Posted by M.I.5� on 01/09/07 08:07

"Joshua Zyber" <joshzyber@comcast.net> wrote in message
news:KYudnSYXZN2mQz_YnZ2dnUVZ_sOknZ2d@comcast.com...
> "M.I.5�" <no.one@no.where.NO_SPAM.co.uk> wrote in message
> news:45a23d0b$1_1@glkas0286.greenlnk.net...
>>> The original resolution of film can't be measured because it's a
>>> photochemical process, not an electronic one. However, once
>>> transferred to DVD, both film and video have the exact same
>>> resolution: 720x480 pixels (or 720x576 for PAL).
>>
>> Yes, but so what?
>
> You claimed that "video" DVDs have more resolution than "film" DVDs, which
> is utter nonsense.
>

Starange how they certainly do.

>>> All DVDs have the same measurable resolution. Don't confuse
>>> resolution with sharpness; they are not the same thing. As far as not
>>> being able to compare the sharpness of two different sources, you
>>> could say the same thing about any two productions photographed
>>> differently than one another, regardless of format. Movie X can be
>>> sharper than Movie Y simply because that's the way they were shot.
>>> Has nothing to do with film or video.
>>
>> But it is an inherent feature of the digitalisation process that when
>> you move from the analogue domain to the digital domain, each pixel of
>> the digital domain gets its infomation from an correspondingly larger
>> area of the analogue source.
>
> Obviously.
>
>> Although the physical number of pixels
>> is fixed, the circle of confusion (to misuse an analogue term) is
>> larger than one pixel
>
> Misuing an analog term is putting it lightly. The film source does not
> contain pixels. Therefore, there is no "circle of confusion" larger than
> pixels in the source.
>

Unfortunately, I couldn't think of a better term considering that we are
discussing two things that cannot really be compared easily.

>> (and is totally dependant mainly on the source
>> material, but also on how good the conversion process is), thus
>> diluting the real resolution.
>
> Diluting the real resolution of the film source, yes. Diluting the
> resolution available on DVD, not at all.
>

The resolution of A DVD can certainly be reduced. I could send you a DVD
with 360x576 resolution (even though the digital information thinks it has
720x576), or anything in between (though interpolation takes place here).

> DVD is capable of diplaying 720x576 pixels (to use PAL since that's your
> country's chosen format). Whether that DVD is mastered from a
> higher-quality source or from a source originally shot at 720x576 pixels,
> the end result will always be a DVD image of 720x576 pixels. What is so
> difficult to understand about that?
>

Nothing, but where those pixels are derived from a larger part of the source
material than those pixels are entitled to represent the real resolution is
reduced somewhat. As you have observed, it is difficult to discus the
subject because the terminology and maesurement techniques are vastly
different between the analogue and digital domain. A totally white screen
has a real resolution of 1x1 exen though the actual pixels recorded on the
disk are still 720x576 (though the MPEG compression will actually treat it
as 1x1 which is when the really huge compression factors are achieved).


> You're arguing that mastering the disc from a higher-quality source will
> result in worse picture quality than if you had mastered from a
> lower-quality source. That's patently absurd!
>
>> Obviously a digital video source will retain its full resolution,
>
> In your argument, it would only retain its full resolution if that
> original resolution were 720x576 pixels. What about content shot on HD
> video and downgraded to DVD?
>
>>> Oh good lord. The DVD may be interlaced, but the fields can be
>>> reconstructed into whole video frames on a progressive scan screen.
>>> That's the whole point of progressive scan. Please read this:
>>>
>>> http://www.hometheaterhifi.com/volume_7_4/dvd-benchmark-part-5-progressive-10-2000.html
>>
>> The arguments in that cite are valid only when the video system is
>> progressive scan from start to finish. If the source material is
>> interlaced, then converting it to progressive, still means that the
>> odd numbered lines occur in a time frame one fiftieth of second
>> earlier than the even numbered lines. Displaying interlaced source
>> material on an LCD TV doesn't look any different than if the DVD
>> player is set to convert to progressive. On a CRT TV there are
>> considerable advantages from the flicker reduction, but this is
>> countered because twice the bandwidth is required for the video
>> content. This is one of the reasons interlaced video replaced
>> progressive (or sequential as it was called when it was abandoned in
>> 1936).
>
> You use a lot of technical jargon that would make a layman think you knew
> what you're talking about, but it's becoming quite clear that your ideas
> are rooted in an antiquated understanding of how video works.
>
> I would recommend that you sit down and actually read the entire article
> that I linked above, specifically the sections titled NTSC Telecine,
> Progressive Scan, and Re-Interleaving 24fps Film.
>

I did. It didn't tell me anything I didn't already know. And it didn't
answer the point, which suggest that you are bolstering up you ignorance of
the subject by citing external sources in the hope of impressing others. It
might work with those others, but not me sunshine. I've been in the
business far too long.

> After that, read this:
>
> http://en.wikipedia.org/wiki/Telecine#Reverse_telecine_.28a.k.a._IVTC.2Finverse_telecine.29
>
> Here's a relevant excerpt:
>
> "On DVDs, telecined material may be either hard telecined, or soft
> telecined. In the hard-telecined case, video is stored on the DVD at the
> playback framerate (29.97 frames/sec for NTSC, 25 frames/sec for PAL),
> using the telecined frames as shown above. In the soft-telecined case, the
> material is stored on the DVD at the film rate (24 or 23.976 frames/s) in
> the original progressive format, with special flags inserted into the
> MPEG-2 video stream that instruct the DVD player to repeat certain fields
> so as to accomplish the required pulldown during playback. Progressive
> scan DVD players additionally offer output at 480p by using these flags to
> duplicate frames rather than fields.
>
> NTSC DVDs are often soft telecined, although lower-quality hard-telecined
> DVDs exist. In the case of PAL DVDs using 2:2 pulldown, the difference
> between soft and hard telecine vanishes, and the two may be regarded as
> equal. In the case of PAL DVDs using 2:3 pulldown, either soft or hard
> telecining may be applied."
>

And you've done it again. This totally fails to address the issue at hand.
How the frames are handled is not the issue. It is the actual process of
digitalisation that is the issue at hand.

I don't have the time to continue this, as you have obviously made your mind
up that you are right and will allow no amount of discussion to enable you
to see the light.

[Back to original message]