|
|
Posted by Joshua Zyber on 01/20/07 16:16
"M.I.5" <no.one@no.where.NO_SPAM.co.uk> wrote in message
news:45af7be9$1_1@glkas0286.greenlnk.net...
>>> However, on an LCD display, it makes no difference *for interlaced
>>> source material* (we were specifically discussing DVD at the time) -
>>> which is what I said.
>>
>> Why would an LCD be any difference than a plasma or
>> progressive-capable CRT in this regard? Are you just pulling stuff
>> out of your ass now or what?
>
> Because of the relatively long time lag of the LCD elements. It makes
> no perceptable visual diffeence if you write to the odd lines of the
> display and then the even, or write to the whole display sequentially.
Once again, you demonstrate the barest hint of understanding on a topic
but pretend to speak authoritatively on it. And once again, you are way
off base.
All LCD displays are inherently progressive scan. There is no such thing
as an LCD that can display interlaced fields. When you feed an
interlaced signal into an LCD display, a deinterlacing chip inside the
set will recombine the fields and display them as progressive frames.
The lag time you speak of (which has been greatly reduced in recent
years to the point where decent sets are barely a couple of
milliseconds) is from one FRAME to the next, not one FIELD to the next.
It is impossible for an LCD to display interlaced fields individually.
>> Film-based content on DVD is stored in 720x480 frames with 2:3
>> reverse field cadence headers that instruct the MPEG decoder how to
>> interlace them for output. The interlaced fields from each original
>> frame do not come from different moments in time. With proper 3:2
>> deinterlacing, the original frames can be reconstructed in their
>> original form.
>
> You are trying to obfuscate the point by switching systems. Only
> Mickey Mouse television systems have to do anything remotely like 3:2
> pulldown just to get film material to work.
What I described is how NTSC works. PAL is even easier, because the
interlaced field pairs are coded in a simple 2:2 cadence that's simpler
to reconstruct. In either case, film-based content can be reconstructed
to display the original progressive frames, with field pairs that come
from the exact same moment in time.
> We were discussing 720x576 pixel DVDs in which film originated
> material is stored in the form that one film frame occupies both the
> odd and even frames, but that is only because the film frame covers a
> time frame of one twenty-fifth of a second*. But in video originated
> frames, the odd lines contains information that occurs one fiftieth of
> a second before the corresponding even line. Converting the output to
> progressive scan cannot change that.
Yes, this is the difference between film-based content and video-based
content. The example you cited was for the end credits of a movie, which
would be film-based content and will work as I described above.
> *Film for TV is shot at 25 frames per second.
Only in PAL-land.
> Cinema film is run 4% fast.
You have that precisely backwards. Film is projected theatrically at
24fps. PAL video speeds this up to 25fps. Hence the term "PAL speed-up".
>> This is of course different than how video-based content is stored on
>> DVD, but we weren't talking about that. You specifically referenced a
>> feature film's end credits.
>
> No, feature film end credits was mentioned but I refered to both
> methods of originating DVD material.
No, you're trying to obfuscate the issue by bringing in video-based
content, which was not the topic of discussion.
You specifically said that end credits on movie content were unreadable.
To which I asked if you were watching on a progressive scan screen. Upon
which you remarked that it couldn't possibly make any difference. In
other words, you were wrong.
>>> Of course for progressive source material then once again there is
>>> an obvious benefit - but we weren't discussing that context.
>>
>> As a matter of fact, we were.
>
> No we weren't.
>
> We're discussing DVDs, which are not progresive scan?
Although interlaced, the separate fields from film-based content on DVD
come from the same moment in time. There is no temporal difference from
one field in a pair to the next. Once properly deinterlaced, you're
seeing the original film frames.
Navigation:
[Reply to this message]
|