Quote:
Originally Posted by etype2
|
Pixel structure is an incorrect/inaccurate term if applied to CRT displays(phosphor dot structure in color and phosphor grain in monochrome are correct). One makes a digital images by creating a grid of boxes (known as pixels) that have a fixed position on the image both in the source and display devices and a specific light value corresponding to the image at that point.
There is no fixed position for a specific pixel of image on any CRT. The scanning in tube sets is not precise enough to achieve that. Any pixel of source info could end up being more than 1/4" off from it's last screen position on the screen between showings (based on things like line voltage, brightness setting, tube aging etc.)... Digital displays and video standards are designed to convey images as pixels all numerically fixed in space and light value, but analog video is designed to opperate as a field of rows of lines and image element position defined by timing relative to the line scanning sync pulses (and there's a lot of fudge factor in the analog system)...you can convert between the two but they are not the same... infact the more conversions one does back and forth the more error/loss introduced to the image.
If this wasn't a rare desirable set and you had the motivation you could modify the chassis to scan 1080i onto that phosphor screen and substantially increase the number of source pixel info placed onto the phosphor dots. In a monochrome set no information is lost doing this as most monochrome phosphor is fine grained enough that with good beam focus and scanning each pixel will be resolved...in color that pixel structure has to be mapped onto the phosphor dot matrix...it can be even if the phosphor dot matrix is courser than the pixel matrix of the source material being scanned onto it (the 15GP22 and 11SP22 in a portacolor both have much courser dot pitch than is ideal for NTSC while larger tubes have fine enough dot pitch to correctly resolve NTSC). When the dot matrix is too course a given pixel may not land on dots of all 3 colors and will not be accurately colored, and in extreme cases a few pixels could be resolved on a dot of a single color phosphor (in that case pixels would be lost in the border around each dot of phosphor) sets like the portacolor they try to bandwidth limit the video to smear pixel info together and hide the more annoying artifacts that this can cause....one could also do the opposite and make the dots much smaller than each pixel so each pixel [ or to use pre-digital video lingo the smallest image element the desiged video bandwidth and scanning system can support] one or more triad of phosphor (which is what CRT design engineers try to accomplish in most designs).