Skip to contentSkip to conversation

auto detection of Flytv duo/hybrid and pci/cardbus confusion

linux-dvb Werner Braun 3 participants 3 messages Feb 20, 2008 – Feb 22, 2008 Media (V4L/DVB)

lore
Share & tools
lore
#0 lore raw
Quoted 9 lines

Am Sonntag, den 17.02.2008, 21:53 +0100 schrieb Peter Missel:

Quoted 8 lines

Greetings all!

Let me clear things up a bit.

Quoted 4 lines
Quoted 4 lines

First clarification, duo versus hybrid.
Are "duo" cards equipped with two independent tuners that can both be used at the same time?
Are "hybrid" cards necessarily equipped with digital and analogue

tuners?

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

Can a two tuner card be both a duo and a hybrid, if one tuner is

digital

Quoted 16 lines
Quoted 16 lines
Quoted 2 lines
Quoted 1 line

the other is analogue and they can both be used at the same time?

LifeView are using two vendor IDs - 4E42h for all (!) their OEMs, and their own one for LifeView branded cards. Hence we need two PCI ID entries for everything, each pair pointing back to the same card data.

Then, card types.

The analog-only and "hybrid" have one single tuner, for DVB-T or analog. The "Duo" cards have two tuner frontends, one for DVB-T and the other for analog.
"Trio" cards add a DVB-S frontend, which cannot be used at the same time as the DVB-T frontend. Like the Duo, these can run one digital and one analog stream in parallel.

Quoted 4 lines
Quoted 4 lines

Finally, card shapes.

Each card type comes in CardBus, PCI, and MiniPCI shape. The flavors are compatible, so that again, the PCI ID data point back to the same card

entry

Quoted 3 lines
Quoted 3 lines

for e.g. the PCI and CardBus Duo.

The card type/shape combinations are distinctly identified by their

subsystem

Quoted 14 lines
Quoted 6 lines

ID. No need to guesstimate anything.

That's the plan at least.

regards,
Peter

Hi Peter!

Your plan is fine so far.

We might add some more comments to group devices obviously together, since those looking first time at it are a bit lost.

Quoted 8 lines

For such i2c IR limits, we have your and Eddi's comments.

Since we can't help it easily, Peter D. should suggest the older version of the MSI A/D for auto detection. It won't make anything more worse on that not fully clear Vivanco stuff, except Hartmut might have ideas.

Cheers,
Hermann

Hermann and Peter,

I guess it was me with the Vivanco card in May/June last year. I took a time out from the mailing list due to time constraints, but am back and willing to help you sorting out the open issues, as far as my limited competences allow for it.

The current status with my Vivanco 21057 card (4e42:3306) is:

- DVB-T works with the patches you suggested last year
- Analog not (but did not bother anyway)
- FM: did not try
- Remote: dto. (I'm sitting in front of the computer anyway)

Funny thing: since Kernel 2.6.22, firmware upload does not work any longer. I have to boot XP before and do a warm reboot, then DVB-T under Linux works.

Best regards
Werner

_______________________________________________
linux-dvb mailing list
[email protected]
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb

#1 lore raw

Am Mittwoch, den 20.02.2008, 23:41 +0100 schrieb Werner Braun:

Quoted 74 lines
Quoted 9 lines

Am Sonntag, den 17.02.2008, 21:53 +0100 schrieb Peter Missel:

Quoted 8 lines

Greetings all!

Let me clear things up a bit.

Quoted 4 lines
Quoted 4 lines

First clarification, duo versus hybrid.
Are "duo" cards equipped with two independent tuners that can both be used at the same time?
Are "hybrid" cards necessarily equipped with digital and analogue

tuners?

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

Can a two tuner card be both a duo and a hybrid, if one tuner is

digital

Quoted 16 lines
Quoted 16 lines
Quoted 2 lines
Quoted 1 line

the other is analogue and they can both be used at the same time?

LifeView are using two vendor IDs - 4E42h for all (!) their OEMs, and their own one for LifeView branded cards. Hence we need two PCI ID entries for everything, each pair pointing back to the same card data.

Then, card types.

The analog-only and "hybrid" have one single tuner, for DVB-T or analog. The "Duo" cards have two tuner frontends, one for DVB-T and the other for analog.
"Trio" cards add a DVB-S frontend, which cannot be used at the same time as the DVB-T frontend. Like the Duo, these can run one digital and one analog stream in parallel.

Quoted 4 lines
Quoted 4 lines

Finally, card shapes.

Each card type comes in CardBus, PCI, and MiniPCI shape. The flavors are compatible, so that again, the PCI ID data point back to the same card

entry

Quoted 3 lines
Quoted 3 lines

for e.g. the PCI and CardBus Duo.

The card type/shape combinations are distinctly identified by their

subsystem

Quoted 14 lines
Quoted 6 lines

ID. No need to guesstimate anything.

That's the plan at least.

regards,
Peter

Hi Peter!

Your plan is fine so far.

We might add some more comments to group devices obviously together, since those looking first time at it are a bit lost.

Quoted 8 lines

For such i2c IR limits, we have your and Eddi's comments.

Since we can't help it easily, Peter D. should suggest the older version of the MSI A/D for auto detection. It won't make anything more worse on that not fully clear Vivanco stuff, except Hartmut might have ideas.

Cheers,
Hermann

Hermann and Peter,

I guess it was me with the Vivanco card in May/June last year. I took a time out from the mailing list due to time constraints, but am back and willing to help you sorting out the open issues, as far as my limited competences allow for it.

The current status with my Vivanco 21057 card (4e42:3306) is:

- DVB-T works with the patches you suggested last year
- Analog not (but did not bother anyway)
- FM: did not try
- Remote: dto. (I'm sitting in front of the computer anyway)

Quoted 6 lines

Funny thing: since Kernel 2.6.22, firmware upload does not work any longer. I have to boot XP before and do a warm reboot, then DVB-T under Linux works.

Best regards
Werner

Hi,

Werner, yes it was your device coming back in mind.

We know now about the new A/D version V1.1, saw only a fuzzy picture so far, but looks like a KS003 something for the remote and not that PIC on the prior, different addresses, IIRC.

That one also has the supposed LNA in the tuner area visible.

Must look it up again, but for what I remember I did not find a difference between your Vivanco card and what Peter D. has.

We also have regspy output from a recent cardbus A/D NB, but Mauro can't test any DVB-T currently and analog is fine on that one with card=94.

However, maybe the differences of the gpio settings in DVB-T mode visible there may help us to figure out a pattern we miss.

analog
SAA7134_GPIO_GPMODE: 8078e700 (10000000 01111000 11100111
00000000)
SAA7134_GPIO_GPSTATUS: 08218000 (00001000 00100001 10000000
00000000)
composite
SAA7134_GPIO_GPSTATUS: 08218000 (00001000 00100001 10000000
00000000)
s-video
SAA7134_GPIO_GPSTATUS: 08218000 (00001000 00100001 10000000
00000000)
radio
SAA7134_GPIO_GPSTATUS: 08018000 (00001000 00000001 10000000
00000000)
dvb-t
SAA7134_GPIO_GPSTATUS: 08e08000 (00001000 11100000 10000000
00000000)
---------------------------------------------------------------------------------
board init: gpio is 8210000 1000 00100001 00000000
00000000

Cheers,
Hermann

--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

#2 lore raw

Hi guys!

Quoted 7 lines

Werner, yes it was your device coming back in mind.

We know now about the new A/D version V1.1, saw only a fuzzy picture so far, but looks like a KS003 something for the remote and not that PIC on the prior, different addresses, IIRC.

If you ask me, that V1.1 card doesn't look anything like any LifeView design I've seen so far. It wouldn't be the first time MSI sell a totally different card than before under the same product name, as "just a newer revision".

The use of MSI's own PCI vendor ID 1462h in the subsystem ID also hints that it isn't a LifeView card as we know them, since they have never ever used an OEM's specific ID. It's always been 4E42h ever since the subsystem ID was invented.

Just my two cents.

regards,
Peter

--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list