Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!zaphod.mps.ohio-state.edu!mips!wyse!vsi1!teraida!netcom!hue
From: hue@netcom.UUCP (Jonathan Hue)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: (Non) Square Pixels?
Message-ID: <7217@netcom.UUCP>
Date: 14 Feb 90 07:12:33 GMT
References: <4687@lmrc.uucp> <3119@cello.UUCP> <4736@lmrc.uucp> <3053@d75.UUCP>
Followup-To: alt.stupidity
Organization: NetCom- The Bay Area's Public Access Unix System {408 249-0290 guest}
Lines: 76

>>In article <3119@cello.UUCP>, robin@sabre.uucp (Robin D. Wilson/1000000) writes:
>>> Most monitors (and TV's do not have "square" pixels.  Most have round or oval
>>> shaped pixels.  In fact, the only monitors that I know to have square pixels 

CRTs don't have pixels, frame buffers do.  If I have a three tube camera
displaying its picture on a television, there aren't any pixels.  Pixel stands
for picture element.  In this case the picture has neither been quantized
nor sampled, so how can there be picture elements?

The useful parameters about a monitor are bandwidth, pitch, horizontal scan
rate, and vertical retrace rate.  The number of pixels in either dimension
is not meaningful as it depends on the device driving it, although recently
monitor vendors have been adding an x by y resolution spec in their literature,
mainly to help non-technical customers figure out whether or not a monitor will
work with their frame buffer.

For example, when I worked at JVC (the inventors of the multiple scan rate
monitor, by the way) we had a monitor with 125MHz bandwidth, which could sync
to horizontal scan rates from 31.5KHz to 75KHz.  The typical frame buffer
that drove it with a 31.5KHz horizontal scan rate was 640x480, at 75KHz
it was 1280x1024.  I guarantee you that we did not make the phosphor dots
("pixels", by your definition) shrink to half their size when we did this.


>>> are the high-end Sony's similar to the one Apple uses for the MacII.  (The
>>> Apple color monitor for the MacII is a modified Sony.)  

The Sony Trinitron CRT uses vertical metal strips for the mask, stabilized by
two (faintly visible) horizontal wires, and a vertical stripe phosphor.  If
you are going to call these "pixels" (which nobody does), then the pixels are
rectangles as they are the entire height of the screen.

When the phosphor is illuminated by the electron beam, it does not expose a
single stripe or single phosphor dot.  The spot size of the beam is usually
larger than the pitch of the tube, meaning the beam shines through multiple
openings in the mask, so multiple stripes or spots are exposed.  BTW, the
profile of the electron beam strength across its diameter approximates a
Gaussian distribution.

Just a bit of trivia - A Sony Trinitron tube has two advantages over the
delta gun/delta mask CRT.  First, a larger percentage of the tube is
covered with phosphor compared to the delta mask or slotted mask type of
CRT.  This is responsible for a 33% increase in brightness.  Second, the
the beam current can be increased without damaging the mask, (I think this
is because it dissapates heat better, being mostly holes) which gives
a 50% increase in brightness.  1.33 x 1.50 equals a 2x increase in
brightness.

>Also, on my NEC 3D there is an adjustment to increase the distance between the
>pixels from right to left (thus making the screen wider, and changing the 
>aspect ratio from that perspective) and another adjustment that changes the 
>ratio from top to bottom.  It sure helps me to get a 1 to 1 aspect ratio on my 
>stuff.  (I'm not sure how NEC does this, but I think it may have something to 
>do with changing the shape of the tension masks.)

This is plainly ridiculous.  The phosphor on the screen is carefully aligned
with the electron guns through the mask, usually by applying the phosphor to
the tube with a photoresist, and exposing the tube through the mask from a
light source located where the guns normally are.  If the geometry of the mask
was somehow changed, this alignment would be lost and it would be impossible to
display a color picture.

The only think the mask does is expand slightly due to heating caused by the
electron beam.  On some monitors the mask frame is held in place by bimetallic
clamps, which moves the mask toward the tube face as it heats up.  This keeps
the mask and phosphor aligned during changes in temperature.

All that the knobs on the NEC do is decrease the current through the
deflection coils, which reduces strength of the magnetic field applied across
the yoke of the tube.  This slows the beam deflection rate across the face of
tube and causes a smaller area of the CRT to be exposed.  A Commodore 1084S
can do this.

Followups to alt.stupidity.

-Jonathan