gatos-devel Mailing List for GATOS

Hi,
Thursday, 30 October, 2003, 1:49:40, Robert Birch wrote:

> In X, do I let it auto-detect which driver to use for the video? Right now
> it uses radeon driver. Do I just add the tv-out option in the Display
> section of X Configuration?
Well, yes, that is what I suppose (I can't try it myself however)
-- 
Best regards,
 Nikolai Zhubr

> Thanks,
> Robert Birch


> -----Original Message-----
> From: Nikolai Zhubr [mailto:s0...@ho...] 
> Sent: Wednesday, October 29, 2003 3:17 PM
> To: Robert Birch
> Cc: gat...@li...
> Subject: Re: [GATOS]Compiling the ATI.2 w/tv-output


> Hi,
> Wednesday, 29 October, 2003, 21:18:51, Robert Birch wrote:

>> I installed RH9 on a different machine and did what you said. 
>> Everything worked perfectly. Thanks.

>> My only question is, I wrote down what files it moved into the proper 
>> directories. I see it is replacing a fair bit of files with different 
>> ones. I'm guessing these are all updated by the ati.2 code. Which 
>> driver do I pick for my card now in X?
> Hmm, not sure what you mean exactly. I think you need to replace (by your
> freshly compiled ones) ati_drv.o, atimisc_drv.o, r128_drv.o and radeon_drv.o
> (they are in modules/) and probably bt829_drv.o and friends (in
> multimedia/). As for X configuration, you'll only need to add appropriate
> tv-out option(s) or otherwise tv will probably be disabled.

In X, do I let it auto-detect which driver to use for the video? Right =
now
it uses radeon driver. Do I just add the tv-out option in the Display
section of X Configuration?

Thanks,
Robert Birch


-----Original Message-----
From: Nikolai Zhubr [mailto:s0...@ho...]=20
Sent: Wednesday, October 29, 2003 3:17 PM
To: Robert Birch
Cc: gat...@li...
Subject: Re: [GATOS]Compiling the ATI.2 w/tv-output


Hi,
Wednesday, 29 October, 2003, 21:18:51, Robert Birch wrote:

> I installed RH9 on a different machine and did what you said.=20
> Everything worked perfectly. Thanks.

> My only question is, I wrote down what files it moved into the proper=20
> directories. I see it is replacing a fair bit of files with different=20
> ones. I'm guessing these are all updated by the ati.2 code. Which=20
> driver do I pick for my card now in X?
Hmm, not sure what you mean exactly. I think you need to replace (by =
your
freshly compiled ones) ati_drv.o, atimisc_drv.o, r128_drv.o and =
radeon_drv.o
(they are in modules/) and probably bt829_drv.o and friends (in
multimedia/). As for X configuration, you'll only need to add =
appropriate
tv-out option(s) or otherwise tv will probably be disabled.
--=20
Best regards,
 Nikolai Zhubr

> Thanks,
> Robert Birch

> -----Original Message-----
> From: Nikolai Zhubr [mailto:s0...@ho...]
> Sent: Thursday, October 23, 2003 2:29 AM
> To: Robert Birch
> Cc: gat...@li...
> Subject: Re: [GATOS]Compiling the ATI.2 w/tv-output


> Hi
> Wednesday, 22 October, 2003, 20:33:39, Robert Birch wrote:
>> XFree86 was installed by the RH9 installation.

>> Do I need to recompile XFree86 4.3 for this to work? I tried =
searching=20
>> for an xc directory but didn't see it. (Could have missed it)
> rpmbuild --recompile XFree86-4.3.0-2.src.rpm
> If you don't need to really install this full copy (and probably you
don't)
> then you need not actually wait until the compilation is finished
> completely, but ctrl-C it after all of the necessary makefiles are =
done
(it
> will still take several minutes), then
> /usr/src/redhat/BUILD/XFree86-4.3.0-2/xc is your source tree to =
compile
> ati.2 against. After ati.2 is done you may just remove the regular =
tree
from
> /usr/src/redhat/BUILD HTH.

Hi,
Wednesday, 29 October, 2003, 21:18:51, Robert Birch wrote:

> I installed RH9 on a different machine and did what you said. Everything
> worked perfectly. Thanks.

> My only question is, I wrote down what files it moved into the proper
> directories. I see it is replacing a fair bit of files with different ones.
> I'm guessing these are all updated by the ati.2 code. Which driver do I pick
> for my card now in X? 
Hmm, not sure what you mean exactly. I think you need to replace (by
your freshly compiled ones) ati_drv.o, atimisc_drv.o, r128_drv.o and
radeon_drv.o (they are in modules/) and probably bt829_drv.o and
friends (in multimedia/). As for X configuration, you'll only need
to add appropriate tv-out option(s) or otherwise tv will probably be
disabled.
-- 
Best regards,
 Nikolai Zhubr

> Thanks,
> Robert Birch

> -----Original Message-----
> From: Nikolai Zhubr [mailto:s0...@ho...] 
> Sent: Thursday, October 23, 2003 2:29 AM
> To: Robert Birch
> Cc: gat...@li...
> Subject: Re: [GATOS]Compiling the ATI.2 w/tv-output


> Hi
> Wednesday, 22 October, 2003, 20:33:39, Robert Birch wrote:
>> XFree86 was installed by the RH9 installation.

>> Do I need to recompile XFree86 4.3 for this to work? I tried searching 
>> for an xc directory but didn't see it. (Could have missed it)
> rpmbuild --recompile XFree86-4.3.0-2.src.rpm
> If you don't need to really install this full copy (and probably you don't)
> then you need not actually wait until the compilation is finished
> completely, but ctrl-C it after all of the necessary makefiles are done (it
> will still take several minutes), then
> /usr/src/redhat/BUILD/XFree86-4.3.0-2/xc is your source tree to compile
> ati.2 against. After ati.2 is done you may just remove the regular tree from
> /usr/src/redhat/BUILD HTH.

UPDATE:: Sent Nikolai a message that I now had the picture but it was
distorted.. That was because it was set to PAL.. Set to NTSC and all
is now well picture wise.. I'm learning - slow but sure.. :^)..

Appreciate!!

    Hal - in Terra Alta, WV - Slackware GNU/Linux 9.0   (2.4.20)
                Proprietary  Formats  Unacceptable


On 10-29, Nikolai Zhubr wrote:
> Hi,
> Tuesday, 28 October, 2003, 22:17:04, Hal MacArgle wrote:
> [...]
> (**) |   |-->Monitor "My Monitor"
> (**) |   |-->Device "VESA Framebuffer"
> [...]
> IMHO you should be using "ati" driver and not vesa framebuffer
> if you are going to get multimedia stuff.
> Open your XF86Config file and look for 'Section "Device"', there
> should be a '"Driver "...."' somewhere (you might want choose to
> either edit the file manually or use some configurator)
> -- 
> Best regards,
>  Nikolai Zhubr

On 10-29, Nikolai Zhubr wrote:
> Hi,
> Tuesday, 28 October, 2003, 22:17:04, Hal MacArgle wrote:
> [...]
> (**) |   |-->Monitor "My Monitor"
> (**) |   |-->Device "VESA Framebuffer"
> [...]
> IMHO you should be using "ati" driver and not vesa framebuffer
> if you are going to get multimedia stuff.
> Open your XF86Config file and look for 'Section "Device"', there
> should be a '"Driver "...."' somewhere (you might want choose to
> either edit the file manually or use some configurator)
> -- 
> Best regards,
>  Nikolai Zhubr

	Greetings: Thank you.. I just didn't/don't know enough about
XF86.. I should have studied it more before trying avview.. I now
have "activity" in the avview window either tuner or composite ports,
but both are not "locked." I can see a picture but it's loaded with
ladder lines and jumping around.. It's probably something I did
trying to get the picture in the first place..

	After looking into the syntax of XF86Config I learned more
about what I should have done in the first place, rather than rely on
a configurator that either let me down or I couldn't work with..

I've got to spend more time with the GUI shell - too many years with
CLI I guess. <grin>

Appreciate!

    Hal - in Terra Alta, WV - Slackware GNU/Linux 9.0   (2.4.20)
                Proprietary  Formats  Unacceptable
.

I installed RH9 on a different machine and did what you said. Everything
worked perfectly. Thanks.

My only question is, I wrote down what files it moved into the proper
directories. I see it is replacing a fair bit of files with different =
ones.
I'm guessing these are all updated by the ati.2 code. Which driver do I =
pick
for my card now in X?=20

Thanks,
Robert Birch

-----Original Message-----
From: Nikolai Zhubr [mailto:s0...@ho...]=20
Sent: Thursday, October 23, 2003 2:29 AM
To: Robert Birch
Cc: gat...@li...
Subject: Re: [GATOS]Compiling the ATI.2 w/tv-output


Hi
Wednesday, 22 October, 2003, 20:33:39, Robert Birch wrote:
> XFree86 was installed by the RH9 installation.

> Do I need to recompile XFree86 4.3 for this to work? I tried searching =

> for an xc directory but didn't see it. (Could have missed it)
rpmbuild --recompile XFree86-4.3.0-2.src.rpm
If you don't need to really install this full copy (and probably you =
don't)
then you need not actually wait until the compilation is finished
completely, but ctrl-C it after all of the necessary makefiles are done =
(it
will still take several minutes), then
/usr/src/redhat/BUILD/XFree86-4.3.0-2/xc is your source tree to compile
ati.2 against. After ati.2 is done you may just remove the regular tree =
from
/usr/src/redhat/BUILD HTH.
--=20
Best regards,
 Nikolai Zhubr

> Thanks
> ----- Original Message -----
> From: "Vladimir Dergachev" <vo...@mi...>
> To: "Robert Birch" <r....@sh...>
> Sent: Tuesday, October 21, 2003 5:56 PM
> Subject: Re: [GATOS]Compiling the ATI.2 w/tv-output


>>
>>
>> On Tue, 21 Oct 2003, Robert Birch wrote:
>>
>> > So, as I said I am fairly new to this.
>> >
>> > Where should I point to? I don't understand where you mean by =
compiled
>> > XFree86 directory. I thought that would be where the bin was =
installed
> for
>> > XFree86.
>>
>> No, this is where you unpacked and compiled XFree86 source. (it =
should
end
>> with /xc/)
>>
>>                         best
>>
>>                           Vladimir Dergachev
>>
>> >
>> > Sorry for not understanding,
>> > Robert Birch
>> > ----- Original Message -----
>> > From: "Vladimir Dergachev" <vo...@mi...>
>> > To: "Robert Birch" <r....@sh...>
>> > Cc: <gat...@li...>
>> > Sent: Monday, October 20, 2003 7:14 PM
>> > Subject: Re: [GATOS]Compiling the ATI.2 w/tv-output
>> >
>> >
>> > >
>> > >
>> > > On Mon, 20 Oct 2003, Robert Birch wrote:
>> > >
>> > > > I am a definite newbie on this so please forgive me.
>> > > >
>> > > > I used the CVS stuff and downloaded the files. Reading the =
readme
> files,
>> > > > they say to issue a xmkmf /path/to/tree command.
>> > > >
>> > > > I issued xmkmf /usr/X11R6/lib/X11 but get errors.
>> > >
>> > > you should do xmkmf /path/to/compiled/xfree86/tree
>> > > which is *NOT* the same place you installed XFree86 in.
>> > >
>> > > In particular, you need the source of XFree86 4.3.0.
>> > >
>> > >                         best
>> > >
>> > >                            Vladimir Dergachev
>> > >
>> > > >
>> > > > I am running RH9.0. Can someone please give me step by step
> instructions
>> > on
>> > > > how to get this to work? Or is there a pre-compiled version I =
can
>> > download
>> > > > with the tv_output stuff installed?
>> > > >
>> > > > Thanks,
>> > > > Robert Birch
>> > > >
>> >



> -------------------------------------------------------
> This SF.net email is sponsored by OSDN developer relations
> Here's your chance to show off your extensive product knowledge
> We want to know what you know. Tell us and you have a chance to win =
$100
> http://www.zoomerang.com/survey.zgi?HRPT1X3RYQNC5V4MLNSV3E54
> _______________________________________________
> Gatos-devel mailing list
> Gat...@li...
> https://lists.sourceforge.net/lists/listinfo/gatos-devel

On Tue, 28 Oct 2003, Luis A. Montes wrote:

> On Mon, 2003-10-27 at 23:05, Vladimir Dergachev wrote:
> > This is likely something to do with BIOS, and also probably with
> > allocation of PCI resources. What happens to memory apertures when you
> > move the card between PCI slots ?
> X fails before it ever gets to the apertures. I ran a diff between two

The apertures are listed in /proc/pci. Also see below.

> log files, and besides the PCI slot being different (0x09 versus 0x13)
> the only difference was in the line
>
> (II) PCI: Probing config type using method 1
> (II) PCI: Config type is 1
> (II) PCI: stages = 0x03, oldVal1 = 0x80001108, mode1Res1 = 0x80000000
>
> in the configuration that works, while the configuration that prints
>
> (II) PCI: Probing config type using method 1
> (II) PCI: Config type is 1
> (II) PCI: stages = 0x03, oldVal1 = 0x00000000, mode1Res1 = 0x80000000
>
> ends up failing with:
>
> (II) Primary Device is: PCI 00:0d:0
> (II) ATI:  Candidate "Device" section "Card0".
> (II) ATI:  Shared PCI/AGP Mach64 in slot 0:13:0 detected.
> (EE) No devices detected.

XFree86 has its own PCI access routines, this was so since ancient times
and is being kept due to the fact that XFree86 supports many different
operating systems.

I suggest you try running XFree86 -verbose 100 and compare XFree86 idea of
PCI layout with /proc/pci. If they are the same then the problem is that
either BIOS screws up PCI bus setup (not unheard of), or one of linux
kernel or XFree86 screws up PCI - not unheard of either, however there was
much work put into this to get things right.

You might also want to check whether options like PnP bus or some funky
PCI support options are enabled in the kernel.

                          best

                            Vladimir Dergachev

>
> Unfortunately, this machine does not have keyboard, and it's kind of
> awkward to get to, but I guess somehow I'll have to.
>
> Thanks
>

On Mon, 2003-10-27 at 23:05, Vladimir Dergachev wrote:
> This is likely something to do with BIOS, and also probably with
> allocation of PCI resources. What happens to memory apertures when you
> move the card between PCI slots ?
X fails before it ever gets to the apertures. I ran a diff between two
log files, and besides the PCI slot being different (0x09 versus 0x13)
the only difference was in the line

(II) PCI: Probing config type using method 1
(II) PCI: Config type is 1
(II) PCI: stages = 0x03, oldVal1 = 0x80001108, mode1Res1 = 0x80000000

in the configuration that works, while the configuration that prints

(II) PCI: Probing config type using method 1
(II) PCI: Config type is 1
(II) PCI: stages = 0x03, oldVal1 = 0x00000000, mode1Res1 = 0x80000000

ends up failing with:

(II) Primary Device is: PCI 00:0d:0
(II) ATI:  Candidate "Device" section "Card0".
(II) ATI:  Shared PCI/AGP Mach64 in slot 0:13:0 detected.
(EE) No devices detected.

Unfortunately, this machine does not have keyboard, and it's kind of
awkward to get to, but I guess somehow I'll have to.

Thanks

On Tue, 28 Oct 2003, Lourens Veen wrote:

> On Tue 28 October 2003 06:11, Luis A. Montes wrote:
> > Hi there,
> >
> > A bit off topic, but I was wondering if anybody can explain me
> > why is it that moving video card (ATI AIW Pro Mach 64 PCI, if it
> > matters) from one PCI slot to other makes a difference. Is there
> > anything you can do to force the drivers to detect the card in a
> > given slot? I'm using a not-too-terribly-new version of the
> > Retinal-Burn drivers with tv-out enabled, and the card doesn't
> > seem to be recognized but in the first slot.
>
> The BIOS gives out IRQs based on in which slot the cards are.
> Changing them around may change the way IRQs are divided over the
> cards, what IRQs are shared, etc. If for example the card or the
> driver doesn't like sharing an IRQ line, then swapping cards around
> may help to give it its own. Most BIOSes have some options to
> fiddle with IRQs and slots.

The only caveat is that IRQs are not needed for TV-out, so it is likely
something else.

                    best

                       Vladimir Dergachev

>
> Lourens
> --
> GPG public key: http://home.student.utwente.nl/l.e.veen/lourens.key
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: The SF.net Donation Program.
> Do you like what SourceForge.net is doing for the Open
> Source Community?  Make a contribution, and help us add new
> features and functionality. Click here: http://sourceforge.net/donate/
> _______________________________________________
> Gatos-devel mailing list
> Gat...@li...
> https://lists.sourceforge.net/lists/listinfo/gatos-devel
>

Hi,
Tuesday, 28 October, 2003, 22:17:04, Hal MacArgle wrote:
[...]
(**) |   |-->Monitor "My Monitor"
(**) |   |-->Device "VESA Framebuffer"
[...]
IMHO you should be using "ati" driver and not vesa framebuffer
if you are going to get multimedia stuff.
Open your XF86Config file and look for 'Section "Device"', there
should be a '"Driver "...."' somewhere (you might want choose to
either edit the file manually or use some configurator)
-- 
Best regards,
 Nikolai Zhubr

On 10-26, Vladimir Dergachev wrote:
> 
> 
> On Sun, 26 Oct 2003, Hal MacArgle wrote:
> 
> > On 10-25, Vladimir Dergachev wrote:
> > > > > Did you select video port in AVview ?
> > > >
> > > > 	Not sure: In setup menu Video Device Parameters: "Xvport used
> > > > for on-screen display" defaults to "none".. I can't seem to change
> > > > that - only trying left and right clicks to put a "-" over the none..
> > >
> > > This means there is no port on your system that supports TV-in. I.e. you
> > > did not install ati.2 drivers. Take a look at
> > > http://gatos.sf.net/ati.2.php for instructions on installing ati.2
> > > modules.
> >
> > 	Greetings: My biggest problem, to date, was my lack of XF86
> > knowledge.. I thought the XF86 "modules" were the same as the Kernel
> > "modules." I'm learning...
> >
> > 	I followed the above ati.2.php file exactly and start_avview
> > comes up fine.. My ATI chip is BL829 so I thought the XF86 module
> > bl829 would solve the problem - but not so.. In fact I've tried and
> > called all the following modules, one at a time, listed in the
> > following partial print of .0.log:
> 
> There is no need to "call" bt829 module, the driver will load it by itself
> when probing.
> 
> Why don't you e-mail the entire XFree86.0.log, gzipped ?
> 

		OK .gz file attached after I rm'd bt829, etc, back to
where I started..



>                       best
> 
>                         Vladimir Dergachev
> 
> > Diags tell me: "No adapter present" and Video port still says "none."
> > Needless to say the avview window is still "black and blank."
> >
> > Appreciate!!
> >
> >     Hal - in Terra Alta, WV - Slackware GNU/Linux 9.0   (2.4.20)
> >                 Proprietary  Formats  Unacceptable
> > .
> >
> >
> >

I have been using gatos for quite some time (excellent job BTW) recently 
upgraded my mobo to a gigabyte ga-7dpxdw+ w/dual athlons.  Since then avview 
(and xawtv) shows a band of video noise on the right side and bottom.  Also 
seems to affect some applications, most notably Firebird, with random bands 
of distortion.  Hardware is ATI Rage 128 pro, AMD-760MPX chipset, AMD-762 
North Bridge AMD-768 South Bridge.  Software is Slack 9.1, kernel vers 2.4.22 
(and also 2.6-test), XFree 4.3 and KDE 3.1.4.  I have tried precompiled 
binaries, compiled X from slackware source and compiled ati.2 from cvs, 
switched to gnome, all with the same behavior.  

Lourens Veen <lo...@ra...> writes:

>> [...]
> I've got the same card, and the same chips on it, although I'm not 
> sure about the eeprom. I've got a small (about 6x6 mm or so) chip 
> on it, but the card is in my machine and I can't get close enough 
> to read the lettering.
>
> Anyway, I've tried attaching a TV before, but when I boot with my TV 
> attached I don't get anything on screen even in text mode. Is that 
> normal or is something broken?

I can't tell. Check your cables.
Text mode display works here (using vesa i can see video on the tv).
And the boot sequence (bios init and co) is also shown on the tv.

-- 
Laurent

	I'm using differnet Debian XFree86 4.3.0 packages that seem to work
fine with my AIW:
XFree86 Version 4.3.0 (Debian 4.3.0-0ds4 20030416150820
dstone@aedificator)
Release Date: 18 March 2003


You wrote:

> 
> I'm trying to run the ati.2 drivers with the Debian XFree86 4.3.0
> prerelease packages. The server is dying with SIGFPE.

> XFree86 Version 4.3.0.1 (Debian 4.3.0-0pre1v3 20030928104247
> ma...@mi...) Release Date: 15 August 2003

On Tue 28 October 2003 06:11, Luis A. Montes wrote:
> Hi there,
>
> A bit off topic, but I was wondering if anybody can explain me
> why is it that moving video card (ATI AIW Pro Mach 64 PCI, if it
> matters) from one PCI slot to other makes a difference. Is there
> anything you can do to force the drivers to detect the card in a
> given slot? I'm using a not-too-terribly-new version of the
> Retinal-Burn drivers with tv-out enabled, and the card doesn't
> seem to be recognized but in the first slot.

The BIOS gives out IRQs based on in which slot the cards are.=20
Changing them around may change the way IRQs are divided over the=20
cards, what IRQs are shared, etc. If for example the card or the=20
driver doesn't like sharing an IRQ line, then swapping cards around=20
may help to give it its own. Most BIOSes have some options to=20
fiddle with IRQs and slots.

Lourens
--=20
GPG public key: http://home.student.utwente.nl/l.e.veen/lourens.key

On Tue, 27 Oct 2003, Luis A. Montes wrote:

> On Sat, 2003-10-25 at 10:11, Vladimir Dergachev wrote:
> > > >Anyway, AFAIK such kind of stuff is hard to support and
> > > >performance gain isn't high enough to care.
> > > >
> > > >
> > >
> > > I think the confusion here is that ATI cards have a hardware scalar, not
> > > a decoder, it will resize your video stream in hardware (fullscreen),
> > > but you have to provide the YUV data.  This means you must first decode
> > > the MPEG stream via the system CPU and then have enough available
> > > PCI/AGP bandwidth to transfer the YUV data to the scalar unit/video memory.
> >
> > The hardware also supports iDCT and motion compensation. The problem is
> > that these features are not straightforward to add support for, and why
> > bother (unless one is interested in how such things work) when a faster
> > cpu will do the job in software perfectly.
> But it would be helpful if trying to *encode* real-time to anything like
> mpeg4. iDCT hardware can be used both ways and most CPU's aren't fast
> enough to encode mpeg4 yet. In my athlon 1700 encoding is about 7-8
> frames a second, or about one fourth of what would be needed. IMHO, iDCT
> is a expensive operation because requires rows*columns*frames operations
> per second. It just seems to me that those kind of repetitive, over the
> whole frame operations are better left to the hardware if available.

AFAIK you need DCT not iDCT (inverse DCT) for encoding, and this is only
supported with newer cards.

                       best

                           Vladimir Dergachev


>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: The SF.net Donation Program.
> Do you like what SourceForge.net is doing for the Open
> Source Community?  Make a contribution, and help us add new
> features and functionality. Click here: http://sourceforge.net/donate/
> _______________________________________________
> Gatos-devel mailing list
> Gat...@li...
> https://lists.sourceforge.net/lists/listinfo/gatos-devel
>

On Tue 28 October 2003 00:41, lau...@li... wrote:
> Vladimir Dergachev <vo...@mi...> writes:
> >> Could you tell me how to check if my Radeon VE/7000 QY (AGP)
> >> has an external or an embedded Rage Theater chip ?
> >
> > Just take a look at the card. If it is an external chip, then
> > you will see a medium-size (12mm x 12mm) square chip which says
> > either Rage Theatre (for RT100) or Rage Theatre 200 (for
> > RT200). The latter is not supported yet. You could also see a
> > chip marked "Impact TV", this is a much earlier model, also not
> > supported yet.
> > I would not expect a card with embedded chip to have any
> > medium-size chips onboard besides GPU and memory.
>
> As I was writing to Federico, there is no external RT chip on it.
> There're only the gpu and his radiator, a chip that looks like an
> eeprom (with PML Flash written on it), and the memory.

I've got the same card, and the same chips on it, although I'm not=20
sure about the eeprom. I've got a small (about 6x6 mm or so) chip=20
on it, but the card is in my machine and I can't get close enough=20
to read the lettering.

Anyway, I've tried attaching a TV before, but when I boot with my TV=20
attached I don't get anything on screen even in text mode. Is that=20
normal or is something broken?

Lourens
--=20
GPG public key: http://home.student.utwente.nl/l.e.veen/lourens.key

This is likely something to do with BIOS, and also probably with
allocation of PCI resources. What happens to memory apertures when you
move the card between PCI slots ?

                      best

                        Vladimir Dergachev

On Tue, 27 Oct 2003, Luis A. Montes wrote:

> Hi there,
>
> A bit off topic, but I was wondering if anybody can explain me why is it
> that moving video card (ATI AIW Pro Mach 64 PCI, if it matters) from one
> PCI slot to other makes a difference. Is there anything you can do to
> force the drivers to detect the card in a given slot? I'm using a
> not-too-terribly-new version of the Retinal-Burn drivers with tv-out
> enabled, and the card doesn't seem to be recognized but in the first
> slot.
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: The SF.net Donation Program.
> Do you like what SourceForge.net is doing for the Open
> Source Community?  Make a contribution, and help us add new
> features and functionality. Click here: http://sourceforge.net/donate/
> _______________________________________________
> Gatos-devel mailing list
> Gat...@li...
> https://lists.sourceforge.net/lists/listinfo/gatos-devel
>

Luis:

Thanks for answering my post, you are the only one
that did and I am sure that there are others that know
more about the problem/solution than either of us!

Likewise, I ended up putting "modprobe videodev" in
rc.local too and then doing the sync and inserting the
km modules. However, after manually loading the
modules and doing a depmod -a, my expectation was that
all three would automaticly load knowing the km's
require videodev first; but it doesn't! I suspect know
that it may be that the requirement for running the
'sync' command may be the problem. I am not sure what
the sync command does or why it is required.

This issue is one that should cause quite a few
problems for us "not-yet" but "soon-to-be" linux
experts!

I do have AVview working nicely now with everything
installing automaticly and an icon on my desktop "to
boot"! However, getting it working was much more work
than I think most people will bother to do! 

Regards,
Lee

--- "Luis A. Montes"
<lui...@wo...> wrote:
> On Sat, 2003-10-25 at 10:11, Vladimir Dergachev
> wrote:
> > > >Anyway, AFAIK such kind of stuff is hard to
> support and
> > > >performance gain isn't high enough to care.
> > > >
> > > >
> > >
> > > I think the confusion here is that ATI cards
> have a hardware scalar, not
> > > a decoder, it will resize your video stream in
> hardware (fullscreen),
> > > but you have to provide the YUV data.  This
> means you must first decode
> > > the MPEG stream via the system CPU and then have
> enough available
> > > PCI/AGP bandwidth to transfer the YUV data to
> the scalar unit/video memory.
> > 
> > The hardware also supports iDCT and motion
> compensation. The problem is
> > that these features are not straightforward to add
> support for, and why
> > bother (unless one is interested in how such
> things work) when a faster
> > cpu will do the job in software perfectly.
> But it would be helpful if trying to *encode*
> real-time to anything like
> mpeg4. iDCT hardware can be used both ways and most
> CPU's aren't fast
> enough to encode mpeg4 yet. In my athlon 1700
> encoding is about 7-8
> frames a second, or about one fourth of what would
> be needed. IMHO, iDCT
> is a expensive operation because requires
> rows*columns*frames operations
> per second. It just seems to me that those kind of
> repetitive, over the
> whole frame operations are better left to the
> hardware if available.
> 
> 
> 
>
-------------------------------------------------------
> This SF.net email is sponsored by: The SF.net
> Donation Program.
> Do you like what SourceForge.net is doing for the
> Open
> Source Community?  Make a contribution, and help us
> add new
> features and functionality. Click here:
> http://sourceforge.net/donate/
> _______________________________________________
> Gatos-devel mailing list
> Gat...@li...
>
https://lists.sourceforge.net/lists/listinfo/gatos-devel


=====
--
God is not a republican or a democrat nor is His government a democracy! - Lee Parmeter

__________________________________
Do you Yahoo!?
Exclusive Video Premiere - Britney Spears
http://launch.yahoo.com/promos/britneyspears/

On Sat, 2003-10-25 at 10:11, Vladimir Dergachev wrote:
> > >Anyway, AFAIK such kind of stuff is hard to support and
> > >performance gain isn't high enough to care.
> > >
> > >
> >
> > I think the confusion here is that ATI cards have a hardware scalar, not
> > a decoder, it will resize your video stream in hardware (fullscreen),
> > but you have to provide the YUV data.  This means you must first decode
> > the MPEG stream via the system CPU and then have enough available
> > PCI/AGP bandwidth to transfer the YUV data to the scalar unit/video memory.
> 
> The hardware also supports iDCT and motion compensation. The problem is
> that these features are not straightforward to add support for, and why
> bother (unless one is interested in how such things work) when a faster
> cpu will do the job in software perfectly.
But it would be helpful if trying to *encode* real-time to anything like
mpeg4. iDCT hardware can be used both ways and most CPU's aren't fast
enough to encode mpeg4 yet. In my athlon 1700 encoding is about 7-8
frames a second, or about one fourth of what would be needed. IMHO, iDCT
is a expensive operation because requires rows*columns*frames operations
per second. It just seems to me that those kind of repetitive, over the
whole frame operations are better left to the hardware if available.

On Fri, 2003-10-24 at 07:54, Lee Parmeter wrote:
> Question:
> 
> After installing the km drivers to /lib/.../misc, the
> drivers do not load at boot I think because 'videodev'
> is not installed first. How may this problem be
> resolved?
I just load it at boot time in rc.local. But I'm sure that if you want
to get fancy the right approach is to use the pre-install option on your
/etc/modules.conf file. See the man page
> 
> Also, ati-remote.o does load from /lib/.../misc, but
> there are unresolved externals reported. What may I do
> to resolve this issue?
> 
> Please advise!
> 
> Lee
> 
> =====
> --
> God is not a republican or a democrat nor is His government a democracy! - Lee Parmeter
> 
> __________________________________
> Do you Yahoo!?
> The New Yahoo! Shopping - with improved product search
> http://shopping.yahoo.com
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: The SF.net Donation Program.
> Do you like what SourceForge.net is doing for the Open
> Source Community?  Make a contribution, and help us add new
> features and functionality. Click here: http://sourceforge.net/donate/
> _______________________________________________
> Gatos-devel mailing list
> Gat...@li...
> https://lists.sourceforge.net/lists/listinfo/gatos-devel

Hi there,

A bit off topic, but I was wondering if anybody can explain me why is it
that moving video card (ATI AIW Pro Mach 64 PCI, if it matters) from one
PCI slot to other makes a difference. Is there anything you can do to
force the drivers to detect the card in a given slot? I'm using a
not-too-terribly-new version of the Retinal-Burn drivers with tv-out
enabled, and the card doesn't seem to be recognized but in the first
slot.

Vladimir Dergachev <vo...@mi...> writes:


>> Could you tell me how to check if my Radeon VE/7000 QY (AGP) has an
>> external or an embedded Rage Theater chip ?
> Just take a look at the card. If it is an external chip, then you will see
> a medium-size (12mm x 12mm) square chip which says either Rage Theatre
> (for RT100) or Rage Theatre 200 (for RT200). The latter is not supported yet.
> You could also see a chip marked "Impact TV", this is a much earlier
> model, also not supported yet.
> I would not expect a card with embedded chip to have any medium-size chips
> onboard besides GPU and memory.

As I was writing to Federico, there is no external RT chip on it.
There're only the gpu and his radiator, a chip that looks like an
eeprom (with PML Flash written on it), and the memory.

-- 
Laurent

On Mon, 27 Oct 2003 lau...@li... wrote:

>
> Hi
>
> Could you tell me how to check if my Radeon VE/7000 QY (AGP) has an
> external or an embedded Rage Theater chip ?

Just take a look at the card. If it is an external chip, then you will see
a medium-size (12mm x 12mm) square chip which says either Rage Theatre
(for RT100) or Rage Theatre 200 (for RT200). The latter is not supported yet.

You could also see a chip marked "Impact TV", this is a much earlier
model, also not supported yet.

I would not expect a card with embedded chip to have any medium-size chips
onboard besides GPU and memory.

                            best

                               Vladimir Dergachev

Hi

Could you tell me how to check if my Radeon VE/7000 QY (AGP) has an
external or an embedded Rage Theater chip ?

I'm trying to test the tv_output branch of ati.2 with no success:

  Bus  1, device   0, function  0:
    VGA compatible controller: ATI Technologies Inc Radeon RV100 QY [Radeon 7000/VE] (rev 0).
      Master Capable.  Latency=64.  Min Gnt=8.
      Prefetchable 32 bit memory at 0xd0000000 [0xd7ffffff].
      I/O at 0x9800 [0x98ff].
      Non-prefetchable 32 bit memory at 0xdfef0000 [0xdfefffff].


[...]
(II) ATI: ATI driver (version 6.4.18) for chipsets: ati, ativga
(II) R128: Driver for ATI Rage 128 chipsets:
[...]
(II) Primary Device is: PCI 01:00:0
(II) ATI:  Candidate "Device" section "ATI RADEON 7000".
(--) Chipset ATI Radeon VE/7000 QY (AGP) found
[...]
(II) Loading sub module "radeon"
(II) LoadModule: "radeon"
(II) Loading /usr/X11R6/lib/modules/drivers/radeon_drv.o
(II) Module radeon: vendor="The XFree86 Project"
        compiled for 4.3.0.1, module version = 4.0.1
        Module class: XFree86 Video Driver
        ABI class: XFree86 Video Driver, version 0.6
[...]
(II) Setting vga for screen 0.
(**) RADEON(0): RADEONPreInit
(II) Loading sub module "vgahw"
(II) LoadModule: "vgahw"
(II) Loading /usr/X11R6/lib/modules/libvgahw.a
(II) Module vgahw: vendor="The XFree86 Project"
        compiled for 4.3.0.1, module version = 0.1.0
        ABI class: XFree86 Video Driver, version 0.6
(II) RADEON(0): vgaHWGetIOBase: hwp->IOBase is 0x03d0, hwp->PIOOffset is 0x0000
(II) RADEON(0): PCI bus 1 card 0 func 0
(**) RADEON(0): Depth 16, (--) framebuffer bpp 16
(II) RADEON(0): Pixel depth = 16 bits stored in 2 bytes (16 bpp pixmaps)
(==) RADEON(0): Default visual is TrueColor
(==) RADEON(0): RGB weight 565
(II) RADEON(0): Using 6 bits per RGB (8 bit DAC)
(II) Loading sub module "int10"
(II) LoadModule: "int10"
(II) Reloading /usr/X11R6/lib/modules/linux/libint10.a
(II) RADEON(0): initializing int10
(II) RADEON(0): Primary V_BIOS segment is: 0xc000
(--) RADEON(0): Chipset: "ATI Radeon VE/7000 QY (AGP)" (ChipID = 0x5159)
(--) RADEON(0): Linear framebuffer at 0xd0000000
(--) RADEON(0): MMIO registers at 0xdfef0000
(--) RADEON(0): BIOS at 0xdfec0000
(II) RADEON(0): Video RAM override, using 32768 kB instead of 32768 kB
(**) RADEON(0): VideoRAM: 32768 kByte (64-bit DDR SDRAM)
(II) RADEON(0): CloneDisplay option not set -- defaulting to auto-detect
(II) RADEON(0): Primary Display == Type 5
(II) Loading sub module "ddc"
(II) LoadModule: "ddc"
(II) Reloading /usr/X11R6/lib/modules/libddc.a
(II) Loading sub module "i2c"
(II) LoadModule: "i2c"
(II) Loading /usr/X11R6/lib/modules/libi2c.a
(II) Module i2c: vendor="The XFree86 Project"
        compiled for 4.3.0.1, module version = 1.2.0
        ABI class: XFree86 Video Driver, version 0.6
(II) RADEON(0): I2C bus "DDC" initialized.
(**) RADEON(0): Header at 0x0128; PLL Information at 0x067c
(II) RADEON(0): PLL parameters: rf=2700 rd=60 min=12000 max=35000; xclk=16600
(==) RADEON(0): Using gamma correction (1.0, 1.0, 1.0)
(II) RADEON(0): Clone modes validation ------------ 
(II) RADEON(0): DDC detection (type 3) for clone modes
(II) RADEON(0): I2C device "DDC:ddc2" registered at address 0xA0.
(II) RADEON(0): I2C device "DDC:ddc2" removed.
(II) RADEON(0): I2C device "DDC:ddc2" registered at address 0xA0.
(II) RADEON(0): I2C device "DDC:ddc2" removed.
(II) RADEON(0): I2C device "DDC:ddc2" registered at address 0xA0.
(II) RADEON(0): I2C device "DDC:ddc2" removed.
(WW) RADEON(0): targa: Using default hsync range of 28.00-33.00kHz
(WW) RADEON(0): targa: using default vrefresh range of 43.00-72.00Hz

hsync and vrefresh do not match the ones I've configured for my targa
monitor.

(II) RADEON(0): Not using default mode "640x350" (hsync out of range)
[bunch of hsync out of range messages]
(II) RADEON(0): Not using mode "1024x768" (no mode of this name)
(II) RADEON(0): Not using mode "800x600" (no mode of this name)
(**) RADEON(0): Virtual size is 3200x768 (pitch 3200)
(**) RADEON(0): *Mode "1024x768"
(**) RADEON(0): *Mode "800x600"
(**) RADEON(0): *Default mode "640x480": 25.2 MHz, 31.5 kHz, 60.0 Hz
(II) RADEON(0): Modeline "640x480"   25.20  640 656 752 800  480 490 492 525 -hs
ync -vsync
(II) RADEON(0): Valid Clone Mode: 640x480
(II) RADEON(0): Total of 1 clone modes found ------------ 

(II) RADEON(0): Validating modes on Primary head (DDCType: 2) ---------
(II) RADEON(0): I2C device "DDC:ddc2" registered at address 0xA0.
(II) RADEON(0): I2C device "DDC:ddc2" removed.
(II) RADEON(0): I2C device "DDC:ddc2" registered at address 0xA0.
(II) RADEON(0): I2C device "DDC:ddc2" removed.
(II) RADEON(0): I2C device "DDC:ddc2" registered at address 0xA0.
(II) RADEON(0): I2C device "DDC:ddc2" removed.
(II) RADEON(0): Total number of valid DDC mode(s) found: 0
(WW) RADEON(0): Mode 1024x768 is out of range.
(WW) RADEON(0): Valid modes must be between 320x200-0x0
(WW) RADEON(0): Mode 800x600 is out of range.
(WW) RADEON(0): Valid modes must be between 320x200-0x0
(WW) RADEON(0): Mode 640x480 is out of range.
(WW) RADEON(0): Valid modes must be between 320x200-0x0
(II) RADEON(0): Total number of valid FP mode(s) found: 0
(EE) RADEON(0): No valid mode found for this DFP/LCD
(**) RADEON(0): RADEONFreeScreen
(II) UnloadModule: "ati"
(II) UnloadModule: "i2c"
(II) Unloading /usr/X11R6/lib/modules/libi2c.a
(II) UnloadModule: "ddc"
(II) UnloadModule: "int10"
(II) UnloadModule: "vgahw"
(II) Unloading /usr/X11R6/lib/modules/libvgahw.a
(II) UnloadModule: "radeon"
(II) Unloading /usr/X11R6/lib/modules/drivers/radeon_drv.o
(EE) Screen(s) found, but none have a usable configuration.

Could you tell me what's going on please ?
It seems theatre modules are never loaded and I can't see any TVOUT
related informations in my log.

(tv is plugged on the radeon card, I can see the console output on the
TV, atitvout detect reports "CRT is attached. TV is attached via S-Video.")

Any advice ?
Thanks in advance.

-- 
Laurent