounds substantially better than either
using Intertia Player. I had IDO and panning enabled, in 16-bit 44 KHz
stereo. For what it's worth, even in surround mode, IDO, panning, 16-bit
44 KHz stereo playing a 32-channel MTM, it always took less than 20% of
CPU time and never skipped on my Pentium 100 MHz system. :)

But Intertia Player (and all native UltraSound players) still sounds better.

Michael Burns-nscst-tnt

unread,
Oct 12, 1994, 12:07:34 PM10/12/94
to
In article r...@clihost.cli.creaf.com, Craig Lemas <cle...@creaf.com.> () writes:

[Valid concerns by Jason Fourier deleted]

> Yes, we ARE reading these messages. I pass all information on to the head of
> the Developer Support department and he does read the info.

Good. Nice to know we have some direct input!

> As for the information being useless, several "professional" developers have
> written drivers for the AWE32 using just the tools that you are condemning as
> usless. These developers include Miles Design (AIL), HMI (SOS), Ratcliff &
> Associates (MIDPAK), Digital Expressions (DOOM), LucasArts, PC Music
> (Psygnosis), & Accolade.

Well, apologies for being blunt, but SO WHAT? As a computer music fan,
I really don't have much use for the above (only if I play games). I would
like to play with the AWE at the bare-metal level, to see what it can do, and
not always through MIDI or someone's else's interface. Oh, I'm sorry, I guess
you though everyone bought the card for games, didn't you? :-)

> Anway, how many registers do you think there are on the EMU8000 chip? How
> about 4096? I'm sure that would be loads of fun to play with.

This is a terrible, terrible argument. It's akin to saying somthing like:

How many parts to a car engine is there? 4000? Guess you can't learn to be
a mechanic.

How many different '486 addressing modes are there? 2000? (I have no real idea,
it's a lot!) Guess you can't learn assembly language.

How many different bones, muscles, nerves are there in the human body? 10,000?
Guess you can't learn to be a doctor.

Just because the information might seem arcane and voluminous to you, what do you
think happens when you have many people looking at the same info at once? Bingo!
They can understand it, and form their own specialty. It's very likely for a
particular application, you only need a small subset of those registers. At least
with documentation, I can read it and EVENTUALLY find the registers I need. Without
the documentation, I can't, and there goes interest in developing ANY application
for the soundcard.

Since your boss is listening: Remember Sony Betamax? IBM MicroChannel? These
were technically superior products at the time. However, due their proprietary
nature, they fell by the wayside. The AWE has some nice features that I would
like to play with that STILL aren't documented (DSP, EMU Synth Chip, Digital
Recording Pin), and I don't want to wait on CL's software to give me those
capabilities. I bought the card for its features (still pretty much promised,
and I've now had my card since March) and I want to get the full use of the card.
If you want to see more software develop, release the info. I will bet that
you will sell a lot more AWE`s when you have independent shareware & freeware
developers developing applications for it.

Also, a question to Craig: Is there enough information in the DIP to develop
a FULL MOD player? I would think that without the EMU register list, some
effects would either not be done at all, or would not be done efficiently.
If CL's internal MOD player developer had to sneak a look at that internal
EMU-register document, then the DIP is not comprehensive enough.

Michael Burns

Jensi

unread,
Oct 12, 1994, 5:37:51 AM10/12/94
to
In article <37bg7t$9...@tierra.santafe.edu>,

David Masten <dma...@spy.org> wrote:
>John Stephens (jo...@ccnet.com) wrote:
>: John D. Palmer (jd...@watt.seas.Virginia.EDU) wrote:
>
>First, sorry this is going to all the damn csip.soundcard groups, haven't
>figured out how to change that. Then again, not really up to me to redirect
>the thread...
>
>: : On my old 486DLC-40 the Mod4Win (or similar) showed 17% CPU usage
>: : when playing MODs in Windows... and since it's expired I haven't tested
>: : it on my 486DX-66, but I'd guess around 10-15% CPU usage. This is not a
>: : big percentage to lose to pitch-bend digital samples!
>
>: What type of MOD were you playing (how many channels) and were you using
>: 16-bit 44 KHz stereo output?
>
>FWIW, I used Mod4Win with a dx2/66 and the Ensoniq card and scored about 15%
>with a standard mod with it set at 16-bit 44 KHz stereo. I too generally
>wan't doing anything else. In fact, when I tried, it choked. But I could
>see others wanting to have them as background music.

Should be much better now. The 2.00 was a 100% speed increase over the 1.xx
and the new 2.10 is another 20-50% increase. For plain old 4 channel MODs
you shouldn't see more than 6-7% (without panning and IDO [oversampling]).

>: Software mixing really degrades sound quality. This is why MODs sound so
>: much better on an UltraSound than a Sound Blaster 16 or other sound
>: cards. If a MOD player for the AWE32 that uses independent hardware
>: channels is released, you will notice a dramatic improvement in sound
>: quality.
>
>Certainly sounded better on my GUS than on an SBPro :-)
>But can't honestly say they sounded better than on the Ensoniq. Then again,
>I was only using standard mods. Does mod4win play the mods with more
>channels?

Sure does. The new 2.10 plays .MOD, .NST, .WOW, .OKT, .STM, .S3M, .669,
.FAR, and .MTM with up to 32 channels. And BTW, with an Ultrasound you'll
only get ~20 kHz for 32 channels. With Mod4Win you'll get the full 48 kHz
if you can handle it ;-)

>And can someone clear up another point. What is the bandwidth of the
>standard 4-channel mod? Someone said something like 16 kHz, but is
>that 16 KHz audio (thus 32 KHz sampling) or 16 K sample, thus 8 K audio?
>
>--
>Dave Masten
>dma...@spy.org

Jensi

Jensi

unread,
Oct 12, 1994, 5:51:02 AM10/12/94
to
In article <4iZBpLS00...@andrew.cmu.edu>,

Jason G Fourier <jf...@andrew.cmu.edu> wrote:
>Excerpts from netnews.comp.sys.ibm.pc.soundcard.games: 6-Oct-94 Re: MOD
>player(s) for the A.. by John D. Pal...@watt.seas
>> While a MOD player that loads the samples into the AWE's ram and
>> plays them as instruments -- like the GUS -- would be nice, why is it
>> such a big deal to people??? (GUS and AWE owners)
>>
>> How much work does everybody do while listening to MODs?! When I
>> play 'em, I like to listen to 'em and watch the 'scopes or whatever the
>> current modplayer shows for graphics...
>>
>> On my old 486DLC-40 the Mod4Win (or similar) showed 17% CPU usage
>> when playing MODs in Windows... and since it's expired I haven't tested
>> it on my 486DX-66, but I'd guess around 10-15% CPU usage. This is not a
>> big percentage to lose to pitch-bend digital samples!
>>
>> It must be the fact that it's (currently) _something_ that the
>> GUS can do that the AWE can't do that makes it so "important."
>
>While it may seem that the desire for a MOD player for the AWE32 is just
>a childish race for superiority, this truly isn't the case. When you
>were using 17% CPU on your older 486, what sound card was the MOD
>playing through? I would expect that it was a mono 8-bit card, but
>mixing samples can actually tie up a significant portion of the CPU.
>For example, trying to play a standard 4-channel MOD through a Sound
>Blaster 16 in stereo at 44KHz can eat up as much as 33% of the CPU
>cycles.

Wrong. 8 bit and 16 bit mixing takes the same amount of time, as
8 bit mixing is essentially 16 bit mixing with discarding the lower
8 bits afterwards. And mono and stereo mixing is the same too.
Whether you mix 4 channels into one or 2 into one twice is
computationally really the same. The only thing that does matter is
the mixing speed (sample rate). It's directly proportional to the
CPU usage. And of course if you bring in panning and/or IDO (T)
[Oversampling] the CPU load will increase. When you get 33% CPU
load for a 4 channel MOD you must have a 386/33 or something like
that.

>-Jason Fourier

Jensi

Jensi

unread,
Oct 12, 1994, 5:40:34 AM10/12/94
to
In article <37dirt$l...@cc.tut.fi>,

;-) hehe, but for those 85% you get some damn good music, not?

>Anssi

Jensi

David E. Weekly

unread,
Oct 10, 1994, 5:22:35 PM10/10/94
to
In article <Fr9ckerQ...@netcom.com> con...@netcom.com (Lord Soth) writes:
>
>In article <373nep$f...@proffa.cc.tut.fi>, s10...@cc.tut.fi (Anssi Saari) wrote:
>> :> On my old 486DLC-40 the Mod4Win (or similar) showed 17% CPU usage
>> :>when playing MODs in Windows... and since it's expired I haven't tested
>> :>it on my 486DX-66, but I'd guess around 10-15% CPU usage. This is not a
>> :>big percentage to lose to pitch-bend digital samples!
>>
>> Mod4Win managed to choke on a 26-channel mtm on my 66, i.e. over
>> 100% CPU usage.
>
>Really? I've never had any problems with any file, including 26+ channel
>S3M's and MTM's.
>

Funny, on *my* DX/2 66 w/8MB RAM Mod4Win chokes badly (110%) trying to
mix a 12-channel .FAR @ 40MHz, 16-bit (Surround, IDO, Pan, Preamp on)!
For your average .MOD or .S3M, I can only get away with (65%) about 35KHz
with all options on! This 26-channel stuff seems sort of odd,
considering I've got the same processor as you guys...Perhaps you
people are mixing @ 30MHz, 8-bit with all the bells and whistles off?
Please reply with your settings when running these mods. 13% sounds
ridiculous to me, unless all the settings are way down...

Thanx,
Dave

--
David Weekly | GenRad's Webmaster
d...@genrad.com | 15-year-old -- 11th grade
mail for PGP key | multiplatform programmer
QB,VB,Think C,Turbo C,TASM,Think Pascal,csh...

Samuel Marshall

unread,
Oct 15, 1994, 4:33:54 PM10/15/94
to
I think you're allowed any sample frequency, well, up to some limit anyway, you
just have to choose the 'wrong' note names (eg if you double the sample frequency
you must type C4 when you want C5). I'm not at all sure about this. But there
definitely are differences between MODs. On a recent magazine cover cd, the MOD
with Project X sounded great. But those with Pinball Dreams were rubbish (like
the game). Obviously, i'm referring to technical quality not music, which is a
matter of opinion anyway.

Summary: you can use a variety of frequencies.

On the other point, I think the quoted 16k is sample freq, because computer
people almost never talk about actual bandwidth.

Sam

Josh Rodman

unread,
Oct 18, 1994, 10:26:12 PM10/18/94
to
In article <37gbi7$b...@newstand.syr.edu> jpuc...@gamera.syr.edu (Jensi) writes:
>From: jpuc...@gamera.syr.edu (Jensi)
>Subject: Re: MOD player(s) for the AWE32
>Date: 12 Oct 1994 09:51:02 GMT

> And mono and stereo mixing is the same too.
>Whether you mix 4 channels into one or 2 into one twice is
>computationally really the same. The only thing that does matter is
>the mixing speed (sample rate). It's directly proportional to the
>CPU usage. And of course if you bring in panning and/or IDO (T)

Well, some (myself included) would argue that "stereo" implies panning.

I mean, you've got a pretty feeble stereo if a sound can only exist in the
left channel OR the right channel.. but of course you're righy, _technically_.

>Jensi


Jensi

unread,
Oct 21, 1994, 2:22:12 AM10/21/94
to
In article <wuti.15....@panix.com> wu...@panix.com (Josh Rodman) writes:
>In article <37gbi7$b...@newstand.syr.edu> jpuc...@gamera.syr.edu (Jensi) writes:
>>From: jpuc...@gamera.syr.edu (Jensi)
>>Subject: Re: MOD player(s) for the AWE32
>>Date: 12 Oct 1994 09:51:02 GMT
>
>> And mono and stereo mixing is the same too.
>>Whether you mix 4 channels into one or 2 into one twice is
>>computationally really the same. The only thing that does matter is
>>the mixing speed (sample rate). It's directly proportional to the
>>CPU usage. And of course if you bring in panning and/or IDO (T)
>
>Well, some (myself included) would argue that "stereo" implies panning.

No, it does not. The whole set of ProTracker modules (Amiga) is in stereo
but without panning. Also many S3M's and MTM's don't use panning although
they easily could. It requires some additional skills to mix the song
right with panning, but anyway, stereo does not imply panning.

>I mean, you've got a pretty feeble stereo if a sound can only exist in the
>left channel OR the right channel.. but of course you're righy, _technically_.

That's why Mod4Win can add surround sound to modules that are stereo but
don't use panning.

>>Jensi
>
>


David E. Weekly

unread,
Oct 22, 1994, 5:41:28 PM10/22/94
to
In article <38a7hr$e...@newstand.syr.edu> jpuc...@rodan.syr.edu (Jensi) writes:
[chopped my comments about CPU usage]
>There is something wrong with your system. Something is hogging it down.
>You should get way lower CPU utilization than what you describe. I've
>heard about system where the DMA can't keep up and slows down the whole
>system. Anyway, a regular 4 channel MOD should give no more than 7% on
>your system (or < 14% with IDO) at 48 kHz 16 bit stereo.

Wow. This hurts. (I use a LOT MORE than 7% on 48Khz 16-bit MODs, even
with all bells off...) Any clue as to how this can be resolved? This
really sucks, as I cannot play anything in the background. (too much
CPU munching...:-/)

>Jensi

Bummed and Confused,

Jensi

unread,
Oct 26, 1994, 12:49:06 AM10/26/94
to

No, I don't know WHAT's wrong, but I know there IS something wrong. You can
try a different DMA channel, or go from an 8 bit DMA channel to a 16 bit one
or vice versa. If you get different CPU usages for 8 and 16 bit or for
mono and stereo there is something wrong. This should not matter.

Jensi

0 new messages