[mythtv-users] System specifications review...

[Alan Young]

Michael T. Dean wrote:
> On 11/16/2009 01:15 PM, Patrick Doyle wrote:
>> On Mon, Nov 16, 2009 at 12:37 PM, Yeechang Lee wrote:
>>  
>>> Patrick Doyle says:
>>>    
>>>> So if I go with this processor, which I selected based on reports of
>>>> it's idle and max power consumption (6.8W and 31W), is there any point
>>>> in adding a VDPAU graphics card later?
>>>>       
>>> Yes.
>>>
>>> Let me reword that: No matter how much CPU horsepower you have, there
>>> is no point in *not* getting a VDPAU-capable card *now*. For $30-50,
>>> how can you lose?
>>>     
>> I could lose if it increased the overall power consumption of my box.
>> That one could be tough to measure, as there is a certain "idle" power
>> consumption that, almost certainly must be higher just by adding a
>> card to the box.  But that would, ideally, be offset by the total
>> reduction in power of using the VDPAU decoder instead of using CPU
>> cycles.
>>   
> 
> Agreed.
> 
>> Hmmm... as I think about this more, it occurs to me that it is more
>> complex than I first thought... after all, when I turn off the TV, the
>> MythTV box is likely to stay powered on, still tuned to and displaying
>> whatever channel it was last tuned to and displaynig.  (At least
>> that's the mode that my family and I have been trained to expect from
>> the TiVo.)
> 
> Myth does not do this.  Myth uses the capture cards /only/ when 
> recordings are in progress (including LiveTV recordings).  Myth does 
> /not/ record random garbage when you're not using it.  This means that 
> Myth actually /does/ benefit from idle CPU/GPU power savings.

There is one case that I know of where it does use the capture card when 
it is not recording.  If you have a ATSC capture card and have EIT 
scanning turned on then it does use the card while not recording to 
download OTA program data.  It shouldn't use the same amount of 
resources as decoding, but it does cause activity.

Alan