🚀 go-pugleaf

In article <1993Jan27.064552.43267@ns1.cc.lehigh.edu> ry01@ns1.cc.lehigh.edu (ROBERT YUNG) writes:
>I remember reading that the GUS can't load new patches into memory while it's
>playing music. I'm just wondering if the LAPC-1 can accept new music patches
>while it's playing. If not, then if this doesn't really affect Roland sound
>quality, then ditto for GUS, right?
>
>Anyway, doesn't the LAPC-1 have only 128K of ROM to hold all its music? They
>sound great and I think they're 16bit. Now why does the GUS need so much more
>memory to hold music (ie: up to 1mb)??
>

The LAPC-1 only stores the attack transient as samples... the sustain
is handled via FM synthesis.  The GUS uses samples for the sustain
as well as the attack, and thus, requires more memory.

>And has anyone heard the GUS try 32 voices at once? I jst want to know how
>much of a strain, if any, does this have on the system ( 486/66 users
>excluded... :-)

There will be no strain on the CPU, since all the voices are handled by
the GF1 chip.

Phat.

In article <C1J0r1.Fr1@watserv2.uwaterloo.ca> ptran@sciborg.uwaterloo.ca (Phat H Tran) writes:
>In article <1993Jan27.064552.43267@ns1.cc.lehigh.edu> ry01@ns1.cc.lehigh.edu (ROBERT YUNG) writes:
>>Anyway, doesn't the LAPC-1 have only 128K of ROM to hold all its music? They
>>sound great and I think they're 16bit. Now why does the GUS need so much more
>>memory to hold music (ie: up to 1mb)??
>>
>
>The LAPC-1 only stores the attack transient as samples... the sustain
>is handled via FM synthesis.  The GUS uses samples for the sustain
>as well as the attack, and thus, requires more memory.

I have an LAPC-1, while I keep hearing about the GUS.
Can anyone tell me from experience:  in native mode, how does the GUS sound
compared to the LAPC-1 (or SCC-1)?  If it's as good or better, how does it
arrange to be half the price?  If it's not, why/how is it better than a
boring Sound Blaster Pro?

-- 
This is NOT a .sig virus do NOT copy me into your .sig file  This is NOT
* Daniel Starr                   *   Ta twn thewn erga                 * 
* dstarr@minerva.cis.yale.edu    *   Hoi psukhwn twn anthrwpwn logoi.  *
a .sig virus do NOT copy me into your .sig file This is NOT a .sig virus

In article <C1J0r1.Fr1@watserv2.uwaterloo.ca> ptran@sciborg.uwaterloo.ca (Phat H Tran) writes:
>In article <1993Jan27.064552.43267@ns1.cc.lehigh.edu> ry01@ns1.cc.lehigh.edu (ROBERT YUNG) writes:
>>I remember reading that the GUS can't load new patches into memory while it's
>>playing music.

BZZZT! wrong. I did it two hours ago. The SDK doesn't have a function for that,
but you can definitely load patches dynamically while playing.

>>I'm just wondering if the LAPC-1 can accept new music patches
>>while it's playing. If not, then if this doesn't really affect Roland sound
>>quality, then ditto for GUS, right?

The GUS can play real samples. The LAPC-1 uses LA synthesis, so it has nothing
to do. The GUS sounds better than LAPC-1 and MT-32. With the proper patches
it can also sound better than a SCC-1.
I'm working on a modplayer with a full graphical interface and it will be able
to medley MOD files (play one mod while loading another, play the other, load
another etc...). And it will use less than 256K, even when playing two 500K
mod files. You cant get that with any other card.

>>Anyway, doesn't the LAPC-1 have only 128K of ROM to hold all its music? They
>>sound great and I think they're 16bit. Now why does the GUS need so much more
>>memory to hold music (ie: up to 1mb)??
>
>The LAPC-1 only stores the attack transient as samples... the sustain
>is handled via FM synthesis.  The GUS uses samples for the sustain
>as well as the attack, and thus, requires more memory.

Exactly. And try having the LAPC playing a note which plays voice (i'll upload
today a patch with the phrase James Brown is dead, and another with 
"techno music". 

>>And has anyone heard the GUS try 32 voices at once? I jst want to know how
>>much of a strain, if any, does this have on the system ( 486/66 users
>>excluded... :-)

Even a XT can handle it.

>There will be no strain on the CPU, since all the voices are handled by
>the GF1 chip.

But you still need the CPU to tell the GF1 to play a note. But as i said
unless you have a 8000 beat per minute song, even a XT can handle that.

Ciao,
--
Francois Dion
    '  _   _   _ 
 CISM (_) (_)  _) FM       Montreal , Canada       Email: CISM@ERE.UMontreal.CA
      (_)  / . _)             10000 Watts          Telephone no: (514) 343-7511
_______________________________________________________________________________
Audio-C-DJ-Fractals-Future-Label-Multimedia-Music-Radio-Rave-Video-VR-Volvo-...

In article <1k6sriINNntn@MINERVA.CIS.YALE.EDU> starr-daniel@yale.edu (Daniel Starr) writes:
>
>I have an LAPC-1, while I keep hearing about the GUS.
>Can anyone tell me from experience:  in native mode, how does the GUS sound
>compared to the LAPC-1 (or SCC-1)?  If it's as good or better, how does it

Better than LAPC-1 with actual patches and better/equivalent/less than SCC1
depending on the patches. So with the new patches coming i would say it
should be better. Anyway it is better since you can always change any patch
that doesn't fit the bill.

>arrange to be half the price?  If it's not, why/how is it better than a

There are only 2 expensive chips for playback: the GF1 (a modified Ensoniq
DOC II chip) and a stereo DAC. The GF1 does all the work and sends the
final serial stream to the DAC. For recording they cut corners by putting
an 8 bit ADC. The option board has a 16 bit ADC. But then the Rolands dont
have recording capabilities nor digital channels for audio.

>boring Sound Blaster Pro?

Sound quality is far better. It totally free the CPU of sound processing.
It has real sounds, not 2 or 4 op FM. It has 16 bit DACs. A good example
is that Play669 runs fine on a 286 while you would need a 486-66 to achieve
the same sampling rate with a SoundBlaster (and you still dont get the same
sound quality. By far.)

Ciao,
--
Francois Dion
    '  _   _   _ 
 CISM (_) (_)  _) FM       Montreal , Canada       Email: CISM@ERE.UMontreal.CA
      (_)  / . _)             10000 Watts          Telephone no: (514) 343-7511
_______________________________________________________________________________
Audio-C-DJ-Fractals-Future-Label-Multimedia-Music-Radio-Rave-Video-VR-Volvo-...

dionf@ERE.UMontreal.CA (Francois Dion) writes:

>In article <1k6sriINNntn@MINERVA.CIS.YALE.EDU> starr-daniel@yale.edu (Daniel Starr) writes:
>>boring Sound Blaster Pro?

>Sound quality is far better. It totally free the CPU of sound processing.
>It has real sounds, not 2 or 4 op FM. It has 16 bit DACs. A good example
>is that Play669 runs fine on a 286 while you would need a 486-66 to achieve
>the same sampling rate with a SoundBlaster (and you still dont get the same
>sound quality. By far.)

Well, depends on playing software... My player system (Digital Sound & Music
Interface = DSMI) can also handle up to 32 channels on a SB/SBPro/PAS+/PAS16
and I must say that the sound quality on PAS 16 (using 16-bit accuracy in
mixing) is the same as with GUS. And playing a four channel mod at 22kHz
takes only about 4% of processing power on my 486/33... Not too bad.
We software developers cannot forget that the installed base for eg. PAS 16
is much greater thatn GUS (for now), so it's important to develop support
for these 'old' systems too. Of course I'll add support for GUS when I get
one of those cards, because DSMI is very easily expandable, but in the mean
time I will enjoy superb quality MODs playing through my PAS 16 :-)

>Francois Dion

 _   _
(_).(_.

In article <lotter.142.728223006@cc.und.ac.za> lotter@cc.und.ac.za (Eddie Lotter) writes:
>ptran@sciborg.uwaterloo.ca (Phat H Tran) writes:
>
>>The LAPC-1 only stores the attack transient as samples... the sustain
>>is handled via FM synthesis.  The GUS uses samples for the sustain
>>as well as the attack, and thus, requires more memory.
>
>That's "LA synthesis". The LAPC does not use FM synthesis at all.

From what I understand, "LA synthesis" is just Roland's name for their
metd of combining digital and analog techniques.  ThThe use of "partials"
in the attack of an instrument sound gives a nice bite, but it's combined
with FM synthesis stuff to round it out  You'll find the same kind of
operator arrangement is available, though with considerably more flexibility.
-- 
Dave Seah (seah@ee.rochester.edu, AFCDaveS@aol.com)
That's Rochester, New York!