<div dir="ltr"><br><div class="gmail_extra"><div class="gmail_quote">On Thu, May 8, 2014 at 7:46 AM, Mike Perkins <span dir="ltr"><<a href="mailto:mikep@randomtraveller.org.uk" target="_blank">mikep@randomtraveller.org.uk</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">On 07/05/14 23:57, Matt Emmott wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
4) Lack of a decent ARM or $100 solution - We see all of these threads on<br>
"would this work with Myth" and "why not this cheap box", but nobody can<br>
ever get them to work. Meanwhile a Roku device can render 1080p video for<br>
$50 with a remote! Why isn't there a low-end version of Myth FE tailored to<br>
some specific devices? If somebody came out with a Myth front end device<br>
with a custom build of MythTV for $199 or lower, I'd buy it in a second.<br>
Why hasn't this happened yet? Resources? Funding? Lack of interest? Surely<br>
somebody can build a no-frills FE interface similar to a Roku or Boxee Box,<br>
no?<br>
<br>
</blockquote></div>
There's a very good reason for this, and that is that almost every $100 box has been designed and made to be as cheap as possible - that may mean less RAM or a processor which can only /just/ manage the media formats it claims to support.<br>
<br>
The big point is that most of these devices are made with non-standard chips and/or non-standard architectures which means that you would need to have the complete specs of *every* device on the board in order to write (or port) a front end to them.<br>
<br>
You won't get that without signing a NDA and probably forking out $bucks for the privilege. By the time you've gotten your wad of specs and made a start on adapting the frontend logic to this device it will be out of production and the manufacturer will have moved on the the next Ooooh! Shiny! he thinks the punters will want to buy.<br>
<br>
Short answer: diminishing returns. Too much effort and cash for what would be a small number of users. It makes more sense in writing a front-end for architectures that are easy for your potential users to obtain and program. Note that ARM /isn't/ a single architecture: there are significant differences between many of the CPU models.<span class="HOEnZb"><font color="#888888"><br>
<br>
-- <br>
<br>
Mike Perkins</font></span><div class="HOEnZb"><div class="h5"><br><br></div></div></blockquote><div> </div><div>I get your point with the countless devices out there today, however why not standardize on the Raspberry Pie.. they just recently open sourced their graphics stack as well... XBMC runs like a champ on the 512 meg version, so you know it can be done... Yadda, Yadda, Yadda...</div>
<div><br></div><div>Just my 2cents.</div><div><br></div><div>Greg </div></div><br></div></div>