Question regarding current ESS Maestro 3M soundcard support

Question regarding current ESS Maestro 3M soundcard support

[Ivica Bukvic]

Hi all!

I have an Dell Inspiron 8K and my question is whether the current
Maestro M3 chip is fully supported or does it currently only furnish
SB16 compatibility mode (I might be talking nonsense here, but this is
what I heard a while ago, or something like it and my understanding is
that the current driver support is only partial due to incomplete
disclosure of the hw specs by the ESS, again, please correct me if I am
wrong). Basically, from my understanding it seems that the M3 soundcard
is being classified as a "sound accelerator" (whatever that means), and
as such I would expect it to be capable of mixing audio streams in
hardware (just as emu10k1 chip does, even though emu10k1 is much more
powerful). So far, I was only able to get one stream out of the card in
the latest ALSA driver, while I got 2 streams with the OSS-Free drivers
(third stream would garble the two existing streams and return an i/o
error).

So, is this the default behavior?

I would assume that most of the modern-day cards should be capable of
down-mixing at least 16 or 32 streams of sound, since that is what they
are usually capable of reproducing in Windows at any given time (unless
in Windows the down-mixing is done in software, which is certainly a
possibility and in many cases a known fact).

So, my questions are:

Is this a hardware limitation?

or

Is this current driver's limitation?

If the second is the case, then what are the chances of improving this
issue in the nearby future?

Finally, what are the chances of getting the ESS to talk and give us the
exact specs for this soundcard so that the driver can be finished?

Any help would be greatly appreciated!

Ivica Ico Bukvic, composer, multimedia sculptor, 
programmer, webmaster & computer consultant 
http://meowing.ccm.uc.edu/~ico/ 
============================
"To be is to do"   - Socrates
"To do is to be"   - Sartre
"Do be do be do"   - Sinatra
"I am"             - God



_______________________________________________
Alsa-devel mailing list
Alsa-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-devel

Re: Question regarding current ESS Maestro 3M soundcard support

[Takashi Iwai]

Hi,

At Mon, 11 Feb 2002 21:28:53 -0500,
Ivica Bukvic wrote:
> 
> Hi all!
> 
> I have an Dell Inspiron 8K and my question is whether the current
> Maestro M3 chip is fully supported or does it currently only furnish
> SB16 compatibility mode (I might be talking nonsense here, but this is
> what I heard a while ago, or something like it and my understanding is
> that the current driver support is only partial due to incomplete
> disclosure of the hw specs by the ESS, again, please correct me if I am
> wrong). Basically, from my understanding it seems that the M3 soundcard
> is being classified as a "sound accelerator" (whatever that means), and
> as such I would expect it to be capable of mixing audio streams in
> hardware (just as emu10k1 chip does, even though emu10k1 is much more
> powerful). So far, I was only able to get one stream out of the card in
> the latest ALSA driver, while I got 2 streams with the OSS-Free drivers
> (third stream would garble the two existing streams and return an i/o
> error).

So it is on ALSA, too.
Theoretically more could be, but it seems that the (only one) DSP code
provided from ESS has a bug.

 
> So, is this the default behavior?
 
Yes.


> I would assume that most of the modern-day cards should be capable of
> down-mixing at least 16 or 32 streams of sound, since that is what they
> are usually capable of reproducing in Windows at any given time (unless
> in Windows the down-mixing is done in software, which is certainly a
> possibility and in many cases a known fact).
> 
> So, my questions are:
> 
> Is this a hardware limitation?
 
No.

> or
> 
> Is this current driver's limitation?

Yes.
 
> If the second is the case, then what are the chances of improving this
> issue in the nearby future?
 
Only if we receive detailed information of DSP engine or better DSP
codes from ESS.


> Finally, what are the chances of getting the ESS to talk and give us the
> exact specs for this soundcard so that the driver can be finished?
 
Unfortunately ESS is famous to provide the only minimum information,
so far.  I hope this will change in future..


ciao,

Takashi

_______________________________________________
Alsa-devel mailing list
Alsa-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-devel

Re: Question regarding current ESS Maestro 3M soundcard support

[Steve Harris]

On Mon, Feb 11, 2002 at 09:28:53 -0500, Ivica Bukvic wrote:
> Hi all!
> 
> I have an Dell Inspiron 8K and my question is whether the current
> Maestro M3 chip is fully supported or does it currently only furnish
> SB16 compatibility mode (I might be talking nonsense here, but this is

I have an 8100 (Maestro 3i). The ALSA support is basicly OK, but things
like jack cause some problems with CPU use, I haven't been able to track it
down, but it might be something simple.

The OSS emaultion layer and simplex playback seem to work fine however.

I suspect that the output only alsa driver for jack would solve a lot of
my problems, and the mic/line input on the M3 is terrible anyway.

My medium term plan is to write a raw ethernet/IEEE-1394 driver+client
pair for jack that allows you to shunt float audio data over to the laptop
from a desktop (with some useful audio hardware in it) and back again.  I
will probably start with raw ethernet packets as I know how to do that, it
may have latency problems though. I have tried IP packets (with ll patched
linux 2.4 and 3com 905's) and that was no good.

The driver would run on the laptop and provide sync'd audio to the jack
clients and the client would run on the desktop machine as a sync slave of
the audio hardware.

- Steve

_______________________________________________
Alsa-devel mailing list
Alsa-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-devel

RE: Question regarding current ESS Maestro 3M soundcard support

[Ivica Bukvic]

Hi, I want to thank you very much for your comprehensive response. I've
written to the ess support e-mail pleading for more Linux support and
got no response (no surprise there). Is there any way that we can start
a petition or somehow force them to release the specs (even if needed
under NDA agreement to a few persons such as yourself who would be
capable of improving the driver)? Also, what are the chances for
reverse-engineering the Windows drivers to build a better Linux driver?

Ivica Ico Bukvic


> -----Original Message-----
> From: Takashi Iwai [mailto:tiwai@suse.de]
> Sent: Tuesday, February 12, 2002 6:14 AM
> To: Ivica Bukvic
> Cc: alsa-devel@alsa-project.org; alsa-user@alsa-project.org
> Subject: Re: [Alsa-devel] Question regarding current ESS Maestro 3M
> soundcard support
> 
> Hi,
> 
> At Mon, 11 Feb 2002 21:28:53 -0500,
> Ivica Bukvic wrote:
> >
> > Hi all!
> >
> > I have an Dell Inspiron 8K and my question is whether the current
> > Maestro M3 chip is fully supported or does it currently only furnish
> > SB16 compatibility mode (I might be talking nonsense here, but this
is
> > what I heard a while ago, or something like it and my understanding
is
> > that the current driver support is only partial due to incomplete
> > disclosure of the hw specs by the ESS, again, please correct me if I
am
> > wrong). Basically, from my understanding it seems that the M3
soundcard
> > is being classified as a "sound accelerator" (whatever that means),
and
> > as such I would expect it to be capable of mixing audio streams in
> > hardware (just as emu10k1 chip does, even though emu10k1 is much
more
> > powerful). So far, I was only able to get one stream out of the card
in
> > the latest ALSA driver, while I got 2 streams with the OSS-Free
drivers
> > (third stream would garble the two existing streams and return an
i/o
> > error).
> 
> So it is on ALSA, too.
> Theoretically more could be, but it seems that the (only one) DSP code
> provided from ESS has a bug.
> 
> 
> > So, is this the default behavior?
> 
> Yes.
> 
> 
> > I would assume that most of the modern-day cards should be capable
of
> > down-mixing at least 16 or 32 streams of sound, since that is what
they
> > are usually capable of reproducing in Windows at any given time
(unless
> > in Windows the down-mixing is done in software, which is certainly a
> > possibility and in many cases a known fact).
> >
> > So, my questions are:
> >
> > Is this a hardware limitation?
> 
> No.
> 
> > or
> >
> > Is this current driver's limitation?
> 
> Yes.
> 
> > If the second is the case, then what are the chances of improving
this
> > issue in the nearby future?
> 
> Only if we receive detailed information of DSP engine or better DSP
> codes from ESS.
> 
> 
> > Finally, what are the chances of getting the ESS to talk and give us
the
> > exact specs for this soundcard so that the driver can be finished?
> 
> Unfortunately ESS is famous to provide the only minimum information,
> so far.  I hope this will change in future..
> 
> 
> ciao,
> 
> Takashi



_______________________________________________
Alsa-devel mailing list
Alsa-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-devel

Re: Question regarding current ESS Maestro 3M soundcard support

[Steve Harris]

On Sat, Feb 16, 2002 at 10:46:05 -0500, Ivica Bukvic wrote:
> capable of improving the driver)? Also, what are the chances for
> reverse-engineering the Windows drivers to build a better Linux driver?

If all we need is the DSP code it could be possible to find it in the
Windows driver.

It may not help unless the DSP code is similar to what we have though.

- Steve

_______________________________________________
Alsa-devel mailing list
Alsa-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-devel