[mythtvnz] Help spending my brothers $5K on MythTV

[Steve Hodge]

On 2/23/06, Steven Ellis <mail_lists@stevencherie.net> wrote:
> Steve Hodge wrote:
> > 576x576 MPEG4 @ 4400 bitrate (quality min/max 15/2). Interlaced DCT
> > and motion estimation on. Audio is 44.1kHz MP3 quality 3.

> Whats the CPU loading like with on the fly MPEG4 encoding. I did live
> MPEG4 capture a couple of years ago on an athlon 2k, and would get
> occasional frame drops which really annoyed me.

At the above resolution on my Sempron 2800+ (actually a 2400+
overclocked to 2GHz) I get about 45-50% idle just recording (~40% CPU
usage by mythbackend). It varies slightly with content and this is for
sport which is about the worst. I'm also watching top via vnc so there
is probably a bit of usage there. If I watch the recording at the same
time I get 25-30% idle (~12% CPU for another mythbackend thread). This
is a combined frontend/backend. I set it up so that I could record
from two dumb capture cards at once, but can't quite also watch one of
them. Now we've moved house and the new place has no UHF aerial so I'm
only using one of the capture cards so I've got CPU cycles to spare.

> Done some recent comparisons of Analog FTA on the PVR-150 on the Default
> settings of 4.5M average and 8M max against some FTA DVB-T streams that
> average around the 4-5M mark and I must admit they are reasonably close
> quality wise. The biggest benefit with the DVB-T stream is the lack of
> ghosting. Mind due my TV1 & 2 reception is excellent as I have great
> line of sight of the transmitter.

Anyone know how the TVNZ DVB-T stream compares to TVNZ DVB-S stream?
Is it the exact same stream?

Steve