[mythtv-users] MacOSX frontend & 1080i

[Todd Ignasiak]

On 3/13/07, Andrew Kimpton <awk at awkward.org> wrote:
> Todd Ignasiak wrote:
> > On 3/13/07, Andrew Kimpton <awk at awkward.org> wrote:
> >
> >> Quoting Todd Ignasiak <ignasiak at gmail.com>:
> >>
> >>
> >>> On 3/13/07, Andrew Gallatin <gallatin at cs.duke.edu> wrote:
> >>>
> >>> Also, I doubt the LCD is being run at 150Hz, where does Myth show
> >>> that?   I think LCDs are set to 60Hz by default, although it's hard to
> >>> tell for sure.   I installed SwitchResX control panel & output the DDC
> >>> data for the display, and it showed it was running 1440x900 @ 60Hz.
> >>>
> >> It's a very long time since I looked at it - but there is code in
> >> Myth's videoout 'portions' which attempts to get the refresh rate for
> >> the display but when that fails (there is no refresh rate for a
> >> digitally connected LCD Panel) defaults to 60Hz. If I recall there is
> >> a second place where this reported value is then multiplied by a
> >> constant (I thought 4 - but that doesn't make sense given the reported
> >> value of 150Hz).
> >>
> >> Some of this may be different if you're using the CoreVideo output
> >> mechanism which I wrote some months ago but I don't think was ever
> >> committed - is it possible that Andrew G & Todd are not running the
> >> same videout device (CoreVideo vs. 'standard'). The reported refresh
> >> rates may be different between these two approaches (see para. 1 above).
> >>
> >
> > Hi Andrew,
> >
> >
> > Also -- I am using the AC3 passthrough, based on your patch, and that
> > works great.  DD5.1 audio from broadcast HD, and DVDs via the internal
> > player.
> >
> >
> Todd - when you performed your comparitive analysis with Andrew G's
> results were you using the AC3 pass through or a stereo analog connection ?

On my Mini, I am using only AC3 passthrough.   But, on my MacBook Pro,
I just used the analog output.


There have been several reports of poor video playback when using high
resolution displays, like 1080p.   This evening I dug out an old CRT
display to try higher resolutions and see if I could replicate the
problem.

I found that everything worked fine at 1600x1200, but when I switched
to 1920x1080p, the video became very choppy.  So, it seems to be
related to the resolution being used.

> The two different audio output mechanisms have different clocking
> methods for the audio and it may well be that there's a bug in the
> analog implementation and that's what's causing Andrew G's issues.
>
> Basically the video frame presentation rate is driven from the current
> audio play position - if audio advances 'too fast' in relation to video
> then video has to drop a frame or more to 'catch up'. Equally if audio
> is 'behind' then video has to display the same frame twice in order to
> let audio catch up.
>
> There is an alternative approach to keeping things in sync involving
> sample rate converting the audio on playback to an effective sample rate
> that is not 44.1 (or 48 etc.) but more closely matches the relationship
> between video frame display rate and audio hardware clock rate (ie
> something a little less or a little more - and it can change over time).
> There's some evidence in the Myth code that this has at least been
> considered - but it can't be applied to any encoded audio format (like
> AC3 or DTS) since there's no access to the raw underlying samples - if
> you're going to support those types of audio encoding you have to 'slew'
> the video to match the audio.
>
> Andrew 8-)
>
> _______________________________________________
> mythtv-users mailing list
> mythtv-users at mythtv.org
> http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
>