[mythtvnz] Multicast your noisy frontend or backend to a quiet box?

[Nick Rout]

On Mon, Feb 6, 2012 at 1:31 PM, tortise <tortise at paradise.net.nz> wrote:
> On 4/02/2012 5:02 p.m., Nick Rout wrote:
>>>
>>> End user desirement.
>>
>>
>> I am surprised everyone in your house wants to watch the same thing at
>> the same time, with pauses governed by one user.
>
>
> Seems a bit of a jump from the subject, which was intended to be more about
> multi-casting and less about noisy frontends.  With the benefit of hindsight
> it might have been better to not refer to noisy frontends, my apologies for
> the noise.
>
>
>>The very antithesis
>> of what mythtv is all about.
>
> For me mythtv is an open source effort that evolves with user / dev input.
>  I'd need persuasion that using mythtv in a way it does not currently do,
> yet is capable of doing, is not what its all about, seems quite the contrary
> reality for me.  To add to my case I also observe the paradigm shift
> "requirement" that mythtv is, that mythtv is designed that one does not
> watch live tv exemplifies that paradigm shift is what mythtv is about, and
> for me it is not necessarily limited to that single mind change.  There are
> some things best viewed live.  World cup finals come to mind. Of course its
> best to record the final, watch it largely live (with minimal delay) yet
> with the ability to pause/rewind etc....While this is a significant aspect
> of mythtv its, for me anyway, not the holy grail or conclusion of mythtv's
> path forward, which is (for me) not closed to other paradigm shifts.
>
> The concept is not new and has been considered, even to the extent of
> enjoying initial dev coding for this some years ago.  Its just, like many
> desirements, not publicly implemented successfully yet.
>
> When using the older SD mythtv I did broadcast a mythtv frontend around the
> house using a modulator that was relatively affordable.  This worked very
> well, requiring a tuner device with each display to access. (that generally
> meant a SD TV! However in the DVB-T HD world a HD modulator and HDHomerun
> would enable any LAN connected PC to view the stream, should one shell out
> the dosh for such a beast... some modulators also include IPTV LAN outputs.)
> I don't plan to detail use scenarios, except to note they proved to be more
> than I initially expected as it was more flexible than I'd expected, and
> having lost something I had remains an itch that demands a cure.

But if you want to multicast a video file, why not simply do that,
rather than trying to multicats a desktop?

>
>
>> Perhaps if you told us the *point* of the exercise it might make some
>> sense.
>
>
> As per the title I've been trying to play the same thing on a few frontends
> all synced up, (+/- a couple of seconds lag acceptable, - without
> necessarily pushing this on everyone in the house) given the costs, e.g. the
> price of DVB-T modulators and HDMI splitter implementation problems, not to
> mention an existing LAN, multicasting a frontend seemed a viable, if not the
> best path forward.   If it were a cheaper noisy frontend in the garage that
> might have also been fine.
>

Seems to me if you really want to do this you run the hdmi output of
your "master" frontend to your various screens around the home. I
believe this can be done with cat5 and reasonably priced baluns, eg
http://www.amazon.com/Sabrent-Extension-Extender-adapter-200-Feet/dp/B001PT1A7C%3FSubscriptionId%3DAKIAI6J7BJOOUGGBRIBA%26tag%3Dhdoc5-20%26linkCode%3Dxm2%26camp%3D2025%26creative%3D165953%26creativeASIN%3DB001PT1A7C

> VLC can be easily run to multicast a file and shows / tests whether the LAN
> is multicast capable (or not), however getting it to do the display was not
> so out of the box, and as has been pointed out there may well be significant
> clone quality display issues with HD video.
>
> I've since considered the frontend control socket, so it may be I can use
> that to instruct a few frontends, while minimising LAN traffic.
>
> http://www.mythtv.org/wiki/Frontend_control_socket gives an overview
> including apps already accessing the socket.  Any of those apps might be
> adapted to do the above.  http://nowsci.com/mythmobile/ is notably absent
> from the list of apps.
>

I think we are all familiar with the frontend control socket. Are you
suggesting writing a program to tell X number of frontends to all
start playing the same file at the same time? And then querying the
frontends to make sure they are all approximately in sync, and making
adjustments to keep them in sync?

> So far the best way I can see to do this for me would be writing a webpage
> to link in with mythweb to do this.  One could then access this from
> wherever, particularly an android for roaming control.  Pausing etc would
> have to be done from that page to keep the sync up.  If there was too much
> time drift a resync with the chosen frontend could be easily included.
>
> An intial play with the socket (LAN IP's subs with x.x.x.x) shows it does
> things like:
>
> # query location
> Playback Recorded 00:37:16 of 01:04:09 1x 1001 2012-02-05T21:29:00 55904
> myth://x.x.x.x:6543/1001_20120205212900.mpg 25
>
> # play file myth://x.x.x.x:6543/1001_20120205212900.mpg
> OK
>

Still not sure of the endgame here. Do you want to be able to wander
from room to room and keep watching the same programme? I can't see
any other use for this in a home setting.

If that is what you want I would suggest looking at how bar video
systems are set up. The TVs all appear to be in sync, and the music
videos at least seem to be played form a computer (Sky Sport etc is
different. I suspect (given the absolute crap quality you seem in some
bars) that they indeed modulate the signal and tune into it on an
otherwise unused tv channel. Same with Sky setups in motels etc

Someone must be doing it better though.