[mythtv-users] Modern ION replacement?

John Morris jmorris at beau.org
Thu May 15 19:26:04 UTC 2014


On Wed, 2014-05-14 at 18:58 +0100, Paul Gardiner wrote:

> It's not that synchronisation that causes problems. The problem is
> ensuring that the decoded images are written to the display memory at
> the correct time. The software doing this sees two sync signals per
> frame. If it picks the wrong one, the interlaces get reordered which
> looks awful.

If the hardware/driver/Myth/whatever can't detect the difference between
fields and frames then it is obviously broken and you are right that
doing the deinterlace in software would be the only option.  But if it
is working then it would be worth comparing the job Myth does to what
the TV itself is capable of which is all I was saying.  Don't think many
(any?) cable boxes convert 1080i to 1080p and they look good on most
modern sets.

This is an idea whose time is coming.  If we would adopt the notion of
just passing video out 'as is' it would really help the efforts to run
Myth on lower end hardware.  And as 4K comes to mass deployment do we
really want to be rescaling everything to 4K on tiny set-top boxes?  And
that is what would be required since continuing to lock Myth's
framebuffer to 1080p would be bad for 720p in that case.  Better to let
the TV evenly scale 720p60 3X to 2160p vs 1.5x to 1080p and then 2x in
the TV.  Yes this change would likely blow up many/most current
installs, which is why the idea needs exploring now so the bugs can be
located and killed before it becomes an even bigger issue in the future.
Better a slow peaceful migration than a rushed one when the issue
becomes a pressing one.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part
URL: <http://www.mythtv.org/pipermail/mythtv-users/attachments/20140515/afc1c227/attachment.sig>


More information about the mythtv-users mailing list