[mythtv-users] looking for Craig Eaton

R. G. Newbury newbury at mandamus.org
Sun Jan 22 22:58:39 UTC 2006


Brian Wood wrote:

>On Jan 22, 2006, at 3:13 AM, R. G. Newbury wrote:
>
>  
>
>>RMH wrote:
>>
>>    
>>
>>>back in March, Craig sent email with the subject "Conflict resolution
>>>between HD3000's NTSC and ATSC tuners?" but his email address is now
>>>bouncing.  here's the archived message:
>>>
>>>http://mythtv.org/pipermail/mythtv-users/2005-March/077576.html
>>>
>>>he apparently got his HD3000 working in both ATSC and NTSC modes,
>>>which is something i would like to do as well.  for some reason, how
>>>to do this is not explained anywhere on the web that i can find.   
>>>if i
>>>could get both modes to work using my one HD3000 card, i think i  
>>>could
>>>program a conflict resolution into mythtv.
>>>      
>>>
>>Buried somewhere on the pcHDTV site, is an explanation of how the card
>>works. The cx88_dvb driver handles ATSC, the cx8800 driver handles  
>>NTSC.
>>Only one driver at a time can talk to the hardware.
>>So you *should* be able to "rmmod cx88_dvb, modprobe cx8800" and watch
>>NTSC content, then "rmmod cx8800, modprobe cx88_dvb" for ATSC.
>>But that may or may not work. You might have to reboot in  between.  
>>And
>>even if you can do that, there is no  guarantee that myth could handle
>>it....I suppose that an "external command" such as a channel change
>>could actually be a module changeover, but you would also have  
>>problems
>>with the database, and the backend would surely choke on not being  
>>able
>>to see one of the cards at startup...
>>
>>    
>>
>The upcoming kernel "merger" of v4l and DVB should take care of this  
>problem. At least I hope it does, as we will all deserve *something*  
>for the inevitable pain that kernel changes cause.
>

I doubt that the kernel merge will help in this situation. Fact is, that 
the hardware works in 2 distinct modes and requires 2 distinct driver 
modules to work properly. If the hardware were in fact built in 
parallel, with 2 of each of the required sections then it might work... 
but that means that the card would in effect need to be 2 tuners, behind 
the actual radio wave level tuner..
Although having the 2 modules in the kernel implies a fair bit of 
stability and ability to play nice, it does not mean that switching 
hardware modes will necessarily be easy. IF myth had some mechanism to 
ignore a missing input, and to be warned to resurrect that input upon a 
change, we might be able to do something, but none of that exists, and 
probably cannot until the actual api which the kernel modules will use, 
is established..
But it would be nice....then you could have an extra available NTSC 
tuner in cases of conflict, with that piece of hardware prioritized to 
recording ATSC first.

Geoff



More information about the mythtv-users mailing list