Posted by Gene E. Bloch on 04/15/06 23:20

On 4/14/2006, Jeff Rife posted this:
> Gene E. Bloch (spamfree@nobody.invalid) wrote in alt.tv.tech.hdtv:
>> I took the lens elements out of a Wollensak Rapax leaf shutter (almost
>> as old as I am) so I could look through it at several LCD devices (an
>> LCD TV, a desktop monitor, and a laptop monitor) and one CRT device (a
>> TV). I triggered the shutter at 1/400 sec while looking towards each of
>> the respective screens in turn.
>>
>> On the CRT TV I saw a fraction of a picture (a horizontal band taken
>> from the whole). The band was in a different location each time,
>> because I couldn't synchronize my trigger finger with the vertical
>> pulse :-)
>>
>> In each LCD device I saw the entire picture every time. Every time.
>> Always. On the LCD TV, that was true for both standard def an high def.
>
> This will *always* be true with such a fast shutter speed. It takes
> an LCD element 1/30th of a second to change from one state to another.
> At 1/400 of a second, you will always see what appears to be the same
> frame.

The point you are missing is that all of the pixels are illuminated all
of the time on an LCD, in contrast to what happens on a CRT. If I used
a slower shutter speed I could not distinguish between pixels that were
illuminated for a millisecond and pixels that were illuminated for 15
or 30 msec. (And of course, we have to make the proper - and obvious -
allowance for pixels whose current information content is black or very
dark.)

BTW, I didn't say that I saw pixels from a single frame, but that I saw
all pixels illuminated at the time of each "snapshot". That obviously
could be from more than one field or frame, depending on where in the
screen-drawing cycle I clicked the shutter.

> A CRT progressively-scanned at 60Hz with a phosphor that took 1/30th of a
> second to decay would result in the same image as on an LCD. Do the
> math for a moment to figure out why.

Obviously I have done that math. That's why I posted what I did, in a
second attempt to show you that each pixel on an LCD changes when its
next value comes in, not shortly after being illuminated.

I know this won't help,
Gino

--
Gene E. Bloch (Gino)
letters617blochg3251
(replace the numbers by "at" and "dotcom")

[Back to original message]