"VIA Announces Strategic Open Source Driver Development Initiative"

[Jeffrey 'jf' Lim]

http://www.via.com.tw/en/resources/pressroom/pressrelease.jsp?press_release_no=2088

would this be good news for the community? This is really mainly
Linux-related, but i'm hoping that their mention of "technical
documentation" will be good enough for Open to be able to support them...

-jf

--
In the meantime, here is your PSA:
"It's so hard to write a graphics driver that open-sourcing it would not
help."
-- Andrew Fear, Software Product Manager, NVIDIA Corporation
http://kerneltrap.org/node/7228

[Stuart Henderson]

On 2008-04-09, Jeffrey 'jf' Lim <jfs.world@...> wrote:
> http://www.via.com.tw/en/resources/pressroom/pressrelease.jsp?press_release_no=2088
>
> would this be good news for the community?

Too early to say, they haven't released anything yet.

[Theo de Raadt]

> http://www.via.com.tw/en/resources/pressroom/pressrelease.jsp?press_release_no=2088
>
> would this be good news for the community? This is really mainly
> Linux-related, but i'm hoping that their mention of "technical
> documentation" will be good enough for Open to be able to support them...

Developers don't need web sites.  They need pdf files documenting the
chips.

Contrast Via's web site to the following:

         http://wikis.sun.com/display/FOSSdocs/Home 

It took us a very long time to get Sun to do this, and it was totally
worth it.  It is kind of strange to us to have Sun suddenly be the
perfect example of openness.

Pay close attention to how VIA is only talking about their newest
flashiest chips, too.

[frantisek holop]

hmm, on Wed, Apr 09, 2008 at 11:25:25AM -0600, Theo de Raadt said that
> It took us a very long time to get Sun to do this, and it was totally
> worth it.  It is kind of strange to us to have Sun suddenly be the
> perfect example of openness.

a bit OT, but
i just had the pleasure of meeting and ex-sun employee, working
mostly on kernel stuff.  i dont know how similar the opensolaris
and solaris kernels are, but he said the solaris kernel code is
a beauty to read, and simplicity and readibility are adhered
to fanatically...  that reminds me another dev community :o)

as i read some of the sun employees blogs, i think there might be
quite some similarities between the two dev cultures (hope this
doesnt insult too much people), it's just that sun is a company...
and that alone ties a lot of hands (as we all know)...

-f
--
suicidal twin kills sister by mistake!

[BOFH-5]

On Wed, Apr 9, 2008 at 3:07 PM, frantisek holop <minusf@...> wrote:

> a bit OT, but
> i just had the pleasure of meeting and ex-sun employee, working
> mostly on kernel stuff.  i dont know how similar the opensolaris
> and solaris kernels are, but he said the solaris kernel code is
> a beauty to read, and simplicity and readibility are adhered
> to fanatically...  that reminds me another dev community :o)


Sun learnt a lot of lessons when it tried to merge sparc and x86 code bases
together around the solaris 2.4 time, iirc.  That's why things like zfs are
endian neutral.  OpenBSD started in the multi cpu world to begin with.



--
http://www.glumbert.com/media/shift
http://www.youtube.com/watch?v=tGvHNNOLnCk
"This officer's men seem to follow him merely out of idle curiosity." --
Sandhurst officer cadet evaluation.
"Securing an environment of Windows platforms from abuse - external or
internal - is akin to trying to install sprinklers in a fireworks factory
where smoking on the job is permitted." -- Gene Spafford
learn french: http://www.youtube.com/watch?v=j1G-3laJJP0&feature=related

[frantisek holop]

hmm, on Wed, Apr 09, 2008 at 03:35:18PM -0400, bofh said that
> Sun learnt a lot of lessons when it tried to merge sparc and x86 code bases
> together around the solaris 2.4 time, iirc.  That's why things like zfs are
> endian neutral.  OpenBSD started in the multi cpu world to begin with.

i might be wrong, but i thought as of yet, not everything
is endian neutral in openbsd (carp?)

-f
--
you don't have to be a cannibal to get fed up with people.

[Zbigniew Baniewski]

On Wed, Apr 09, 2008 at 09:07:08PM +0200, frantisek holop wrote:

> > It took us a very long time to get Sun to do this, and it was totally
> > worth it.  It is kind of strange to us to have Sun suddenly be the
> > perfect example of openness.

So, perhaps the best audio-option would be something using VIA Envy24(HT) -
which is reportedly better than Audigy(2)? Time to swap?
--
                                pozdrawiam / regards

                                                Zbigniew Baniewski

[j4nKy]

On Wed, Apr 09, 2008 at 10:57:05PM +0200, Zbigniew Baniewski wrote:
> On Wed, Apr 09, 2008 at 09:07:08PM +0200, frantisek holop wrote:
>
> > > It took us a very long time to get Sun to do this, and it was totally
> > > worth it.  It is kind of strange to us to have Sun suddenly be the
> > > perfect example of openness.
>
> So, perhaps the best audio-option would be something using VIA Envy24(HT) -
> which is reportedly better than Audigy(2)? Time to swap?

envy(4) already exists in -current (and will be in 4.3).  doesn't support
the HT version though.

--
jakemsr@...
SDF Public Access UNIX System - http://sdf.lonestar.org

[Zbigniew Baniewski]

On Wed, Apr 09, 2008 at 10:49:07PM +0000, Jacob Meuser wrote:

> > So, perhaps the best audio-option would be something using VIA Envy24(HT) -
> > which is reportedly better than Audigy(2)? Time to swap?
>
> envy(4) already exists in -current (and will be in 4.3).  doesn't support
> the HT version though.

Yes, I noticed it's there - but does the driver support all of the available
capabilities? The "VIA opening" won't be of any help in this particular case?
--
                                pozdrawiam / regards

                                                Zbigniew Baniewski

[j4nKy]

On Thu, Apr 10, 2008 at 01:11:47AM +0200, Zbigniew Baniewski wrote:

> On Wed, Apr 09, 2008 at 10:49:07PM +0000, Jacob Meuser wrote:
>
> > > So, perhaps the best audio-option would be something using VIA Envy24(HT) -
> > > which is reportedly better than Audigy(2)? Time to swap?
> >
> > envy(4) already exists in -current (and will be in 4.3).  doesn't support
> > the HT version though.
>
> Yes, I noticed it's there - but does the driver support all of the available
> capabilities?

according to BUGS in envy(4), no.  but emu(4) doesn't support all
the features of the emu10k1 chips, either.

> The "VIA opening" won't be of any help in this particular case?

at least some datasheets are/have been available:

http://envy24.svobodno.com/datasheets/

--
jakemsr@...
SDF Public Access UNIX System - http://sdf.lonestar.org

[Zbigniew Baniewski]

On Thu, Apr 10, 2008 at 12:08:26AM +0000, Jacob Meuser wrote:

> > Yes, I noticed it's there - but does the driver support all of the available
> > capabilities?
>
> according to BUGS in envy(4), no.  but emu(4) doesn't support all
> the features of the emu10k1 chips, either.

I understand - but the mentioned "VIA opening" is suggesting, that perhaps
completing the envy driver can be much easier, if VIA will release the docs;
Creative Labs, unfortunately, still doesn't seem to be willing to.

I'm not sure, nevertheless, if that envy24-related docs is enough; there are
some other chips on the envy-fitted cards, anyway.

> > The "VIA opening" won't be of any help in this particular case?
>
> at least some datasheets are/have been available:
>
> http://envy24.svobodno.com/datasheets/

I think, I'll have to make a comparison with Audigy soon...  ;) as I can
see, there are even (semi?)professional cards built using Envy; like f.e.
this one: http://www.ixbt.com/multimedia/esi-juli@...
--
                                pozdrawiam / regards

                                                Zbigniew Baniewski

[Jeffrey 'jf' Lim]

On Thu, Apr 10, 2008 at 9:41 AM, Zbigniew Baniewski <zb@...> wrote:

> On Thu, Apr 10, 2008 at 12:08:26AM +0000, Jacob Meuser wrote:
>
> > > Yes, I noticed it's there - but does the driver support all of the
> available
> > > capabilities?
> >
> > according to BUGS in envy(4), no.  but emu(4) doesn't support all
> > the features of the emu10k1 chips, either.
>
> I understand - but the mentioned "VIA opening" is suggesting, that perhaps
> completing the envy driver can be much easier, if VIA will release the
> docs;
> Creative Labs, unfortunately, still doesn't seem to be willing to.
>

oh it's more than that! Creative: the company that sues you for your
drivers. And gets to decide which features it will want to enable its
drivers for you, the consumer. How's that for a creative perspective on the
rights of the customer!

http://hardware.slashdot.org/hardware/08/03/29/046201.shtml


-Jeff

--
In the meantime, here is your PSA:
"It's so hard to write a graphics driver that open-sourcing it would not
help."
-- Andrew Fear, Software Product Manager, NVIDIA Corporation
http://kerneltrap.org/node/7228

[Peter_APIIT]

Jeffrey 'jf' Lim wrote:

http://www.via.com.tw/en/resources/pressroom/pressrelease.jsp?press_release_no=2088

would this be good news for the community? This is really mainly
Linux-related, but i'm hoping that their mention of "technical
documentation" will be good enough for Open to be able to support them...

-jf

--
In the meantime, here is your PSA:
"It's so hard to write a graphics driver that open-sourcing it would not
help."
-- Andrew Fear, Software Product Manager, NVIDIA Corporation
http://kerneltrap.org/node/7228

Good news. I will support VIA. Keep up the good works.

[Hannah Schroeter]

Hi!

On Wed, Apr 09, 2008 at 10:12:49PM +0200, frantisek holop wrote:
>hmm, on Wed, Apr 09, 2008 at 03:35:18PM -0400, bofh said that
>> Sun learnt a lot of lessons when it tried to merge sparc and x86 code bases
>> together around the solaris 2.4 time, iirc.  That's why things like zfs are
>> endian neutral.  OpenBSD started in the multi cpu world to begin with.

>i might be wrong, but i thought as of yet, not everything
>is endian neutral in openbsd (carp?)

FFS itself (the on-disk layout). In contrary, ext2 *is*.

>-f

Kind regards,

Hannah.

[Alexandre Ratchov-2]

On Thu, Apr 10, 2008 at 03:41:54AM +0200, Zbigniew Baniewski wrote:

> On Thu, Apr 10, 2008 at 12:08:26AM +0000, Jacob Meuser wrote:
>
> > > Yes, I noticed it's there - but does the driver support all of the available
> > > capabilities?
> >
> > according to BUGS in envy(4), no.  but emu(4) doesn't support all
> > the features of the emu10k1 chips, either.
>
> I understand - but the mentioned "VIA opening" is suggesting, that perhaps
> completing the envy driver can be much easier, if VIA will release the docs;
> Creative Labs, unfortunately, still doesn't seem to be willing to.
>

besides the MIDI port and the world clock the envy24 chip support
is quite complete. Unfortunately that doesn't mean that all
envy-based cards are fully usable.

 - first, envy24 is a generic digital only chip; it's connected to
   up to 4 codecs that do the analog<->digital conversions and that
   hold the gain knobs. So to add support for a new cards we must
   add support for its codecs, and we need to know how these codecs
   are wired to the envy24 chip, how gpio pins are used, etc...
   (this may require docs from the sound card manufacturer, not
   via)

 - second, there are limitations in most audio apps and in our
   audio(4) device that makes envy24-based cards hard to use (eg.
   lack of 24/32-bit encoding or 10/12 channel support). IMO, this
   is the most urgent to solve.

> I'm not sure, nevertheless, if that envy24-related docs is enough; there are
> some other chips on the envy-fitted cards, anyway.
>
> > > The "VIA opening" won't be of any help in this particular case?
> >
> > at least some datasheets are/have been available:
> >
> > http://envy24.svobodno.com/datasheets/
>
> I think, I'll have to make a comparison with Audigy soon...  ;) as I can
> see, there are even (semi?)professional cards built using Envy; like f.e.
> this one: http://www.ixbt.com/multimedia/esi-juli@...

afaik, these cards are based on envy24ht, not envy24.

-- Alexandre

[Zbigniew Baniewski]

On Thu, Apr 10, 2008 at 09:36:51PM +0200, Alexandre Ratchov wrote:

>  - first, envy24 is a generic digital only chip; it's connected to
>    up to 4 codecs that do the analog<->digital conversions and that
>    hold the gain knobs. So to add support for a new cards we must
>    add support for its codecs, and we need to know how these codecs
>    are wired to the envy24 chip, how gpio pins are used, etc...
>    (this may require docs from the sound card manufacturer, not
>    via)

That's I was afraid of.

> afaik, these cards are based on envy24ht, not envy24.

What do you think about (much cheaper) Chaintech AV-710? There's a version
with envy24... perhaps someone's using this under OpenBSD?

http://icrontic.com/articles/chaintech_av710_71_audio_card_review
http://techgage.com/article/chaintech_av-710_71_sound_card/
http://www.sudhian.com/index.php?/articles/show/654
--
                                pozdrawiam / regards

                                                Zbigniew Baniewski

[Alexandre Ratchov-2]

On Thu, Apr 10, 2008 at 09:47:37PM +0200, Zbigniew Baniewski wrote:

> On Thu, Apr 10, 2008 at 09:36:51PM +0200, Alexandre Ratchov wrote:
>
> >  - first, envy24 is a generic digital only chip; it's connected to
> >    up to 4 codecs that do the analog<->digital conversions and that
> >    hold the gain knobs. So to add support for a new cards we must
> >    add support for its codecs, and we need to know how these codecs
> >    are wired to the envy24 chip, how gpio pins are used, etc...
> >    (this may require docs from the sound card manufacturer, not
> >    via)
>
> That's I was afraid of.
>

well, if both codecs and the digital chip are well documented, how
they are connected is not too hard to guess. There's an EEPROM that
gives hints.

> > afaik, these cards are based on envy24ht, not envy24.
>
> What do you think about (much cheaper) Chaintech AV-710? There's a version
> with envy24... perhaps someone's using this under OpenBSD?
>
> http://icrontic.com/articles/chaintech_av710_71_audio_card_review
> http://techgage.com/article/chaintech_av-710_71_sound_card/
> http://www.sudhian.com/index.php?/articles/show/654
>

according to the second link, it uses envy24HT so it will not work
with the current envy(4) driver. FYI envy24 is also known as VT1712
or ICE1712. Esi-julia and AV-710 seem to use the VT1721.

For a sound card (beside being supported) the most important is the
analog part, that will determine the sound quality, the esi julia
seems quite promising in this respect.

Nevertheless, if i one day I get one, I'll happily work on the "HT"
driver ;)

-- Alexandre

[Zbigniew Baniewski]

On Thu, Apr 10, 2008 at 10:25:50PM +0200, Alexandre Ratchov wrote:

> well, if both codecs and the digital chip are well documented, how
> they are connected is not too hard to guess. There's an EEPROM that
> gives hints.

You're right: "if".  ;)

But found some more info about the other chips:
http://wiki.hydrogenaudio.org/index.php?title=Chaintech_AV-710

> > > afaik, these cards are based on envy24ht, not envy24.
> >
> > What do you think about (much cheaper) Chaintech AV-710? There's a version
> > with envy24... perhaps someone's using this under OpenBSD?
> >
> > http://icrontic.com/articles/chaintech_av710_71_audio_card_review
> > http://techgage.com/article/chaintech_av-710_71_sound_card/
> > http://www.sudhian.com/index.php?/articles/show/654
> >
>
> according to the second link, it uses envy24HT so it will not work
> with the current envy(4) driver. FYI envy24 is also known as VT1712
> or ICE1712. Esi-julia and AV-710 seem to use the VT1721.

Perhaps I misunderstood that test at "icrontic" - there was a comparison of
the chips, and this was suggesting, that there are four versions of the
card; probably wrong conclusion.

The testers are publishing a bit contradictory informations: f.e. on the
page:  http://techgage.com/article/chaintech_av-710_71_sound_card

First you'll find: "VIA ENVY 24PT", several verses down a remark: "The heart
of the card is the Envy24 HT-S Chipset" - with a photo on the side. A photo
of... ENVY 24PT. Immediately below - image of ENVY 24HT-S.   :-O

What a pity; the card has quite good reviews. OK, must look further...
--
                                pozdrawiam / regards

                                                Zbigniew Baniewski

[Zbigniew Baniewski]

Found it - looks good, but it's an expensive one  :/  what do you think about
that other chips? Are they supported presently?

http://www.digit-life.com/articles/maudioaudiophile/

* main chip - multichannel PCI controller ENVY24 from IC Ensemble;
* I2S stereo codec AKM AK4528VF with the 24bit/96kHz DAC and ADC;
* CS8427 digital transceiver;

Also: http://www25.big.jp/~jam/audiocard/audiophile/
--
                                pozdrawiam / regards

                                                Zbigniew Baniewski

[Zbigniew Baniewski]

I would to ask about the issue to be found under Linux - is it valid for
OpenBSD's "audio" too?

http://www.diyaudio.com/forums/showthread.php?threadid=93315

"The latest versions of ALSA which are included with Ubuntu Edgy, and I
think Dapper Drake as well, will resample all audio to 48kHz if your
soundcard does not support hardware mixing. This is also true if the driver
doesn't support hardware mixing. As far as I can tell, there is absolutely
no support for hardware mixing with any of the Envy24 chips in Linux. The
problem with this resampling is that by default ALSA uses a poor resampling
algorithm to save CPU usage, and destroys the quality of everything played
back. ALSA uses this software mixing and resampling in order to let more
than one application play audio at the same time. I have found a solution to
the audio quality issue however. [..]"
--
                                pozdrawiam / regards

                                                Zbigniew Baniewski