[mythtv-users] High end, state of the art Myth Frontend
Michael T. Dean
mtdean at thirdcontact.com
Wed Sep 18 17:33:29 UTC 2013
On 09/18/2013 12:40 PM, Gary Buhrmaster wrote:
> On Wed, Sep 18, 2013 at 4:16 PM, Joseph Fry wrote:
> ...
>> Agreed! They should have done away with interlaced with all HD
>> formats, there was no technical reason to keep it.
> Selective memory there? There were some technical
> justification to keep it at the time that the standard was
> being decided (the '80s), although you may not agree
> with the reasoning (or the requirements), and I would
> agree that today some of those reasons are less
> convincing (partially as a reflection of 30 years of
> Moore's law and what you can put inside the box).
Yeah, the main justification for interlaced formats (for both NTSC and
ATSC) was bandwidth. They had 6MHz to use. For ATSC, that meant
19.39Mbps for broadcast TV. With the MPEG-2 encoders available at the
time, that meant good picture quality up to 720p60 or 1080p30 or
1080i60. Now, ask Andre (who works in Sports TV) if he'd rather watch a
sports broadcast in 1080p30 or 1080i60 (with good deinterlacing). I'll
bet he'll choose 1080i60. Would 720p60 be better? Sure, if you're OK
with having less than half as many pixels (spatial resolution). (And
the subjectivity of deciding which is better--720p60 vs 1080i60 vs
1080p30--is why the standard allows all of them.)
Interlacing is simply a means of increasing the temporal resolution of a
video at a given spatial resolution without a requirement for increased
bitrate. (And don't say, "You can't invent the pixels that are
missing," because doing so from an interlaced signal is /much/ easier
(and much more founded in reality) than the, "Let's invent the 3 frames
that would have come between these 2 frames we're given and display at
120Hz that LCDs do to make up for LCD technology's inherent limitations
when used for video.)
Mike
More information about the mythtv-users
mailing list