<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Wed, Jan 21, 2015 at 9:46 AM, Simon Hobson <span dir="ltr"><<a href="mailto:linux@thehobsons.co.uk" target="_blank">linux@thehobsons.co.uk</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Jay Ashworth wrote:<br>
<br>
> My response will probably go into the permanent-moderation file*, but...<br>
><br>
> Why? The current model takes note of the fact that devices are often<br>
> grouped, and the scheduler needs to know about it, and also that grouped<br>
> devices often share configuration parameters.<br>
><br>
> If neither of those things has changed in the physical model, then the<br>
> software model probably oughtn't change either.<br>
><br>
> Why is it?<br>
<br>
Which is exactly the question I was pondering asking.<br>
The current model was developed for a reason, and I don't see that the primary reason has gone away - unless something fundamental has changed.<br></blockquote></div><br></div><div class="gmail_extra">If you read the post on the dev list (<a href="http://www.gossamer-threads.com/lists/mythtv/dev/581886">http://www.gossamer-threads.com/lists/mythtv/dev/581886</a>), David explains the main reason--it's because capture cards and the way we use them have changed. The original setup with separate cards and inputs was to support cards/encoders with multiple mutually exclusive (analog) inputs such as S-video connected to a VCR, composite connected to a cable box, etc. Only one of the inputs could be used at a time. The indication is that this arrangement is quite uncommon now, but he plans to still support it (automatically/transparently?) via input groups. Also this flattens the virtual tuners (which were never considered in the original incarnation) and solves some scheduling issues with them. I recommend reading his post.<br><br></div><div class="gmail_extra">Karl<br></div></div>