Mailing List Archive: MythTV: Users

Re: Request for sticky topic - low power consumption contest

Index | Next | Previous | View Threaded

myLC at gmx

May 19, 2012, 6:44 AM

Post #1 of 4 (421 views)
Permalink

Re: Request for sticky topic - low power consumption contest

Jeremy Jones wrote:
===================
> ... the users list is just an email distribution, but the
> wiki has pages for 'sticky' type of information. ...

Agreed! It's a much better place for full content including
pictures or even videos.

> ... I did a quick search in google and
> found this page:
> http://www.mythtv.org/wiki/Power_saving

Yes, thank you. That's the "do it yourself" section most
Linux aficionados can deal with. I'm not sure if every
Myth-TV user can be counted in on this.
Definitely very useful tips though! :-)

Nick Rout wrote:
================
> Nice Idea but I foresee a stagnant page in a few months
> once the hardware info becomes outdated.

Could be, but I don't agree.
We are currently seeing more and more small and independent
companies bringing out ARM machines. Their number should
grow in the next few years. ARM boxes should then be capable
of running a full featured Myth-TV with power consumptions
below 5W.
There already is a small "hype" about tiny boxes like the
Rasberry Pi or the CuBox (to name but two – both not capable
[CuBox currently not capable?] of running Myth-TV).
These companies cannot afford ads and thus have to "actively
feed" those hypes to sell their mini-boards. Hence it
wouldn't surprise me if they would partake on that Wiki.
Their content/results would have to be – as anybody’s –
reproducible, of course.

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://www.mythtv.org/mailman/listinfo/mythtv-users

myLC at gmx

May 20, 2012, 3:34 AM

Post #2 of 4 (390 views)
Permalink

Re: Request for sticky topic - low power consumption contest [In reply to]

John Drescher wrote:
===================
> I believe that accurately measuring this is out of the
> scope of most users since they do not have $1000+ setups
> to get accurate power usage readings. It has been shown on
> quite a few review sites that "Kill A Watt" meters are not
> accurate enough to measure this.

True - let's put it this way: You can look up what a certain
combination of board and processor is going to consume under
which circumstances. It's being published by companies using
these (even 5K+) setups.
While it might be hard to tell in the area between 20-35W,
new ARM boards are coming out consuming even less than 3W
(max). All I'm saying is - if you can report one of those
systems running Myth-TV without limitations - put it on top
of the list and see if anybody can do it better.
It doesn't hurt to list a specific manufacturer. Say you got
a "ZOTAC IONITX-T-U NVIDIA ION - Intel Atom D525 -
Motherboard" (I don't) and it fulfills its Myth-TV duties (I
can't tell), put it on the list. Also list the price
(automated via Google-Shopping?), which in this case would
currently be $92.50.
This way, users can look up what systems to pick for Myth-TV
if they want low power consumption at a decent price.

> Also Full load and Idle do not show the total picture as
> i3 systems (depending on the work load) can be more energy
> efficient than Atom systems since i3s get the same amount
> of work done much faster and thus return to idle much
> faster.

In case of watching/recording TV the job is "always the
same" though. The system has to be ON while watching or
recording and it has to be SOFT-OFF when waiting. If a
system has a DSP handling the playback and it's being
supported by Myth-TV this can be a killer argument which
affects the power consumption greatly. So, no matter how
fast your i3-7 is - if the integrated GFX can handle the job
you're better off - and that power consumption would be what
matters in the end.

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://www.mythtv.org/mailman/listinfo/mythtv-users

mikep at randomtraveller

May 20, 2012, 8:06 AM

Post #3 of 4 (390 views)
Permalink

Re: Request for sticky topic - low power consumption contest [In reply to]

On 20/05/12 11:34, myLC [at] gmx wrote:
> John Drescher wrote:
>> Also Full load and Idle do not show the total picture as
>> i3 systems (depending on the work load) can be more energy
>> efficient than Atom systems since i3s get the same amount
>> of work done much faster and thus return to idle much
>> faster.
>
> In case of watching/recording TV the job is "always the
> same" though. The system has to be ON while watching or
> recording and it has to be SOFT-OFF when waiting. If a
> system has a DSP handling the playback and it's being
> supported by Myth-TV this can be a killer argument which
> affects the power consumption greatly. So, no matter how
> fast your i3-7 is - if the integrated GFX can handle the job
> you're better off - and that power consumption would be what
> matters in the end.
>
Er, no. If your system is on but sitting at any menu I guarantee that it will be
using a great deal less power than if it is decoding a video and displaying it.
I would even venture to suggest that running an SD video consumes far less power
than running an HD video (video = recorded material of some kind).

Even the bit rate/frame rate is going to make a difference to the power
consumption. You can't say that it is an all or nothing proposition.

--

Mike Perkins

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://www.mythtv.org/mailman/listinfo/mythtv-users

myLC at gmx

May 21, 2012, 4:22 AM

Post #4 of 4 (382 views)
Permalink

Re: Request for sticky topic - low power consumption contest [In reply to]

Mike Perkins wrote:
===================
> Er, no. If your system is on but sitting at any menu I
> guarantee that it will be using a great deal less power
> than if it is decoding a video and displaying it.

I got an i5 and playing back HD content via Quick-Sync is a
joke. My guess is that it's somewhere around 0.5 - 1W.

> I would even venture to suggest that running an SD video
> consumes far less power than running an HD video (video =
> recorded material of some kind).
> Even the bit rate/frame rate is going to make a difference
> to the power consumption.

I also got a mobile phone capable of display 1080p via HDMI
cable; should be around 0.5W too. Again � due to dedicated
hardware. SD video has to be up-scaled and interpolated. In
the end the difference is probably small.

> You can't say that it is an all or nothing proposition.

I don't, but doing it (Soft-Decoding) via CPU consumes at
least 10x more. If I activate Soft-Decoding on my mobile
phone I can kiss goodbye to the batteries... ;-)

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://www.mythtv.org/mailman/listinfo/mythtv-users

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact Gossamer Threads