Proper Aspect Ratio Measurements on an RGB Monitor

[Great Hierophant]

The monitor in question is an IBM 5153 Color Display, the standard for digital RGBI monitors. This display has a 13" viewable tube. On the back of the monitor, there is a control for vertical size and vertical hold. No other controls are present externally that will change the picture size or position. If the horizontal size is fixed, what should the vertical size be? I would say that the picture size should be proportional to the 4:3 screen dimensions. An active picture on my 5153 measures 9.5625", which I derived using a paper tape measure from IKEA. As that is the long measurement, the short measurement must be 3/4 of that, or 7.1718. So, I adjusted my vertical size knob until I got a active display of roughly that height and I was done.

 

[reenigne]

4:3 is the usually quoted number, and this gives you the usually quoted 5:6 pixel aspect ratio of CGA. However, if you crunch the numbers in the TV standard SMPTE 170M you actually get about 1.37:1. The tolerance on the horizontal burst period actually makes it between 1.362:1 and 1.375:1 if my calculations are correct. But the hardware isn't accurate enough for it to be meaningful to say the ratio is 1.37:1 rather than 1.33:1 - the width of the picture will change by more than 3% depending on what's being displayed (brighter images show up wider than darker ones).

 

[VileR]

the width of the picture will change by more than 3% depending on what's being displayed (brighter images show up wider than darker ones).

is this in any way analogous to the general phenomenon of CRT "blooming", caused by dips in voltage regulation when high-brightness images are shown? I'm curious since that tends to make brighter images both wider *and* taller, which doesn't change the aspect ratio, at least not in a consistent / meaningful way. Or is it an entirely different process at work in the 5153?

 

[reenigne]

Yeah, bloom is the effect I was thinking of. Thinking about it, though, you may be right - if it's caused by just the electron beam current taking away power from the deflection coils then maybe it does affect the vertical as well as the horizontal. The horizontal deflection works at much higher frequencies, though, so I'd have thought that it would be more susceptable to effects like that. But even if bloom doesn't affect the aspect ratio significantly, I think there will be many other affects such as temperature and voltage change that would make it impossible to keep the aspect ratio constant to that degree of accuracy.