Re: LCD vs CRT resolution

A whiff of hypocrisy is making my nostrils twitch ...

On Sat, 23 Aug 2008 12:02:27 +0100, Andy Champ <no.way@xxxxxxxxxx>
wrote:

Most CRTs are running in the native resolution of the broadcast
pictures. Most LCDs aren't.

Interesting to see you and Roderick Stewart agreeing with each other
in this subthread, unaware or forgetful of the fact that 2.5 years ago
he wrote something that directly contradicts the above.
http://tinyurl.com/6fscju
.... standing in for ...
http://groups.google.com/group/uk.tech.digital-tv/tree/browse_frm/thread/79d29213846ee51b/62e708e140626c65?rnum=41&_done=%2Fgroup%2***.tech.digital-tv%2Fbrowse_frm%2Fthread%2F79d29213846ee51b%2F3b9f1922522df523%3Flnk%3Dst%26q%3D%26#doc_62e708e140626c65

You can't have it both ways, guys ...

In fact, CRTs and LCDs are very similar in this respect. They both
have 'native' resolutions that may or may not bear any meaningful
relationship to the resolution being displayed. LCDs have a 'cell'
density, CRTs have a dot or stripe pitch and a corresponding shadow
mask pitch. This post is quite old, but it makes the latter point
well enough ...
http://tinyurl.com/5kvw3j
.... standing in for ...
http://groups.google.com/group/alt.video.laserdisc/tree/browse_frm/thread/26c662d4e0c10a32/24719a0dfde95aea?rnum=21&q=crt+dot+pitch+group%3A*tv*&_done=%2Fgroup%2Falt.video.laserdisc%2Fbrowse_frm%2Fthread%2F26c662d4e0c10a32%2F24719a0dfde95aea%3Flnk%3Dst%26q%3Dcrt%2Bdot%2Bpitch%2Bgroup%253A*tv*%26#doc_24719a0dfde95aea

The CRT & LCD I used for the CRT vs LCD investigation that I
previously linked ...
http://tinyurl.com/5ccryd
.... standing in for ...
http://www.cemh.eclipse.co.uk/JavaJive/AudioVisualTV/CRTvsLCD/CRTvsLCD.html
.... can both be shown by measurement to have 'native' resolutions that
have nothing to do with any broadcast one.

I'll deal with the LCD first as that is the least controversial. It
has a 'cell' pitch of 0.390mm both vertically and horizontally,
equating to a 'native' resolution of about 867x492.

The CRT has phosphor in triple vertical stripes with a horizontal
pitch 0.614mm. Given the dimensions of the screen, this equates to a
horizontal resolution of around 516, nothing to do with broadcast
anything!

Vertical resolution is more difficult, because presumably it's fixed
by the scan resolution. In the ancillary experiment which follows the
above linked, the CRT picture *appears* to have pixels, and I can see
this effect in various other shots I took then and more taken today.
Yet the phosphor is laid out in vertical stripes, so the only sense in
which these might be 'pixels' is if they are showing where alternate
lines of the previous field are fading in the picture before being
refreshed.

Straight away, a problem with this idea is that they do not appear, as
might be expected, uniformly across the picture, with uniform
gradation down its length, but appear only 'pseudo randomly' in
mid-tone patches.

Nevertheless, to test the possibility further, I directly measured
their spacing today, and by knowing the screen width and height in the
earlier pictures can estimate the spacing then, and they're all
different:
0.822mm,0.812mm,0.686mm,0.691mm,0.662mm,0.616mm,0.613mm,0.650mm

As an experimenter, I was always taught that I hadn't measured
anything until I'd estimated the potential error in the measurement:
+-1.5mm in measuring the width and height of the screen = 0.8%
worst case
+-1 in 45 *camera* pixel measuring any dimension on the
photographs = 2.2% worst case
+-0.1 in 15mm allowance for the set square rulings = 0.67% worst
case

Yet the average of above figures is 0.694 and the maximum variation
from average is 0.128 or 18.4%, and the standard deviation is 0.076 or
10.9%. Both are well outside the likely maximum error.

So they're not 'pixels', in the sense above. My best guess is that
they're compression artifacts of some sort.

But wider point proved anyway, by the horizontal resolution.

These pictures not only were the basis of the above measurements, but
also include a demonstration of jaggies on both TVs, which also show
that, as suggested in my previous post, they are in the source video
rather than an artifact on or other type of display (to save some
up/download time, they are cut down to the bottom LH corner of the
originals, which covered the central area of each screen).
http://tinyurl.com/69lvqh
.... standing in for ...
http://s28.photobucket.com/albums/c219/JavaJive/TV/?action=view&current=JaggiesCRT.png
http://tinyurl.com/5bjs4b
... standing in for ...
http://s28.photobucket.com/albums/c219/JavaJive/TV/?action=view&current=JaggiesLCD.png

Notes:
KV-16WT1U CRT TV:
Width of screen: 317 +-1.5mm
Height of screen: 179 +-1.5mm
Measured 'Dot' Pitch: 0.614 x ?mm
TX-15LT2 LCD TV:
Width of screen: 338 +-1mm
Height of screen: 192 +-1mm
Measured Cell Pitch: 0.390 x 0.390mm

Other conditions as for the CRT vs LCD experiment(s) linked, except
that the pictures were taken at maximum zoom.
.