[opendtv] Re: 625 video quality is good enough....
- From: jeroen.stessen@xxxxxxxxxxx
- To: opendtv@xxxxxxxxxxxxx
- Date: Wed, 20 Oct 2004 09:02:53 +0200
Hello,
Craig Birkmaier wrote:
> As we move from scanning CRTs to progressive displays where the
> entire screen is active all the time, the issues related to frame
> rates disappear. A DLP or LCD display can do an equally good job with
> 50 or 60P; they can even display 24P without any 3:2 garbage.
and
> As we move to displays that present entire frames this issue too
> will go away. DLP, LCD and plasma display can present 24P without
> any pulldown.
Not true. If such display accepts 24 Hz refresh at all (and some
do) then it is only because internally the frames are repeated 2
or 3 times to generate a more standard refresh rate again.
LCDs need a high enough refresh rate because the voltage on each
pixel cell must be inverted every frame. Since the result for each
polarity is not equal, this causes some flicker at half the frame
rate. This can only be masked by a high enough refresh rate.
(Because of this problem you can actually see interlace-like line
flicker on some -progressive- LCD projection displays.)
PDPs and DLPs work internally with sub-fields or bit-planes.
In some fields pixels will be on, in other fields they will be off.
Especially the large-weight sub-fields will cause frame flicker.
Again this can only be masked by a high enough refresh rate.
E.g. for 60 Hz input the MSB fields may not be repeated, for 50 Hz
input some MSB fields may be produced twice, and for 24 or 25 Hz
input (if supported at all) the MSB fields may be produced 3 times.
The end result may be marginally better (e.g. quantisation noise)
than if the video source had applied 2x or 3x frame repetition,
but for motion portrayal the result will be largely identical.
And any way you want to render it, 24 Hz motion portrayal sucks !
> The only displays I have seen that deals adequately with interlaced
> HD are the ALIS plasma panels.
Surely you did not mean to exclude all those CRT projectors ?
> In essence these panels are designed to simulate the way an
> interlaced display works, shifting each field up or down
> appropriately.
For all practical purposes they ARE interlaced displays.
The plasma is alternately ignited between odd-even row electrodes
or between even-odd row electrodes. This is not just a simulation.
> IMHO, even these displays do not provide the quality of a good
> progressive display.
That depends on which progressive display you wish to compare
them with... Fujitsu(-Hitachi) found that they could make a 1024i
display for the price of a 512p display (with the same amount of
row electrodes and row drivers). The vertical spot size is fairly
large, so there is not too much resolution gain to be made, but
for that the interlace artefacts are very mild too, of course.
You can choose to operate such display also as a 512p display,
and that gives a visibly inferior picture. So the choice is easy...
In the mean time they have discovered that for the same cost
(more or less) they can also operate this same construction as
a 1024p display, so then the next choice is also easy... ;-)
(Assuming that you can feed it twice the amount of samples...)
Greetings,
-- Jeroen
+-------------------------------+----------------------------------------+
| From: Jeroen H. Stessen | E-mail: Jeroen.Stessen@xxxxxxxxxxx |
| Building: SFJ-5.22 Eindhoven | Philips Digital Systems Laboratories |
| Phone: ++31.40.2732739 | Visiting & mail address: Glaslaan 2 |
| Mobile: ++31.6.44680021 | NL 5616 LW Eindhoven, the Netherlands |
| Pager: ++31.6.65133818 | Website: http://www.pdsl.philips.com/ |
+-------------------------------+----------------------------------------+
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at
FreeLists.org
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.
Other related posts:
