[mythtv-users] What hardware do I need to be able to dual record using Comcast cable?

[David Madsen]

On Thu, Oct 16, 2008 at 7:12 AM, Michael T. Dean
<mtdean at thirdcontact.com> wrote:

> And plenty more examples out there on the 'net.  I wouldn't be surprised
> if those who are transcoding HDTV are spending more money on increased
> power usage from the additional effort of transcoding than they would
> spend just buying some new big HDD's.  I'm pretty certain that those who
> record/transcode/delete do.  Those who record/transcode/archive (where
> legal to keep recordings long term, of course) may also.


This got me thinking a bit and I did some back of the envelope
calculations to put it into perspective.  Here's what I've come up
with.  By transcoding, we are essentially 'buying' extra disk space
with electricity.  Since electricity is not free, we can calculate how
much it costs to 'buy' disk space by use of transcoding.  I'm only
calculating raw space gained by transcoding, this does not factor in
whether the recordings will eventually be deleted or archived
permanently.  This will also be based more or less on my MythTV
install and my recorded HD recordings and file sizes, which may not be
applicable to everyone.


Usage delta between idle/encoding on a Q6600: ~50W
Local cost for 1kW of electricity: ~$0.09
Time to transcode 1hr episode from MPEG2 -> H.264 (2 pass encoding): ~4hrs
Original file size (commercial cut): ~5GB
Encoded file size: 1.5GB

Power used while encoding     = 50W/hr * 4hrs   = 200W
Cost of power used to encode = .200kW * $0.09 = $0.018
Space saved by encoding       = 5GB - 1.5GB     = 3.5GB

Cost of space gained by encoding = $0.018/3.5GB = $0.005/GB


Since the lowest price of new media is around $0.10/GB these days,
transcoding (given this situation) appears to be an order of magnitude
cheaper still.  This little experiment also does gloss over many other
variables one could throw into the mix, but nothing (I could think of)
that would skew the results far enough to change the result.  Other
people with vastly different energy prices, lower efficiency
processors or a host of other things could also end up with quite
different results.

Anyway, thanks Mike for passing along the idea.  I had never
considered transcoding from that perspective before and I always find
it interesting to evaluate the opportunity cost of different things.

--Dave