Errors by Those Who Should Know Better

Interlace can be simple to understand if explained properly. However, much of the stuff on the net and elsewhere is wrong or at least misleading. For what it's worth, here's my little list.

The Educator

Professor Kelin J. Kuhn in his 'Conventional Analog Television - An Introduction' page says that even fields are composed of even lines and odd fields of odd lines. He doesn't see the contradictions resulting from this (or chooses to ignore them). His 21 line raster diagram (lines wrongly numbered) has line number zero not labelled at all (it is ideosyncratic to have a line zero anyway and as this is in both odd and even fields I infer that Prof Kuhn thinks that zero has the unique property of being both an odd and even number). Line 21 is also absent.

The Prof isn't consistent though and in the text just before the picture of the raster he says (talking about the 525 line NTSC system):

"However lines number 248 to 263 and 511 to 525 are typically blanked to provide time for the beam to return to the upper left hand corner for the next scan."

Now implying (correctly) that lines are numbered in time (the line numbers that he says are blanked are not entirely correct though). His figure immediately following the 21 line raster, showing a 525 field scan, is also correct (but lines misnumbered as before).

The Manufacturers

 Toshiba says:

"Traditional Interlace scanning mode reproduces scanning lines of either odd number or even number in each field which creates the following two problems.

• Noise caused by a frame being composed of two fields in a stripe pattern with a time lag.
• Flickering due to many pictures appearing and disappearing in a second."

It's not clear to me what Mr Tosh is trying to say.. it sounds as though he's saying that odd fields are composed of odd lines and even lines comprise the even field and futhermore that odd fields have one more line than even ones (which, if his initial assumption were true would be a necessary consequence; if this is what he's saying then he's the only one who's spotted this inconsistency - without recognising that it is an inconsistency because he goes on to say that a field consists of 312½ lines).

Why Mr Tosh thinks that interlace increases flicker and noise he doesn't say. In fact, interlace reduces flicker and since it conserves bandwidth it also reduces noise. Further illustration of Mr Tosh's grasp of the subject can be seen a few sentences on when he says:

"For viewing not illusion of grille images of interlace but continuation of frames like a movie film, the following effects can be gained:...."

Doesn't inspire confidence in Tosh's products, does it?

Elantec's 1881 datasheet has a figure reproduced, it says, from an EIA document - this shows the field beginning at the start of the equalising pulses (page 7) and on page 10 tells us that 'there is an odd field that includes all the odd lines, and an even field that consists of the even lines'; oh dear - that sounds very odd!

In Conclusion

Perhaps some might think that I am being pedantic in highlighting these errors. Maybe I am, but for someone seeking information and hoping for a coherent exlanation from these people will, not only be disappointed, but may end up confused rather than enlightened. Educators and international electronics manufacturers ought to be able to get simple things right.