Fri, 23 Jun 2006 12:58:12 +0400
Hayova <hayova (AT) gmail (DOT) comwrote:
Hello all!
This is my angry letter :(
Some time ago I wrote this message for list:
But I have 0 (ZER) answers :(((
Isn't there anyone who can help, or there is another problem?
I have a AverMedia Hybrid FM PCMCIA card with XC3028 tunner which
doesn't work besides the composite input at the moment. There was some
discussions few days ago, and I guess developers here are working very
hard, and I assume everyone has the same problem. So I think the only
thing we can do is wait and see, or help with the testings.
-W

Hello all!

This is my angry letter :(

Some time ago I wrote this message for list:

But I have 0 (ZER) answers :(((

Isn't there anyone who can help, or there is another problem?

I have hardware, I have the time and expirience for trying to solve this
problem, but I need a little support

Even Avermedia ask me with just few words: "Sorry, we have double check this
information with our RD, according to company pravicy, we can not provide
this information to you, we are sorry about this, and please understand.
Thank You".
My for theme questions was:
*I need to know some technical features of the Curdbus Plus card for making
it work under linux: 1. what is i2c-address of ALPS TSZE1 tuner? (Im
suggested it's 0xc0/0x60). 2. what i2c-address of Philips TEA5767 tuner? (Im
suggested it's 0xc2/0x61). 3. what is gpiomask ? (Im suggested it's
0x0C440003) 4. what is VideoPin, AudioPin and gpio for TV? (Im suggested
there are: 1, TV, 0x0C400001). 5. what is VideoPin, AudioPin and gpio for
Composite? (Im suggested there are: 3, Line1, 0x0C400002). 6. what is
VideoPin, AudioPin and gpio for S-Video? (Im suggested there are: 6, Line1,
0x0C400002). 7. what is AudioPin and gpio for Radio? (Im suggested there
are: Line2, 0x04400001). 8. what gpio port should I use for activating the
card?*

That's realy sad situation :(((

Is there anyone who can help me a little. I need konw at least the
i2c-addresses of the devices on the board

This is what is detected with i2c_scan=1 option of saa7134:
saa7134[0]: i2c scan: found device @ 0x80 [?]
saa7134[0]: i2c scan: found device @ 0x84 [?] - TDA9887 ?
saa7134[0]: i2c scan: found device @ 0xa0 [eeprom]
saa7134[0]: i2c scan: found device @ 0xc0 [tuner (analog)] -- ALPS
TSEZ1 ?
saa7134[0]: i2c scan: found device @ 0xc2 [?] - TEA5767 ?

More info and some logs can be found at

Thank you.

Hayova,

6/23/06, Hayova <hayova (AT) gmail (DOT) comwrote:
Hello all!

This is my angry letter :(

Some time ago I wrote this message for list:

But I have 0 (ZER) answers :(((

Isn't there anyone who can help, or there is another problem?

Most likely a combination of both.

I cannot speak for the developers here, however, it is most likely that:
1. your concerns are addressed elsewhere.
2. the card is already supported in the Mercurial repository
(http://linuxtv.org/hg/v4l-dvb) - you are using the latest one right?

Again, I cannot speak for the developers, but:
- (as far as I know) everyone here does development for this driver in
their spare time.
- Spare time is limited.

Most days I see messages on this list from people I would like to
help, but am unable to - mostly due to my lack of experience, time and
hardware.

I have hardware, I have the time and expirience for trying to solve this
problem, but I need a little support

Even Avermedia ask me with just few words: "Sorry, we have double check this
information with our RD, according to company pravicy, we can not provide
this information to you, we are sorry about this, and please understand.
Thank You".

Avermedia seems to not be interested in helping Linux driver development.

My for theme questions was:
*I need to know some technical features of the Curdbus Plus card for making
it work under linux: 1. what is i2c-address of ALPS TSZE1 tuner? (Im
suggested it's 0xc0/0x60). 2. what i2c-address of Philips TEA5767 tuner? (Im
suggested it's 0xc2/0x61). 3. what is gpiomask ? (Im suggested it's
0x0C440003) 4. what is VideoPin, AudioPin and gpio for TV? (Im suggested
there are: 1, TV, 0x0C400001). 5. what is VideoPin, AudioPin and gpio for
Composite? (Im suggested there are: 3, Line1, 0x0C400002). 6. what is
VideoPin, AudioPin and gpio for S-Video? (Im suggested there are: 6, Line1,
0x0C400002). 7. what is AudioPin and gpio for Radio? (Im suggested there
are: Line2, 0x04400001). 8. what gpio port should I use for activating the
card?*

All of these can be found out by merely testing various combinations
of pins. - tracing the board after obtaining a pinout of the driver
chip. (this exists for BTTV - I'm meaning to do this with my card once
I have the time.)

You are unlikely to blow up your board by putting in the wrong gpio
values. And there are windows tools to find out which ones are used.
Check out RegSpy which is included with DScaler.

That's realy sad situation :(((

No, this is the reality of working on drivers for open source. Every
week I hear about some manufacturer or vendor refusing to give out
details for Linux. And everyone's time is limited.

Is there anyone who can help me a little. I need konw at least the
i2c-addresses of the devices on the board

You have the board - you *can* find these out.

This is what is detected with i2c_scan=1 option of saa7134:
saa7134[0]: i2c scan: found device @ 0x80 [?]
saa7134[0]: i2c scan: found device @ 0x84 [?] - TDA9887 ?
saa7134[0]: i2c scan: found device @ 0xa0 [eeprom]
saa7134[0]: i2c scan: found device @ 0xc0 [tuner (analog)] -- ALPS
TSEZ1 ?
saa7134[0]: i2c scan: found device @ 0xc2 [?] - TEA5767 ?

See?

This is what is detected with i2c_scan=1 option of saa7134:
saa7134[0]: i2c scan: found device @ 0x80 [?]
saa7134[0]: i2c scan: found device @ 0x84 [?] - TDA9887 ?
saa7134[0]: i2c scan: found device @ 0xa0 [eeprom]
saa7134[0]: i2c scan: found device @ 0xc0 [tuner (analog)] -- ALPS
TSEZ1 ?
saa7134[0]: i2c scan: found device @ 0xc2 [?] - TEA5767 ?

>See?

I don't see anything, the real output was:
saa7134[0]: i2c scan: found device @ 0x80 [?]
saa7134[0]: i2c scan: found device @ 0x84 [?]
saa7134[0]: i2c scan: found device @ 0xa0 [eeprom]
saa7134[0]: i2c scan: found device @ 0xc0 [tuner (analog)]
saa7134[0]: i2c scan: found device @ 0xc2 [?]

signs was only my suggestions

>Check out RegSpy which is included with DScaler.

I've allreadyu done it. I got the .gpiomask and .gpio values, but radio
still don't work.
Like I allredy wrote it's all strange TEA5767 usualy located on i2c-addr
0xc0(0x60), and when I try to force this address to TEA5767 on this card
it's seem to work but there are no sound. BUT APLS tuner absolutly don't
work on i2c-addr 0xc2 (0x61) in this case :((( When I force i2c-addr
0xc0(0x60) for APLS tuner - it's work (I have both image and sound), BUT
TEA5767 on i2c-addr 0xc2 (0x61) absolutly don't work too :((((

6/23/06, Hayova <hayova (AT) gmail (DOT) comwrote:
Like I allredy wrote it's all strange TEA5767 usualy located on i2c-addr
0xc0(0x60), and when I try to force this address to TEA5767 on this card
it's seem to work but there are no sound. BUT APLS tuner absolutly don't
work on i2c-addr 0xc2 (0x61) in this case :((( When I force i2c-addr
0xc0(0x60) for APLS tuner - it's work (I have both image and sound), BUT
TEA5767 on i2c-addr 0xc2 (0x61) absolutly don't work too :((((

Well, either the TEA5767 is on another address. you have to do
something weird to get it going - I have no idea what this might be -
you'll have to figure it out yourself unless someone else knows -
which is unlikely as nobody responded.

Thanks,

>Well, either the TEA5767 is on another address. you have to do
>something weird to get it going - I have no idea what this might be -
>you'll have to figure it out yourself unless someone else knows -
>which is unlikely as nobody responded.

Well Could it be that i2c_scan shows wrong results? If not, then all
that I have to do is try to use each found address BUT could it be that
ALPS or TEA can work on more that one address I thing there are not :((
So If the ALPS works on 0xc0 then all that I can do - try all other
addresses for TEA

saa7134[0]: i2c scan: found device @ 0x80 [?]
saa7134[0]: i2c scan: found device @ 0x84 [?]
saa7134[0]: i2c scan: found device @ 0xa0 [eeprom]
saa7134[0]: i2c scan: found device @ 0xc0 [tuner (analog)]
saa7134[0]: i2c scan: found device @ 0xc2 [?]

Hi,

Am Freitag, den 23.06.2006, 15:33 +0400 schrieb Hayova:
This is what is detected with i2c_scan=1 option of saa7134:
saa7134[0]: i2c scan: found device @ 0x80 [?]
saa7134[0]: i2c scan: found device @ 0x84 [?] - TDA9887 ?
saa7134[0]: i2c scan: found device @ 0xa0 [eeprom]
saa7134[0]: i2c scan: found device @ 0xc0 [tuner (analog)] -- ALPS
TSEZ1 ?
saa7134[0]: i2c scan: found device @ 0xc2 [?] - TEA5767 ?

>See?

I don't see anything, the real output was:
saa7134[0]: i2c scan: found device @ 0x80 [?]
saa7134[0]: i2c scan: found device @ 0x84 [?]
saa7134[0]: i2c scan: found device @ 0xa0 [eeprom]
saa7134[0]: i2c scan: found device @ 0xc0 [tuner (analog)]
saa7134[0]: i2c scan: found device @ 0xc2 [?]

signs was only my suggestions

>Check out RegSpy which is included with DScaler.

I've allreadyu done it. I got the .gpiomask and .gpio values, but radio
still don't work.
Like I allredy wrote it's all strange TEA5767 usualy located on i2c-addr
0xc0(0x60), and when I try to force this address to TEA5767 on this card
it's seem to work but there are no sound. BUT APLS tuner absolutly don't
work on i2c-addr 0xc2 (0x61) in this case :((( When I force i2c-addr
0xc0(0x60) for APLS tuner - it's work (I have both image and sound), BUT
TEA5767 on i2c-addr 0xc2 (0x61) absolutly don't work too :((((

You _saw_ the TEA5767HN ?

According to the available documentation it can only be at 0x60,
but the pll is there.

Also that sort of tda9887 at 0x84 is undocumented, we found it on the
Cinergy400 cardbus without radio once by experimentations.

That is the only reason why you can see anything at all ;)

Cheers,
Hermann

Em Sex, 2006-06-23 16:08 +0400, Hayova escreveu:
>Well, either the TEA5767 is on another address. you have to do
>something weird to get it going - I have no idea what this might be -
>you'll have to figure it out yourself unless someone else knows -
>which is unlikely as nobody responded.
According with Philips datasheet, tea5767 is only available at addr 0xc0
(8-bit notation). This board might have a i2c bridge, with one gpio pin
used to change from TI PLL to tea5767 (this is not common on analog
tuners, but this kind of architecture is being widely used on hybrid
designs).

The tea5767/5768 autodetection works fine. At least, there is no known
case where this is failing. This chip is very small, and generally
protected by a small metallic can that can't be removed (it is soldered
together with the chip).

Well Could it be that i2c_scan shows wrong results?
Unlikely, but possible.
If not, then all
that I have to do is try to use each found address BUT could it be that
ALPS or TEA can work on more that one address I thing there are not :((
TI chips have two address, by default.
So If the ALPS works on 0xc0 then all that I can do - try all other
addresses for TEA

saa7134[0]: i2c scan: found device @ 0x80 [?]
saa7134[0]: i2c scan: found device @ 0x84 [?]
of those I2C addresses may be an IR or a sound decoder. 0x84 is used
on saa711x chips (0x42 on 7-bit notation).
saa7134[0]: i2c scan: found device @ 0xa0 [eeprom]
saa7134[0]: i2c scan: found device @ 0xc0 [tuner (analog)]
saa7134[0]: i2c scan: found device @ 0xc2 [?]
0xc2 is likely to be a tuner.

strategy to identify those stuff is to look what chips you have
inside your board. It is useful to open the metallic can on your tuner
and check if the tuner have a TI SN 761677. This chip have two
addresses.

Cheers,
Mauro.

Hermann,

Em Sex, 2006-06-23 14:12 +0200, hermann pitton escreveu:
Also that sort of tda9887 at 0x84 is undocumented, we found it on the
Cinergy400 cardbus without radio once by experimentations.

Hmm a tda9887 at 0x84 ?!? This would be harder to make it work
properly, due to saa7115 module probing that address. Are you sure about
that?

Cheers,
Mauro.

>
This board might have a i2c bridge, with one gpio pin
used to change from TI PLL to tea5767 (this is not common on analog
tuners, but this kind of architecture is being widely used on hybrid
designs).

Change from TI PLL to tea5767 on one i2c-addr?

The tea5767/5768 autodetection works fine. At least, there is no known
case where this is failing. This chip is very small, and generally
protected by a small metallic can that can't be removed (it is soldered
together with the chip).

It works on my card NLY if I comment initialization of the card in
saa7134_board_init1 function.

TI chips have two address, by default.

Any examples?

saa7134[0]: i2c scan: found device @ 0xc0 [tuner (analog)]
saa7134[0]: i2c scan: found device @ 0xc2 [?]
0xc2 is likely to be a tuner.

You mean ALPS ?

strategy to identify those stuff is to look what chips you have
inside your board. It is useful to open the metallic can on your tuner
and check if the tuner have a TI SN 761677. This chip have two
addresses.

Well what can I do if it will be so (chip will have two addresses)?

>
Also that sort of tda9887 at 0x84 is undocumented, we found it on the
Cinergy400 cardbus without radio once by experimentations.
Well without tda9887 module tvtuner don't work

Am Freitag, den 23.06.2006, 09:30 -0300 schrieb Mauro Carvalho Chehab:
Hermann,

Em Sex, 2006-06-23 *s 14:12 +0200, hermann pitton escreveu:
Also that sort of tda9887 at 0x84 is undocumented, we found it on the
Cinergy400 cardbus without radio once by experimentations.

Hmm a tda9887 at 0x84 ?!? This would be harder to make it work
properly, due to saa7115 module probing that address. Are you sure about
that?

Yes, Mauro.

Hans Koch, Gerd and me squeezed it until it came out there.
That is the reason why 0x84 was added to the tda9887 scan range.

However, Hans described the chip as somehow merged with the pll chip,
not much related to what is known from tda988/5/6/7. The only hint to
such a chip with only that 0x84 address I found later on the general MK3
documentation Hans received once and is kept in the ivtv tuner
documentation now. (IIRC by Hauppauge through request of Axel Thimm for
Jarod Wilson on MCE500 with dual NTSC MK4 and tea5767hn)

Cheers,
Hermann